creating an economically viable, closed-system, …...dr. richard g. smith, unh assistant professor,...
TRANSCRIPT
CREATING AN ECONOMICALLY VIABLE, CLOSED-SYSTEM, ENERGY-
INDEPENDENT DAIRY FARM THROUGH THE ON-FARM PRODUCTION OF ANIMAL
BEDDING AND HEAT CAPTURE FROM AN AERATED STATIC PILE HEAT RECOVERY
COMPOSTING OPERATION
BY
MATTHEW M. SMITH
B.S. in Environmental Conservation Studies, University of New Hampshire, 2008
M.S. in Resource Administration & Management, University of New Hampshire, 2010
DISSERTATION
Submitted to the University of New Hampshire
in Partial Fulfillment of
the Requirements for the Degree of
Doctor of Philosophy
in
Natural Resources and Environmental Studies
May, 2016
ii
ALL RIGHTS RESERVED
© 2016
Matthew M. Smith
iii
This dissertation has been examined and approved in partial fulfillment of the requirements for
the degree of Doctor of Philosophy in Natural Resources and Environmental Studies by:
Dr. John D. Aber, UNH Professor, Department of Environmental Sciences
Dr. Mark J. Ducey, UNH Professor, Department of Natural Resources
Dr. Theodore E. Howard, UNH Professor, Department of Natural Resources
Dr. Richard G. Smith, UNH Assistant Professor, Department of Natural Resources
Dr. Heather Darby, UVM Extension Associate Professor, Agronomy Specialist
Bruce Fulford, City Soil, Principal Operator
On April 14, 2016
Original approval signatures are on file with the University of New Hampshire Graduate School.
iv
DEDICATION
I would like to dedicate this dissertation to my wife Stephanie for providing me with support and
motivation throughout my years in graduate school.
v
ACKNOWLEDGEMENTS
This dissertation would not have been possible without the generous support from a
private donor, whose donation and insight led to the construction of the Joshua Nelson Heat
Recovery Composting Facility, where a majority of this research was conducted. Funding for
this research was also provided by the University of New Hampshire Agricultural Experiment
Station (AES) through a USDA Hatch grant (NH00605) and a USDA McIntire-Stennis grant
(NH00073-M). Extensive research funding was also provided by a USDA Sustainable
Agriculture Research and Education (SARE) grant. Finally, funding for the last year of my
doctoral program was provided by the UNH Graduate School through a Dissertation Year
Fellowship.
I would also like to acknowledge the undergraduate work-study students that assisted
with this project over the past five years (Sarah J Ehrmentrau, Zach Charewicz, Pat Cota, Pia K
Marciano, Katerina N Messologitis and Charles Simms). Their work and high motivation made
this research project run much more smoothly.
I would also like to acknowledge the dissertation committee members (John Aber,
Richard Smith, Theodore Howard, Heather Darby, Mark Ducey, and Bruce Fulford) for
providing a tremendous amount of support and freedom with this project. In particular, I would
like to acknowledge my good friend and committee chair, John Aber, who brought me into this
amazing project five years ago and has been incredibly supportive in both my academic and
personal life.
Finally, I would like acknowledge my family, who provided tremendous support and
guidance throughout my educational career and were my greatest advocates and editors.
vi
Table of Contents DEDICATION .......................................................................................................................... iv
ACKNOWLEDGEMENTS.........................................................................................................v
LIST OF TABLES .................................................................................................................... xi
LIST OF FIGURES .................................................................................................................. xi
ABSTRACT ............................................................................................................................xiv
CHAPTER 1: INTRODUCTION TO THE AGROECOSYSTEM STUDY .................................1
BACKGROUND .........................................................................................................1
PURPOSE OF THE AGROECOSYSTEM STUDY ....................................................5
RESEARCH SITE AND ORIGINS OF THE STUDY .................................................6
Agroecosystem Site .................................................................................................6
Origins of the Agroecosystem Project ......................................................................7
Baseline Agroecosystem Health Assessment (Bedding, Manure, Energy) ................7
Bedding Usage (Pre-2012) .......................................................................................7
Manure Management (Pre-2012) ..............................................................................8
Energy use for Hot Water Heating (Pre-2012) ........................................................ 11
Baseline Values as a Whole ................................................................................... 11
DISSERTATION CHAPTER SUMMARIES ............................................................ 11
Chapter 2: Animal bedding cost and somatic cell count across New England dairy
farms: Relationship with bedding material, housing type, herd size, and management
system.................................................................................................................... 12
Chapter 3: Financial viability of producing animal bedding with a wood shaving
machine ................................................................................................................. 12
Chapter 4: Heat recovery from composting – A comprehensive literature review ... 13
Chapter 5: Heat recovery from composting: A step-by-step guide on building an
aerated static pile heat recovery composting facility ............................................... 14
Chapter 6: Recovering heat from composting as an innovative approach to reduce
greenhouse gas emissions and fossil fuel consumption in the agricultural sector .... 15
Chapter 7: Conclusion ............................................................................................ 15
REFERENCES .......................................................................................................... 17
CHAPTER 2: ANIMAL BEDDING COST AND SOMATIC CELL COUNT ACROSS NEW
ENGLAND DAIRY FARMS: RELATIONSHIP WITH BEDDING MATERIAL, HOUSING
TYPE, HERD SIZE, AND MANAGEMENT SYSTEM ........................................................... 21
ABSTRACT .............................................................................................................. 21
INTRODUCTION ..................................................................................................... 22
MATERIALS AND METHODS ............................................................................... 24
vii
Survey Tool ........................................................................................................... 24
Sample Size and Selection ..................................................................................... 25
Questionnaire Mailings .......................................................................................... 26
Statistical Analysis ................................................................................................. 26
Non-Response Bias ................................................................................................ 27
RESULTS AND DISCUSSION................................................................................. 28
Response Rate and Survey Demographics .............................................................. 28
Bedding Usage ....................................................................................................... 30
Bedding Preference ................................................................................................ 31
Bedding Avoidance ................................................................................................ 32
Bedding Cost ......................................................................................................... 33
Bedding Rank Compared to Other Farming Expenses ............................................ 33
Relationship between Bedding Material with Cost and SCC .................................. 34
Relationship between Housing Type with Bedding Cost and SCC .......................... 36
Relationship between Dairy Cow Herd Size with Bedding Cost and SCC .............. 37
Relationship between Management System with Bedding Cost and SCC ............... 39
Interest in Producing Bedding with a Wood Shaving Machine ............................... 39
LIMITATIONS OF THE STUDY ............................................................................. 41
CONCLUSIONS ....................................................................................................... 41
ACKNOWLEDGEMENTS ....................................................................................... 42
REFERENCES .......................................................................................................... 43
APPENDICES ........................................................................................................... 48
Appendix 1: IRB Approval Letter .......................................................................... 48
Appendix 2: Questionnaire sent to New England Dairy Farmers ............................ 49
CHAPTER 3: FINANCIAL VIABILITY OF PRODUCING ANIMAL BEDDING WITH A
WOOD SHAVING MACHINE ................................................................................................ 53
ABSTRACT .............................................................................................................. 53
INTRODUCTION ..................................................................................................... 54
METHODS ............................................................................................................... 56
Research Site ......................................................................................................... 56
Wood Shaving Machine and Accessories ............................................................... 57
Wood Shaving Machine Financial Decision Model ................................................ 57
Background Model Calculations ............................................................................ 59
Model Output......................................................................................................... 61
Parameter Values used in this Study ....................................................................... 61
viii
RESULTS AND DISCUSSION................................................................................. 62
Scenario 1: On-Farm Production Value Only ......................................................... 62
Scenario 2: Animal Bedding Production for Profit ................................................. 65
Operational Recommendations Relating to Economic Feasibility ........................... 69
ADDITIONAL CONSIDERATIONS AND LIMITATIONS OF THE STUDY ......... 71
CONCLUSION ......................................................................................................... 73
REFERENCES .......................................................................................................... 76
APPENDICES ........................................................................................................... 77
Appendix 1: Calculation of Expansion Factor between Solid Wood and Shavings . 77
Appendix 2: Log Loading Time ............................................................................. 78
Appendix 3: Machine Output Adjusted for Log Loading ........................................ 80
Appendix 4: Time Requirement to Load Logs onto Debarking Platform ................ 81
Appendix 5: Time Requirement to Debark Logs Manually ..................................... 82
CHAPTER 4: HEAT RECOVERY FROM COMPOSTING – A COMPREHENSIVE
HISTORICAL REVIEW OF SYSTEM DESIGN, RECOVERY RATE AND UTILIZATION . 83
ABSTRACT .............................................................................................................. 83
INTRODUCTION ..................................................................................................... 83
COMPOSTING HEAT RECOVERY PRINCIPLES AND APPLICATIONS ............ 84
HISTORY OF RECOVERING HEAT FROM COMPOSTING ................................. 87
Modern Era Begins 1971-1980 .............................................................................. 88
1981 – 1990 ........................................................................................................... 90
1991-2000.............................................................................................................. 93
2001 – 2010 ........................................................................................................... 96
2011-2016.............................................................................................................. 99
CONCLUDING THOUGHTS ................................................................................. 103
REFERENCES ........................................................................................................ 108
CHAPTER 5: HEAT RECOVERY FROM COMPOSTING: A STEP-BY-STEP GUIDE ON
BUILDING AN AERATED STATIC PILE HEAT RECOVERY COMPOSTING FACILITY
............................................................................................................................................... 113
EXECUTIVE SUMMARY ...................................................................................... 113
TECHNOLOGY OVERVIEW ................................................................................ 114
Aerobic Heat Production vs. Anaerobic Biogas Production .................................. 114
Heat Production from Composting ....................................................................... 115
Heat Recovery from Composting ......................................................................... 117
Acrolab’s Isobar® Heat Pipe Technology ............................................................ 121
ix
The Value of Heat ................................................................................................ 123
ORIGINS OF THE UNH PROJECT ........................................................................ 124
PLANNING AND SIZING THE FACILITY ........................................................... 125
Feedstock Parameters ........................................................................................... 125
Assess Current Hot Water Demand and Location ................................................. 126
Feedstock Residence Time within Facility ........................................................... 127
Sizing the Facility ................................................................................................ 127
Aeration Floor Design .......................................................................................... 129
Facility Location .................................................................................................. 134
BUILDING A HEAT RECOVERY COMPOSTING FACILITY ............................ 135
Site Preparation.................................................................................................... 136
Underground Slab and Concrete Wall Preparation ............................................... 136
Pouring Concrete Walls ....................................................................................... 138
Insulating the Concrete Slab and Setting up the Aeration Ductwork ..................... 144
Structural Support (Joints and Pad Reinforcement) .............................................. 146
Installing the Aeration Channels .......................................................................... 148
Pouring the Slab and Finishing the Composting Floor .......................................... 153
Prepping and Pouring the Internal Concrete Apron .............................................. 159
Prepping and Pouring the External Concrete Apron ............................................. 160
Installing the Leachate Network and Pouring the Mechanical Room Floor ........... 161
Raising the Building ............................................................................................ 162
Setting up the Mechanical Room .......................................................................... 164
Aeration Lines ..................................................................................................... 165
Primary Aeration Supply and Exhaust Lines ........................................................ 168
Installing Agrilab Technologies Isobar Unit ......................................................... 171
Installing the Aeration Control System ................................................................. 178
Setting up the Aeration Schedule for Heat Recovery ............................................ 179
Testing and Insulating the System ........................................................................ 181
COST OF THE UNH HEAT RECOVERY COMPOSTING FACILITY ................. 183
CONCLUDING THOUGHTS ................................................................................. 184
REFERENCES ........................................................................................................ 187
APPENDICES ......................................................................................................... 192
Appendix 1: Materials List & Estimated Cost of a Similarly-Sized Facility .......... 192
Appendix 2: Summary Specs from other Heat Recovery Composting Sites .......... 196
x
Appendix 3: UNH Facility Layout ....................................................................... 199
Appendix 4: Quantity of Concrete used at UNH Facility ...................................... 200
Appendix 5: Diagrams for UNH Pole Barn .......................................................... 201
Appendix 6: Breakdown of the Cost for the UNH Facility ................................... 206
Appendix 7: Recommended Cost-Saving Strategies Found Throughout Report .... 207
Appendix 8: Summary Steps for UNH Facility Construction ................................ 209
CHAPTER 6: RECOVERING HEAT FROM COMPOSTING AS AN INNOVATIVE
APPROACH TO REDUCE GREENHOUSE GAS EMISSIONS AND FOSSIL FUEL
CONSUMPTION IN THE AGRICULTURAL SECTOR ........................................................ 211
ABSTRACT ............................................................................................................ 211
INTRODUCTION ................................................................................................... 211
MATERIALS & METHODS................................................................................... 215
Research Site and Operational Considerations...................................................... 215
Compost Recipe ................................................................................................... 215
Compost Aeration and Vapor Temperature for BTU Estimation........................... 215
Heat Exchange System......................................................................................... 217
Heat Recovery Rate ............................................................................................. 217
Energy Equivalent to Fossil Fuel .......................................................................... 218
Greenhouse Gas Mitigation from Emission Reductions and Carbon Storage ........ 219
Statistical Analysis ............................................................................................... 219
RESULTS & DISCUSSION .................................................................................... 219
Average Compost Vapor and Pile Temperature at Facility ................................... 219
Heat Recovery Rate ............................................................................................. 220
Fossil fuel Equivalent .......................................................................................... 225
Greenhouse Gas Mitigation .................................................................................. 227
System Potential and Room for Improvement ...................................................... 229
CONCLUSION ....................................................................................................... 231
REFERENCES ........................................................................................................ 233
APPENDIX ............................................................................................................. 237
Appendix 1: Average Compost Vapor and Pile Temperatures by Age for Additional
Batches used in the Study .................................................................................... 237
CHAPTER 7: CONCLUDING THOUGHTS ON THE AGROECOSYSTEM STUDY .......... 238
Chapter 2: Animal bedding cost and somatic cell count across New England dairy
farms: Relationship with bedding material, housing type, herd size, and management
system.................................................................................................................. 238
xi
Chapter 3: Financial viability of producing animal bedding with a wood shaving
machine ............................................................................................................... 241
Chapter 4: Heat recovery from composting – A comprehensive literature review . 243
Chapter 5: Heat recovery from composting: A step-by-step guide on building an
aerated static pile heat recovery composting facility ............................................. 246
Chapter 6: Recovering heat from composting as an innovative approach to reduce
greenhouse gas emissions and fossil fuel consumption in the agricultural sector .. 247
Agroecosystem Study as a Whole ........................................................................ 249
REFERENCES ........................................................................................................ 252
LIST OF TABLES
Table 1: Dairy Farm Attributes vs. 2012 USDA Census of Agriculture .......................... 29
Table 2: Bedding Usage % by Number of Dairy Farms across New England ................. 30 Table 3: Reasons for Preferring a Bedding Material ....................................................... 31
Table 4: Reasons for Avoiding a Bedding Material ........................................................ 32 Table 5: Overall Cost of Bedding Materials for New England Dairy Farmers ................ 33
Table 6: Rank of Bedding Cost in Relation to Other Farming Expenses ......................... 34 Table 7: Reported Bedding Cost and SCC for Various Bedding Materials ..................... 35 Table 8: Cost and SCC in Relation to Housing Type ...................................................... 36
Table 9: Cost and SCC in Relation to Farm Size ............................................................ 38 Table 10: Cost and Somatic Cell Count in Relation to Management System .................. 39
Table 11: Primary uses for Dairy Farm Woodlots across New England .......................... 40 Table 12: Parameters for the Economic Decision Model ................................................ 58
Table 13: UNH ODRF Parameter Values ....................................................................... 62 Table 14: Cost Reductions of Producing Bedding for On-Farm Production Only ........... 63
Table 15: On-Farm Bedding Production Breakeven Point .............................................. 64 Table 16: Profit Potential of Producing Bedding using a Wood Shaving Machine .......... 66
Table 17: Profit Potential if Producing Shavings with Employees Receiving a Wage ..... 67
LIST OF FIGURES
Figure 1: Pre-2012 Manure and Farm Waste Residual Storage Site ..................................8
Figure 2: Pre-2012 Manure Pit .........................................................................................9 Figure 3: Survey Response by Herd Size vs. 2012 USDA Census Data .......................... 29
Figure 4: UNH Wood Shaving Machine Being Loaded .................................................. 57 Figure 5: Example of Parameter Values ......................................................................... 59
Figure 6: Example of Parameter Definitions .................................................................. 59 Figure 7: Basic Formula for Aerobic Composting ........................................................ 115
Figure 8: Compost Temperature over Compost Age for UNH Experimental Batch 2 ... 115 Figure 9: Internal Workings of Acrolab's Isobar Heat Pipe (Acrolab 2013) .................. 121
Figure 10: UNH Isobar Heat Exchange System (Loughberry Manufacturing 2012) ..... 122 Figure 11: Flow Diagram of Heat Recovery System (Agrilab 2013) ............................ 123
xii
Figure 13: Aeration Hole Diameter by Pipe Length for Holes 6’’ and 12'' on Center .... 130 Figure 14: Aeration Floor Spacing at UNH Compost Facility ...................................... 131
Figure 14: Floor Spacing for the Aeration Lines at the UNH Compost Facility ............ 132 Figure 15: Aerial View of UNH Organic Dairy Research Farm ................................... 135
Figure 16: Insulated Underground PEX Pipe ............................................................... 137 Figure 17: Compost Leachate Tank at UNH Facility .................................................... 137
Figure 18: Concrete Dimensions at UNH Compost Facility ......................................... 139 Figure 19: Concrete Piers at the UNH Compost Facility .............................................. 139
Figure 20: Back Mechanical Room after first Concrete Pour ........................................ 140 Figure 21: Back Mechanical Room Concrete Dimensions ............................................ 141
Figure 22: Back Push Wall at UNH Composting Facility ............................................. 141 Figure 23: Vapor Barrier on Back Push Wall at UNH Composting Facility.................. 142
Figure 24: High-Tension Fabric Structure with Waste Block Walls (ClearSpan 2013) . 143 Figure 25: Insulation below Main Composting Floor at UNH Composting Facility ...... 144
Figure 26: Thermal Break Installation against Internal Walls ....................................... 146 Figure 27: Concrete Slab-Connecting Dowels at UNH Composting Facility ................ 146
Figure 28: Expansion Joint between Compost Floor and External Apron ..................... 147 Figure 29: Aeration Line Form Setup Prior to Concrete Pour ....................................... 149
Figure 30: Aeration Lines through Back Push Wall with Hose Bibbs ........................... 150 Figure 31: Aeration Line Forms for Cover Plates ......................................................... 151
Figure 32: Alternate Cover Plate Form Setup (Jerose 2013) ......................................... 152 Figure 33: Alternate Cover Plate Form Setup (Jerose 2013) ......................................... 152
Figure 34: Alternate Cover Plate Form Setup (Jerose 2013) ......................................... 153 Figure 35: Concrete Pour for Main Composting Floor at UNH Facility ........................ 154
Figure 36: Profile of Compost Pile and Floor at UNH Facility ..................................... 155 Figure 37: Drilling of Aeration Holes .......................................................................... 155
Figure 38: Drilling Location for Leachate Holes .......................................................... 156 Figure 39: Profile of a Cover Plate over an Aeration Line ............................................ 157
Figure 40: Profile of the UNH Aeration Floor and Subfloor ......................................... 158 Figure 41: Cover Plate Aeration Hole with Chipping ................................................... 158
Figure 42: Aeration Line with Welded Wire Mesh Impeding Proper Cover Plate Fit.... 159 Figure 43: Specifications when Pouring the Internal Concrete Pad ............................... 160
Figure 44: Specifications when Pouring the External Concrete Pad.............................. 161 Figure 45: Specifications when Pouring Back Mechanical Floor .................................. 162
Figure 46: Specifications for Mechanical Room Ceiling .............................................. 165 Figure 47: Butterfly Valve Connecting Aeration Line to Aeration Header ................... 165
Figure 48: Setup for Aeration Headers ......................................................................... 166 Figure 49: Exhaust Vapor Recirculation Connection Points ......................................... 167
Figure 50: Leachate Hookup Specifications for Each Pair of Aeration Lines ................ 168 Figure 51: Biofilter Processing Exhaust Vapor from UNH Composting System ........... 170
Figure 52: Pipe Stands for Aeration System at UNH Facility ....................................... 171 Figure 53: Delivery of Agrilab Technologies Heat Exchange Unit ............................... 171
Figure 54: Support Structure for Bulk Storage Tank of Water ...................................... 172 Figure 55: Aeration Line Hookup with Heat Exchange Unit ........................................ 173
Figure 56: Connections to and from Primary Blower ................................................... 174 Figure 57: Hot Water Supply and Return Lines from Isobar Unit ................................. 174
xiii
Figure 58: Underground PEX Hot Water Supply and Return Lines .............................. 175 Figure 59: Water Lines between Composting Facility and Milk Room......................... 175
Figure 60: Plate Exchanger in Milk House ................................................................... 176 Figure 61: Hot Water Heater Receiving Compost-Heated Water .................................. 176
Figure 62: Isobars within the Bulk Storage Tank (unfilled) .......................................... 177 Figure 63: Agrilab's Aeration Control System .............................................................. 178
Figure 64: Compressor Running Pneumatic Lines in the Aeration System ................... 179 Figure 65: UNH Aeration System following Insulation ................................................ 182
Figure 66: CHRS Diagram for UNH Facility ............................................................... 216 Figure 67: Typical Composting Temperature over Time at the UNH Facility ............... 220
Figure 68: Compost Heat Recovery Rate vs. Vapor Temperature ................................. 221 Figure 69: Heat Recovery Rate by HST Temperature at the UNH Facility ................... 222
Figure 70: GHRS Energy Saving Equivalent by Average Vapor Temperature.............. 225 Figure 71: Greenhouse Gas Emission Avoidance in Oil and Propane from the CHRS .. 227
xiv
ABSTRACT
CREATING AN ECONOMICALLY VIABLE, CLOSED-SYSTEM, ENERGY-
INDEPENDENT DAIRY FARM THROUGH THE ON-FARM PRODUCTION OF ANIMAL
BEDDING AND HEAT CAPTURE FROM AN AERATED STATIC PILE HEAT RECOVERY
COMPOSTING OPERATION
by
Matthew M. Smith
University of New Hampshire, May, 2016
This study explored two innovative approaches to assist dairy farmers in reducing costs
associated with animal bedding and waste accumulation on their farms. This was accomplished
by finding value in underutilized farm resources and waste streams and recycling them back
through the system, or converting them into revenue generating products to be exported off the
farm. The University of New Hampshire Organic Dairy Research Farm was used as a case study
for this agroecosystem project.
The first innovative approach was the production of animal bedding using a wood
shaving machine and low-grade eastern white pine (Pinus strobus L) harvested from the farm’s
woodlot. This approach was selected because a pre-study survey of 129 New England dairy
farmers conducted as part of this research, indicated that bedding costs increased by 89% from
2003 – 2013. This substantial cost increase raised the obvious question, is there a possible cost
saving alternative to outright purchasing of bedding material? After two years of using the wood
xv
shaving machine, a financial decision model was developed, allowing individuals to input
operator-specific parameters to determine the feasibility of producing wood shavings at their site.
Model output indicated that the average-sized organic or conventional dairy farm in New
England would not have a favorable payback period, requiring up to 11 years to pay for the
machine, if producing bedding for only on-farm consumption. Large farms, or a cooperative of
smaller farms using a trailer-mounted machine were found to have the greatest likelihood of
positive returns from this type of venture.
The second innovative approach involved the construction and operation of a novel
composting system, capable of processing the farm’s wastes, while recovering thermal energy
from the process for on-farm hot water heating. A detailed review of 45 compost heat recovery
systems was used to assess which system to build at the UNH site. Based on this review, the
aerated static pile (ASP) composting method with Agrilab Technologies’ heat exchange system
was selected. During the construction phase of this facility, a detailed report on how to build an
ASP heat recovery composting facility was developed for use by compost practitioners
contemplating use of this technology. Upon completion of this facility, research trials began on
heat extraction and greenhouse gas (GHG) mitigation potential of the system. Results indicated
that the UNH system was capable of recovering 293,909 - 811,332 BTU/day with compost vapor
temperatures in the 121 – 133°F range. With regard to fossil fuel offsetting, the compost heat
exchange unit would reduce oil consumption by 911 – 2515 gallons per year. From a GHG
mitigation standpoint, this is equivalent to an offset of 10 – 28 TCO2/yr.
1
CHAPTER 1: INTRODUCTION TO THE AGROECOSYSTEM STUDY
BACKGROUND
In the last few decades the overall number of dairy farms in New England has decreased
considerably. This trend has also been seen on a national level, with an 88% drop in dairy farms
from 1970 – 2006 (MacDonald et al. 2007). What has become problematic for the dairy industry
is the fact that U.S. milk consumption has been declining every decade since 1940 (Steward et al.
2013). However, when looking at the organic dairy component of the market, the number of
organic farms and animals has been increasing (McBride and Greene 2009), with 2,265 certified
cows in 1992, compared to 254,771 certified organic cows in 2011 (USDA ERS 2014).
Although it is important to make a distinction between the different trends occurring in
the organic and conventional dairy markets, one trend they do have in common is consolidation
of livestock, with an increasing number of farmers moving toward larger operations. From 2000
– 2006, the number of conventional operations with ≥ 2000 cows doubled, while farms with <
200 cows shrunk rapidly (MacDonald et al. 2007). A major reason for this trend was cost
reductions associated with economies of scale. Conventional dairies with fewer than 50 cows had
production costs per hundredweight (cwt) almost double that of operations with 200-499 cows
(MacDonald et al. 2007).
When looking at organic dairies, there has also been a growing trend toward large-scale
production, especially in the western portion of the United States, which accounts for 80% of the
dairies with more than 200 cows. This is in stark contrast to the Northeast, where the average
organic herd size is 53 cows, and 87% of operations milk fewer than 100 cows (McBride and
Greene 2009). However, few of these small-scale organic dairies produce a positive net return
2
over long run capital costs. McBride and Greeene (2009) also found that increasing the size of
the operation increased the likelihood of generating enough return to cover capital and labor
costs. Should a large percent of smaller organic dairies continue to have a negative return, it is
likely that there will be a shift toward consolidation similar to that occurring in the conventional
market. However, the degree to which consolidation occurs is limited by the farmer’s ability to
graze their livestock. This is due to the USDA National Organic Program ruling, requiring a
grazing season of at least 120 days, with at least 30% of the dry matter intake coming from
pasture (USDA 2010).
The important message regarding conventional and organic dairy operations pertinent to
this research is that consolidation of livestock on fewer acres of land has a greater propensity to
create problems with nutrient management. This is because the land base on which such
operations exist cannot absorb the quantity of manure, spent animal bedding and other farm
wastes being produced on the farm. Controlling these waste streams is important, as leaching of
nutrients can lead to eutrophication and hypoxia of waterways (Bittman and Mikkelsen 2009,
Galloway et al. 2003, Vitousek et al. 1997). The leaching of nutrients and sediment is also why
agriculture is the leading contributor to the impairment of rivers and lakes in the United States.
More specifically, 60% of the impaired river miles and 50% of the impaired lakes by acreage are
due to agriculture (EPA 2012). Agriculture is also the third leading contributor to the impairment
of estuaries in the US. Together, these impairments result in 40% of the surveyed rivers, lakes
and estuaries in the U.S. not being clean enough for fishing or swimming (EPA 2012).
In addition to water pollution, improper management of agricultural wastes also results in
increased greenhouse gas (GHG) emissions, with agriculture representing 9% of anthropogenic
sources in the U.S. (US EPA 2014c). Additionally, the U.S. has seen a 17% increase in GHG
3
emissions from the agriculture sector since 1990 (US EPA 2014c), mainly from nitrous oxide
(N2O) and methane (CH4) coming from waste products stored under anaerobic conditions
(Hellebrand 1998). When looking at N2O, the agriculture sector contributes 75% of the U.S.
emissions (US EPA 2014b). This is troubling as N2O has a global warming potential 310 times
that of CO2 (Utami et al. 2011) and is expected to increase by 5% by 2020, primarily from the
agricultural sector (US EPA 2014b). Agriculture is also the leading contributor of CH4 emissions
to the environment, representing 36% of emissions (EPA 2014a). CH4 is also 25 times more
potent as a greenhouse gas than CO2 (US EPA 2014a).
In addition to GHG emissions, improper management of agricultural wastes increases the
occurrence of foul odors (hydrogen sulfide - H2S and volatile fatty acids), originating from
anaerobic conditions of decomposing material (Chiumenti et al. 2005, Wright 2001, Rynk et al.
1992). Improper management of agricultural wastes also increases emissions of reactive nitrogen
(Nr) to the atmosphere. In the Northeast, agriculture contributes 76% of ammonia (NH3)
emissions (Smith et al. 2006). This is especially important for farming regions, as NH3 is often
deposited within a close proximity to its source. Increased emissions of NH3 and other forms of
Nr are problematic, as they can cause: decreased forest productivity in areas with Nr saturation
(Aber et al. 1995), increased acidity in precipitation leading to acidic surface water and soil
(Bittman and Mikkelsen 2009, Galloway and Cowling 2002), increased ozone in the troposphere,
decreased ozone in the stratosphere (Galloway and Cowling 2002), decrease in visibility
(Bittman and Mikkelsen 2009), and increase in respiratory illness, cancer and cardiac disease
(Bittman and Mikkelsen 2009, Galloway et al. 2003). Compounding these environmental issues
is the fact that Nr can have a cascading effect, where a single atom of N can travel through
ecosystems causing multiple negative consequences along the way (Galloway et al. 2003). For
4
instance, a molecule of nitrogen oxide (NOx) can cause photochemical smog, and after being
oxidized in the atmosphere to nitric acid (HNO3), can then cause acidification or eutrophication
of waterways (Grubber and Galloway 2008).
With an ever-increasing population, the negative impacts associated with improper
management of agricultural wastes, especially on dairy farms, will only increase. This is because
the increase in food production necessary to feed the growing population will come from
intensification on existing land, which will further concentrate wastes. Land expansion through
extensification will also bring agriculture closer to non-farm neighbors. Either path has the
potential to increase the negative interactions between farmers, their adjacent neighbors, and
those living downstream. This will likely set the stage for new and more stringent regulations
regarding farm waste streams. Currently, a best management approach is used to handle
agricultural wastes in the U.S. However, this approach was recently challenged by a coalition of
environmental groups, who brought two lawsuits (Civil Action No. 15-cv-0141 and Civil Action
No. 15-cv-139) against the EPA for not regulating the nation’s largest farms. The plaintiffs argue
that the largest farms should be considered as pollution sources and therefore regulated under the
Clean Air Act. More specifically, the lawsuits aim at addressing NH3 and H2S on the grounds
that both gases damage humans and the environment.
While a regulatory approach is a plausible method to curb the negative externalities from
the agriculture sector, this approach will hurt the bottom line of the dairy industry, which is
already struggling. Additionally, it is not likely that smaller farms will be regulated, meaning
they will continue to be managed under a voluntary approach. What is problematic with the
volunteer approach is that it is only as good as the financial stability of the farmer. During times
of economic hardship, recommended practices that reduce environmental damage but may be
5
more costly will not be adhered to as closely, unless there are economic incentives to manage
differently. As a consequence, new and innovative approaches are needed to assist dairy farmers
in better managing their inputs and outputs coming to and from their agroecosystem (agricultural
ecosystem). More importantly, these approaches need to be scalable and economically feasible
for farmers to actually pursue on their farms.
PURPOSE OF THE AGROECOSYSTEM STUDY
This research explores new and innovative approaches to assist dairy farmers in reducing
externalities associated with waste accumulation on their farm. This is accomplished by finding
value in underutilized farm resources and waste streams and recycling them back through the
system or by converting them into revenue generating products that can be exported off the farm.
The two innovative approaches are producing animal bedding from the farm’s low-grade
woodlot using a wood shaving machine, and building and operating a novel composting system
capable of processing all the farm’s waste streams, while also recovering the thermal energy for
on-farm hot water heating. Both approaches are viewed from an agroecosystem and economic
approach.
An agroecosystem approach is used in this research out of recognition that the most
effective method in controlling pollution from leaving the farm system is through input/output
management, where internal and external flows are assessed (Odum 1989). Bohlen and House
(2009) also described how the single greatest step to increase economic sustainability,
environmental compatibility and biological resilience of agroecosystems is to design manure
management systems that are more consistent with the nutrient cycles of natural systems. They
also stressed the importance of developing systems that are more energy efficient, less dependent
on external inputs and instead rely on internal ecosystem control processes. These ideas are in
6
line with the agroecosystem principles outlined in Gitau et al. (2009), who described how proper
management of agroecosystems results in a level of productivity based on the rate of resource
regeneration, the rate the environment is able to absorb waste and byproducts, and by developing
technology that converts non-renewable resources to renewable forms. The focus of the present
research is in line with these agroecosystem concepts, as producing animal bedding onsite and
composting with heat recovery both reduce farm inputs, while increasing outputs. These
strategies also result in a more sustainable agroecosystem capable of recirculating and exporting
nutrients vs. accumulating them until they leave the farm as a form of pollution. Importantly, the
more internal cycling within the agroecosystem, the more it resembles a natural ecosystem, and
the more sustainable it tends to be in the long run (Gliessman 2001).
An economic component is also presented throughout this research, to address whether it
is possible to create high enough economic returns or cost reductions to make the more
sustainable approach the more practical and profitable option for the dairy farmer. This concept
was described in Gliessman (2001), who described how an agroecosystem can only be
sustainable if economic and cultural systems encourage the sustainable practices and not
undermine them.
RESEARCH SITE AND ORIGINS OF THE STUDY
Agroecosystem Site
The research site for this study was the University of New Hampshire Organic Dairy
Research Farm (ODRF) in Lee, NH. The farm has 300 acres of land with 140 in forage, 120 in
woodlot, and 40 in pasture. The organic Jersey herd is maintained around 100 head, with 50
cows and 50 replacement animals. During the grazing season, the 40 acres of pasture are
managed under intensive rotational grazing. During the winter season, the herd is fed organic
7
baleage harvested from the 140 acres of forage land. Additional feed in the form of round bales
(40 bales/year) and grain (91 tons/yr) are also imported to the farm.
Origins of the Agroecosystem Project
The focus area for this agroecosystem research originated from conversations between
UNH researchers and organic dairy producers (Stoneyfield Organic, Organic Valley, Horizon
Organic, and Aurora Organic Dairy). The consensus from these discussions was that new and
innovative approaches were needed to reduce input costs for feed, energy, and bedding on dairy
farms, while also reducing the farm’s ecological footprint through more effective waste
management. Of these research topics, this study focused on bedding, energy, and the waste
management system of the agroecosystem.
Baseline Agroecosystem Health Assessment (Bedding, Manure, Energy)
In order to gauge the success of the project, a general agroecosystem health assessment
was conducted prior to the initiation of the study in 2012. As described in Gitau et al. (2009), this
baseline compares a set of indicators between the pre and post-management time period. The
specific indicators used in this study were bedding usage on the farm, manure management, and
hot water heating needs.
Bedding Usage (Pre-2012)
Prior to the purchase of the wood shaving machine, the UNH ODRF used kiln-dried
eastern white pine (Pinus strobus L) shavings for its bedded pack system. The bedded pack was
managed by applying a 2’’ layer of fresh bedding once a week in the summer and 2-3 times a
week during the winter. Manure was also removed from the pack twice daily, while the cows
were being milked. Manure removal is necessary in these systems, as it reduces the nutrient
source for disease causing bacteria (McFarland 2009). Removal of manure also reduces the
8
frequency in which the pack has to be bedded, saving cost in material purchases. Before the
purchase of the wood shaving machine, the farm purchased 700 yd3 of bedding per year, at a cost
of $1800-$2100 per 100 yd3 truckload. This input represented an annual cost of $12,600 -
$14,700 to the farm each year.
Before the agroecosystem study, the bedded pack was cleaned out once a year in October,
before the cows came in from pasture for winter housing. Emptying the pack was a two day
process, where material was loaded into a dump truck with a tractor. Spent bedding would then
be transported to the farm’s primary manure and waste storage area, where it would be piled and
left to decompose (Figure 1).
Figure 1: Pre-2012 Manure and Farm Waste Residual Storage Site
Manure Management (Pre-2012)
Prior to the agroecosystem study, the farm’s manure was collected and stored in a lower
field in unmanaged piles. Roughly 270 wet tons of manure were collected during the winter
months (November – April) from the feed yard, bedded pack, and alleyways. All of the manure
deposited during the winter was collected and put into a manure pit for temporary storage.
Because of the pasture-based system, manure was not collected during the summer months, as it
was scattered across the pasture. Manure totals for the winter months were based on production
9
estimates from the literature by cow breed, with Jersey cows excreting 74 lbs of wet feces/day, or
12.5 lbs dry feces/day (Knowlton et al. 2010). The Nr content of this material is roughly 10-11
lbs of Nr/ton of manure, with a daily total of 0.7 lbs Nr/day/cow (Knowlton et al. 2010, Jokela
and Peters 2009). Total annual N was calculated at roughly 223-260 lbs N/year/cow, with the
range in N depending on feed (Pennington et al. 2013, Knowlton et al. 2010, Jokela and Peters
2009).
Prior to this study, manure was scraped into the small manure pit with a skid steer.
Unfortunately, the area above this manure pit was sloped in a manner that allowed water from
the bedded pack roof and yard to drain into the pit. Only a small lip and channel prevented water
from entering the pit. However, the undersized lip and shallow channel would often become
clogged with manure or snow, allowing rain water to inundate the pit and cause the semi-solid
manure to become a slurry (Figure 2).
Figure 2: Pre-2012 Manure Pit
After being saturated with water, the manure posed a serious challenge with regard to transport.
Often, several buckets of spent bedding would have to be blended into the mix to be able to
transport the material, as the farm did not have equipment to handle the manure as a slurry. A
10
study from Pattey et al. (2005) also found that manure stored as slurry has the greatest GHG
potential when compared to manure stored in a stockpile or composted. The primary contributors
to the elevated GHG emissions were CH4 and N2O (Pattey et al. 2005). In addition to serving as a
source of GHG emissions, the past management also resulted in elevated nitrate (NO3-) in the
groundwater in areas where the manure was stockpiled. High rain events would also cause the
manure pit to overflow, carrying manure into the heifer field.
If following the slope of the land, the nutrient plume from the frequent wash-outs likely
ended up in the farm wetland. Interestingly, ground and surface water sampling from
collaborating researchers did not find nutrient loading in the wetland, suggesting that a majority
of the nitrogen was likely being denitrified. However, continual nutrient loading would likely
result in pollution concerns with phosphorus, as the persistent supply of exported P to the system
could surpass the system’s internal cycling. Hashemi et al. (2014) reported that 60-85% of the P
in feed is excreted in manure. For dairy operations, this translates to an excretion of 4, 3, and 2
lbs/ton of P2O5 for milking cows, dry cows, and calf/heifers, respectively (Hashemi et al. 2014,
Beegle et al. 2014). Importantly, the farm wetland drains into the Lamprey River, which is
designated as wild and scenic. Fixing the manure pit was an immediate priority for this research
study.
After the manure pit was filled (roughly 20 yd3 capacity) the material would be loaded
into a manure spreader and unloaded into the farm’s primary manure storage area. The area of
manure storage prior to this study was located in the back corner of a field in a low spot, which
was frequently flooded with water. This did not pose a problem during the winter months, when
the ground was frozen, but made removing the material in the spring and fall almost impossible.
Often times, the area was too wet to remove the material, resulting in manure accumulation. As a
11
consequence, there was almost a 3-yr period were manure was stockpiled and not spread (Figure
1).
As with the manure pit, the saturated manure storage area was a source of both air and
water pollution from the farm. Additionally, the long residence time between manure unloading
and field application resulted in tremendous weed growth on the piles. Unfortunately, this likely
resulted in the spread of weed seeds on the pasture and hayfields following manure application.
Energy use for Hot Water Heating (Pre-2012)
Prior to the agroecosystem study, Pellissier and Benander (2011) tracked the farm’s hot
water heating from past energy bills and fuel oil deliveries. In their study, they found that the
ODRF consumed 212,000,000 BTU annually for hot water heating in the milk parlor. Together,
the electricity and fuel costs amounted to $6869/yr using 2011 oil and electricity prices
(Pellissier and Benander 2011).
Baseline Values as a Whole
When looking at the baseline values for the inputs (bedding and energy) and outputs
(spent bedding, manure, and waste feed) at the onset of this agroecosystem study, it was clear
that a nutrient and financial imbalance was occurring and needed to be addressed immediately.
The following section describes the various approaches used to correct these imbalances, by
finding value in underutilized or waste products within the agroecosystem.
DISSERTATION CHAPTER SUMMARIES
Each dissertation chapter covers a separate component of the agroecosystem, which are
connected by the nutrients and embedded energy that flow from one component of the system to
the other. By way of example, carbon harvested in the farm woodlot for the production of
shavings is used in the bedded pack barn for bedding. It then becomes a carbon source for
12
microbes in the composting facility, and is finally returned to the farm fields where it is used as a
soil amendment. Below are chapter summaries for this dissertation.
Chapter 2: Animal bedding cost and somatic cell count across New England dairy farms:
Relationship with bedding material, housing type, herd size, and management system
Chapter 2 is a survey study that explores why New England dairy farmers use a specific
bedding material and whether they are interested in producing wood shavings from their farm
woodlot using a wood shaving machine. The study was conducted by sending a questionnaire to
607 New England dairy farmers (25% of the regional dairy farm population). The primary
research questions addressed in this study were:
1. What bedding materials do New England dairy farmers use and why?
2. What percent of dairy farmers experienced increased bedding costs over the last decade,
and how were those costs managed?
3. What is the current annual bedding material cost per cow?
4. Does the bedding type, housing system, farm scale, or management system have an effect
on farmers’ self-reported SCC or bedding cost?
5. Are regional dairy farmers interested in producing animal bedding using a wood shaving
machine as a potential cost-saving and revenue-generating alternative?
Information from this chapter was used to guide the research questions on producing
wood shavings with a wood shaving machine presented in Chapter 3.
Chapter 3: Financial viability of producing animal bedding with a wood shaving machine
Chapter 3 explores the production cost of producing wood shavings using a wood
shaving machine and eastern white pine trees harvested from the farm woodlot. Data obtained
from operating the wood shaving machine were used to develop a model that farmers and
13
foresters can use to determine the economics of producing wood shavings using their site-
specific variables. The model has 23 input variables, and provides individuals with the necessary
economic data to determine whether the venture is economic when using their unique machine
and site-specific parameter values. The primary research questions for this study were:
1. Is this type of venture economically feasible for the average-sized organic and
conventional dairy farm?
2. What is the minimum size farm operation that could support such a venture?
3. How much money could an individual make when using the machine for 10, 20, 30,
or 40 hours per week for an entire year?
4. What are operational considerations that make or break this type of venture?
As detailed in this chapter, the feasibility of using a wood shaving machine to produce
animal bedding was absent from the literature, other than unreliable information from various
manufacture’s websites selling the machines. The lack of non-biased information regarding the
economics of producing animal bedding from this type of venture justified this study.
Chapter 4: Heat recovery from composting – A comprehensive literature review
Chapter 4 details compost heat recovery systems, dating from ancient China 2000 years
ago to the present (2016). The basis for this review originated from questions raised during the
onset of the agroecosystem study about how to best process the spent animal bedding and
manure produced at the UNH ODRF. The primary research questions for this study were:
1. What is the difference in heat recovery by system type (convective, conductive and
latent) and scale (lab, pilot, and commercial)?
2. Are there economies of scale with regard to heat recovery?
3. Are some systems more suitable for commercial applications than others?
14
4. Is the technology beyond the prototype phase, and can it serve as a mainstream
process to reduce energy costs on farms?
This detailed review was warranted as the last comprehensive review on the topic was
conducted over 30 years ago by Fulford (1983). A detailed review was also necessary to
determine what type of compost heat recovery system would be implemented on the farm as part
of the agroecosystem study.
Chapter 5: Heat recovery from composting: A step-by-step guide on building an aerated static
pile heat recovery composting facility
Chapter 5 is an in-depth discussion on how to construct an aerated static pile (ASP) heat
recovery composting facility. The research was conducted through observations during the
construction of the UNH heat recovery composting facility from October 2012 – June 2013.
Specific attention was given toward design, cost structure, and alternative building and material
options to achieve the same outcome. The primary research questions for this study were:
1. What methods and materials were used to construct the UNH facility?
2. Are alternative methods and materials available and at what cost?
3. What cost-saving strategies could be used at future composting sites?
Prior to this research, only one short article from Tucker (2006) addressed facility design
for a commercial-scale composting facility. The chapter fills a much needed information gap on
compost facility design, cost, and tradeoffs. The research is also presented in a manner
appropriate for compost practitioners, who may save thousands of dollars by following the
recommendations outlined in the report. This chapter also describes how the facility was
designed for replicated research trials, which are covered in the following chapter.
15
Chapter 6: Recovering heat from composting as an innovative approach to reduce greenhouse
gas emissions and fossil fuel consumption in the agricultural sector
Chapter 6 explores the heat extraction and utilization rate from the UNH heat recovery
composting facility when using the spent animal bedding, manure, and other waste residuals
produced on the farm. The specific research questions asked were:
1. What is the heat extraction and utilization rate from a commercial-scale heat recovery
composting facility?
2. What is the fossil fuel equivalent of the captured heat?
3. To what degree does the heat recovery system offset GHG emissions for the purpose
of carbon pricing?
This study is unique in that it presents heat recovery rates from an active commercial-
scale facility and presents a range in heat recovery values based on incoming vapor temperature
and utilization efficiency. Prior to this research, only Allain (2007) presented compost heat
recovery rates in a range based on utilization. All previous studies on this subject, which are
described in detail in Chapter 4, present a static heat recovery rate based on optimal composting
conditions existing in perpetuity. This assumption is not realistic for an active composting site,
meaning compost heat recovery rates are largely overestimated in the literature. This study is
also the first to assess the GHG mitigation potential of a compost heat recovery system using
actual versus modeled results.
Chapter 7: Conclusion
The final chapter of the dissertation summarizes the findings from Chapters 2 – 6. The
chapter also addresses the degree to which the various waste management and nutrient recycling
16
strategies worked at the UNH ODRF. A holistic approach is used in this discussion, connecting
all of the various components of the agroecosystem.
17
REFERENCES
Aber, J.D., Magill, A., Mcnulty, S.G., Boone, R.D., Nadelhoffer, K.J., Downs, M., and Hallett,
R. 1995. Forest biogeochemical and primary production altered by nitrogen saturation.
Water, Air, and Soil Pollution. 83(3): 1665-1670.
Allain, C. 2007. Energy recovery at biosolids composting facility. BioCycle 48(10): 50-53.
Beegle, D.B., and Durst, P.T. 2014. Managing phosphorus for crop production. PennState
University Extension. Agronomy Facts 13.
Bittman, S., and Mikkelsen, R. 2009. Ammonia emissions from agricultural operations:
Livestock. Better Crops, 93(1): 28-31.
Bohlen, P.J., and House, Gar. 2009. Sustainable agroecosystem management. Boca Raton, FL:
CRC Press. 301 p.
Chiumenti, A., Chiumenti, R., Diaz, L.F., Savage, G.M., Eggerth, L.L., and Goldstein, N. 2005.
Modern Composting Technologies. Emmaus, PA: JG Press. 97 p.
Fulford, B. 1983. Biothermal energy: Cogenerants of thermophylic composting and their
integration within food producing and waste recycling systems. In E. Stentiford (ed.),
Composting of solid wastes and slurries; Proc. Intern. Symp., Leeds, England, 28-30
September 1983. Leeds, England: University of Leeds.
Galloway, J.N., Aber, J.D., Erisman, J.W., Seitzinger, S.P., Howarth, R.W., Cowling, E.B., and
Cosby, B.J. 2003. The nitrogen cascade. BioScience. 53(4): 341-356.
Galloway, J.N., and Cowling, E.B. 2002. Reactive nitrogen and the world: 200 years of change.
Ambio. 31(2): 64-71.
Gitau, T., Gitau, M.W., and Waltner-Towers, D. 2009. Integrated assessment of health and
sustainability of agroecosystems. Boca Raton, FL: CRC Press. 227 p.
Gliessman, S.R. 2001. Agroecosystem sustainability: Developing practical strategies. Boca
Raton, FL: CRC Press. 210 p.
18
Gruber, N, and Galloway, J.N. 2008. An earth-system perspective of the global nitrogen cycle.
Nature. 451(17): 293-296.
Hashemi, M., Herbert, S., Chickering-Sears, C., Weis, S., Gradil, C., Purdy, S., Huyler, M.,
Prostak, R., and Carlevale, J. 2014c. Manure as a nutrient resource. UMass Extension.
CDLE Pub. 11-37. [Internet]. Available from:
<http://extension.umass.edu/cdle/sites/extension.umass.edu.cdle/files/fact-
sheets/pdf/ManureasaNutrientResource%2811-37%29.pdf>.
Hellebrand, H. J. 1998. Emission of nitrous oxide and other trace gases during composting of
grass and green waste. Journal of Agricultural Engineering Research. 69: 365-375.
Jokela, B., and Peters, J. 2009. Dairy manure nutrients: Variable, but valuable. Forage Focus,
United States Department of Agriculture, Agricultural Research Service.
Knowlton, K.F., Wilkerson, V.A., Casper, D.P., and Mertens, D.R. 2010. Manure nutrient
excretion by Jersey and Holstein cows. J. Dairy Sci. 93(1): 407-412.
MacDonald, J.M., O’Donoghue, E.J., McBride, W.D., Nehring, R.F., Sandretto, C.L., and
Mosheim, R. 2007. Profits, costs, and the changing structure of dairy farming. United
States Department of Agriculture Economic Research Service, Economic Research
Report Number 47.
McBride, W.D., and Greene, C. 2009. Characteristics, costs, and issues for organic dairy
farming. United States Department of Agriculture Economic Research Service, Economic
Research Report Number 82.
McFarland, D. (2009). Allow organic bedding to work with good management. The Dairy
Focus, 10(4), 1-3. Penn State Cooperative Extension Dairy Program.
Odum, E.P. 1989. Input management of production systems. Science. 243: 177-182.
Pattey, E., Trzcinski, M.K., and Desjardins, R.L. 2005. Quantifying the reduction of greenhouse
gas emissions as a result of composting dairy and beef cattle manure. Nutrient Cycling in
Agroecosystems. 72: 173-187.
19
Pellissier, P., and Benander, M. 2008. Application of Geothermal Energy at the UNH Organic
Dairy. Research Poster presented as part of project in Environmental Science at the
University of New Hampshire
Pennington, J.A., VanDevender, K., and Jennings, J.A. 2013. Nutrient and fertilizer value of
dairy manure. University of Arkansas Division of Agriculture Cooperative Extension
Services, FSA4017.
Rynk, R., van de Kamp, M., Willson, G.B., Singley, M.E., Richard, T.L., Kolega, J.J., Couin,
F.R., Laliberty, L., Kay, D., Murphy, D.W., Hoitink, H.A.J., and Brinton, W.F. 1992. On-
Farm Composting Handbook. Ithaca, NY: Northeast Regional Agricultural Engineering
Service (NRAES-54). 186 p.
Smith, A.M., Keene, W.C., Maben, J.R., Pszenny, A.P., Fischer, E., and Stohl, A. 2006.
Ammonia sources, transport, transformation, and deposition in coastal New England
during summer. Journal of Geophysical Research. 112: 1-12.
Stewart, H., Dong, D., and Carlson, A. 2013. Why are Americans consuming less fluid milk? A
look at generational differences in intake frequency. USDA Economic Research Service,
Economic Research Report Number 149.
Tucker, M.F. 2006. Extracting thermal energy from composting. BioCycle. 47(8): 38.
United States Department of Agriculture. 2014. Organic Production Overview. Economic
Research Service. [Internet]. Available from: <http://www.ers.usda.gov/data-
products/organic-production.aspx#.UxifPvldWuI>.
United States Department of Agriculture. 2010. National Organic Program (NOP) Final Rule, 7
CFR Part 205.
United States Environmental Protection Agency. 2014a. Greenhouse gas emissions: Methane
emissions. [Internet]. Available from:
<http://epa.gov/climatechange/ghgemissions/gases/ch4.html>.
United States Environmental Protection Agency. 2014b. Greenhouse gas emissions: Nitrous
oxide emissions. [Internet]. Available from:
<http://epa.gov/climatechange/ghgemissions/gases/n2o.html>.
20
United States Environmental Protection Agency. 2014c. Sources of greenhouse gas emissions:
Agricultural. [Internet]. Available from:
<http://epa.gov/climatechange/ghgemissions/sources/agriculture.html>.
United States Environmental Protection Agency. 2012. Nonpoint source pollution: The
Nation’s largest water quality problem. US EPA, EPA 841-F-96-004A.
Utami, T.S., Hermansyah, H., and Nasikin, M. 2012. Biofiltration of nitrous oxide using cow-
manure based compost as medium filter. Journal of Environmental Protection. 3: 584-
588.
Vitousek, P.M., Aber, J.D., Howarth, R.W., Likens, G.E., Matson, P.A., Schindler, D.W.,
Schlesinger, W.H., and Tilman, D.G. 1997. Human alteration of the global nitrogen
cycle: sources and consequences. Ecological Applications. 7:737-750.
Wright, P. 2001. Overview of anaerobic digestion systems for dairy farms. Natural Resource
Agriculture and Engineering Service, RAES-143.
21
CHAPTER 2: ANIMAL BEDDING COST AND SOMATIC CELL COUNT
ACROSS NEW ENGLAND DAIRY FARMS: RELATIONSHIP WITH
BEDDING MATERIAL, HOUSING TYPE, HERD SIZE, AND
MANAGEMENT SYSTEM
ABSTRACT
Research on bedding material for dairy farmers is typically focused on microbial growth
and associations with somatic cell count (SCC). With few exceptions, information on cost or
why farmers select specific bedding materials is absent from the literature. This study of New
England dairy farmers addresses these gaps, by exploring the relationship between bedding
material, SCC, and cost. Farmer perception of various bedding materials and interest in
producing bedding with a wood shaving machine were also examined. A questionnaire was sent
to conventional and organic dairy farmers across the six New England states, obtaining a
response rate of 22% (129 dairy farms). Survey analysis showed that the primary bedding
materials used by New England dairy farmers were sawdust, sand, and wood shavings. From
2003 – 2013 the regional cost of bedding increased by 89%. Sand and manure solids were less
expensive than other materials ($115/cow/yr and $39/cow/yr respectively), while wood shavings
were the most expensive ($250/cow/yr). Farmers using freestalls had lower bedding costs
($99/cow/yr) than other housing types, especially those using tie stalls, who reported the highest
material costs ($229/cow/yr). No difference in bedding cost was found between organic and
conventional dairies. However, bedding cost decreased significantly as herd size increased.
When looking at producer-reported SCC, no relationship was found with housing type (freestalls,
tie stalls, and compost bedded packs), herd size, or management system (organic and
conventional). However, farmers using sawdust reported elevated SCC (152,695 cells/mL) when
compared to producers using other bedding materials. Dairy farmers with a herd size ≤ 100 head
22
showed the greatest interest in the idea of local cooperatives to produce their own bedding with a
wood shaving machine. With 96% of New England dairy farmers using some quantity of woody
bedding on their farms, this type of venture could prove to be a cost-saving strategy to address
rising bedding costs.
INTRODUCTION
One of the crucial factors required to maintain a healthy dairy herd is having sanitary
animal bedding. This is because of the frequency and duration of contact between the cow and
bedding material. Dairy cows will often lie down 8-16 hrs/day if given the opportunity (Tucker
et al., 2009). Regardless of housing system, extended periods of time spent standing or lying
down on unsanitary bedding increases the risk of cow stress, illness, and disease. This is
especially true when the bedding is moist and/or contaminated with the animal’s waste (Smith et
al., 1985; Zadoks et al., 2011). A study by Fregonesi et al., (2007) found that dairy cows lie
down five hours less per day when on wet bedding vs. dry bedding. This has important
implications, as increased lying time is believed to increase milk production (Dyck et al., 2009).
Moist and contaminated bedding also promotes a favorable living medium for disease causing
bacteria and fungi that are harmful to dairy cows. More specifically, increased exposure to
contaminated bedding increases the occurrence of environmental mastitis (Hogan et al., 1989).
This particular disease is caused when pathogenic bacteria enter the cow’s teat orifice and move
into the teat canal, causing an infection of the mammary gland (Pennington, 1994). Depending
on the severity of the infection, mastitis can result in reduced milk yield, reduced milk quality,
and increased quantity of discarded milk (Grohn et al., 2004). Additionally, many dairy
producers are paid a milk premium for higher quality milk, which is lower in both somatic cell
count (SCC) and standard plate count. The occurrence of mastitis increases both the somatic cell
23
and standard plate counts of milk, reducing the economic bonus many producers rely on. From
an animal welfare standpoint, mastitis can cause fever, reduced appetite, weight loss, and even
result in the need to cull the animal (Smith et al., 1985; Nickerson et al., 1995; Munoz et al.,
2008). With bedding being one of the primary sources of exposure to environmental mastitis
pathogens (Ruegg, 2006), the management of this material is important in maintaining herd
health and the economic vitality of the entire farm operation.
Dairy farmers across the six New England states have several bedding options that they
can use exclusively or in combination. Bedding selections often reflect what materials are locally
available, can be produced or grown on site, are inexpensive, and are compatible with farm
infrastructure. In the highly forested region of New England, bedding from mill waste (sawdust
and planer shavings) has historically been an inexpensive and easy to find product. However, the
last few decades has seen a continual decrease in the number of mills operating regionally. This
problem accelerated in 2005 due to the collapse in the new home construction market and further
in 2007-2009 due to the recession (Woodall et al., 2012). In addition to fewer mills, increased
mill efficiency and modernization reduced the amount of waste byproduct available for animal
bedding. Some mills have even redirected their entire wood waste stream to power their dry
kilns. Transportation costs to deliver bedding to farms has also increased, as fewer mills result in
longer travel distances between mill and farm. Compounding these problems is the expanding
wood pellet industry in the region, which now competes with farmers for this limited resource.
The result is more expensive woody bedding that many dairy farmers cannot afford.
While not all New England farmers use woody bedding, the high cost of this resource is
putting pressure on dairy farmers to use other forms of bedding, which inevitably increases
demand and drives up costs of imported alternatives. Farmers either have to pay more for
24
bedding, allocate more land to grow bedding crops (straw, hay), invest in new infrastructure to
accommodate new bedding alternatives, or use less bedding in general. For many regional dairy
farmers, none of these options are positive. As a consequence, researchers from the University of
New Hampshire and University of Vermont designed a survey study for regional dairy farmers to
determine what the current state of bedding usage and cost are across the six New England
states. The objectives of this study were to assess the following: 1) what bedding materials are
New England dairy farmers using and why, 2) what percent of dairy farmers experienced
increased bedding costs over the last decade and how were those costs managed, 3) determine
the current annual bedding material cost per cow, 4) assess whether bedding type, housing
system, farm scale, or management system has an effect on farmers’ self-reported SCC or
bedding cost, and 5) gauge dairy farmer interest in the on-farm production of animal bedding
using a wood shaving machine as potential cost-saving and revenue-generating alternative.
Results from this study will be used to guide future investigations on how to reduce farmer’s
bedding cost.
MATERIALS AND METHODS
Survey Tool
This survey study was conducted at the University of New Hampshire (UNH) and was a
joint project between UNH and the University of Vermont (UVM). The survey tool used for this
study was a mailed questionnaire, followed by an online version sent via email to those not
responding to the paper mailing. The questionnaire was developed over a six-month period, with
the assistance of experts in the field of dairy and natural resource management. Research
questions and the cover letter for the questionnaire were pre-tested using a focus group of
university dairy farm managers and researchers. Cognitive interviewing was used during the
25
focus group, to fully understand how individuals were interpreting each question and whether the
group was interpreting questions consistently. This same group also validated the content of the
questionnaire as a whole, to ensure it accurately addressed the specific research questions being
asked. The focus group was also asked to carefully analyze the content of the cover letter, which
described: the aim of the study, who was conducting it, how the information would be used, the
respondent’s rights as a human subject, assurance of their confidentiality, and informed consent
(right to participate or not). Upon completing this process, the completed questionnaire was
provided to the University of New Hampshire Institutional Review Board for the Protection of
Human Subjects in Research (IRB), which approved the study under IRB exempt status
(Appendix 1). Following IRB approval, the refined questionnaire was pilot tested by a small
sample of the target population (three organic farmers and two conventional farmers) to
determine the ease, quality, and time requirement of the questionnaire. This process led to the
omission of two questions for a final questionnaire of 28 questions (Appendix 2). The online
version of the questionnaire was also pilot-tested by three members of the research team to
ensure all links worked and the visual presentation made sense.
Sample Size and Selection
The target population for this study was conventional and organic dairy farm managers
with active operations in the six New England states. A total of 2457 dairy farms across New
England (2207 conventional and 250 organic) was used as the population for sample size
determination (USDA, 2012, 2014). A goal of 93 respondents was selected for the study,
corresponding to a margin of error of 10%, a confidence level of 95%, and a response
distribution of 50%. Sample size for the questionnaire was based on the respondents needed (93)
divided by the expected response rate (15%) for a 620 target number of questionnaires.
26
Addresses for dairy farms, both physical and email, were found through state and
national online directories. The primary directories used to obtain addresses were the New
England States Holstein Association, American Jersey Cattle Association, U.S Ayrshire
Breeder’s directory, and the American Guernsey Association. Additional participants for the
study were obtained from Organic Valley and Moo Milk, who mailed questionnaires to their
constituents on behalf of the research team. A total of 607 participants (25% of the target
population) was obtained from these sources. All participants were used in the study to increase
the odds of obtaining the target sample size of at least 93 respondents.
Questionnaire Mailings
Questionnaires were sent by first class mail on March 17, 2014 to 607 dairy farms (395
conventional and 212 organic) across the six New England states. A late winter mailing was
deliberately selected to increase the response rate, as spring, summer, and fall are typically busier
times of the year for dairy farmers. On May 9, 2014 the online version of the questionnaire was
sent to dairy farm managers that did not respond to the mailed questionnaire. Farm managers
were contacted via email with a link to the questionnaire, which was developed in
SurveyMonkey. One week following the first email, a reminder email was sent with a link to the
questionnaire to those who had not responded to the first request. Only 35 dairy farmers were
contacted for the online questionnaire, due to the lack of publically available email addresses.
Statistical Analysis
Raw data from both the mailed and online questionnaires were compiled and entered into
Microsoft Excel. A completion rate of 90% was used to determine whether a questionnaire
would be included in the data set. Four questionnaires (all email-based) were omitted due to this
cutoff limit. Data from questionnaires passing this cutoff were entered into excel by one member
27
of the research team, with every entry being verified by a second member to remove any type of
response error. Missing data from questionnaires with < 100% completion but ≥ 90% were dealt
with on an individual question basis, where that particular respondents’ results for that one
question were omitted from data analysis. Data analysis was conducted using IBM SPSS. Linear
regression analysis and ANOVA were used to explore the relationship of SCC and bedding cost
per cow against bedding material selections, housing type, farm size, and management system.
Data for the question asking dairy farmers about their top five farming expenses was
analyzed using a scoring system to give weight to the number of responses by rank
corresponding to that expense. The following scoring system was used to determine which
expenses were greatest for the sample population: rank 1 (10 points), rank 2 (8 points), rank 3 (6
points), rank 4 (4 points), rank 5 (2 points). By way of example, if the top five expenses for
Respondent 1 were feed, labor, fuel, bedding and repairs, while Respondent 2 reported labor,
feed, fuel, bedding and repairs, the ranking system would attribute 18, 18 12, 8 and 4 points for
feed, labor, fuel, bedding, and repairs, respectively.
Response rate was calculated one month following the last email reminder to complete
the online questionnaire. Adjustments were made to the total sample population sent
questionnaires and to those returning them. The original sample population of 607 was adjusted
to reflect any return to sender questionnaires, while the total number of viable questionnaires was
adjusted to reflect the 90% completion cutoff.
Non-Response Bias
Bias occurring from non-response, where survey respondent answers differ from non-
respondents, was addressed before and after the mailing of the questionnaire. Non-response bias
was considered from the beginning of the study by using the Tailored Design Method (TDM)
28
(Dillman, 2000). Components of this method that were used were: 1) created a short respondent-
friendly questionnaire developed with industry experts, 2) pre-tested the questionnaire multiple
times, 3) included a return envelope with a first class stamp, 4) contacted respondents multiple
times over a several month period through two modes of contact (mail and email), 5) provided
contact information (phone and email) for the lead researcher in the cover letter, 6) described
how results of the questionnaire could indirectly benefit the respondent financially, 7) provided a
description of university sponsorship (UNH and UVM), and 8) provided a token of appreciation
(free copy of finished survey report). Non-response error post-questionnaire was considered by
using a modified version of the Comparison of Demographic and Socioeconomic Difference
(CDSD) method (Sivo et al., 2006), where farmer age, gender and working experience were
compared to the target population. Demographics (farmer age and gender), along with farm
characteristics (number of head in the dairy herd and farm acreage) were compared to 2012
USDA Census of Agriculture data to determine whether a similar distribution existed between
the research study and that of the agricultural census.
RESULTS AND DISCUSSION
Response Rate and Survey Demographics Of the 607 questionnaires sent via first class mail, 114 dairy farmers responded, with
another 17 questionnaires coming back as return to sender. Of the 35 questionnaires sent via
email, 19 dairy farmers responded with four of the online questionnaires being removed for not
meeting the 90% completion cutoff. A total of 129 completed questionnaires from 99
conventional, 25 organic and 5 unspecified dairy farms was obtained from a sample population
of 590 potential participants (22% response rate). The total population response rate when
29
looking at all active dairy farms in New England, was 10% for organic farms, 4.5% for
conventional farms and 5.3% when combining the two (Table 1).
Table 1: Dairy Farm Attributes vs. 2012 USDA Census of Agriculture
Dairy Farms in New
England1
Completed Questionnaires Total Population
Response Rate
(N) Distribution (%) (N) Distribution (%) %
Maine 581 24 18 14 3.1
Vermont 1075 44 58 45 5.4
New Hampshire 251 10 17 13 6.8
Massachusetts 278 11 14 11 5.0 Connecticut 242 10 9 7 3.7
Rhode Island 30 1 2 1 6.7
Unspecified 0 0 11 9 0
Total 2457 100 129 100 5.3 1(USDA, 2012)
The distribution of dairy farm responses by state between the sample and target
population were similar, with five of the six sampled states having a distribution within 3% of
each other (Table 1). Only Maine dairy farmers were slightly under-represented when comparing
the distributions on a regional level. However, response rate in relation to herd size (lactating
cows) did not match closely between sample and target populations (Figure 3).
Figure 3: Survey Response by Herd Size vs. 2012 USDA Census Data
0
10
20
30
40
50
60
≤ 49 Cows 50 - 99 Cows 100-199 Cows ≥ 200 Cows
% o
f Fa
rms
New England Herd Size Distribution Between Sample and Target Populations
Target PopulationDistribution*
Sample Population Distribution
30
While the distribution of farms between the sample and target population were similar for the
two larger herd size classes, smaller dairies (≤ 49 cows) were slightly underrepresented in this
study.
Bedding Usage
The quantity of bedding used regionally averaged 10 m3/cow/yr (± 1.15 m3) or 5
m3/head/yr (± 0.54 m3). Of the bedding materials used regionally, sawdust was the most
common, with 23% of surveyed dairy farmers using the material exclusively and 81% using
some quantity on their farm (Table 2). Sand was the second most popular bedding material with
3% using it exclusively and 33% using some fraction on their farm. Wood shavings were a close
third, with 4% using the material exclusively and 31% using some fraction on their farm. Hay
and straw were moderately popular choices, with a majority of dairy farmers using <10% of each
to supplement one of the more common bedding selections. The less commonly used materials
(woodchips, recycled manure solids (MS), recycled paper fiber, and other) were typically used in
very small amounts, serving as a supplement to other bedding materials.
Table 2: Bedding Usage % by Number of Dairy Farms across New England
Bedding
Usage (%)
Sand Sawdust Shavings Hay Woodchip
s
MS Straw Paper
Fiber
Other1
1-10 9 9 10 14 6 1 18 1 3
11-20 2 14 2 3 1 . 1 1 1 21-30 3 6 5 1 1 . . 1 2
31-40 2 3 3 1 . . . . .
41-50 2 13 5 2 . 1 . 2 .
51-60 5 6 . 1 . . . . .
61-70 4 3 1 . . . . . 1
71-80 9 7 3 . . . . . .
81-90 3 6 1 . 2 3 1 . .
91-100 4 38 10 4 . . . 1 .
Farms (N) 43 105 40 26 10 5 20 6 7 1Other includes leaves, horse litter, ground corn husks, and Casella Organics Clean Cow Bedding
When looking further into bedding material selections, only five dairy farmers (4%) did
not use any woody bedding in their mix. Of those five, four used hay exclusively and the other
31
used a combination of sand and straw. This was not surprising given that the region is
predominantly forested. Geography is also responsible for the lack of straw bedding used in New
England, which produces minimal quantities of cereal crops for straw byproduct (USDA, 2015).
Of the 20 dairy farms in the sample population using straw, 90% reported straw as < 10% of total
bedding usage (Table 2).
Bedding Preference
When respondents were asked why they preferred their primary bedding material,
responses varied by bedding type (Table 3). Dairy farmers using sand cited low bacteria as a top
reason, which is a characteristic well supported in the literature (Hogan et al., 1989; Godden et
al., 2008). Farmers using sawdust cited the ease of material handling and absorbency as primary
benefits. The characteristic of high absorbency for sawdust has also been reported by Zehner et
al., (1986). Farmers using wood shavings reported dryness and ease of handling as primary
benefits, while those using hay found the ability to grow it on site as a top preference. Dyck et
al., (2009) also reported the advantage of growing hay on site as a bedding source, especially for
organic dairy farmers, who are required to use more costly organic certified bedding. Dairy
farmers using MS reported high absorbency as their top preference, which is a characteristic
reported by Zehner et al., (1986) in a study of five different bedding materials.
Table 3: Reason for Preferring a Bedding Material and Number of Farms Supporting that Reason
Bedding Preference Sand Sawdust Shavings Hay Straw MS
Dry 1 . 9 . . .
Clean 3 . 3 . 1 1
Comfortable 2 2 1 . . 1 Absorbent . 6 2 . . 3
Low bacteria 10 3 3 . . 1
Can be purchased locally 3 . 4 1 . .
Can be grown on site . . . 4 1 .
Can be produced on site 4 . . . . 2
Easy handling . 13 4 1 . .
Low cost 4 4 1 1 . 2
Visually appealing . 1 . . . .
Manure system compatibility . 1 2 1 . 1
32
Bedding Avoidance
Dairy farmers were also asked if they avoided any particular type of bedding material. Of
the 71 dairy farmers (55%) reporting a particular bedding avoidance, 59% avoided sand, with
incompatibility with the manure system and wear on equipment being the top reasons for not
using the material (Table 4). The incompatibility of sand with manure systems has also been
reported in the literature (Dyck et al., 2009; Endres, 2012). The next most avoided bedding
material was straw, which was cited as being too costly, difficult to handle, and not compatible
with liquid manure systems. A study of Vermont bedded pack dairy facilities also found that
straw was regionally expensive (Gilker et al., 2012). Hay was the third most avoided material,
with respondents reporting similar issues to straw, only without the high cost. Sawdust was the
fourth most avoided bedding material, with bacteria concerns being the main reason for
avoidance. The characteristic of sawdust supporting mastitis-causing bacteria, especially
Klebsiella spp., is well supported in the literature (Hogan et al., 1989; Zdanowicz et al., 2004).
High bacteria counts were also the primary reason respondents avoided MS. This concern has
been reported in the literature with numerous studies finding that the material can support high
microbial populations (Zehner et al., 1986; Godden et al., 2008; Harrison et al., 2008).
Interestingly, none of the surveyed organic farmers used any quantity of MS.
Table 4: Reason for Avoiding a Bedding Material and Number of Farms Supporting that Reason
Bedding Preference Sand Sawdust Shavings Hay Straw MS
Not absorbent . . 2 1 . .
Dirty . . . 1 1 1
High bacteria . 7 2 1 . 3
Hard on equipment 12 . . . . .
Unsuited with housing system 6 . . . . 1 Expensive 1 1 2 3 10 .
Not readily available 1 2 1 . 1 1
Difficulty with handling 4 . . 5 6 .
Smells bad when wet . . . 1 . .
Avoid for no particular reason . 2 . . . .
Incompatible with manure system 18 . . 11 10 .
33
Bedding Cost
Bedding cost for New England dairy farm respondents averaged $194/cow/yr or
$86/head/yr (Table 5). From 2003 – 2013 average regional bedding material costs increased 89%
(50% when accounting for inflation). The high SE and large difference between mean and
median values speaks to the diversity of dairy operations across New England.
Table 5: Overall Cost of Bedding Materials for New England Dairy Farmers
Annual herd
bedding costs 2013
(n = 117)
Annual herd
bedding costs 2003
(n = 80)
Cost/cow
2013
Cost/head
2013
Median $10,000 $4000 $143 $66
�̅� $14,281 $7494 $194 $86
SE $1688 $1185 $20 $6
When looking at bedding, only 9 farmers (7%) did not have increased costs over the ten
year period (2003 - 2013). This finding is consistent with that of Laughton et al., (2014), who
also reported the increasing cost of dairy supplies throughout New England, which included
bedding. Of the dairy farmers not reporting increased cost, five were using sand, two sawdust,
one purchased a wood shaving machine and one was growing hay onsite. Of the farmers using
woody bedding, 98% experienced increased costs from 2003 to 2013. The higher cost of woody
bedding was also reflected when respondents were asked whether they changed their bedding
source over the past ten years. During this period, 34 farmers (26%) reported switching primary
bedding materials, with 82% of respondents converting from sawdust or shavings to a non-
woody bedding type. The primary reasons cited for switching from woody bedding were mill
closings (12 farmers), switched to sand (7 farmers) and increased cost (6 farmers).
Bedding Rank Compared to Other Farming Expenses
In comparing bedding cost in relation to other farming expenses, conventional dairy
farmers ranked bedding 4th most expensive, while organic dairy farmers ranked bedding 5th
34
(Table 6). Cost of imported feed and labor were ranked 1st and 2nd for both conventional and
organic dairy farmers.
Table 6: Rank of Bedding Cost in Relation to Other Farming Expenses
Rank of costs Conventional dairies (N = 102) Organic dairies (N = 23)
Cost Score1 N Cost Score1 N
1 Feed 970 100 Feed 220 23
2 Labor 364 50 Labor 70 9
3 Repairs 226 47 Fuel 62 12
4 Bedding 224 46 Repairs 58 12
5 Fuel 194 41 Bedding 54 15 1Score based on number of responses (N) and the associated rank (Rank 1 = 10, Rank 2 = 8, Rank 3 = 6, Rank 4 = 4
and Rank 5 = 2)
A study of 142 dairy farms in New England by Laughton et al., (2014) reported a similar
ranking of costs for conventional dairy farmers, with the following costs/cow/yr: feed ($1948),
labor ($802), repairs ($343), supplies ($255), and gasoline, fuel and oil ($254). However,
bedding cost was not a specific line item in their economic analysis, but was contained within the
supplies category and was often the highest cost within that category (Laughton, 2015 Personal
Communication).
Relationship between Bedding Material with Cost and SCC
Of the various bedding materials used by the New England dairy farmer respondents, the
cost of using sand ($115/cow/yr ± $20) and MS ($39/cow/yr ± $7)) was significantly lower (P=
0.03 for sand and P = 0.04 for MS) than the costs of other bedding materials (Table 7). This was
in contrast to the cost of shavings ($250/cow/yr ± $35), which was more expensive than the other
bedding materials (P = 0.03).
35
Table 7: Reported Bedding Cost and SCC for Various Bedding Materials
Primary bedding
material
Per cow Per head
Bedding cost SCC (cells/mL) Bedding cost
�̅� SE �̅� SE �̅� SE
Sawdust (70) $ 194 $ 21 152,695 7059 $ 91 $ 9
Sand (26) $ 115 $ 20 133,560 8486 $ 58 $ 10
Shavings (19) $ 250 $ 35 144,200 17,277 $ 120 $ 15
Hay (7) $ 137 $ 33 112,500 31,192 $ 88 $ 30
MS (4) $ 39 $ 7 120,750 23,178 $ 20 $ 5
Woodchips (2) $ 71 $ 63 182,500 17,500 $ 42 $ 38
Straw (1) NA NA 100,000 NA $ 83 $ 0
Few studies have specifically focused on the cost comparison between various bedding
materials to serve as a comparison to this study. However, a study by Harrison et al., (2008) did
report similar findings regarding MS, and how dairy farmers switching to MS reported savings
($0.01 – $0.26 per cwt) in both bedding purchases and in manure handling. A second study by
Panivivat et al., (2004) also mentioned bedding costs briefly for dairy calves in Arkansas. In their
study, long wheat straw and sand were the most expensive bedding materials, followed by
granite fines, wood shavings and rice hulls.
When looking at bedding materials, sawdust was the only bedding type exhibiting a
significant relationship (P = 0.02) with SCC (152,695 cells/mL ± 7059 cells/mL) (Table 7).
While significant, the SCC value was still below the National Mastitis Council’s threshold of
200,000 cells/mL for “normal” measures. Because bacteria count by bedding type was not
measured in this study, it is difficult to pinpoint the true cause of the elevated SCC reported by
farmers using sawdust bedding. However, numerous studies have reported that sawdust harbors
more mastitis-causing bacteria, especially Klebsiella spp., when compared to other bedding types
(Hogan et al., 1989; Zdanowicz et al., 2004; Dyck et al., 2009). With bacterial counts in bedding
being related to rates of clinical mastitis (Hogan et al., 1989), it is likely that the elevated SCC
reported on farms using sawdust bedding was due to the bedding material itself and not some
36
form of management. This is especially true as sawdust usage in this study was reported across
all housing types, farm scales, and in both management systems.
In contrast to the findings of this study, a review by Dufour et al., (2011) found that sand
bedding was associated with lower SCC. Wenz et al., (2007) also found that mattresses, sand and
newspaper were all associated with lower BTSCC, when compared to composted manure, which
was 2.9 times more likely to have elevated BTSCC. Additionally, Fulwider et al., (2007) found
no difference in SCC with bedding type when using rubber-filled mattresses, sand, water beds,
and compost packs. Harrison et al., (2008) reported similar findings when comparing digested
manure solids and sand bedding. However, none of these studies included sawdust in their
comparison of bedding types, making it difficult to discern whether the relationship between
elevated SCC and sawdust bedding is unique to our study of New England dairies.
Relationship between Housing Type with Bedding Cost and SCC
Dairy farmers using freestalls were found to have lower bedding costs than farmers not
using that type of housing system on their farm (P = 0.003). This was in contrast to farmers using
tie stalls, who reported elevated bedding costs (P = 0.02) (Table 8).
Table 8: Cost and SCC in Relation to Housing Type
Housing type1 Per cow Per head
Bedding cost SCC (cells/mL) Bedding cost
�̅� SE �̅� SE �̅� SE
TS (40) $ 229 $ 35 159,727 9625 $ 108 $ 15
FS (29) $ 99 $ 17 130,583 7833 $ 56 $ 10
BP (5) $ 243 $ 84 135,000 15,811 $ 96 $ 27
FS + BP (17) $ 132 $ 29 136,941 13,548 $ 70 $ 14 TS + BP (12) $ 233 $ 35 154,917 19,212 $ 116 $ 18
TS + FS (11) $ 136 $ 21 132,800 11,068 $ 65 $ 9
TS + FS + BP (9) $ 257 $ 38 147,429 23,663 $ 100 $ 12 1TS = Tie Stall, FS = Free Stall, BP = Bedded Pack
Few studies have explored the relationship between housing type and bedding cost to
serve as comparison to this study. However, a study by Endres (2012) reported that the cost of
bedding was significantly higher for dairy farmers using bedded packs ($146 - $347/cow/yr) than
37
farmers using freestalls ($33 - $55/cow/yr). The higher cost of bedded packs was also reported
by Gilker (2012), who found that bedded packs use four times as much bedding as freestall
barns, costing $756 – 2657/cow/yr. Barberg et al., (2007) also reported that the cost of bedding
for packs ($128 - $310/cow/yr) was the greatest concern for those using that housing type.
However, Barberg et al., (2007) also reported a reduced occurrence of lameness and increased
cow comfort on bedded packs, which may offset the extra cost of more bedding material.
While the relationship between compost bedded packs and cost was not significant in this
study, three of the farmers using that system exclusively were using hay from their own land,
significantly reducing their bedding costs and causing the high SE reported in Table 8. A larger
sample size containing more farms with compost bedded packs would have been more effective
at isolating any potential relationships for that housing type.
When looking at housing type in relation to SCC in the present study, no relationship was
found (P > 0.05). These findings are in disagreement with Rodrigues et al. (2005) and Dufour et
al., (2011), who reported that freestall systems are associated with lower SCC, when compared to
other housing types. Barberg et al., (2007) also reported that farmers switching from other
housing types to compost bedded packs saw a significant decrease in BTSCC.
Relationship between Dairy Cow Herd Size with Bedding Cost and SCC
Economies of scale were found for bedding cost for both conventional (P = 0.01) and
organic (P = 0.01) dairies. On conventional dairies, bedding cost increased from $111/cow/yr ±
$28 on farms with ≥ 200 cows to $272/cow/yr ± $46 on farms with ≤ 49 cows. Likewise, the cost
of bedding on organic dairies increased from $101/cow/yr ± $4 on farms with 100 – 199 cows to
$160/cow/yr ± $22 on farms with ≤ 49 cows (Table 9).
38
Table 9: Cost and SCC in Relation to Farm Size
Herd size Per cow Per head
Bedding cost SCC (cells/mL) Bedding cost
�̅� SE �̅� SE �̅� SE
Conventional
≤ 49 Cows (34) $ 272 $ 46 132,958 11,533 $ 116 $ 17 50 – 99 Cows (36) $ 190 $ 21 160,829 20,849 $ 94 $ 10
100 - 199 Cows (15) $ 129 $ 20 169,714 15,246 $ 64 $ 11
≥ 200 Cows (14) $ 111 $ 28 145,929 10,474 $ 57 $ 15
Organic
≤ 49 Cows (12) $ 160 $ 22 132,833 10,864 $ 80 $ 11
50 - 99 Cows (11) $ 124 $ 20 146,727 12,438 $ 61 $ 10
100 – 199 Cows (2) $ 101 $ 4 130,000 10,000 $ 75 $ 19
The relationship between increasing bedding cost with decreasing herd size is a
consistent trend in the literature. In a study of Northeast dairy farms, Laughton et al., (2014)
reported that larger farms had lower production costs (including bedding) from economies of
scale. Reduced costs associated with economies of scale were also reported by MacDonald et al.,
(2007) for conventional dairies and by McBride and Greene (2009) for organic dairies. A review
of the literature by Tauer and Mishra (2006) also concluded that the higher cost of production on
smaller dairies in the U.S. was associated with inefficiency, rather than varying technology.
When looking at the relationship between SCC and herd size in the present study, no
significant effect was observed for either conventional (P = 0.48) or organic (P = 0.56) dairy
farms. However, these results are in conflict with the findings reported in the literature. A study
of Wisconsin dairy farms by Ingham et al., (2011) found that small farms (≤ 118 cows) had
significantly higher SCC (369,000 cells/mL) when compared to the large farms (119-713 cattle)
or confined animal feeding operations (≥714 cattle) which had SCC of 273,000 cells/mL and
240,000 cells/mL respectively. A similar trend of decreasing SCC with increasing herd size was
also reported in other studies (Allore et al., 1997; Oleggini et al., 2011; and Archer et al., 2013).
39
Relationship between Management System with Bedding Cost and SCC
Management system (organic vs. conventional) did not have an effect on bedding cost (P
= 0.136) (Table 10). This is in contrast to the findings of McBride and Greene (2009), who found
that organic dairies tended to have higher operating costs than conventional farms. However,
they also reported that organic dairies tended to be smaller than conventional dairies, with
economies of scale resulting in reduced cost for both management systems. With the strongest
relationship between bedding cost being farm size in the present study, it is likely that the higher
material costs reported in McBride and Greene (2009), were a result of farm size and not
specifically the management system.
Table 10: Cost and Somatic Cell Count in Relation to Management System
Management System Per Cow Per Head
Bedding
Cost
SCC
Cells/mL
Bedding
Cost
�̅� SE �̅� SE �̅� SE
Organic (24) $ 136 $ 14 139,478 7966 $ 70 $ 7
Conventional (96) $ 191 $ 18 147,000 5854 $ 92 $ 8
All Farms (129) $ 179 $ 15 145,018 4887 $ 86 $ 6
Like cost, no relationship was found between management system and SCC in the present
study (Table 10). This suggests that milk quality is not dependent on management system.
Stiglbauer et al., (2013) reported similar results between conventional (213,000 cells/mL) and
organic (221,000 cells/mL) dairy farms in New York, Wisconsin and Oregon. These findings are
also supported by Sato et al. (2005), Pol and Ruegg (2007), and Haskell et al., (2009), who all
reported no difference in SCC between organic and conventional dairy farms.
Interest in Producing Bedding with a Wood Shaving Machine
Of the dairy farmers surveyed, 99 (77%) reported owning a woodlot, with an average
ownership of 64 hectares. Of this acreage, 77% reported owning mixed woodlots, 13% softwood
and 10% hardwood. The primary uses for these woodlots were firewood and timber production
(Table 11).
40
Table 11: Primary uses for Dairy Farm Woodlots across New England
Forest activity Number of farms and %
reporting the activity
Firewood 64
Timber 54
Maple sugar production 12
Unutilized 6
Wildlife habitat 5 Public recreation 4
Silvopasture 3
Financial safety net 1
Wood shavings for bedding 1
When asked whether they would be interested in participating in local farmer
cooperatives to produce their own wood shavings, 13% of respondents said yes, 45% said maybe
and 42% said no. Of the respondents saying yes, 82% were owners of farms with ≤ 99 cows,
18% owned 100-199 cows and 0% said yes for the largest size class (≥ 200 cows). When asked
whether they would purchase wood shavings from local cooperatives, 20% of the sampled
farmers said yes, 17% no and 63% said maybe. Of those saying yes, 79% were owners of farms
with ≤ 99 cows. Dairy farmers were also asked whether the wood shavings would have to be kiln
dried. Of those surveyed, 63% said yes, 19% said no and 18% were unsure.
The greater interest in producing bedding with a wood shaving machine from dairy farm
owners with smaller herds was not surprising. This study showed that the cost of bedding per
cow in New England is almost twice as expensive for farms with ≤ 49 cows as for farms with ≥
200 cows. A primary reason for this is due to economies of scale, where owners of small dairy
farms are less likely to order bulk purchases of bedding. This is especially true for the most
common bedding sources in New England – sawdust and shavings. Many of the mills in the
region that supply this bedding source will blow the sawdust or mill shavings into a tractor
trailer, delivering over 75 m3 at a time. For many smaller dairy farms in the region, this type of
purchase would result in storing large quantities of bedding for an extended period of time,
increasing the risk of contamination.
41
LIMITATIONS OF THE STUDY
One of the limitations of this study relates to the cost of the various bedding materials,
which represent a snapshot of the current market. Had this study been conducted pre-2005,
before the collapse of the new home construction market and subsequent closing of many of the
regions sawmills, the conclusions regarding woody bedding may have been different. A second
limitation of this study relates to non-response error. Stiglbauer et al., (2013) reported a similar
limitation, where non-respondents may have avoided the questionnaire because of a lack of
management standards, or not adhering to organic standards. A follow up questionnaire to non-
respondents with a comparison between early and late respondents could have been used to
isolate potential reasons for non-response, as described in Sivo et al., (2006). A final limitation
was the use of producer-reported SCC. While the focus of this study was on cost, the values for
SCC could have been stronger if a subset of the sample population had BTSCC tested by the
research team for cross-reference. Wenz et al., (2007) conducted a questionnaire in this manner,
and found that a majority of producers across the 21 surveyed states, did not underestimate SCC
and that the producer-reported SCC was an accurate representation.
CONCLUSIONS
The primary bedding materials used by New England dairy farm respondents were
sawdust, sand, and wood shavings. From 2003 – 2013 bedding costs increased by 89% for the
surveyed population. In relation to other farming expenses, conventional and organic dairy
farmers ranked bedding 4th and 5th most costly. The cost of sand ($115/cow/yr) and MS
($39/cow/yr) were significantly less than other bedding selections. This was in contrast to wood
shavings, which were significantly more expensive ($250/cow/yr). In terms of housing type,
farmers using freestalls realized significant cost savings in bedding purchases ($99/cow/yr) while
42
farmers using tie stalls experienced higher bedding material costs ($229/cow/yr). No difference
in bedding cost was observed when comparing management system (organic vs. conventional).
However, economies of scale were observed between small and large dairies, with bedding cost
decreasing as herd size increased. On conventional operations, bedding cost decreased from
$272/cow/yr on farms with ≤ 49 cows to $111/cow/yr on farms with ≥200 cows. Organic
operations had a similar trend with cost decreasing from $160/cow/yr on farms with ≤ 49 cows to
$101/cow/yr on farms with ≥ 200 cows.
When looking at producer-reported SCC, no relationship was found with housing type
(freestalls, tie stalls, and compost bedded packs), herd size or management system (organic and
conventional). However, farmers using sawdust reported elevated SCC (152,695 cells/mL) when
compared to producers using other bedding materials.
When dairy farmers were asked about their interest in participating in local cooperatives
to produce their own bedding with a wood shaving machine, the majority showing interest were
small-scale dairy producers. Likewise, operators of small-scale dairies were the most interested
in purchasing bedding from those cooperatives. With 96% of New England dairy farmers using
some quantity of woody bedding on their farms, this type of venture could prove to be a cost-
saving strategy to reduce regional bedding costs.
ACKNOWLEDGEMENTS
The authors of this survey study would like to thank the 129 New England dairy farmers
who completed the questionnaire that made this study possible. We would also like to express
gratitude to the USDA, for providing the Sustainable Agriculture Research and Education
(SARE) grant that funded this research.
43
REFERENCES
Allore, H. G., P. A. Oltenacu, and H. N. Erb. 1997. Effects of season, herd size, and geographic
region on the composition and quality of milk in the northeast. J. Dairy Sci. 80:3040-
3049.
Archer, S. C., F. M. Coy, W. Wapenaar, M. J. Green. 2013. Association of season and herd size
with somatic cell count for cows in Irish, English, and Welsh dairy herds. Vet J. 196:515-
521.
Barberg, A. E., M. I. Endres, J. A. Salfer, and J. K. Reneau. 2007. Performance and welfare of
dairy cows in an alternative housing system in Minnesota. J. Dairy Sci. 90:1575-1583.
Dillman, D. A. 2000. Internet and Mail Surveys: The Tailored Design Method. John Wiley, New
York, NY.
Dufour, S., A. Frechette, H. W. Barkema, A. Mussell, and D. T. Scholl. 2011. Invited review:
Effect of udder health management practices on herd somatic cell count. J. Dairy Sci.
94:563-579.
Dyck, E., S. Flack, K. Hoffman, E. Maltby, K. Mendenhall, R. Perry, S. Richards, L. Tikofsky,
and B. Wallis. 2009. The organic dairy handbook: A comprehensive guide for the
transition and beyond. K. Mendenhall, ed. Northeast Organic Farming Association of
New York, Inc., New York, NY.
Endres, M. I. 2012. Bedding options for dairy cows. WCDS Adv Dairy Tech. 24:361-369.
Fregonesi, J. A., D. M. Veira, M. A. G. von Keyserlingk, and D. M. Weary. 2007. Effects of
bedding quality on lying behavior of dairy cows. J. Dairy Sci. 90:5468-5472.
Fulwider, W. K., T. Grandin, D. J. Garrick, T. E. Engle, W. D. Lamm, N. L. Dalsted, and B. E.
Rollin. 2007. Influence of free-stall base on tarsal join lesions and hygiene in dairy cows.
J. Dairy Sci. 90:3559-3566.
Gilker, R. E. 2012. Bedded pack in Vermont: Five stories. University of Vermont-Extension,
Burlington.
44
Godden, S., R. Bey, K. Lorch, R. Farnsworth, and P. Rapnicki. 2008. Ability of organic and
inorganic bedding materials to promote growth of environmental bacteria. J. Dairy Sci.
91:151-159.
Grohn, Y. T., D. J. Wilson, R. N. Gonzalez, J. A. Hertl, H. Schulte, G. Bennett, and Y. H.
Schukken. 2004. Effect of pathogen-specific clinical mastitis on milk yield in dairy cows.
J. Dairy Sci. 87:3358-3374.
Harrison, E., J. Bonhotal, and M. Schwarz. 2008. Using manure solids as bedding. Cornell Waste
Management Institute. Ithaca, NY.
Haskell, M. J., F. M. Langford, M. C. Jack, L. Sherwood, A. B. Lawrence, and K. M. D.
Rutherford. 2009. The effect of organic status and management practices on somatic cell
counts on UK dairy farms.
Hogan, J. S., K. L. Smith, K. H. Hoblet, D. A. Todhunter, P. S. Schoenberger, W. D. Hueston, D.
E. Pritchard, G. L. Bowman, L. E. Heider, B. L. Brockett, and H. R. Conrad. 1989.
Bacterial counts in bedding materials used on nine commercial dairies. J. Dairy Sci.
72:250-258.
Ingham, S. C., Y. Hu, and C. Ane. 2011. Comparison of bulk-tank standard plate count and
somatic cell count for Wisconsin dairy farms in three size categories. J. Dairy Sci.
94:4237-4241.
Laughton, C., N. Flint, and S. Petig. 2014. 2013 Northeast dairy farm summary. Farm Credit
East, ACA and Yankee Farm Credit, ACA. 56p.
MacDonald, J. M., E. J. O’Donoghue, W. D. McBride, R. F. Nehring, C. L. Sandretto, and R.
Mosheim. 2007. Profits, costs, and the changing structure of dairy farming. Economic
Research Report Number 47. United States Department of Agriculture Economic
Research Service, Washington, DC.
McBride, W. D., and C. Greene. 2009. Characteristics, costs, and issues for organic dairy
farming. Economic Research Report Number 82. United States Department of
Agriculture Economic Research Service, Washington, DC.
45
Munoz, M. A., G. J. Bennett, C. Ahlstrom, H. M. Griffiths, Y. H. Schukken, and R. N. Zadoks.
2008. Cleanliness scores as indicator of Klebsiella exposure in dairy cows. J. Dairy Sci.
91:3908-3916.
Nickerson, S. C., W. E. Owens, and R. L. Boddie. 1995. Mastitis in dairy heifers: Initial studies
on prevalence and control. J. Dairy Sci. 78:1607-1618.
Oleggini, G. H., L. O. Ely, and J. W. Smith. 2001. Effect of region and herd size on dairy herd
performance parameters. J. Dairy Sci. 84:1044-1050.
Panivivat, R., E. B. Kegley, J. A. Pennington, D. W. Kellogg, and S. L. Krumpelman. 2004.
Growth performance and health of dairy calves bedded with different types of materials.
J. Dairy Sci. 87:3736-3745.
Pennington, J. A. 1994. Reducing somatic cell count in dairy cattle. University of Arkansas
Cooperative Extension Service, FSA-4002-PD-3-06R.
Pol, M., and P. L. Ruegg. 2007. Treatment practices and quantification of antimicrobial drug
usage in conventional and organic dairy farms in Wisconsin. J. Dairy Sci. 90:249-261.
Rodrigues, A. C. O., D. Z. Caraviello, and P. L. Ruegg. 2005. Management of Wisconsin dairy
herds enrolled in milk quality teams. J. Dairy Sci. 88:2660-2671.
Ruegg, P. L. 2006. The role of hygiene in efficient milking. WCDS Adv Dairy Tech. 18:285-
293.
Sato, K., P. C. Bartlett, R. J. Erskine, and J. B. Kaneene. 2005. A comparison of production and
management between Wisconsin organic and conventional dairy herds. Livest Prod Sci.
93:105-115.
Sivo, S. A., C. Saunders, Q. Chang, and J. J. Jiang. 2006. How low should you go? Low response
rates and the validity of inference in IS questionnaire research. J. Assoc. Inf. Syst. 7(6):
351-414.
Smith, K. L., D. A. Todhunter, and P. S. Schoenberger. 1985. Environmental mastitis: Cause,
prevalence, prevention. J. Dairy Sci. 68:1531-1553.
46
Stiglbauer, K. E., K. M. Cicconi-Hogan, R. Richert, Y. H. Schukken, P. L. Ruegg, and M.
Gamroth. 2013. Assessment of herd management on organic and conventional dairy
farms in the United States. J. Dairy Sci. 96:1290-1300.
Tauer, L. W., and A. K. Mishra. 2006. Can the small dairy farm remain competitive in U.S.
agriculture? Food Policy. 31:458-468.
Tucker, C. B., D. M. Weary, M. A. G. von Keyserlingk, and K. A. Beauchemin. 2009. Cow
comfort in tie-stalls: Increased depth of shavings or straw bedding increases lying time. J.
Dairy Sci. 92:2684-2690.
USDA (US Department of Agriculture). 2012. Census of Agriculture. Washington, DC.
USDA (US Department of Agriculture). 2015. Crop Production 2014 Summary. National
Agriculture Statistics Service. Washington, DC.
USDA (US Department of Agriculture). 2014. Organic Survey (2014). AC-12-SS-4.
Washington, DC.
Wenz, J. R., S. M. Jensen, J. E. Lombard, B. A. Wagner, and R. P. Dinsmore. 2007. Herd
management practices and their association with bulk tank somatic cell count on United
States dairy operations. J. Dairy Sci. 90:3652-3659.
Woodall, C. W., W. G. Luppold, P. J. Ince, R. J. Piva, and K. E. Skog. 2012. An assessment of
the downturn in the forest products sector in the northern region of the United States.
For. Prod. J. 61:604 – 613.
Zadoks, R. N., H. M. Griffiths, M. A. Munoz, C. Ahlstrom, G. J. Bennett, E. Thomas, and Y. H.
Schukken. 2011. Sources of Klebsiella and Raoultella species on dairy farms: Be careful
where you walk. J. Dairy Sci. 94:1045-1051.
Zdanowicz, M., J. A. Shelford, C. B. Tucker, D. M. Weary, M. A. G. von Keyserlingk. 2004.
Bacterial populations on teat ends of dairy cows housed in free stalls and bedded with
either sand or sawdust. J. Dairy Sci. 87:1694-1701.
47
Zehner, M. M., R. J. Farnsworth, R. D. Appleman, K. Larntz, and J. A. Springer. 1986. Growth
of environmental mastitis pathogens in various bedding material. J. Dairy Sci. 69:1932-
1941.
48
APPENDICES
Appendix 1: IRB Approval Letter
49
Appendix 2: Questionnaire sent to New England Dairy Farmers
New England Dairy Farm Survey:
Low-cost alternatives for animal bedding materials
Thank you for taking the time to consider participating in this short survey. The goal of our
study is to assess which bedding materials regional dairy farmers use primarily, as well as the
associated costs and benefits of each available option. We also aim to gauge farmers’ interest in
the on-farm production of wood-shavings as a low-cost alternative for animal bedding material.
The anonymous information gathered here will be used by the University of New Hampshire and
the University of Vermont to identify and evaluate low-cost alternatives for bedding materials
available to dairy farmers in our region. This short survey will take only about ten minutes of
your time to complete, yet the goal of the project is to provide you and other dairy farmers with
significant savings in materials costs for years to come.
Informed Consent Information: This project is being conducted by the University of New
Hampshire (UNH) and the University of Vermont (UVM), with the approval of the Institutional
Review Board (IRB) at UNH. Please note that your participation is purely voluntary; you are
free to refuse to answer any question, and you are free to withdraw your consent and discontinue
participation at any time. All data collected will be completely anonymous. The results of this
research may be published or reported to scientific bodies, and any such reports or publications
will be reported in a group format. Thus, no individual identity will be determinable through
demographic variables such as age or gender. Participation in this study is expected to present
minimal risk to you and other participants. Our anticipated number of participants is 100
regional dairy farmers. You are not expected to receive any direct benefits from your
participation; however, the investigator hopes that the information gained here will benefit
society indirectly.
If at any time you have questions or concerns about any procedure in this project, you may e-
mail the primary investigator, Charles Louis Simms, at [email protected] or speak with
him directly by calling 207-409-0056. Additional researchers include Matt Smith, Dr. John
Aber, Dr. John Halstead, Jen Colby and Dr. Juan Alvez. You should also understand that you
can select to be provided a summary of the project’s findings at the end of this survey. If you
have any questions about your rights as a research subject, you may contact Julie Simpson in
UNH Research Integrity Services at 603-862-2003 or email at: [email protected]
50
New England Dairy Farm Survey: Low-cost alternatives for animal bedding materials
1: As of today, how many cattle (include calves and non-lactating) do you manage at your farm?
______
1a: As of today, how many lactating cows are you managing at your farm? ____________
2: What type of housing system(s) do you use for your cows (circle all that apply)?
a. Tie-Stall b. Free-Stall c. Compost Bedded Pack
d. Bedded Pack e. Other (please specify here) _________________________
3: What % of your TOTAL bedding materials consists of the following: (Please be sure %’s add
up to 100)
a. Sand __________________ % b. Sawdust ________________%
c. Hay ____________________ % d. Straw ___________________ %
e. Woodchips _______________ % f. Wood-shavings ____________ %
g. Composted Manure Solids __________% h. Other (please specify below)
__________ %
4: Why do you prefer the bedding materials mix you currently use? (please indicate if you use
different materials for lactating and non-lactating cows – for additional comments, please use
back page)
5: Do you explicitly avoid any of the bedding materials listed above? (please specify which and
why)
6: Please provide a total estimate of how much bedding material you purchased in calendar year
2013 (please estimate in cubic yards OR tons): _______________ cubic yards OR
_____________ tons
7: Please provide a total estimate of your annual expenses for bedding materials in 2013:
$_________
8A: Of all your farming expenses, where does bedding material rank? (1st=most costly, 2nd, 3rd,
etc.) ___
8B: Please list your five largest farming expenses, from most costly to least costly:
51
9A: Have you changed your primary source of bedding material in the past 10 years? YES /
NO
9B: IF YES, what was your primary bedding material before, and why did you change?
10: In the past 10 years, have your bedding materials costs per ton or cubic yard: (circle one
answer) (A) Increased (B) Decreased (C) Stayed the
same
11: Please provide a total estimate of how much bedding material you purchased in 2003 (10
years ago) (estimate in cubic yards OR tons): _______________ cubic yards OR
_________________ tons
12: Estimate your total annual bedding materials expenses for 2003 (10 years ago):
$___________
13A: If you have been purchasing bedding materials for less than 10 years, please estimate your
total annual bedding costs for your first full year of operation: year of operation ______ and
$___________ 13B: Please provide a total estimate of how much bedding material you
purchased in that first full year of operation (estimate in cubic yards OR tons):
_______________ cubic yards OR _____________tons
14A: Does your farm have woodlots? YES / NO 14B: Are they mostly: Hardwood /
Softwood / Mixed
14C: Please specify your estimated wooded acreage: ______________________ acres
14D: For what purposes/values do you manage your woodlots? (ex: firewood, timber, etc.)
15: Are you familiar with the application of wood-shaving machines for on-farm production of
animal bedding materials? A. NO B. Yes, I have seen one in operation C. Yes, I have
heard/read of them
16: Would you be interested in participating in a local farmer cooperative project to produce
your own wood-shavings as a low-cost alternative for animal bedding materials? YES /
NO / MAYBE
17: Would you purchase wood-shavings from local farmer co-ops? YES / NO /
MAYBE
18: Would you require that those wood-shavings be kiln-dried? YES / NO
52
19: What maximum moisture-content would be acceptable for your use of the shavings?
________ %
20: What breed(s) of cow do you manage?
______________________________________________
21: What is your farm’s total annual milk-production? (please specify in pounds)
_______________
22: What is your herd’s average somatic cell count? (please specify in cells/ml)
________________
23: How would you describe your farm: A. Conventional B. Organic
24: For how many years have you been managing cattle? _____________ years
25: What is your age? __________ years 26: What is your gender? Male /
Female
27. Would you be willing to participate in further interviews concerning this project? YES /
NO
28. Would you like to receive the results of this survey? YES / NO
If you would like to receive the results of this survey and/or participate in further interviews,
please provide your contact information below, as well as any additional comments that you
would like to convey for this project (feel free to use space on back of these pages, as well) –
THANK YOU VERY MUCH for your time and assistance in this project!
53
CHAPTER 3: FINANCIAL VIABILITY OF PRODUCING ANIMAL
BEDDING WITH A WOOD SHAVING MACHINE
ABSTRACT
This study explores the feasibility of producing animal bedding using a wood shaving
machine. A financial decision model was created to allow individuals to input their own site and
machine-specific parameters into the model. Model output provides financial data for two
scenarios. Scenario 1 explores whether a farmer could support this type of venture by producing
bedding for on-farm consumption only. Scenario 2 explores the profit potential an individual
(farmer or non-farmer) could gain if running the wood shaving machine for 10, 20, 30, or 40
hours per week throughout the year. Model output for Scenario 1 found that the average-sized
organic and conventional dairy farm in New England would not be capable of supporting this
type of venture if only producing bedding for on-farm consumption. For organic dairies,
operating and capital costs would result in an annual loss of $6,844 per year during a five year
financing period, with a machine payback period of 11 years. Farms likely to find a reasonable
return from this type of venture would be large-scale, or a cooperative of locally clustered small
farms using the machine collectively. For Scenario 2, annual revenue during the loan period
ranged from $36,553 - $186,703 when operating 10 - 40 hours per week, with an implicit hourly
wage of $70 - $90 per hour for one worker. Outside the loan period, annual revenue increased to
$50,050 - $ 200,200, with an implicit wage of $96 per hour for one worker operating 10 - 40
hours per week. However, if operating for 40 hours per week throughout the year, supply may
outweigh demand, as enough bedding could be produced for 30 organic dairies or 25
conventional dairies of average size in the New England region. Regardless of the scenario,
54
economies of scale were found, as increasing bedding output spread the capital cost of the
machine over more units of bedding.
INTRODUCTION
From 2003 - 2013, the cost of animal bedding increased by 89% for organic and
conventional dairy farmers across the New England region (Smith et al. 2016). Cost increases
over this period were largely driven by a shortage of woody bedding from regional sawmills.
Historically, these sawmills provided inexpensive woody bedding (sawdust and planer shavings)
as a byproduct of the milling process. However, the collapse in the new home construction
market in 2005 and the recession of 2007-2009 resulted in the closing of a large fraction of the
region’s sawmills (Woodall et al., 2012). Bedding supply was further limited due to increased
mill efficiency and modernization, which reduced the amount of waste in the milling process.
Increased efficiency and modernization for some mills also meant redirecting their entire wood
waste stream to a wood-fired boiler to power their dry kilns, further reducing bedding
availability. The shortage of woody bedding also resulted in increased transportation cost to
deliver bedding, as the distance between operating mills selling bedding and dairy farms
increased. With 96% of New England dairy farmers using some fraction of woody bedding
(Smith et al. 2016), few were able to avoid the increased costs.
While other forms of animal bedding are utilized in New England, the cost of alternative
bedding also increased, as dairy farmers switching to other forms of bedding drove up demand
for those resources. Smith et al. (2016) found that 26% of surveyed New England dairy farmers
switched their primary bedding material from 2003 - 2013. Of those switching, 82% moved
away from woody bedding, with cost being the primary motivator for using an alternate material.
55
Beyond switching bedding materials, farmers can find economic relief by producing
bedding onsite with currently available resources. One method used in New England is to grow
hay for bedding. However, of the 129 New England dairy farmers surveyed in Smith et al.
(2016), only 20% reported using hay, with 65% of those surveyed using < 20% of the material
when compared to other bedding materials. While cost savings can be achieved by growing hay
for bedding, the high nutrient content of the material can cause the spread of disease-causing
bacteria that are associated with mastitis (Dyck et al. 2009). Hay is also incompatible with most
liquid manure systems (Smith et al. 2016), and growing the material for bedding competes with
land for forage and pasture. This is of particular concern for organic farms, which require
expensive organic feed. Hay is also nitrogen demanding, meaning more fertilizer inputs will
have to be allocated toward non-food producing crops.
Another bedding material that can be produced onsite is composted manure solids
(CMS). This material is beginning to gain popularity, due to low cost and the ability to sell the
final product as compost. However, this material is still relatively uncommon in New England,
with only 4% of the surveyed dairy farmers in Smith et al. (2016) using the material. CMS are
also associated with high bacterial counts in the bedding (Dyck et al. 2009).
A third option for creating bedding onsite is to use the farm woodlot for the production of
shavings using a wood shaving machine. Of those surveyed in Smith et al. (2016), one dairy
farmer was using this strategy. Unfortunately, information on the economics, management
requirements, and general feasibility of this type of venture is absent from the academic and
extension literature. With the cost of a wood shaving machine starting at roughly $50,000, there
is risk in starting this type of venture, due to the lack of literature and resources regarding the
actual feasibility of producing animal bedding with this strategy. Filling this knowledge gap is
56
especially important for New England dairy farmers, who are in need of lower cost bedding
options. Regional dairy farmers also have the necessary acreage to supply wood for this type of
venture, with those being surveyed in Smith et al. (2016) reporting an average size woodlot of
159 acres. For these reasons, researchers at the University of New Hampshire (UNH) purchased
a wood shaving machine for use at their Organic Dairy Research Farm (ODRF) to determine the
feasibility of this type of venture. The objectives of this study were to 1) develop a financial
decision model that allows individuals to input site and machine-specific information to
determine the economics of producing bedding at their site, 2) use the model to predict the
economics of producing bedding for the average sized New England organic dairy farm, 3) use
the economic model to forecast the revenue an individual (farmer or non-farmer) could generate
if using the wood shaving machine for 10, 20, 30, and 40 hrs per week and 4) provide
operational recommendations to those interested in pursuing this type of venture.
METHODS
Research Site
The UNH ODRF was selected as the location for the research project, as the farm milks
50 cows, which is very close to the New England average of 53 cows for an organic dairy
(McBride and Green 2009). The farm also has a 120 acre woodlot, with large quantities of low
quality eastern white pine that could be used to produce animal bedding. Additionally, the farm
was already using eastern white pine shavings as bedding, allowing for an easy transition from
purchased bedding to material being produced onsite.
57
Wood Shaving Machine and Accessories
After review of several different wood shaving machine companies and models, the
research team decided to purchase a trailer-mounted Tremzac 248T with a built-in diesel engine
(Figure 4).
Figure 4: UNH Wood Shaving Machine Being Loaded
Machine cost was $52,900. This particular manufacturer and model were selected because of the
close proximity to the distributer and service team, an output of 19 yd3/hr, the ability to have a
diesel-powered and trailer-mounted machine, and the eagerness of the company to have their
machine tested under a research setting. A built-in diesel engine was required, due to a lack of
tractors on the research farm that could drive a PTO machine during the summer months when
most of the tractors are used for forage management. In addition to the wood shaving machine, a
six tine Sidney Gorilla Grapple skid steer attachment was purchased to load the 8 ft. logs into the
machine hopper. Cost for the grapple was $2350.
Wood Shaving Machine Financial Decision Model
The financial decision model was created using Microsoft Excel. Excel was used for ease
of creating the model, and the ease at which it could be distributed and used by farm and forestry
58
stakeholders. Model parameters were developed from operational experience from two years of
using the wood shaving machine (2014 and 2015), ensuring all possible costs were included. The
two years of operational experience also allowed for the creation of a model capable of handling
site and machine-specific variabilities.
The final economic decision model had 23 input parameters, which are to be filled in by
the individuals considering this type of venture (Table 12).
Table 12: Parameters for the Economic Decision Model
Parameter # and Descriptions
1 Machine Make/Model 13 Fuel Consumption for Material Handling &
Log Loading (gallon/hr)
2 Power Source 14 Hourly Wage per Worker
3 Machine Output Adjusted for Log
Loading (yd3/hr)
15 Number of Workers Operating the Machine
4 Bedding Requirement for Farm
(yd3/yr)
16 Average Cost of Wood per Cord
5 Current Cost of Bedding on
Farm/Year
17 Average Cost of Wood per Ton
6 Cost of Wood Shaving Machine +
Associated Expenses
18 Local Cost of Knife Sharpening per
Sharpening
7 Interest Rate if Financing the Wood
Shaving Machine
19 Cost of Knife Replacement
8 Loan Period (Yrs) 20 Cost of Routine Maintenance per Day
9 Fuel Consumption to Run Shaving
Machine (gallons/hr)
21 Depreciation Period (Yrs)
10 Current Cost of Diesel/Gallon 22 Depreciation Salvage Value (residual value)
11 Energy Consumption if Using an
Electric Machine
23
Local Cost of 1 yd3 of Air-Dried Green
Shavings 12 Cost/kWh for Electricity in Local
Area
When looking at the 23 parameters, the model requires some values to be left blank. For
instance, if using a PTO-powered machine, the model informs the individual to leave parameters
11 (energy consumption if using an electric machine) and 12 (cost/kWh for electricity) blank. A
second example is if the individual is purchasing wood by the ton. In this scenario, parameter 16
(average cost of wood per cord) would be left blank (Figure 5).
59
Figure 5: Example of Parameter Values
As depicted in Figure 5, the model also has specific instructions below each parameter,
guiding the individual on how to properly fill in the information. The model also comes with a
“Parameter Definitions” and “Model Calculations” tab within the excel file, providing further
information on: 1) conversion factors to get into the right units for use in the model, 2)
background calculations for how the model works, 3) troubleshooting procedures should
something not work, and 4) links to various information sources to assist in filling out the model
(Figure 6).
Figure 6: Example of Parameter Definitions
Background Model Calculations
After inputting the 23 parameter values, background equations in excel provide a range of
output data. Of the background calculations, three necessitate explanation, due to the fact that
60
alternative methods could have been used in calculating the model output. The first background
calculation relates to how depreciation was calculated. In this model, machine hour rate
depreciation was selected, as it best represents how a machine wears over its productive life. The
following equation was used:
Depreciation per Hour = (cost of machine – salvage value)/total machine hours
The second background calculation relates to how the capital expenses (wood shaving
machine, grapple, log loader, etc.) are financed. In this model, fixed-rate financing was assumed,
primarily due to the inability to forecast rate changes of a variable rate loan. The following fixed-
rate formula was used:
Fixed-Rate Monthly Loan c = r/ [1-(1+r)-N] * P0
Where:
r = monthly interest rate expressed as a fraction (= r/100/12)
N = loan's term (# years of loan * 12)
P0 = Loan Principal (shaving machine, log loader, etc.)
c = Monthly Payment
The final background calculation of note is the expansion factor between solid wood and
shavings. In this study, the expansion factor was determined experimentally, by comparing the
average weight between 1 ft3 of 1/8’’ shavings from fresh cut eastern white pine logs to that of 1
ft3 of solid fresh cut eastern white pine wood. The weight of the wood shavings was based on
filling a 1 ft3 box (30 replicates) and obtaining an average weight per 1 ft3 of shavings. A
shavings value of 7.4 lbs/ft3 was found and used for the model (Appendix 1). The weight of solid
eastern white pine wood was based on values from the literature. A value of 37.1 lbs/ft3 was used
for the model (Coder 2011). Based on these weights, an expansion factor of 5:1 was found
61
between solid wood and shavings. The expansion factor resulted in a solid wood to shaving
conversion factor of 10.0 yd3 of shavings for every ton of fresh cut pine and 15.7 yd3 of shavings
per cord of solid wood.
Model Output
Output for the model is delivered for two different scenarios simultaneously, both of
which report values on an annual basis. The first scenario is for farmers looking to produce
enough bedding for their farm, with no intention of selling material beyond that point. Output for
this scenario includes: hours of machine operation needed for farm demand, wood requirement to
meet farm demand (cords or tons), wood cost (cords or tons), labor cost for running the machine,
labor cost associated with setup, labor cost associated with removing knives, fuel cost for log
loading, fuel cost for machine operation (PTO, electric, or diesel), cost to sharpen knives, cost to
purchase new knives, maintenance cost, depreciation, total operating cost and capital costs
amortized over the loan period. The model also provides the total cost (operating and capital) to
produce 1 yd3 of shavings, revenue, and the implicit wage potential for the loan and post loan
period.
The second scenario explores the potential revenue an individual (farmer or non-farmer)
could make if running the machine for 10, 20, 30, or 40 hours per week. Output is presented in
this manner to allow for situations where individuals have another full time occupation, and the
wood shaving venture is for supplemental income, as is the likely case for most farm or forestry
operations. Output categories from this scenario are the same as Scenario 1.
Parameter Values used in this Study
While the output from the decision model will vary between operators, the following
parameters were used in this case study to illustrate 1) how the model works, 2) the economics of
62
using a wood shaving machine to support an average-sized New England organic dairy farm, and
3) the economics of using a wood shaving machine based on machine usage (Table 13).
Table 13: UNH ODRF Parameter Values
Parameters 1 - 12 Value Parameters 13 - 23 Value
Machine Make/Model Tremzac
248T
Cost/kWh for Electricity in Local
Area
$0.12
Power Source Diesel Fuel Consumption for Material
Handling & Log Loading (gallon/hr)
1
Machine Output Adjusted for Log
Loading (yd3/hr)
10
Hourly Wage per Worker $0
Bedding Requirement for Farm
(yd3/yr)
700 Number of Workers Operating the
Machine
2
Current Cost of Bedding on
Farm/Year
$10,500 Average Cost of Wood per Cord $0
Cost of Wood Shaving Machine +
Associated Expenses
$60,850 Average Cost of Wood per Ton $40
Interest Rate if Financing the
Wood Shaving Machine
4.15% Local Cost of Knife Sharpening per
Sharpening
$100
Loan Period (yrs) 5 Cost of Knife Replacement $800
Fuel Consumption to Run Shaving
Machine (gallons/hr)
2 Cost of Routine Maintenance per Day $10
Current Cost of Diesel/Gallon $2.07 Depreciation Period (Yrs) 15
Energy Consumption if Using an
Electric Machine
0
Depreciation Salvage Value (residual
value)
$5000
Local Cost of 1 yd3 of Air-Dried
Green Shavings
$15
RESULTS AND DISCUSSION
Scenario 1: On-Farm Production Value Only
The use of a wood shaving machine to produce animal bedding for on-farm purposes
exclusively, would not provide a positive economic return for a majority of New England dairy
farms, which do not have a high enough bedding demand to pay for the capital and operating
costs of the machine. When using the case study farm, which represents the average size New
England organic dairy, bedding import costs would be offset by $10,500/yr, but would cost
$17,344/yr in capital and operating costs to produce. This would result in an annual loss of
$6,844 during the five year financing period. Post loan period, the farm would save roughly
63
$6,653 per year with a 76 hour labor requirement, earning an implicit wage of $88/hr for one
worker (Table 14). The machine would not pay for itself for 11 years.
Table 14: Cost Reductions of Producing Bedding for On-Farm Production Only
Tremzac 248T
Annual
Parameter
Value
Machine Hours Needed to Meet Farm Bedding Demand 76
Wood Required to Meet Farm Demand (tons) 70
Wood Cost (tons) $ 2,800
Fuel Cost for Log Loading $ 158
Fuel Cost for PTO or Diesel Mills $ 315
Cost of Knife Sharpening $ 190
Cost of Knife Replacement $ 152
Maintenance Cost (oil, chain and bearing replacement, etc.) $ 95
Depreciation $ 136
Operating Costs (sum of the above costs) $ 3,847
Amortized Capital Costs (shaving machine, log loader,
conveyor, etc.) $13,497
Loan Period Cost (capital costs, operating cost) $ 17,344
Loan Period Revenue $ (6,844)
Implicit Hourly Wage/Person During Loan Period $ (45)
Post Loan Period Cost, or Cost after Year 1 if not Taking out a
Loan (operating cost) $ 3,847
Post Loan Period Revenue $ 6,653
Implicit Hourly Wage for One Worker $88
If using the same parameter values, the model calculates that the breakeven volume of
wood shavings during the 5 year loan period would be 1420 yd3 at a cost of $21,300 ($15/yd3).
Farms with bedding cost and volume equivalents less than these values would not cover capital
and operating costs during the loan period. However, post loan period, the farm would save
$13,497 per year and receive an implicit wage of $88/hr for one worker after all costs are
removed (Table 15). In this scenario, the machine would pay for itself in five years.
64
Table 15: On-Farm Bedding Production Breakeven Point
Tremzac 248T
Annual
Parameter
Value
Machine Hours Needed to Meet Farm Bedding Demand 154
Wood Required to Meet Farm Demand (tons) 142
Wood Cost (tons) $ 5,680
Fuel Cost for Log Loading $ 320
Fuel Cost for PTO or Diesel Mills $ 639
Cost of Knife Sharpening $ 386
Cost of Knife Replacement $ 309
Maintenance Cost (oil, chain and bearing replacement, etc.) $ 193
Depreciation $ 276
Operating Costs (sum of the above costs) $ 7,803
Amortized Capital Costs (shaving machine, log loader,
conveyor, etc.) $ 13,497
Loan Period Cost (capital costs, operating cost) $ 21,301
Loan Period Revenue $ 0
Implicit Hourly Wage/Person During Loan Period $ 0
Post Loan Period Cost, or Cost after Year 1 if not Taking out a
Loan (operating cost) $ 7,803
Post Loan Period Revenue $ 13,497
Implicit Hourly Wage for One Worker $ 88
While these parameters are specific to the university farm, this scenario illustrates a very
important point – economies of scale exist with this type of venture. With the capital cost spread
over five years, the actual cost of shavings decreases when output increases, as the capital costs
are spread over more units of bedding. For this reason, smaller farms like the one used in this
case study, will likely have a negative return during the loan period if they are only producing
wood shavings for themselves. This is of concern, as Smith et al. (2016) found that the average
bedding usage per cow in New England was 13 yd3 per year. With the average number of cows
per organic and conventional farm in New England being 53 (Mcbride and Green 2009) and 65
cows (USDA 2012) respectively, the average bedding demand of 689 yd3 for organic and 845
yd3 for conventional dairies would not cover cost during the loan period.
65
Scenarios where on-farm production of bedding make sense are on larger farms, where
the payback period is less than five years. Alternatively, a group of smaller but locally clustered
farms could combine resources in a wood shaving cooperative, to gain the benefits of economies
of scale. With many of the small to mid-range wood shaving machines being trailer-mounted, the
machine could be transported from one farm to the other, where farmers could process logs from
their own woodlots. Due to the high output from even the smaller machines, a small-scale dairy
could produce enough bedding for a month in a 1-3 day period. By way of example, the UNH
ODRF has a 125 yd3 storage bay in the barn for wood shavings, which could be filled in 12 hours
of machine operation. With that in mind, a fairly large number of farms could be in the
cooperative without running into use conflicts.
When looking at the feasibility and interest in wood shaving cooperatives, Smith et al.
(2016) found that 58% of the New England dairy farmers in their survey, showed interest in
wood shaving cooperatives. Of those interested 82% were owners of farms with ≤ 99 cows.
Interestingly, farmers with ≥ 200 cows had absolutely no interest in the cooperatives (0%
interest).
Scenario 2: Animal Bedding Production for Profit
While producing bedding exclusively for the farm is a feasible option for larger
operations or a cooperative of smaller farms, Scenario 2 illustrates the more likely situation
where the venture is profit-driven and bedding is sold. Table 16 illustrates the economics of a
wood shaving venture based on machine use per week, using the same 23 parameter values in
Scenario 1.
66
Table 16: Profit Potential of Producing Bedding using a Wood Shaving Machine
Potential Revenue if Running the Machine for 10, 20, 30, or 40 hours per week
Machine Hours/Week 10 20 30 40
Machine Hours/Yr 520 1040 1560 2080
Bedding Production/Yr (yd3) 5,200 10,400 15,600 20,800
Wood Requirement/Yr (tons) 520 1,040 1,560 2,080
Wood Cost/Yr (tons) $ 20,800 $ 41,600 $ 62,400 $ 83,200
Fuel Cost/Yr for Material Handling &
Log Loading $ 1,076 $ 2,153 $ 3,229 $ 4,306
Fuel Cost/Yr for PTO or Diesel
Machines $ 2,153 $ 4,306 $ 6,458 $ 8,611
Cost of Knife Sharpening/Yr $ 1,300 $ 2,600 $ 3,900 $ 5,200
Cost of Knife Replacement/Yr $ 1,040 $ 2,080 $ 3,120 $ 4,160
Maintenance Cost/Yr $ 650 $ 1,300 $ 1,950 $ 2,600
Annual Depreciation $ 931 $ 1,862 $ 2,793 $ 3,723
Total Annual Operating Costs (includes depreciation) $ 27,950 $ 55,900 $ 83,850 $ 111,800
Annual Loan Payment for Capital Costs $ 13,497 $ 13,497 $ 13,497 $ 13,497
Annual Costs During Loan Period (capital costs, operating cost, and depreciation) $ 41,447 $ 69,397 $ 97,347 $ 125,297
Cost Per 1 yd3 Shavings During Loan
Period $ 7.97 $ 6.67 $ 6.24 $ 6.02
Revenue During Loan Period $36,553 $ 86,603 $136,653 $186,703
Implicit Hourly Wage for One Worker $ 70.29 $ 83.27 $ 87.60 $ 89.76
Annual Cost Post Loan Period, or after
Year 1 if not Taking out a Loan (operating cost) $ 27,950 $ 55,900 $ 83,850 $ 111,800
Cost Per 1 yd3 Shaving $ 5.38 $ 5.38
$
5.38 $ 5.38
Revenue Post Loan Period $ 50,050 $100,100 $150,150 $ 200,200
Implicit Hourly Wage for One Worker $ 96.25 $ 96.25 $ 96.25 $ 96.25
As depicted in Table 16, the annual revenue during the loan period ranges from $36,553 -
$186,703 when operating 10 - 40 hours per week. The implicit hourly wage for one worker is
$70 - $90 per hour respectively. The average cost per 1 yd3 of shavings also decreases from $8 to
$6 when running the venture full time vs part time. Post loan period, annual revenue increases to
$50,050 - $ 200,200, with an implicit wage of $96 per hour when operating 10 - 40 hours per
week. An important point to mention is that the output from this scenario provides an implicit
hourly wage, which is suited for a family run business, where a specific wage is not paid to
67
employees. However, Table 17 illustrates the economics of this type of venture if two employees
are hired to run the wood shaving venture and are each paid $15/hr.
Table 17: Profit Potential if Producing Shavings with Employees Receiving a Wage
Potential Revenue if Running the Machine for 10, 20, 30, or 40 hours per week
Machine Hours/Week 10 20 30 40
Machine Hours/Yr 520 1040 1560 2080
Bedding Production/Yr (yd3) 5,200 10,400 15,600 20,800
Wood Requirement/Yr (tons) 520 1,040 1,560 2,080
Wood Cost/Yr (tons) $ 20,800 $ 41,600 $ 62,400 $ 83,200
Labor Cost/Yr for Running the Machine $ 15,600 $ 31,200 $ 46,800 $ 62,400
Labor Cost/Yr for Setup & Takedown &
Knife Replacement and Sharpening $ 1,365 $ 2,730 $ 4,095 $ 5,460
Fuel Cost/Yr for Material Handling &
Log Loading $ 1,076 $ 2,153 $ 3,229 $ 4,306
Fuel Cost/Yr for PTO or Diesel
Machines $ 2,153 $ 4,306 $ 6,458 $ 8,611
Cost of Knife Sharpening/Yr $ 1,300 $ 2,600 $ 3,900 $ 5,200
Cost of Knife Replacement/Yr $ 1,040 $ 2,080 $ 3,120 $ 4,160
Maintenance Cost/Yr $ 650 $ 1,300 $ 1,950 $ 2,600
Annual Depreciation $ 931 $ 1,862 $ 2,793 $ 3,723
Total Annual Operating Costs (includes depreciation) $44,915 $ 89,830 $134,745 $ 179,660
Annual Loan Payment for Capital Costs $ 13,497 $ 13,497 $ 13,497 $ 13,497
Annual Costs During Loan Period (capital costs, operating cost, and depreciation) $ 58,412
$103,327
$148,242 $ 193,517
Cost Per 1 yd3 Shaving During Loan
Period $ 11.23 $ 9.94 $ 9.50 $ 9.29
Revenue During Loan Period $ 19,588 $ 52,673 $ 85,758 $ 118,843
Implicit Hourly Wage for Business
Owner $ 37.67 $ 50.65 $ 54.97 $ 57.14
Annual Cost Post Loan Period, or after
Year 1 if not Taking out a Loan (operating cost) $ 44,915 $ 89,830
$134,745 $ 179,660
Cost Per 1 yd3 Shaving $ 8.64 $ 8.64 $ 8.64 $ 8.64
Revenue Post Loan Period $33,085 $ 66,170 $ 99,255 $ 132,340
Implicit Hourly Wage for Business
Owner Post Loan Period $ 31.81 $ 31.81 $ 31.81 $ 31.81
If hiring two individuals to run the machine, the annual revenue during the loan period ranges
from $19,588 - $118,843 when operating 10 - 40 hours per week. The implicit hourly wage for
the business owner of the venture would be $38 - $57 per hour respectively. The average cost of
68
shavings also increases due to the workers’ wages, with a cost of $ 9 - $11 per 1 yd3 for a 10 – 40
hr per week operation during the loan period. Post loan period, annual revenue increases to
$33,085 - $132,340, which is equivalent to an implicit wage of $32 per hour for the business
owner, when operating 10 - 40 hours per week.
When looking at both examples under Scenario 2, where profit is the primary objective, it
is important to note that just because the venture is capable of producing large volumes of wood
shavings, does not mean there is sufficient demand locally to absorb that level of supply. If
running the machine for 40 hours per week throughout the year, roughly 20,800 yd3 of wood
shavings could be produced. If the UNH ODRF represents the average-sized organic dairy farm
in New England, it would take roughly 30 dairy farms to absorb this quantity of wood shavings.
If looking at conventional dairies, it would require 25 average-sized farms. However, if the
bedding were sold to other farm operations (equine, poultry, swine, etc.), the feasibility of
supporting a full-scale wood shaving venture becomes more realistic.
A second important consideration for those interested in selling wood shavings is that
material being sold to dairy farms requires the logs to be debarked if the bedding is being used
by lactating cows. This is because bark has been associated with supporting the spread of disease
causing bacteria that can cause coliform mastitis (Thomas 2009, Maroney 2005). While
debarking machines exist, most are well over $30,000 used. The purchase of such a machine,
combined with the fuel and labor to run it, would increase the cost of producing the shavings
substantially. Additionally most dairy farms require the animal bedding to be kiln-dried. Of the
129 New England dairy farmers surveyed in Smith et al. (2016), 63% indicated that they require
woody bedding to be kiln-dried. While this leaves 37% that may purchase green shavings, this
low number could be problematic for an individual forecasting the sale of large volumes of wood
69
shavings. If the business venture is geared toward the dairy industry, a drying system would have
to be considered. However, the equine industry uses pine shavings as a primary bedding source,
and having kiln-dried shavings is not as important to have as in the dairy industry. In some cases,
air-dried shavings are preferred to kiln-dried, as there is reduced dust, which is more of a concern
for horses than having slightly elevated moisture content (kiln-dried at 10-12% vs. air dried at
20-30%). In this situation, the wood shavings producer could sell to the equine industry without
the need of a drying system.
When looking at the economic feasibility of Scenario 2, the potential for large capital
expenses beyond the wood shaving machine (debarker and drying system) may prove to be
financially unrealistic for most individuals. However, this type of venture may prove to be suited
toward regional milling operations. Milling operations are suited for a wood shaving venture, as
they already have industrial dry kilns capable of drying the bedding. This type of operation
would also have the ability to remove the bark off the logs, either through a debarker or by
sawing off the outer layer. Some mills may even find it advantageous to remove the outer layers
of knot-free wood for lumber and run the knotty core through the wood shaving machine.
Regional milling operations also have access to the necessary supply of pine needed to feed the
machine.
Operational Recommendations Relating to Economic Feasibility
After two years of producing wood shavings at the UNH ODRF, the authors of this study
have several operational recommendations on how to make the system more financially viable.
The first recommendation relates to material handling. During the first few months of operation,
a tractor fork was used to load logs into the wood shaving machine hopper. This method was
inefficient, as removing individual logs from the large pile of stored pine proved to be labor
70
intensive. Additionally, loading multiple logs at a time without them entering the hopper slightly
crooked, proved to be difficult. Importantly, crooked logs create airspace in between the logs in
the hopper, reducing machine output. The solution to this problem was purchasing a skid-steer
grapple attachment. This made removing logs from the large pile very efficient. Logs could also
be quickly and strategically placed within the hopper box, requiring only 3.5 minutes to load a
hopper box with 5-8 logs (Appendix 2). The greater control of log placement when using the
grapple also reduced the internal airspace in between logs, increasing machine output efficiency
(Appendix 3). From an economic standpoint, a grapple, or hydraulic log loading deck is worth
the expense. For reference, the UNH ODRF purchased a grapple over a log loading deck, as the
grapple attachment could be utilized for other purposes on the farm.
A second recommendation relating to material handling pertains to how the shavings are
discharged from the end of the wood shaving machine. In most cases, the shavings will simply
discharge out the rear of the machine into a pile, unless a separate conveyor or blower
attachment is purchased. At the UNH ODRF, a separate blower or conveyor attachment was not
purchased initially. Instead, a tractor with a large bucket was placed at the rear of the machine,
where it would capture the shavings, which were then moved and piled away from the machine.
While this method saved capital cost initially, it increased labor cost by requiring two individuals
to run the machine vs. one. As with the grapple, having a separate conveyor or blower to stack
the shavings, or load them for transport, is worth the expense. For reference, the UNH ODRF
will be purchasing a conveyor over a blower, as the conveyor can be utilized for other purposes
on the farm.
A third recommendation regarding the wood shaving venture relates to debarking the
logs. While a debarking machine can be purchased, they are as expensive as a wood shaving
71
machine. An alternative method used at the UNH ODRF, which may be suitable for some
operators, is to remove the bark manually. While this sounds like an unrealistic task, the
following methods proved to be feasible at the research farm. Logs are stored for a 4-6 month
period, where wood shrinkage from drying causes the bark of the pine logs to separate. Roughly
6-8 logs are then loaded onto a raised I-beam platform, requiring 3 minutes 25 seconds to load
(Appendix 4). The logs are then debarked with a sharpened sidewalk ice scraper, requiring an
average of 6 minutes 50 seconds per hopper load (Appendix 5). The logs are then loaded into the
hopper box. Once loaded, the machine runs itself, while the operator waits up to 25 minutes until
more logs need to be loaded. For some operators, this down time could be utilized for debarking,
saving capital and operating costs associated with using a mechanical debarker. This scenario
may be especially feasible for smaller-scale operations, where the quantity of shavings produced
does not justify buying another specialized piece of equipment. For reference, a number of
debarking tools were tested, with a sharpened sidewalk ice scraper being the most effective due
to ease of use and cost.
ADDITIONAL CONSIDERATIONS AND LIMITATIONS OF THE STUDY
While the model presented in this study is robust, allowing for variation between wood
shaving machines, site variables, etc. the output is only as good as the input parameter values.
This can become problematic, as some of the input variables (output, maintenance cost, knife
longevity, fuel consumption), will come directly from the various wood shaving machine
manufacturers. Unfortunately, many of the manufacturers only report best case scenarios, which
means the output from the model will likely underestimate the true cost of production.
A second limitation of the study relates to input variables that are not programed into the
model, but have a significant impact on the overall financial viability of the wood shaving
72
venture. The first variable not addressed in the model is tax incentives. Tax incentives were not
incorporated into the model, as there are too many variables to calculate for every individual’s
specific tax situation. However, it should be noted that the overall economics of the wood
shaving venture, and the numbers reported in the earlier scenarios, would look more positive
after considering this variable. For instance, the model has a specific depreciation rate related to
machine life, but does not include the annual depreciation deduction one can take on machinery.
Likewise, the model does not consider farmer tax incentives that include: machinery/equipment
deduction, farm wage deduction, farm expense deduction, repayment of loans, and others listed
in IRS Publication 225. Additional tax incentives also exist if the individual is harvesting wood
themselves. Forest management and harvesting qualifies for a number of tax credits and
deductions that include: timber depletion, reforestation, depreciation, forest management and
protection, etc. (Wang 2012). If harvesting from farm woodlots, even more tax deductions and
incentives can be found under IRS publication 225. Importantly, the degree to which these tax
credits and deductions affect the financial viability of the operation are specific to the individual
filling out the model and therefore cannot be modeled easily.
While tax incentives will likely increase the financial viability of the wood shaving
venture, the potential need to kiln dry the shavings will add substantial cost to the operation. This
model did not incorporate a drying system, due to the variability in systems and fuel sources.
Additionally, some farmers will purchase air-dried bedding if the moisture content is below 30%.
At the UNH ODRF, bedding is air-dried by placing the shavings on the paved driveway. After a
day in the sun, the moisture content drops below 30%. During a dry day in the summer, the
moisture content drops into the mid-teens, just above kiln-dried levels. However, while this
73
method may be suitable for on-farm production, it would not likely satisfy farmers wanting to
purchase kiln-dried bedding with a moisture content below 12%.
A final variable worth mentioning, which was not incorporated into the model, relates to
how the bedding material will be sold with regard to the expansion factor between solid wood
and shavings. Wood shavings are sold on a volume basis, meaning the individual selling the
material is also selling air. In this study, experimental results found an expansion factor of 5:1
between solid wood and shavings (Appendix 1). However, this expansion factor was 48% less
than the expansion factor reported by Salsco, a manufacturer of wood shaving machines in
Connecticut, USA. On their website (www.salsco.com), they report that 1 ft3 of shavings from
their shaving machine weighs 5 lbs/ft3, resulting in 1 ton of solid wood producing 14.8 yd3 of
shavings (Salsco 2016). If using eastern white pine, this would translate to an expansion factor of
7.42., which is 48% more shavings per unit of solid wood than the background calculation used
in the model. The variability in the expansion factors likely relates to the physical nature of the
shavings themselves. If producing more thick shavings, there is less surface area per unit of solid
wood, which would likely reduce the expansion factor. Likewise, shavings with more of a curl
will stack less efficiently on one another, increasing airspace and the expansion factor. What is
important about this discussion is that the expansion factor has a tremendous effect on the overall
economics of the operation, as larger expansion factors result in more profit, as output per unit of
purchased wood increases.
CONCLUSION
An economic decision model for forecasting the feasibility of using a wood shaving
machine to produce animal bedding was developed and tested using parameter values from the
UNH ODRF. It was found that the average size organic or conventional dairy farm in New
74
England does not have the bedding demand to justify the purchase of a wood shaving machine, if
only producing bedding for on-farm consumption. In this scenario, operating and capital costs
would result in an annual loss of $6,844 per year during the five year financing period, with a
machine payback period of 11 years. If using the same 23 parameter values, costs were covered
during the 5 year loan period if the farm utilized 1420 yd3 of bedding per year, costing $21,300
($15/yd3). Post loan period, the farm would save $13,497 per year and receive an implicit wage
of $88/hr for one worker. Economies of scale were found, as increased bedding output spread the
capital cost of the machine over more units of bedding. This indicates that larger farms with
greater bedding demand will be able to cover costs more effectively than smaller farms.
However, a cooperative of locally clustered small farms using a trailer-mounted wood shaving
machine could pool resources to gain the benefits of economies of scale by bringing the machine
from farm to farm when bedding is needed.
If the wood shaving venture is profit-driven, where bedding is exported, the same 23
parameter values result in an annual revenue during the loan period ranging from $36,553 -
$186,703 when operating 10 - 40 hours per week. The implicit hourly wage for one worker
ranges from $70 - $90 per hour. Outside the loan period, annual revenue increases to $50,050 - $
200,200, with an implicit wage of $96 per hour for one worker operating 10 - 40 hours per week.
However, if operating for 40 hours per week throughout the year, supply may outweigh demand,
as enough bedding could be produced for 30 organic dairies or 25 conventional dairies of
average size in the New England region.
When looking at operational considerations, tax incentives will likely increase the
economics of the venture and profit margin in each of the scenarios presented in this study. In
contrast, the potential need for a debarker or drying system, and the reliance on parameter values
75
from machine manufacturers, will all increase the cost of the venture and the cost values
presented in this study. Importantly, the output from this study should be used on a case by case
basis, where the individual uses the output from this model and combines it with the localized
market for animal bedding in their region. Only then can an informed decision be made on
whether using a wood shaving machine to produce bedding is an economically feasible venture.
76
REFERENCES
Barnett, B.J. 2000. The U.S. farm financial crisis of the 1980s. Agricultural History. 74(2):366-
380.
Coder, K.D. 2011. Estimating tree stem & branch weight. Warnell School of Forestry,
Biomechanics Series, WSFNR11-22.
Dyck, E., S. Flack, K. Hoffman, E. Maltby, K. Mendenhall, R. Perry, S. Richards, L. Tikofsky,
and B. Wallis. 2009. The organic dairy handbook: A comprehensive guide for the
transition and beyond. K. Mendenhall, ed. Northeast Organic Farming Association of
New York, Inc., New York, NY.
Maroney, M. 2005. Coliform mastitis. University of Wisconsin Milk Money Fact Sheet 04.
McBride, W. D., and C. Greene. 2009. Characteristics, costs, and issues for organic dairy
farming. Economic Research Report Number 82. United States Department of
Agriculture Economic Research Service, Washington, DC.
Salsco. 2016. Shaving processing equipment and knife grinder. [Internet]. Available from:
<http://www.salsco.com/products/shavings-processing-equipment.html.>.
Thomas, H.S. 2009. The cattle health handbook: Preventive care, disease treatments &
emergency procedures for promoting the well-being of your beef or dairy herd. Storey
Publishing, North Adams, MA.
USDA (US Department of Agriculture). 2012. Census of Agriculture. Washington, DC.
Wang, L. 2012. Federal income tax on timber: A quick guide for woodland owners. United
States Department of Agriculture Forest Service. FS-1007.
Woodall, C. W., W. G. Luppold, P. J. Ince, R. J. Piva, and K. E. Skog. 2012. An assessment of
the downturn in the forest products sector in the northern region of the United States.
For. Prod. J. 61:604 – 613.
77
APPENDICES
Appendix 1: Calculation of Expansion Factor between Solid Wood and Shavings
Sample
#
Loose Shavings
+ Bucket Weight
(lbs.)
Compacted
Shavings +
Bucket Weight
(lbs.)
Five Gallon
Bucket Weight
(lbs.)
Loose
Shavings
Weight with
1/3 Expansion
Factor (lbs/ft3.)
Compacted
Shavings with
1/3 Expansion
Factor (lbs/ft3.)
1 6.6 9 1.6 7.5 11.1
2 7.2 11.2 1.6 8.4 14.4
3 6 9.8 1.6 6.6 12.3
4 6.2 9.8 1.6 6.9 12.3
5 5.6 8.6 1.6 6.0 10.5
6 6.4 9 1.6 7.2 11.1
7 7.8 12.2 1.6 9.3 15.9
8 7.2 10 1.6 8.4 12.6
9 6.6 10.4 1.6 7.5 13.2
10 8 12 1.6 9.6 15.6
11 6.4 9.4 1.6 7.2 11.7
12 7.2 10.4 1.6 8.4 13.2
13 6.4 9 1.6 7.2 11.1
14 7.8 10 1.6 9.3 12.6
15 6.2 8.2 1.6 6.9 9.9
16 6.8 9.2 1.6 7.8 11.4
17 4.6 7.8 1.6 4.5 9.3
18 4.8 8.2 1.6 4.8 9.9
19 6.4 9.4 1.6 7.2 11.7
20 5.8 8.2 1.6 6.3 9.9
21 6.8 9.2 1.6 7.8 11.4
22 6.8 9.4 1.6 7.8 11.7
23 7.4 9 1.6 8.7 11.1
24 7.4 9 1.6 8.7 11.1
25 6.4 8.6 1.6 7.2 10.5
26 6 8.6 1.6 6.6 10.5
27 6.8 9.4 1.6 7.8 11.7
28 7 10 1.6 8.1 12.6
29 5.8 8.2 1.6 6.3 9.9
30 6 8.2 1.6 6.6 9.9
Average 7.4 11.7
STDEV 1.2 1.6
78
Appendix 2: Log Loading Time
Rep
Logs
Loaded Time Required (min)
1 5 4
2 4 3.25
3 5 3
4 4 2.5
5 5 2.5
6 5 3
7 4 3
8 5 4
9 5 2.5
10 4 3
11 5 3
12 6 4.5
13 4 3
14 5 4.75
15 6 4
16 5 4
17 4 2.25
18 5 3.5
19 4 2.75
20 6 4.5
21 5 3
22 6 3.5
23 5 5.5
24 6 4
25 4 3
26 5 3.5
27 6 5
28 5 3.5
29 4 3
30 6 4.5
31 5 4
32 7 4.75
33 4 3.75
34 5 4
35 4 3.5
36 5 3
37 5 3
38 6 3
39 6 3.5
79
40 6 4
41 8 4.25
42 6 3.5
43 4 3
44 6 4
45 5 4
46 5 3.75
47 5 3.5
48 4 2.5
49 5 2.5
50 4 2.5
51 5 4
52 5 3.25
53 5 4.25
54 7 4
55 5 4
56 4 2.75
57 5 2.25
58 6 4.25
59 4 4
60 5 3.5
61 5 3.5
62 5 4
63 6 3
64 6 3.5
65 6 4
Average 5.11 3.53
STDEV 0.87 0.71
80
Appendix 3: Machine Output Adjusted for Log Loading
Rep
Output/15 min
Interval (yd3)
Output/Machine
Hr (yd3)
Output/min
(yd3)
Output Adjusted for Log
Loading (yd3)
1 3.75 15 0.3 12.4
2 2.5 10 0.2 8.3
3 4 16 0.3 13.2
4 3 12 0.2 9.9
5 3.25 13 0.2 10.7
6 1 4 0.1 3.3
7 3 12 0.2 9.9
8 4 16 0.3 13.2
9 3.5 14 0.2 11.6
10 1.5 6 0.1 5.0
11 3.25 13 0.2 10.7
12 3 12 0.2 9.9
13 3.75 15 0.3 12.4
14 2.5 10 0.2 8.3
15 3.25 13 0.2 10.7
16 0 0 0.0 0.0
17 3.25 13 0.2 10.7
18 3.5 14 0.2 11.6
19 3.5 14 0.2 11.6
20 4 16 0.3 13.2
21 3 12 0.2 9.9
22 3 12 0.2 9.9
23 3 12 0.2 9.9
24 3.5 14 0.2 11.6
25 2 8 0.1 6.6
26 3.5 14 0.2 11.6
27 4 16 0.3 13.2
28 3 12 0.2 9.9
29 1.5 6 0.1 5.0
30 4 16 0.3 13.2
31 4 16 0.3 13.2
32 3.25 13 0.2 10.7
33 2.5 10 0.2 8.3
34 2.5 10 0.2 8.3
35 3 12 0.2 9.9
36 4 16 0.3 13.2
37 2.5 10 0.2 8.3
Average 3.02 12.08 0.2 10.0
SD 0.90 3.62 0.1 3.0
81
Appendix 4: Time Requirement to Load Logs onto Debarking Platform
Rep
Logs
Loaded
Time Required
(min)
1 6 3.5
2 6 3
3 5 3
4 6 4
5 6 3.75
6 7 4.5
7 6 3
8 5 3
9 6 2.75
10 6 3
11 6 4.25
12 5 3.5
13 6 4
14 5 3
15 6 3
16 6 4
17 6 3.25
18 6 3.5
19 5 3
20 5 2.5
21 6 3
22 6 3.25
23 6 4
24 5 3.75
25 5 3
26 6 3.25
27 6 4
28 6 3.5
29 6 3
30 6 4
31 5 2.5
32 5 3.5
Average 5.7 3.4
STDEV 0.5 0.5
82
Appendix 5: Time Requirement to Debark Logs Manually
Rep Logs Debarked
Time Required
(min)
1 6 7
2 6 6.5
3 5 5
4 6 6
5 6 5.75
6 7 8.25
7 6 6.5
8 5 6
9 6 9
10 6 6.75
11 6 7
12 5 6
13 6 5.75
14 5 6.25
15 6 7.75
16 6 9
17 6 10
18 6 6.5
19 5 7.75
20 5 6
21 6 10
22 6 7
23 6 5.5
24 5 6
25 5 6
26 6 6.5
27 6 8
28 6 5.5
29 6 6
30 6 6.75
31 5 7
32 5 6
Average 5.7 6.8
STDEV 0.5 1.3
83
CHAPTER 4: HEAT RECOVERY FROM COMPOSTING – A
COMPREHENSIVE HISTORICAL REVIEW OF SYSTEM DESIGN,
RECOVERY RATE AND UTILIZATION
ABSTRACT
It has long been recognized that composting yields a large quantity of thermal energy, which is
normally lost to the surrounding environment as heat. Efforts to recover this heat using compost
heat recovery systems (CHRSs) have been sporadic. Literature on the subject is also disjointed.
To summarize the research that has been conducted, the authors performed an extensive
literature review, covering publications in scientific journals, trade magazines, books, theses, and
published reports. A focus on CHRS design and heat recovery rates is presented. The review
covers 45 CHRSs in 16 different countries, ranging from simple hotbeds used in China 2000
years ago, to advanced super-thermal conductor heat pipe systems in 2016. Heat recovery rates
varied significantly, with no predictable trend among the 45 systems. Recovery rates averaged
1895 kJ/hr (1159 kJ/kg DM) for lab-scale systems, 20,035 kJ/hr (3899 kJ/kg DM) for pilot-scale
systems, and 204,907 kJ/hr (7084 kJ/kg DM) for commercial systems.
INTRODUCTION
It has long been recognized that aerobic composting liberates a great deal of thermal
energy, which is normally lost to the surrounding environment as heat. There have been many
sporadic and varied efforts to capture this heat for positive use. Some efforts have been
scientific, attempting to estimate the potential amount of energy available using compost heat
recovery systems (CHRS). Other explorations have focused on developing CHRS for
commercial or academic interest. The scale of these applications ranges from compost piles of a
few cubic meters (Brown 2014) to large in-vessel composting facilities (Irvine et al. 2010,
84
Winship et al, 2008, Tucker 2006). Many CHRS advances have been made through independent
projects, carried out by enterprising individuals seeking inexpensive energy for their homes or
farms.
The available literature on the topic is correspondingly varied. The authors reviewed over
45 publications representing scientific journals, professional trade publications, student theses,
books, published reports and articles in popular media. The body of peer-reviewed literature is
modest and does not adequately tell the story of CHRS development. To tell that story,
information must be gleaned from professional and popular publications. Fortunately, some
independent projects have been documented, although typically in a manner that is difficult to
characterize scientifically. Likewise, reports of commercial ventures often lack the information
needed to make standard comparisons between various CHRS applications. Nevertheless,
advances in CHRSs have been and continue to be made. The purpose of this paper is to present a
comprehensive status report of CHRS technology and its expected development.
COMPOSTING HEAT RECOVERY PRINCIPLES AND APPLICATIONS
Heat recovery from composting can be considered in three stages: heat production, heat
capture, and heat utilization. These stages are highly interdependent. The ultimate energy
available from a composting substrate is the same as that available from combustion of the
substrate, which can be found in existing data or analysis, such as calorimetry. However,
composting falls far short of completely oxidizing the organic compounds and liberating the
inherent energy. The actual amount of heat produced is determined by factors such as feedstock
energy content, feedstock degradability, duration of composting, and the conditions prevailing
during composting (e.g. moisture, temperature, substrate consistency and particle size).
Estimates of energy release vary with the feedstock and how investigators characterize the
85
energy. A literature review of various compost feedstocks from Adams (2005) found the average
heat production rate to be 19.44 MJ/kg dry matter (DM).
As decomposition liberates heat, the surrounding composting substrate and the air and
water within increase in temperature. In addition, some of the liquid water evaporates, creating
water vapor. Thus, the heat liberated during composting essentially takes two forms – sensible
heat (energy associated with an increase in temperature) and latent heat (energy associated with
an increase in water vapor). The total energy of a substance is characterized by its enthalpy (h) in
units of joules per gram (J/g) or kilojoules per kilogram (kJ/kg). Enthalpy accounts for the both
sensible heat and latent heat of a mass of air.
As air moves through a composting substrate, either by passive or forced convection, it
removes heat released during composting. The air gains sensible heat as it increases in
temperature and it gains latent heat as it increases in humidity. Overall its enthalpy increases.
The amount energy (Q) that air carries away depends on the change in enthalpy (Δh) and the
mass flow rate of the air (ṁ). It can be calculated by Equation 1.
Q = ṁ (Δh) = ṁ (h2 – h1) (eq. 1)
h1 = the enthalpy of the air entering the system (e.g. ambient air)
h2 = the enthalpy of the air leaving the system.
In a typical composting operating, the increase is enthalpy is predominately due to the
increase in latent heat. For example, assume that ambient air at 20ºC and 80% relative humidity
moves through a composting substrate and exits at 50ºC and 99% relative humidity. These
conditions are realistic and representative of composting conditions generally (e.g. exiting air is
commonly saturated with moisture). The enthalpy of the ambient air (h1) is about 50 kJ/kg da
and the enthalpy of the exiting air (h2) is about 272 kJ/kg da. The difference, 222 kJ/kg-da, is the
86
specific amount of composting heat removed by the air. The portion due to an increase in latent
heat is 192 kJ/kg-da, or 86%. As this example suggests, water vapor is the dominant pool of heat
available from composting.
There are three approaches used to extract heat from composting. The simplest method is
direct heat utilization of compost vapor (Fulford 1986, Aquatias 1913). Greenhouses are the
iconic use for these systems, as they can benefit from both the heat and carbon dioxide (CO2)
available in the composting exhaust air. The second approach is hydronic heating through
conduction of within-pile heat exchangers (Brown 2014, Pain and Pain 1972). These systems can
direct the heated water into a hydronic space heating system or use the heated water for a
consumptive water use (e.g. equipment cleaning). Often, a storage tank is included to buffer
temperatures and energy demand. The third and more recent approach is to capture the latent
heat using compost vapor and a condenser-type heat exchanger (Brown 2015, Smith and Aber
2014, Tucker 2006). This approach captures the greatest quantity of thermal energy and is most
commonly used by commercial composting sites.
The efficiency of the various heat recovery systems depends on the flow rate and
temperature of the heat-extracting fluid. In general, greater flow rates and lower entering fluid
temperatures capture more heat but exiting fluid temperatures tend to decrease. To a large extent,
the system flow rates and temperatures are tied to the utilization of the recovered heat. With
CHRSs that circulate water in pipes within a pile, the water flow rate is usually designed to yield
a target exiting water temperature. At a given flow rate, more heat is extracted from the pile as
the temperature of the return water decreases. Hence, the performance of the CHRS depends on
the ultimate use of the energy and its affect upon the return water. For example, a system that
consumes much of the heated water for cleaning equipment might return cold well water to the
87
composting pile and thus extract a greater amount of heat. It is possible to remove too much heat,
and consequently lower the composting temperature by circulating too much cold water (Brown
2014, Viel et al. 1987). For CHRSs that recover heat from the exhaust air, the flow rate of the air
is usually determined by the process aeration needs. The entering air temperature is determined
by ambient conditions. Again, the efficiency of the CHRS depends on how heat is recovered
downstream. The energy in the exhaust air can be wasted if a greenhouse is already warm or if
the temperature differential in a heat exchanger is small because the heat is poorly used.
HISTORY OF RECOVERING HEAT FROM COMPOSTING
While it is unknown when humans first began to utilize the heat from decomposing
organic matter, it is known that rural farmers in northern China were capturing this renewable
heat source over 2000 years ago, with the use of hotbeds (Brown 2014). This compost heat
recovery system (CHRS) was constructed by digging a 1 m trench, filling it with manure, and
covering it with a layer of topsoil for crop production. When planted above the decomposing
manure, plants benefited from the microbially-produced heat being generated below, allowing
for season extension of 1-2 months in the spring and fall.
Extracting heat from compost was further refined in France, starting in the 1600’s, where
hectares of glass-enclosed hotbeds were utilized for winter cultivation and season extension
(Fulford 1983). During this period, the most commonly used feedstock was horse manure. A
mixture of old and fresh horse manure was used to balance the heat released. Too much fresh
manure would create temperatures too hot for optimal plant growth, while too much old manure
would not produce enough heat (Aquatias 1913). The glass-enclosed French hotbeds also
required less manure than the hotbeds used in China. Aquatias (1913) described using only a
depth of 25.4 cm of compost when compacted, or 30.5 – 33 cm when using loose manure.
88
Depths beyond 33 cm created too much heat, causing leggy plants in the winter from an uneven
ratio of favorable heat to unfavorable light levels. This heat recovery method was suitable for
growing winter crops capable of handling soil temperatures below 10-13ºC. Large-scale use of
French hotbeds came to an end in the 1920's, as the horse was replaced with the automobile.
With the primary composting feedstock (horse manure) not being as abundant, large-scale use of
this CHRS disappeared (Fulford 1983).
From the 1940’s to the 1970’s a similar method of direct heat utilization from composting
rose to popularity with Dutch and English farmers. Instead of using horse manure in glass-
enclosed trenches, decomposing straw bales were used as the heat source and growing medium
(Loughton 1977). Straw bale culture involved soaking bales with a nutrient-rich liquid manure,
capping them with compost, and then planting crops on top. The heat and CO2 from the
decomposing straw were commonly used for season extension of tomatoes, cucumbers and
lettuce. Loughton (1977) reported a 21% increase in yield for spring-grown cucumbers on bales,
when compared to those grown in the soil. He also reported that wheat straw was the best
medium for heat production, while hay bales had poor results. This heat recovery method lost
favor, due to the high cost of straw (Fulford 1983).
Modern Era Begins 1971-1980
CHRSs made a significant leap forward, with the publication of Pain and Pain (1972).
This book described the work of Jean Pain and his combined heat and power composting system
in France. Prior to this work, the thermal energy from composting was primarily recovered
passively via convection of heat to the root zone of crops. Jean Pain’s system, called a Pain
mound, was very different, utilizing a 50 Mg heap of chipped brushwood with hundreds of
meters of water-filled tubing imbedded in the compost for heat exchange. The decomposing
89
brushwood warmed water within the tubing via conduction. The ability to warm water increased
the utility of the CHRS considerably, as it could be used for more than just agricultural purposes.
Pain and Pain (1972) reported that a 50 Mg heap of brushwood warmed well water from 10ºC to
60ºC at a rate of 4 liters per minute for 6 months, without interfering with the composting
process. The system supplied domestic hot water and heating to a 100 m2 farmhouse for 6
months, by circulating hot water from within the composting mass to a cast iron heater. When
looking at heat recovery, the system was capable of extracting 50,115 kJ/hr or 4330 kJ/kg DM
over a 6-month period. In addition to using compost heat for the farmstead, the authors described
using a mound with 16,800 kg of feedstock to heat a 105 m2 (211 m3) high tunnel. The authors
reported that they were capable of growing fruits and vegetables in spring-like conditions during
the winter season.
A second type of CHRS was described in Knapp (1978). The system was similar to that
of the French hotbeds, only it used a 7.6 m3 pile of decomposing leaves, household waste, lawn
clippings and manure within a greenhouse for winter heating. During the 30-day trial, it was
found that compost vapor emitting from the pile provided frost protection against killing frosts,
due to a thin layer of hoarfrost that formed on the plants. Knapp (1978) was able to grow onions,
garlic, Bibb lettuce, corn-salad, chervil, cabbage, parsley and rooted cuttings from shrubs,
despite greenhouse temperatures dropping to -7ºC four times during the testing period.
Another CHRS from this period was described by Vemmelund and Berthelsen (1979).
Their system used a small-scale double-walled 1 m3 bin to process agricultural manures. Heat
exchange occurred by filling the inner void between the bin walls with water, which was warmed
via conduction from manures being aerated within the bin. A single bin was capable of warming
water to 40ºC, with a heat recovery rate of 2304 kJ/hr. The authors suggested that a combined 4
90
bin system, with inner heat exchange walls at the cross sections, would be capable of producing
water at 50ºC, as heat exchange would occur in the inner and hottest portions of the pile. The
water could then be used directly in a radiator.
1981 – 1990
From 1981-1990, many CHRS designs evolved from conduction-based recovery systems
to those using the compost vapor stream to capture the latent heat. Replicated research and peer-
reviewed publications on the topic also started to appear in the literature. A primary reason for
the increased interest in extracting heat from composting was due to the volatile energy prices
that occurred during the previous decade, with the oil crises of 1973 and 1979. Authors of
compost heat recovery work described the need to move away from volatile energy inputs and
switch to more renewable forms (Haug 1993, Fulford 1986, Thostrup 1985, Fulford 1983).
In 1981, the Biothermal Energy Center (BTEC) formed in Portland, Maine, USA, with
the mission of developing small-scale composting greenhouses (Fulford 1983). The organization
had a series of publications, detailing how to extract heat from composting. In one of those
publications, White (1982) detailed how 0.6 - 2.2 m2 of commercial greenhouse space can be
heated with 0.9 Mg of externally located compost, if using recirculating water in EPDM heat
exchange mats. It was also reported that EPDM heat exchange mats are more efficient at
transferring compost heat than PVC, copper, or polyethylene.
Schuchardt (1984) took the same BTEC concept and applied it to a large-scale
greenhouse in Germany. The system heated a 110 m2 greenhouse by recirculating water through
2550 m of tubing contained within a 197 m3 pile of chipped brushwood. Water temperatures
within the tubing were maintained between 30-40ºC for 9 months (February-October), while the
central core of the compost pile remained at 60ºC for 20 months. The experiment was cut short,
91
due to compost settling that damaged the exchange tubing. However, over the 9-month period,
the heat recovery rate was 12,222 kJ/hr (2286 kJ/kg DM), saving 3140 L of heating oil and
producing 140 m3 of finished compost.
During the 1981 – 1990 period, small-scale lab experiments were also conducted to
determine the most effective methods of extracting heat from the composting process. Viel et al.
(1987) used a water jacket around a 100 L insulated composting reactor to recover 628-837 kJ/kg
DM. The composting mix was sewage sludge, floating foams, and poplar sawdust. One finding
was that fats were almost completely degraded (85%), indicating that they were capable of
providing much of the thermal energy release. They also found that extracting heat too early
through the exchange tubing impairs microbial activity by suppressing the temperature.
However, if heat recovery was delayed until compost temperatures reached 60ºC, microbial
activity would increase after activating the heat exchanger. Seki (1989) and Verougstraete et al.
(1985) reported a similar finding, where utilization of heat during the composting startup phase
(4-5 days) can inhibit microbial activity and reduce compost temperatures if using a within-pile
heat exchanger.
Verougstraete et al. (1985) found that within-pile piping systems limit energy recovery
potential, due to the poor heat conduction of cellulosic substances found in compost. Heat
exchange tubing is also difficult to install and fix once placed in the pile. The authors
recommended recovering heat from compost vapor, using an air-to-water heat exchanger.
Issues concerning within-compost heat exchangers were also described by Thostrup
(1985), after testing three different compost heat extraction methods in Denmark. In the first
experiment, pig manure was placed in an insulated composting chamber, surrounded by a heat
exchange water jacket. This system was not capable of producing water above 40ºC and the
92
compost could not be mixed, due to the location of the heat exchange unit. The second heat
extraction method directed exhaust vapor from a forced aeration composting system into an air-
to-water heat exchanger. This method had positive results, capable of warming water to 55ºC.
This design was applied to a pilot-scale composting plant, processing manure and bedding waste
from 200 pigs and 35 sows. The system was a tower design, where compost vapor was forced
through a chamber containing heat exchange pipes with a working fluid. Over the 21-day test
period, 5 m3 of compost was capable of producing 16,925 kJ/hr. However, the pilot plant was
mechanically complicated and was taken offline after 6 months, due to high labor costs
(Thostrup 1985).
Fulford (1986) described a viable greenhouse CHRS that successfully utilized heat from
compost vapor for winter crop production. The CHRS was a joint effort between BTEC and the
New Alchemy Institute. The system used a polyethylene-covered greenhouse with a 19 m3
composting chamber attached to the north end of the structure. The composting chamber
contained 10 separate bins, which were aerated by electric blowers that pushed air through the
compost. The heated exhaust was blown through perforated pipe below growing beds, which
served as biofilters. Heat transfer was direct utilization of the exhaust vapor from the composting
chamber to the root zone of crops growing above the filter media. Heat was transferred to the
media through stored latent heat (Fulford 1986). To ensure a constant supply of hot air, fresh
compost was loaded into two composting bins every 4-5 days. When compost temperatures
averaged 54ºC, the upper growing bed was maintained at 24-27ºC, while the temperature of the
lower growing bed was maintained at 16ºC. The compost-heated growing beds proved to be ideal
for starting new seedlings.
93
In addition to supplying heat for growing crops, Fulford (1986) reported additional
benefits of the compost vapor, which included irrigation, supplemental CO2, and nitrogen
fertilizer from ammonia (NH3) that was converted to ammonium (NH4+) and nitrate (NO3
-) in the
biofilter media. However, the fertilization benefits of the NH3 in the vapor stream were limited.
A follow-up report from Schonbeck (1989) described the problem of excessive NH3
accumulation in the soil. Excess nitrogen, combined with low light levels during the winter, also
resulted in unhealthy NO3- accumulation in the leafy vegetables. A peat moss biofilter installed in
the composting chamber was used to fix the problem, scrubbing over 90% of the NH3 from the
vapor stream prior to being sent through the growing beds.
1991-2000
While the heat recovery systems of this decade expanded on past technologies, the
increased activity of peer-reviewed literature helped confirm many of the findings reported from
independent organizations and individuals. Beck et al. (1992) described a CHRS that used the
Carboferm ® process, where liquid from an odor-removing jet washer was used for heat and
nutrient recovery. The system operated by pulling air through compost in a concrete bunker, and
sending it into a scrubber, where the condensing compost vapor exchanged thermal energy and
nutrients to the process liquid. A heat exchanger was used to extract the thermal energy from the
liquid, which was later used as a nitrogen fertilizer source for a greenhouse. The authors reported
a 20% increase in crop production and a 20% savings in energy from using the stored latent heat
at night.
A second study from Jaccard et al. (1993) reported the findings from a pilot-scale CHRS
in Switzerland. The composting plant was a continuous-fed reactor, where 0.5 m3 of yard waste
was loaded and unloaded into the reactor by two tangent screws. The yard waste was aerated
94
with a blower from the top down. Heat recovery occurred through an air-water heat pump.
Composting and heat recovery took place for only 5 days in the reactor, taking advantage of the
highest temperature thermophilic stage. Heat recovery had a power of 1948 kJ/hr, with 81%
coming from latent heat, 8% sensible heat and 12% sensible heat in the extract material. The
authors projected that doubling the reactor size from 0.5 m3 to 1.0 m3 would double the recovery
rate to 3600 kJ/hr, while a 6 m3 reactor, with a feeding rate of 10 kg/hr, would be capable of
providing enough hot water for 4 middle-sized homes.
Over the course of this decade, peer-reviewed research verified the feasibility and utility
of using compost heat directly in growing beds, supporting the findings in the non-academic
literature from previous decades. Hong et al. (1997) conducted a study looking at the effects of
composting on underground soil temperatures within a 55 m2 greenhouse in Korea. The study
was conducted by placing a mixture of cattle manure and rice hulls in three long trenches (60 cm
H * 60 cm W * 8 m L), with plants growing in parallel rows. A 4 cm diameter pipe below each
composting trench supplied forced air to speed up the decomposition process. Researchers found
that heat from the compost greatly affected the temperature of the adjacent growing beds. During
the months of January and February, underground soil temperatures in the greenhouse were
maintained between 17.5 - 32.5ºC, while outside underground temperatures were 6 - 11.9ºC. The
authors concluded that the CHRS was suitable for winter cultivation in a greenhouse. A similar
study from Kostov et al. (1995) compared the growth and production of cucumbers in a 142 m2
greenhouse in Bulgaria, using direct utilization of compost heat. Cucumbers were grown in
trenches containing either a mixture of cattle manure and soil (control) or a nutrient-
supplemented compost mixture of vine branches, flax residues, or grape prunings, husks and
seeds. During the study period, root zone temperatures of cucumbers growing over the compost
95
were higher than the control, reaching 29.6ºC. Compost treatments also had higher CO2 and
microbial biomass than treatments with just manure. Treatments with compost had fruit
production 10-12 days earlier and had a yield 48-79% greater than the control. Compost
treatments also produced 40-44% more profit than the control. The increased yield and profit
was attributed to warmer soils from the compost, elevated CO2, larger microbial biomass, and
the supplemental nutrients added to the compost media.
In addition to pilot-scale experiments, Seki and Komori (1992, 1995) tested CHRS
designs at the lab level. In Seki and Komori (1995), a small cylindrical compost container
containing both a within-pile and vapor condenser heat exchanger were used to recover 16-22%
of the heat generated during the composting of a chicken manure, rice bran and sawdust mixture.
Over the 7-14 day period, 714 – 773 kJ/hr were recovered. However, the energy to run the
composting system was higher than the energy recovered, due to poor insulation of the
composing chamber, and the large pressure drop of the passing fluid through the heat exchanger.
The authors suggested that more insulation and larger heat exchange tubes would solve the issue.
A brief article in BioCycle magazine described an innovative system developed by an
organic lawn care company in Omaha, Nebraska, USA (Anon 1991). This CHRS combined a
geothermal, solar, and composting system to heat water. The integrated system began by
warming city water from 8.3ºC to 12.7ºC through a subterranean geothermal tank. The tempered
water was then sent to a second water tank, which was heated another 11ºC with a solar system.
The warm water was then sent through tubing within a composting pile of leaf and yard waste,
where it was heated another 10ºC to a final temperature of 34ºC. During the winter, when
outside temperatures dropped to -27ºC and a wind chill factor of -44ºC, the water temperature of
the integrated system remained at 33ºC. This system corrected one of the flaws of within-pile
96
CHRSs, where cold water circulating through the composting pile can reduce microbial activity
and future compost temperatures. It was reported that the integrated system worked well in the
winter, but only required the solar heater during the summer months.
A final CHRS reported during this period came from Greer and Diver (2000). In their
organic vegetable production guide, they described how to recover heat from decomposing straw
bales. They reported that 26.5 L of manure tea are required per 23 kg straw bale for optimal heat
production. Following saturation, bales are composted for a short period and are capped with 15
cm of compost-based potting mix once temperatures drop below 43ºC. At this time, root balls of
transplants are planted on top of the decomposing bale, which benefit from CO2 and heat release.
2001 – 2010
The period between 2000-2010 saw an expansion in compost heat recovery technologies,
as small-scale and pilot demonstration projects were scaled up to the commercial level. A
majority of the findings were reported in a combination of student theses and practitioner-
oriented sources, with few peer-reviewed publications.
Of the three theses in this review, all focused on the feasibility of using a CHRS to
support winter greenhouse production. Adams (2005) modeled the effects of a CHRS for winter
heating in Vermont, USA, using compost thermal data from peer-reviewed journals, local
weather data, and operational data from a local composting facility. His findings suggested that if
heating a greenhouse with an internally located compost pile, and using the compost vapor
directly, 27% of the greenhouse floor space would be needed for the compost pile, and up to
50% if accounting for compost handling and storage. A second thesis from Gilson (2009) looked
at the feasibility of year-round crop production using a CHRS in Ontario Province, Canada. The
author proposed using a within-pile heat exchanger made of coiled pipe that would circulate hot
97
water between an externally located compost pile and cast-iron radiator panels located within a
greenhouse. Cast-iron radiators were recommended, due to their inexpensive cost and the ability
to find them at recycling facilities. The author also suggested the benefits of not having to scrub
volatile organic compounds (VOCs) and NH3 from the compost vapor stream, which is necessary
when using a direct-vapor CHRS. A third thesis from Chambers (2009) described a CHRS using
a 435 kg mixture of horse manure, sawdust and woodchips to heat a 2046 L insulated chamber.
Heat recovery occurred via a within-pile heat exchanger made from two arrays of PEX piping,
which were connected to a radiator panel inside a winter high tunnel. The CHRS increased high
tunnel temperatures 2-3ºC above the unheated control, recovering 451 kJ/hr (1584 kJ/kg DM)
over the 25-day test period.
One of the first large-scale commercial CHRSs was described in Tucker (2006). The
system at Diamond Hill Custom Heifers in Vermont, USA, processes 544 - 726 Mg of
agricultural wastes at any one time, using the aerated static pile (ASP) composting method. A
suction fan pulls air through the compost into a network of pipes and into Agrilab Technologies
heat exchange system. The system uses Isobar™ stainless steel super-thermal conductor heat
pipes, which transfer the latent heat from the compost vapor to an 800 gallon tank of water.
Heated water is used for warming milk formula and to provide radiant floor heating in the calf
barn. Heat recovery rates of up to 211,011 kJ/hr were reported. Total project costs for the
composting facility, storage barn, compost mixer and Agrilab’s heat exchange unit, were under
$500,000 (Smith 2016).
A second commercial-sized CHRS was described in Allain (2007). This CHRS was
designed to prevent snow and ice from freezing GORE™ compost covers to the ground at a
biosolids composting facility in New Brunswick, Canada. The system contained a network of
98
pipes cast into the concrete composting pad, with a recirculating water/glycol solution. Central
pipes under the middle of the compost windrows warmed the heat exchange fluid, which was
then pumped to the outer pipes at the edges of the pile, warming the pad and preventing the
GORE™ covers from freezing to the ground. The author found that if the heated fluid was
pumped too frequently, compost pile temperatures decreased. However, a pump interval with 4-6
hrs of down time per cycle allowed the heat exchange fluid to reach maximum temperature,
while not affecting compost pile temperatures, due to heat exchange. A heat recovery rate of
16,353 – 23,000 kJ/hr for a 6 hr and 4 hr time off interval were reported, when using 11,000 Mg
of compost at 60ºC.
During this period, researchers also modeled how much heat could be extracted from
commercial-sized CHRSs. In Winship et al. (2008) a transportable in-vessel composting
container designed by Alpheco Composting in the UK was used to model heat production and
recovery. The vessels, called Aergestors, had a stainless steel interior with a built in aeration
floor and irrigation system. Through a network of supply and exhaust manifold pipes, up to 10
vessels with a capacity of 15 Mg each could be grouped together with a single Aerator. Heat
exchange would occur in the moisture trap of the Aerator, where a heat pump would be located.
The heated water could then be tied into any building’s hot water circuit. A second study from
Irvine et al. (2009) used a commercial in-vessel composting operation in Scotland to model how
much heat could be produced and captured from a CHRS. The proposed heat exchanger would
be a network of water-filled stainless steel pipes hanging above the airspace in the composting
vessel. Modeling suggested that 7000-10,000 kJ/kg DM could be obtained from the system over
a 15-day period, warming water to 47.3 - 60ºC at a cost of $0.78/kWh for domestic hot water and
$0.16/kWh for space heating. A third study from Di Maria et al. (2008) used a model to simulate
99
heat recovery from a proposed commercial ASP composting facility in Italy. The recovery
system was a vapor compression heat pump that had heat transfer coils within the composting
mass. The authors reported that the 55-65ºC compost temperatures could be increased to 80-90ºC
with a heat pump, recovering 4000-5000 kJ/kg of thermal energy from the compost.
2011-2016
Since 2010, there has been significant growth in small, medium and large-scale CHRS.
Of the small-scale systems, Li et al. (2012) looked at developing a prototype CHRS that could be
installed in US. households, utilizing waste from only the home for heating needs. The system
was tested using a lab reactor with a mixture of grass clippings, sludge, leaves, and sawdust. The
composting system used two chambers. One chamber contained 92.4 kg of feedstock, while the
second contained a water tank and an air-to-water heat exchanger. A net energy generation of
4845 kJ/hr was recovered.
Gnanaraj (2012) used a small-scale lab reactor to test a CHRS in India, using 39.6 kg of
sugarcane waste. A copper plate connected to a heat pipe transferred thermal energy from within
the composting mass to an external heat exchanger with a water jacket for the heat sink. An
average heat recovery rate of 1080 kJ/kg over 42 days was reported.
A third small-scale lab experiment was conducted by Seki et al. (2014), to examine the
ability of a CHRS to warm a fishpond for use in rural regions of Japan. The heat recovery system
contained three separate containers, with the first being a 0.157 m3 composting chamber of
bamboo chips. A 0.9 m stainless steel water loop was imbedded in the compost to extract heat
through conduction, and was connected to a 0.0156 m3 water reservoir through a closed loop.
Warm water from the reservoir was circulated through a third container (0.0156 m3) serving as a
fishpond through a 2 m loop of copper pipe. Results from this experiment were input into a
100
model to test the capability of a 50 m3 bamboo chip pile to heat a 5 m3 fishpond. Model
simulations suggested that the CHRS could extract heat for 42 days, and maintain the fishpond at
20ºC.
In addition to small-scale lab experiments, several pilot-scale and mid-scale CHRSs were
reported in the literature, with many of them being modified Jean Pain mounds. Two
organizations, one American (Compost Power Network - CPN) and the other German (Native
Power), led the way in developing these CHRSs. Both organizations provide workshops and
technical details on how to replicate these heat recovery systems. A book titled The Compost-
Powered Water Heater from CPN founder, Gaelan Brown, details how to replicate CHRSs,
including their modified Jean Pain mound, which is capable of recovering 10,550 kJ/hr (5081 –
7500 kJ/kg) from a 31 m3 heap of woodchips over a 12 to 18-month period (Brown 2014).
A second type of mid-scale CHRS was described in Alwell (2014). This system
contained 4 composting bins, with a combined feedstock capacity of 5 m3. Bins were made of
dimensional lumber, foam board insulation, and PVC pipe. Each bin was aerated from below
using a blower, and was connected to a 1041 L tote of water, which served as a heat sink for two
separate heat exchangers. The first heat exchanger was a stainless steel tube within the compost
pile, while the second was an array of copper pipes located within the exhaust line of the system.
Heated water was sent to growing beds within a high tunnel through an underground PEX pipe.
Over a 5-day period, a 28 m2 growing bed within a high tunnel was warmed 3ºC, saving 20 L of
propane (4023 kJ/hr). The cost of the CHRS was $7000, with a majority coming from the 140 W
solar panel used to run the pumps and blower.
Another mid-scale CHRS was described in Brown (2014). The system was installed on a
dairy farm in Vermont, USA, to process agricultural wastes. The system used a rectangular
101
insulated composting box (3.7 m * 3 m * 12.8 m) built over a concrete aeration floor. Heat was
recovered by blowing air through the compost and up into the ceiling, where 366 m of
polyethylene tubing were attached. Water was circulated between the tubing and a radiator in an
adjacent 149 m2 workshop. The system captured 21,101 – 31,652 kJ/hr continuously.
In addition to small and mid-scale CHRSs, there was continued reporting on commercial-
sized systems in the literature. Rada et al. (2014) described a proposed large-scale CHRS that
would be designed to dry sewage sludge with heat generated from an adjacent food and green
waste composting facility. The goal would be to reduce the volume and weight of the sludge and
increase the ease of storage before further processing. The proposed composting plant would be
a collection of concrete bio-cells with an aeration floor cast into the concrete, and a stainless
steel heat exchange tube in the upper portion of each unit. Exhaust gas would warm water in the
heat exchanger through conduction and convection. The heated water would then be circulated
through tubing in the pad below the sludge, promoting evaporation. Modeling suggested that a
composting plant with six biocells and an annual feedstock capacity of 15,000 Mg would have a
heat recovery power of 153,000 kJ/hr.
Day (2014) described an active commercial-scale CHRS at the Hawk Ridge composting
facility in Maine, USA, which processes over 34,405 m3 of municipal solid waste per year. The
facility, which is operated by Casella Organics, uses a geothermal heat pump and a CHRS in a
hybrid geo/biothermal system. During winter, hot compost vapor passing through the facility’s
odor scrubber warms process water to 43ºC. The heated water is then sent through underground
piping to warm a maintenance shop through radiant floor heating. It also warms an office
building by warming the soil around an existing geothermal heat pump. System cost was
$40,000, with an annual energy savings of $10,000 per year.
102
During this period Triea Technologies™ also began selling and refining their
commercial-scale CHRS called the BioMASS HRS™. The system is marketed for the poultry
industry, where the heat is used to warm chicken houses. Initially, their test facility in West
Virginia, USA, extracted heat from a closed water loop between a heat exchanger in the compost
and a heat pump. The heated fluid was then sent to a fluid-to-air heat exchanger in another closed
loop, warming fresh air to 37ºC for the poultry house. Energy recovery rates of 153,862 kJ/hr
(4294 kJ/kg DM) over a 50-day period were reported. However, the company converted their
CHRS from a within-pile conduction system to an exhaust vapor condenser to increase heat
recovery. The new system has a peak recovery output of 395,646 kJ/hr (11,041 kJ/kg DM), when
composting 184 m3 of poultry manure and bedding over a 50-day period (Triea pers. comm.
2015).
Significant expansion and advances were also made in the commercial-scale CHRSs sold
by Agrilab Technologies, with four systems becoming operational during this time period. As
with their first system described earlier in Tucker (2006), all four CHRSs use the ASP
composting method with one of Agrilab’s vapor condensing systems. The first system from this
time period was installed at Sunset View Farm in New York, USA, becoming fully operational
in early 2011. The farm raises 2000 heifers and composts cow manure and bedding. Water
temperature in the heat sink tank is maintained at 46ºC, representing an average heat recovery
rate of 205,736 kJ/hr and a farm energy savings of $9285 (Quinn et al. 2014).
In 2012 Agrilab installed a smaller-scale CHRS at Jasper Hill Farm in Vermont, USA.
Heat recovery comes from composting manure solids and bedding from the 45-head cow barn,
while liquid manure, whey from their cheese making facility and wash water are used in the
anaerobic digester. The heat from the CHRS is used to warm three 26,500 L anaerobic digester
103
tanks and maintain temperatures at 38ºC (Smith 2016). The 27-32ºC compost exhaust vapor post
heat exchange is used in their winter greenhouse beds, which serve as a biofilter. They are
growing tomatoes, peppers, greens, strawberries, pineapples and a banana tree (Brown 2014).
In 2013, a third Agrilab CHRS was installed at the University of New Hampshire’s
Organic Dairy Research Farm in New Hampshire, USA. The system, described in Smith and
Aber (2014), was designed specifically for research trials on compost heat production, recovery,
and utilization. The CHRS is similar in size to Jasper Hill’s, processing manure and bedding
waste from a 100-head dairy herd. However, the aeration lines on the main composting floor
were designed in pairs, allowing for replicated research on compost batches. The heated water is
also used in the milk room for cleaning and sanitizing equipment. A detailed analysis outlining
the cost structure and how to replicate the system for under $300,000 is described in Smith
(2016).
In 2015, a fourth Agrilab system was installed in the urban environment of Boston, MA,
USA. The composting site, operated by City Soil & Greenhouse LLC, uses a mobile Agrilab
system (The Compost Heat Wagon ™) contained within a portable trailer. The portable unit
serves as the heat recovery an aeration source for a 260 m2 composting greenhouse. The
composting greenhouse has a 191 m3 capacity composting floor with an integrated biofilter and
growing bed system. Heated water from Agrilab’s system is used for radiant heating in the
growing beds, allowing for year-round crop production. Heat recovery rates of 63,300 kJ/hr have
been recognized, with a system capability of 295,415 kJ/hr.
CONCLUDING THOUGHTS
This review covered 45 different CHRSs in 16 different countries, from simple systems
used by Chinese farmers 2000 years ago, to advanced systems using super-thermal conductor
104
heat pipes in 2016. A wide variation in project scale was presented, with 11 lab scale, 19 pilot
scale and 15 commercial scale systems described. The 11 lab-scale CHRSs were all reported in
peer-reviewed literature. This was in contrast to the literature from the nine commercial-scale
systems using operational data, which were all published in practitioner-based sources or
conference proceedings. The lack of peer-reviewed literature using actual and not modeled data
for commercial systems represents a significant research gap that needs to be filled.
CHRSs fit into four broad categories, based on how the thermal energy was extracted.
The categories include: direct heat utilization of compost vapor (9 systems), hydronic heating
through conduction of within-pile heat exchangers (17 systems), compost vapor exchange
through latent heat (17 systems), and a combination of several technologies (2 systems). While
direct vapor and within-pile CHRSs are still being used today, academic and practitioner-based
literature indicated preference toward systems extracting heat from compost vapor using
condenser-type heat exchangers. Reasons provided for using compost vapor heat exchangers
over within-pile exchangers were:
1) Within pile heat exchange tubing can be easily damaged, due to compression from
compost settling, loading, or unloading (Schuchardt 1984, Pain and Pain 1972).
2) Recirculating the within-pile heat exchange liquid too early or too fast can inhibit the
composting process, reducing temperatures and future heat exchange (Allain 2007, Seki
1989, Viel et al. 1987, Verougstraete et al. 1985).
3) Compost cannot be mixed once the heat exchanger is placed within the pile (Thostrup
1985).
4) Energy recovery is limited due to poor heat conduction characteristics of composting
feedstocks (Verougstraete et al. 1985).
105
5) Unlike the within-pile approach, vapor heat exchangers can potentially capture the
abundant latent heat energy produced during composting (Jaccard et al. 1993,
Verougstraete et al. 1985).
Preference for vapor exchange was also shown by active commercial sites covered in this
review, with 87% of the facilities using that form of heat exchange.
When looking at the heat recovery capabilities of the 45 CHRSs, no predictable trend of
heat recovery by system type or scale was apparent. Recovery rates were 1895 kJ/hr (s = 1609
kJ/hr) for lab-scale systems, 20,035 kJ/hr (s = 16,505 kJ/hr) for pilot-scale systems, and 204,907
kJ/hr (s = 118,477 kJ/hr) for commercial-scale systems. On an energy per weight basis, recovery
rates were 1159 kJ/kg DM (s = 602 kJ/kg DM) for lab-scale systems, 4302 kJ/kg DM (s = 2003
kJ/kg) for pilot-scale systems, and 7084 kJ/kg DM (s = 3272 kJ/kg DM) for commercial-scale
systems. Heat recovery rates varied significantly and were dependent on a combination of the
following factors: system scale, type of heat exchange system, composting method, composting
feedstocks, continuous vs. batch loading, model vs. operational data, geographic location,
duration of heat recovery, and method of reporting thermal energy recovery. An attempt to
standardize the reported recovery values to a single comparable unit of heat recovery per unit dry
biomass was attempted, but proved to be impossible. A majority of the literature from
practitioner-based sources reported values as BTU/hr, and did not provide information on how
much biomass was used to generate the energy recovery values. If information on biomass
quantity was reported, moisture content values were absent, making it impossible to calculate dry
biomass without making assumptions. Complicating the standardization process further,
commercial systems like those from Agrilab Technologies, reported recovery rates from active
facilities using batch loading, where various ages of compost contributed to the recovery value.
106
As a consequence, the 211,011 kJ/hr value reported in Tucker (2006), underestimates the
maximum recovery from the system, as several compost batches in the facility were older and
had passed the high-heat active phase. This method of reporting also made it difficult to convert
to a heat recovery per unit weight value, as the value would include large quantities of biomass
not contributing significant quantities of heat.
For CHRS data to be more comparable, recovery rates should be presented as heat
recovery per unit time and as specific energy, describing energy per unit mass (kJ/kg DM) or per
unit of volatile solids (kJ/kg VS). It is also necessary to report whether average or peak recovery
values are being used and for the duration those values represent. This is especially important
due to the changing temperatures found within a composting mass over time. If energy recovery
is only reported during the thermophilic phase, recovery rates will be much higher than those
reporting recovery rates over a several week or month period. By way of example, Jaccard et al.
(1993) reported heat recovery over a 5-day period, while Schuchardt (1984) reported heat
recovery over a 9-month period. Ideally, multiple heat recovery values would be reported,
representing the various stages of the composting process. Importantly, this would provide
compost practitioners more useful information on how much thermal energy they may be able to
extract on their site based at their composting period.
While the 45 CHRSs described in this review show tremendous variability in heat
recovery, design, and efficiency, the authors hope that the breadth of systems covered will help
compost practitioners decide what type of system is best for their site. The significant increase in
practitioner-based publications on CHRSs from 2010-2016 also suggests an interest in this field
of composting, especially with commercial-scale operations. This represents a distinct shift from
pre-2010, which was dominated with prototype and lab-based systems, indicating that recovering
107
heat from composting is becoming a viable alternative energy source and is that much closer to
becoming a mainstream process.
108
REFERENCES
Adams, Zak. 2005. “Understanding Biothermal Energy.” Master’s thesis., University of
Vermont, Burlington.
Allain, Conrad. 2007. “Energy Recovery at Biosolids Composting Facility.” BioCycle 48:50-53.
Alwell, Audrey. 2014. “Innovative System uses Composting Process to Heat High Tunnel.”
Organic Broadcaster, July.
Anon. 1991. “Compost Preheats Water.” BioCycle 32:20.
Aquatias, P. 1913. Intensive Cultivation of Vegetables on the French System. London: Upcott
Gill.
Beck, D., Konrad Soyez., M. Prause., and K. Fieback. 1992. “A Combined Process for Fast
Composting, Biotechnological CO2 Production, and Utilization of Heat Loss for
Horticulture.” Acta Hortic 302:257-277.
Brown, Gaelan. 2015. “Heat Recovery, Food Production at Boston Composting Facility.”
BioCycle 56:39-40.
Brown, Gaelan. 2014. The Compost-Powered Water Heater. Woodstock: The Countryman Press.
Chambers, Donal P. 2009. “The Design and Development of Heat Extraction Technologies for
the Utilization of Compost Thermal Energy.” Master’s thesis., Galway Mayo Institute of
Technology, Galway.
Day, Doug. 2014. “Reusing More, Using Less.” Treatment Plant Operator, October: 42-44.
Di Maria, Francesco., Manuel Benavoli., and Mirco Zoppitelli. 2008. “Thermodynamic Analysis
of the Energy Recovery from the Aerobic Bioconversion of Solid Urban Waste Organic
Fraction.” Waste Manage 28:805-812.
109
Fulford, Bruce. 1983. “Biothermal Energy: Cogenerants of Thermophylic Composting and their
Integration within Food Producing and Waste Recycling Systems.” In Composting of
Solid Wastes and Slurries, edited by E. Stentiford. Leeds, England: University of Leeds.
Fulford, Bruce. 1986. The Composting Greenhouse at New Alchemy Institute: A Report on Two
Years of Operation and Monitoring. Research Report 3. Cape Code, MA: New Alchemy
Institute.
Gilson, Courtney. 2009. “Designing a Compost-Heated Greenhouse to Foster Sustainable Food
Security.” Undergraduate Thesis., The University of Waterloo, Waterloo.
Gnanaraj, R. Arul. 2012. “Energy Recovery from Sugarcane Press Mud.” Asian J. Sci. Res
5:185-195.
Greer, Lane., and Steve Diver. 2000. Organic Greenhouse Vegetable Production. Research
Report IPO78. Butte, MT: NCAT.
Haug, Roger T. 1993. The Practical Handbook of Compost Engineering. Boca Raton: CRC
Press.
Hong, J.H., K.J. Park., and B.K. Sohn. 1997. “Effect of Composting Heat from Intermittent
Aerated Static Pile on the Elevation of Underground Temperature.” Appl Eng Agric
13:679-683.
Irvine, G., E.R. Lamont., and B. Antizar-Ladislao. 2010. “Energy from Waste: Reuse of
Compost Heat as a Source of Renewable Energy.” Int. J. Chem. Eng 2010:1–10.
Jaccard, Leon., Pierre Lehmann., Marcello Civilini., and Marco de Bertoldi. 1993. “Yard Waste
Composting with Heat Recovery.” Compost Sci. Util 1:10-14.
Knapp, Dan. 1978. “Composting in a Solar Greenhouse for CO2 and Heat.” In The Solar
Greenhouse Book, edited by James C. McCullagh, 287-292. Emmarus: Rodale Press.
Kostov, Ognian., Y. Tzvetkov., Nadia. Kaloianova., and Oswald V. Cleemput. 1995. “Cucumber
Cultivation on Some Wastes during their Aerobic Composting.” Bioresource Technol
53:237-242.
110
Li, Haorong., Daihong Yu., and Yanshun Yu. 2012. “A Preliminary Study of an Innovative
Biomass Waste Aerobic Degradation System for Hot Water Heating.” In Material
Challenges in Alternative and Renewable Energy II: Ceramic Transactions Volume 239,
edited by George Wicks, Jack Simon, Ragaiy Zidan, Robin Brigmon, Gary S. Fischman,
Sivaram Arepalli, Ann Norris, and Megan McCluer, 69-77. Hoboken, NJ: John Wiley &
Sons.
Loughton, Arthur. 1977. “Straw-Bale Culture of Greenhouse Crops.” In Controlled Environment
Agriculture, edited by Merle H. Jensen, 208-215. Tuscon, AZ: University of Arizona.
Pain, Ida., and Jean Pain. 1972. The Methods of Jean Pain: Another Kind of Garden.
Draguignan: Ancienne Imprimerie NEGRO.
Quinn, Sean., Sandy Quinn, and Brian Jerose. 2014. Sunset View Farm Composting and Thermal
Energy System. NYSERDA Project # 10819 Final Report. New York: Agrilab
Technologies.
Rada, Elena C., Marco Ragazzi., Stefano Villotti., and Vincenzo Torretta. 2014. “Sewage Sludge
Drying by Energy Recovery from OFMSW Composting: Preliminary Feasibility
Evaluation.” Waste Manage 34:859-866.
Schonbeck, Mark. 1989. “Composting Greenhouse Update.” New Alchemy Quarterly 36:16-17.
Schuchardt, Frank. 1984. “Heat Loss during Composting of Sawtimber.” Landbauforschung
Völkenrode 34:189-195.
Seki, Hirakazu., and Tomoaki Komori. 1995. “Experiment of Heat Recovery from Compost by a
Trial Heat Exchanger.” Acta Hort 399:167-174.
Seki, Hirakazu. 1989. “An Investigation of Practical Process Design and Control of a Soil
Warming System with Heat Generated in Compost.” J. Agri. Met 44:259-267.
Seki, Hirakazu., and Tomoaki Komori. 1992. “Packed-Column-Type Heating Tower for
Recovery of Heat Generated in Compost.” J. Agri. Met 48: 237-246.
111
Seki, Hirakazu., Shiro Kiyose., and Shoko Sakida. 2014. “An Experimental System for the
Recovery, Accumulation, and Utilization of Heat Generated by Bamboo Chip
Biodegradation using a Small Scale Apparatus.” J. Agri. Met 70:1-11.
Smith, Matthew M. 2016. “Creating an Economically Viable, Closed-System, Energy-
Independent Dairy Farm through the On-Farm Production of Animal Bedding and Heat
Capture from an Aerated Static Pile Heat Recovery Composting Operation.” PhD diss.,
University of New Hampshire, Durham. (In Review)
Smith, Matthew M., and John D Aber. 2014. “Heat Recovery from Compost.” BioCycle 55:27-
29.
Thostrup, P. 1985. “Heat Recovery from Composting Solid Manure.” In Composting of
Agricultural and Other Wastes, edited by J.R.K. Gasser, 167-180. London: Elsevier
Applied Science Publishers.
Tucker, Molly F. 2006. “Extracting Thermal Energy from Composting.” BioCycle 47:38.
Vemmelund, Niels., and Leif Berthelsen. 1979. “A Note on Heat Recovery from Mechanically
Aerated Farm-Yard Manure.” Agric. Waste 1:157-160.
Verougstraete, A., E.J. Nyns., and H.P. Naveau. 1985. “Heat Recovery from Composting and
Comparison with Energy from Anaerobic Digestion.” In Composting of Agricultural and
Other Wastes, edited by J.R.K. Gasser, 135-146. London: Elsevier Applied Science
Publishers.
Viel, Maurice., Daniel Sayag., Annie Peyre., and Louis Andre. 1987. “Optimizat ion of In-Vessel
Co-Composting through Heat Recovery.” Biol Waste 20:167-185.
White, John W. 1982. Letters to the Editor. BioThermal Update Report 1. Portland, ME:
BioThermal Energy Center.
Winship, E.A.N., Holmes, D., and Notion, D. 2008. “Combined Heat and Composting.” In
Moving Organic Waste Recycling Toward Resource Management and Biobased
Economy; Orbit 6th International Conference, edited by Ljiljana Rodic-Wiersma, Josef
112
Barth, Ing. habil. Werner Bidlingmaier, Marco de Bertoldi, and Luis F. Diaz, 1451-1463.
Wageningen, Netherlands: Orbit.
113
CHAPTER 5: HEAT RECOVERY FROM COMPOSTING: A STEP-BY-
STEP GUIDE ON BUILDING AN AERATED STATIC PILE HEAT
RECOVERY COMPOSTING FACILITY
EXECUTIVE SUMMARY
The focus of this report is to describe the process of building a heat recovery composting
facility using the aerated static pile (ASP) method and Agrilab Technologies Heat Transfer
System. The heat recovery composting facility constructed at the University of New Hampshire
(UNH) will serve as a case study. The report begins with a technology review, followed by
detailed information on facility design, specific materials used, cost, and cost-saving
strategies/considerations for those wanting to install this type of system at their site. While the
facility was built on a university organic dairy farm to process agricultural wastes, a majority of
the structural designs, materials list, and cost-saving strategies will be the same for any farmer
(dairy, equine, poultry, etc.) or compost operator wanting to build this type of facility on their
site. More specifically, this type of system is suitable for any type of waste being composted
aerobically, whether animal, biosolid, digestate, municipal, yard, food, etc.
The ultimate goal of this report is to provide enough detailed information that compost
operators could design their own ASP composting system, reducing the amount of time and
money that would otherwise be spent on engineering and consulting costs. The step-by-step
instructions from the planning phase through project completion, along with the materials and
cost list (Appendix 1) also provide guidance to operators on how to construct and purchase large
portions of the system themselves, leading to substantial cost savings. This report can also be
used to answer many technology and cost questions that are pertinent to policy makers and
investors who may be considering supporting this type of venture. The reader is encouraged to
114
reference this report to hopefully expedite the often timely design and funding phases that exist
for these types of projects.
TECHNOLOGY OVERVIEW
Aerobic Heat Production vs. Anaerobic Biogas Production
An important point to make is that this technology involves composting, where heat, not
biogas (methane - CH4), is being captured and utilized. In this type of aerobic system, CH4
production from lack of oxygen (anaerobic) is an economic loss, representing material that could
have generated heat for capture through composting. Because anaerobic microbes are less
metabolically efficient, there is only a partial breakdown of the material, with a tremendous
amount of chemical energy being left in the bonds of the CH4 compound. Unless the end goal is
anaerobic digestion with biogas production/capture, this situation poses significant problems for
compost operators wanting to explore heat extraction technologies. They include:
1. Reduced compost quality – many intermediate compounds that remain from only
a partial breakdown of the biomass, are phytotoxic to plants (Epstein 2011, Misra
et al. 2003). This is especially true for fatty acids.
2. Foul odors – originate from intermediate compounds and some end products
[volatile fatty acids, ammonia (NH3), and hydrogen sulfide (H2S)] (Chiumenti et
al. 2005, Wright 2001, Rynk et al. 1992)
3. Corrosion - H2S and fatty acids (Chen et al. 2010).
4. Elevated greenhouse gas emissions - CH4 is 21 times more potent as a greenhouse
gas than carbon dioxide (CO2) (US EPA 2013).
5. Reduction in compost sterilization – inefficient low heat composting will not
destroy weed seeds and pathogens within the composting mix (Misra et al. 2003).
115
6. Heat exchange suffers – more energy leaving the system as CH4 results in less
heat production for recovery.
Heat Production from Composting
The bio-oxidation of organic material that occurs during composting is an exothermic
reaction that continually releases heat, and can be represented by the general equation in
Figure 7.
Figure 7: Basic Formula for Aerobic Composting
In a compost pile, temperatures will go from ambient mesophilic thermophilic
mesophilic ambient (Epstein 2011). While the exact range for what is qualified as mesophilic
or thermophilic varies in the literature, a general range is 50-110ºF for mesophilic and > 110ºF
for thermophilic. Following pile formation, temperatures will often increase sharply, and reach
thermophilic temperatures within 48 hours (Chiumenti et al. 2005) (Figure 8).
Figure 8: Compost Temperature over Compost Age for UNH Experimental Batch 2
0
20
40
60
80
100
120
140
160
180
1 2 7 14 16 22 29 36 42 48 55 62
Tem
per
atu
re (F
)
Compost Age (Days)
Compost Temperature over Compost Age for UNH Experimental Batch 2
Bottom of Pile
Middle of Pile
Surface
116
If heat is not removed, temperatures will increase to the point where the microbes start
dying off (> 160ºF). In this thermophilic stage, oxygen demand and heat production are highest,
as the microbes target and metabolize the most easily digestible materials first (starches, sugars
and fats) (Epstein 2011, Rynk et al. 1992). During this stage, the amount of aeration needed for
heat removal can be more than 10x the requirement of microbial oxygenation (Rynk et al. 1992).
As the composting process continues, the quantity of easily digestible compounds decreases,
leaving more difficult substances to consume (proteins, cellulose, and lignin). At this point, the
cumulative metabolic rate (and microbial population) plateaus and begins to decline. As
microbial levels decline, so does the pile temperature (Epstein 2011, Chiumenti et al. 2005).
With heat extraction as a goal, maintaining pile temperatures between 130-150ºF and prolonging
the point of plateau and temperature decline are two strategic goals for the operator. Although
one may think that maintaining pile temperatures in excess of 160ºF would increase heat
recovery from the system that method would actually subject the microbes to inhibitive
temperatures they could not survive. Although the heat exchange system would perform well
during this short phase, long term heat recovery would likely suffer, as the heat producers
(microbes) would be sacrificed for this temporary gain. Achieving maximum heat production
and heat recovery requires the provision of an optimal microbial living environment, where they
can thrive and reproduce. Some basic guidelines for optimal composting are:
1. Organic material – feedstocks are thoroughly mixed and have a combined carbon-
to-nitrogen (CN) ratio of 27:1 - 30:1 (Epstein 2011)
2. H2O - pile moisture content is between 50-60% (Epstein 2011)
3. O2 – compost pile oxygen content is between 10-18% during the active phase
(Epstein 2011), and 1-5% during the maturation phase (Chiumenti et al. 2005).
117
Also ensure the aeriation system does not have any short circuits causing
preferential airflow.
4. Temperature - maintain pile temperatures under 150ºF through aeration and/or
turning (Epstein 2011)
5. Porosity - free air space should be 35-50% (Chiumenti et al. 2005)
6. pH – maintain between 5.5 – 8.0 (Chiumenti et al. 2005)
7. Particle size - no larger than 1-3’’ (Chiumenti et al. 2005)
8. Contaminants - Absence of materials toxic to microbes
9. Drainage – compost leachate is drained away from the pile to prevent an
anaerobic base from forming
Individuals interested in more specific details concerning the science of the composting
should reference Epstein (2011), Chiumenti et al. (2005), Haug (1993) and Rynk et al. (1992).
Heat Recovery from Composting
Numerous compost heat recovery systems have been tested over the years, and are
described in great detail in Smith and Aber (2016). In their review, three primary mechanisms of
how to extract heat from a composting pile were described. The first method originated in
ancient China 2000 years ago, and involved growing crops above a composting mass, which
supplied heat to the root zone of crops through convection (Brown 2014). This system was
further advanced in France during the 1600’s, where acres of glass-enclosed hotbeds were used
for crop production (Aquatias 1913). As with the Chinese system, a trench was filled with
composting manure and was capped with topsoil for crop growth. As the manure composted,
heat rising through convection warmed the roots of the crops, allowing for several months of
season extension in the spring and fall. However, this system lost favorability in France in the
118
early 1900’s when the primary composting feedstock (horse manure) was no longer in large
supply, due to the replacement of the horse with the automobile. A version of this method made
a short comeback in the 1940’s – 1970’s by English and Dutch farmers, who used decomposing
straw bales to extend the growing season of tomatoes, cucumbers and lettuce (Loughton 1977).
While the convection method is the simplest and least costly in extracting heat from the
composting process, it is also the least efficient and is limited to horticultural applications. The
second approach to recover heat from composting is conduction-based, which offers a substantial
advance in both heat recovery and utility. This method was pioneered by Jean Pain in the 1970’s
at his farmstead in France. In his system, a 55 ton pile of chipped brushwood with hundreds of
feet of coiled tubing located within the composting mass, was used to heat water from 50°F to
140°F at one gallon a minute for a six-month period. The water was used to warm a high tunnel
and to heat a farmhouse (Pain and Pain 1972). What made this system a substantial leap forward
was the ability to use the thermal mass of the compost pile to warm water that could be used for
any purpose. This method is still used today and is described in detail in Brown (2014).
Although warming water via conduction of composting feedstocks is more efficient than
convection-based systems, this method is more suitable for backyard operations, where the time
and labor consuming aspects of installing and removing the pipe during pile formation and
breakdown can be absorbed by an enthusiastic homeowner. This method is typically not suitable
for commercial operations, where revenue is the goal and labor/time is accounted for. Problems
can also arise if too much heated water is removed from the pile, and/or the replacement water is
too cold (Smith and Aber 2015). This scenario can inhibit microbial growth and even crash the
microbial population, causing putrid conditions. However, if managed properly, this can be a
successful option for small-scale hobby operations.
119
An improvement to using pipe embedded within the composting mass is to imbed the
recirculating heat exchange pipe within concrete below the composting feedstocks. This type of
system is more suited for commercial operations, as the time and labor aspects of installing and
dismantling the pipe are avoided. While the addition of concrete increases the cost of the
operation, it is a more realistic option for commercial operations processing large quantities of
biomass. A commercial composting facility in New Brunswick, Canada uses this type of system
to prevent snow and ice from freezing GORETM compost covers to the ground (Allain 2007).
However, as with the within-pile heat recovery systems, one has to be careful with how much
heated water is extracted, in addition to carefully monitoring the temperature of the makeup
water. These details were not considered at a separate Canadian composting facility, when trying
to extract heat from water-filled pipe below composting fish waste. The plant operators
circulated the heat exchange liquid too fast, removing heat until it caused the compost pile to
cool down and crash. The result was putrefaction of the feedstocks and odor complaints from
adjacent neighbors that resulted in the facility shutting down. If less water were pulled from the
system, this approach would have worked. However, there is risk when extracting heat from a
within-slab system, as the slab is part of the thermal mass of the pile. Pulling too much heat from
one section of the pile (in this case the bottom) risks anaerobic and odorous conditions.
The final approach to recover heat from the composting process is to extract the thermal
energy from the exhaust vapor using an aeration system. By mechanically moving air through
the pile, the aerobic microbes receive needed oxygen, while removing excess heat that can
inhibit their growth and reproduction (Epstein 2011, Rynk et al. 1992). Importantly, heat
recovery does not interfere with the composting process like conduction-based recovery systems.
The simplest method under this approach is to use the heated compost vapor directly. These
120
systems usually involve placing perforated PVC pipe below composting feedstocks and forcing
air through the composting pile (positive aeration) with a fan. As air is forced through the
decomposing material, hot compost vapor is forced out. Because of high levels of NH3 and
volatile organic compounds (VOCs) in the compost vapor stream, a biofilter is necessary to
scrub out the contaminants. Early research utilizing this technology came from the New
Alchemy Institute, where a winter greenhouse was warmed through compost vapor, which had
been sent through a biofilter (Fulford 1986). More recent research from Adams (2005) explored
this technology for heating greenhouses in Vermont.
While positive aeration systems serve as a valuable tool for season extension and reduced
heating costs for greenhouse and high tunnel growers in cooler climates, it has limited
applications, due to the difficulty in capturing the diffused heat across the pile. The amount of
available heat is also limited, as only 13.4% of the heat generated within a compost pile is
contained in the air (Themelis 2005). Composting systems using positive aeration also risk
corroding any type of building or structure they are housed in, as highly corrosive vapor is being
blown into the airspace. A more effective method in capturing heat from composting is through
negative aeration, where air is pulled through the composting mass and into a single chamber
where the heated vapor can be directed. In some systems, this heated vapor is sent through a
biofilter, where the contaminants are scrubbed and the heat and CO2 are diffused into a
greenhouse (Smith and Aber 2016). However, a more efficient system is to direct the heated
vapor into a chamber containing an air-to-water heat exchanger. By using a heat exchanger, the
thermal energy contained within the water molecules of the vapor stream can be extracted. This
is important as 63% of the energy balance within a composting pile is contained in the water
vapor (Themelis 2005). Furthermore, if an air-to-water heat exchanger is used, heat recovery
121
does not impact the composting process and the heated water can be used for more than just
horticultural applications. Agrilab Technologies Inc., developed such a system, by using
Acrolab’s Isobar® Heat Pipe technology and the ASP composting method (Agrilab
Technologies is a U.S.A. based vendor of the Acrolab Isobar system).
Acrolab’s Isobar® Heat Pipe Technology
Acrolab’s Isobar Heat Pipe is a two-phase super-thermal conductor that provides thermal
uniformity across the pipe by immediately transferring heat evenly across the entire unit
(Acrolab 2013). The heat exchanger uses an extremely high-grade stainless steel evacuated pipe
filled with a working refrigerant. When heat is applied to the evaporator side of the pipe, the
refrigerant inside heats up and vaporizes. That vapor travels the length of the pipe and condenses
on the cooler side, releasing the latent heat of condensation. After condensing, the condensate is
returned to the warm end of the pipe through capillary action in a metallic wick contained within
the isobar (Acrolab 2013). The beauty of the system is that there are no mechanical parts within
the isobar (Figure 9).
Figure 9: Internal Workings of Acrolab's Isobar Heat Pipe (Acrolab 2013)
122
The Isobar Composting and Thermal Energy System developed by Agrilab Technologies,
uses this technology, by utilizing the metabolic heat generated during the composting process for
heat exchange. The system uses 6-12 Isobar, 30-60’ in length, contained within a single unit that
has a vapor chamber and a highly insulated bulk storage tank of water (number and length of
Isobar depends on monthly feedstock tonnage). The Isobar run the length of the unit, with
roughly ten feet contained within the sealed tank of water, which serves as a thermal battery
(Figure 10).
Figure 10: UNH Isobar Heat Exchange System (Loughberry Manufacturing 2012)
The system operates by pulling heated vapor from composting feedstocks through the
aeration network and into the vapor chamber containing the array of Isobar. This vapor (120-
165ºF) is pulled across the portion of the Isobar located within the vapor chamber. When the
vapor condenses on the cooler pipe surface, it transfers the latent heat condensation to the pipe
(≈2260 kJ/kg), which is used to vaporize a refrigerant. The vapor within the Isobar travels up the
pipe into the section of the unit contained within the highly insulated storage tank of water. The
cooler water in the tank causes the vapor within the pipe to condense, once again transferring the
123
latent heat of condensation, only this time it is transferred from inside the pipe to the water
(Figure 11). The heated water (typically 100-140ºF) can be used for any application requiring hot
water (radiant floor heating, aquaculture, greenhouse, preheater for an anaerobic digester,
preheater for a standard hot water system, etc.). Current uses for this type of system can be found
in Appendix 2.
Figure 11: Flow Diagram of Heat Recovery System (Agrilab 2013)
The Value of Heat
Based on the four farms currently utilizing this system, it has been demonstrated that heat
capture of over 1,000 BTU/ton/hr over a 120 day composting period is possible (Agrilab
Technologies 2013). If managing the system more intensively, heat capture of over 1,500
BTU/ton/hr over a 60 day composting period is possible. From a research perspective, UNH
plans on utilizing Agrilab’s system and perfecting the methods of heat production, recovery, and
utilization. Data will be made available in a series of follow-up reports and publications.
124
ORIGINS OF THE UNH PROJECT
The idea for the Heat Recovery Composting Facility at the UNH Organic Dairy Research
Farm developed during conversations about how to make the farm a more closed agroecosystem.
To reach this goal, all proposals needed to be profitable and replicable for farmers in the region.
A large component of making the farm a more closed agroecosystem involved reducing energy
imports to the farm. A second major component was improving the farm’s manure management
system, which involved storing manure in large anaerobic piles that would occasionally be
spread on the fields. Although this is a common practice on many dairy farms throughout the
region, it posed a significant environmental problem. Because the stockpiled manure was not
being composted aerobically, odors from anaerobic decomposition were emitted from the piles.
The manure piles were also emitting high levels of CH4, and leaching nitrate (NO3-) into an
adjacent waterway. The quality of the manure being spread on the fields was also poor, due to
losses of nitrogen from leaching and volatilization. Finally, the unmanaged piles of manure were
the primary breeding ground for the biting flies on the farm, which are known to pose health
concerns for dairy cows (Campbell et al. 1993).
The initial solution to the manure management problem was to develop a passive aeration
windrow system to process the manure and spent bedding on the farm. This type of system is
known to be inexpensive and has proven to be successful in composting animal manures (Rynk
et al. 1992). Three windrows were created with the dimensions of 30’L * 8’W * 4’H. Cost
savings in fuel and labor were immediately recognized from a reduction in material to be spread
on the fields. The final product was also more stabilized and solved the runoff issue at the farm.
However, after a year of composting, UNH researchers and a private donor began discussing the
possibility of building a heat recovery composting facility using Agrilab’s Isobar heat pipe
125
technology. The idea was that a more advanced composting system would help address both the
manure management and energy goals of the farm in one step. At the time, only one other
facility in the world (Diamond Hill Custom Heifers) was using this technology on a commercial
level. Their facility (built in 2005) had 2000 heifers and processed 150 tons of feedstock every
month (Tucker 2006). However, the UNH Organic Dairy Research Farm was much smaller,
producing a fraction of the waste. As with most composting projects, economies of scale have to
be considered. While the technology was proven to work for a large-scale dairy operation, it was
yet to be tested on a small dairy farm with under 100 head. However, the research team and
private donor decided that it was worthwhile to construct the facility, as it could be used to refine
compost heat production and extraction methods, and also determine the economics of the
composting system on a smaller-scale. Designs for the UNH facility began in May 2012,
construction began August 2012 and the facility was completed in June 2013.
PLANNING AND SIZING THE FACILITY
Feedstock Parameters
The first step in designing an ASP composting facility with Agrilab’s heat recovery unit
is to determine feedstock quantity, along with the corresponding feedstock chemical and physical
properties. In assessing feedstock quantity, the smallest of the heat recovery systems from
Agrilab Technologies, require 60 yd3 of mixed feedstock per month (Agrilab Technologies
2013). It is important to note that the feedstock requirement is based on a mixture of waste
materials that collectively result in conditions that are optimal for microbial growth. The three
most important factors to consider are CN ratio, moisture content, and bulk density. An online
compost recipe builder can be used to determine whether one has enough waste materials in the
right proportion to generate the 60 yd3 minimum. In situations where there is a deficit of
126
material, feedstock can either be imported, or in some cases, stockpiled from other times of the
year when that material is in excess. For instance, the carbon source for the UNH facility comes
from the bedded pack barn, which is cleaned out only twice a year (May and November).
Because it is only cleaned twice a year, the spent bedding has to be stockpiled. Likewise, during
the summer months, manure is in shortage because the cows are out at pasture for > 8 hours a
day. Excess manure from the winter months is stored in small windrows to supplement the
summer composting recipes. However, stockpiling is not ideal, as microbes will consume the
stockpiled material, reducing heat recovery potential. In assessing feedstocks, it is important to
realize that a deficit in nitrogen will slow down the composting process, reducing heat recovery,
while too much nitrogen will increase temperatures too quickly and result in increased NH3
volatilization and lower quality compost (Chiumenti et al. 2005, Rynk et al. 1992).
Assess Current Hot Water Demand and Location
After determining whether the farm has enough biomass (or can obtain enough biomass)
in the optimal proportions, the next step is to assess the hot water demand of the farm, and
whether the heat recovery unit is economical. From a practical standpoint, if one is already
planning on building an ASP composting facility, the added cost of the heat exchange unit is
likely economical. Regardless, assessing the current farm energy demand is valuable as the heat-
exchange unit can be sized accordingly, ensuring it is not overbuilt. For UNH, the cost of the
farm’s heating needs is roughly $8,300/year. A majority of this cost originates from heating
water to 180-190ºF for the various milk sanitization processes that occur on the farm. As a
consequence, the primary function of the heat recovery unit for the UNH facility is to preheat the
50ºF well water entering the primary water boiler, which is heated by both oil and electricity.
127
Feedstock Residence Time within Facility
The next step in the planning phase is to determine the residence time the composting
materials will remain in the facility. In making this decision, it is important to consider whether
the compost is to be cured in the facility, as that decision will require more space, due to slower
turnover. In the case of the UNH facility, we decided to cure the compost in the facility, which
was a decision made for research purposes (120 day residence time within facility). Although
there are advantages to curing within a facility (faster due to forced aeration, less chance of
contamination from seed, will not get saturated by rain), it requires a larger building and a much
higher initial capital cost. An alternative is to have a much shorter residence time within the
facility, and cure the compost outside under a compost cover, which allows the material to
breathe, shed rain, prevent seed from entering, and is a fraction of the cost. Managing the system
under a shorter residence time (if one has enough biomass) is also strategic from a heat recovery
and economic standpoint, as compost temperatures under this type of system peak during the
first few weeks of composting, and gradually decrease afterward. A shorter residence time would
allow for heat extraction to continually occur during the highest heat producing periods of the
composting process.
Sizing the Facility
With information on feedstock quantity, and the length of time it will spend in the
facility, a total composting volume within the facility can be estimated. With UNH as an
example, facility size was based on 250 yd3 of feedstock (manure, waste feed hay, and spent
animal bedding) per month, and a compost residence time of 120 days. Because the facility
would be loaded in monthly batches, the resulting facility would have to have four bays each
accommodating 250 yd3/month. Assuming a pile height of 9 feet, the length and width of the
composting floor can be determined based on a combination of site conditions and building
128
design to get the needed volume. In our scenario, each bay was 32’L*20’W*9’H (215
yd3/month). Although this is 35 yd3 short of the theoretical monthly maximum of feedstock, the
facility was reduced in size from the original proposal to save cost (≈ $20,000).
After calculating the dimensions of the composting floor, extra footage has to be added
for the mechanical room and walkways within the composting room for exits (code). For UNH,
the isobar unit going into the mechanical room had dimensions of 30’L * 34.5’’W * 30’’H. In
addition to the Isobar unit, extra space has to be added for the aeration pipe and the leachate
system. In our scenario, the mechanical room ended up being 10’W * 96’L. The composting
floor also had an additional 8’ concrete apron for an internal walkway to the exits. In sizing a
non-research facility, both the width of the mechanical room and the width of the internal apron
can be reduced by several feet. The UNH facility was intentionally built with extra width at these
locations to better accommodate large groups visiting the facility.
With information on the size of the composting floor, mechanical room, walkways, and
all other needed internal space, a total square footage can be estimated. For UNH, the resulting
facility was 96’L * 50’W * 22’H. Of the 50’ width, 10’ along the entire length of the facility
represents the mechanical room, while another 8’ represents the internal apron for access to the
exits. The remainder of the building represents the composting floor. Each bay (aeration zone)
accommodates a compost pile 32’L * 20’W * 9’H (215 yd3/month) (Appendix 3). The height of
the building was based on the height of the tallest machine to go within the facility. For UNH,
we needed a clearance of 22 feet to accommodate the dumping of material from the farm’s
primary dump truck.
Although the facility was built to process 215 yd3/month with a 16 week residence time, a
facility of the same size only housing the compost during the active phase (≈ 3 - 4 weeks), could
129
go from processing 215 yd3/month (2,580 yd3/yr) to 860 yd3/month (10,320 yd3/yr). As
mentioned earlier, the residence time the compost stays in the facility greatly affects the amount
of biomass that can be processed. As a side note, the UNH facility will switch to a much shorter
residence time after the economic analysis has been completed for a facility handling just 215
yd3/month.
Aeration Floor Design
When designing the aeration floor, the successes and failures of past ASP floor designs
were considered, to ensure the piles would receive an optimal level of aeration across the entire
pile. It is important to note that there is a decrease in oxygen provided to the pile as the length
from the blower increases. For this reason, piles should not exceed 50-75’ (Rynk et al. 1992). At
the UNH facility, aeration lines were 30’ in length and were made of 4’’ PVC pipe, which fit
within the general recommendation of aeration lines being 4-6’’ in diameter (Epstein 2011, Rynk
et al. 1992). Each line had ½’’ diameter holes drilled 6’’ on center to serve as the aeration holes.
The specific size of the holes was based on the diameter of the pipe and the length of the run. A
common equation used to calculate aeration hole size is:
Hole diameter = √[(D2*S)/(L*12)]
D = pipe diameter in inches
L= pipe length in feet
S= hole spacing (in)
From a graphical standpoint, this is represented in Figure 13.
130
Although pipe with holes pre-drilled is available, it is best to purchase pipe and drill the
holes on-site following the cement pour. The purpose of drilling the holes after the pour is to
have the ability to fill the pipes with water to prevent them from moving or floating during the
concrete pour, and prevent cement or other debris from getting into the aeration network during
the construction process (both issues will be discussed in detail later).
After deciding on pipe diameter, length, hole spacing, and hole size, the next step is to
determine the spacing between the aeration pipe. Based on past ASP facilities, the general
recommendation is to have aeration lines 3-4 feet apart (Epstein 2011). Closer spacing is
recommended for materials with a higher bulk density (manures, sludge, etc.), where higher
oxygenation is required. At the UNH facility, aeration lines were set up with research trials in
mind, and were spaced to accommodate treatment walls (Appendix 3). In a non-research facility,
a uniform spacing within the 3-4’ range would have been used for all the aeration lines,
compared to our facility, which had varying spacing.
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
20 30 40 50 60 70 80
Aer
atio
n H
ole
Dia
met
er (i
n)
Pipe Length (ft)
Recommended Aeration Hole Diameter by Pipe Length for Aeration Holes 6 and 12'' on Center
4'' Pipe Diameter w/holes 6'' on Center
4'' Pipe Diameter w/holes 12'' on Center
6'' Pipe Diameter w/holes 12'' on Center
6'' Pipe Diameter w/holes 6'' on Center
Figure 6: Recommended Aeration Hole Diameter by Pipe Length for Aeration Holes 6 and 12'' on Center Figure 12: Aeration Hole Diameter by Pipe Length for Holes 6’’ and 12'' on Center
131
In addition to having 3-4’ between each pair of aeration lines, the two externally located
lines on either side of the facility should be cast 3-4’ from the side walls (Figure 14).
The reason to cast the first aeration line 3-4’ away from any wall is to prevent preferential air
channels from the Coanda Effect, which is the tendency of moving air or liquid to attach itself to
a nearby surface, and flow along it. In a composting pile, walls too close to an aeration channel
can serve as this surface and result in preferential airflow on the edges, causing more
oxygen/faster decomposition on the sides and less oxygen/slower decomposition in the middle
(Chiumenti et al. 2005). As the pile continues to decompose under this condition, the problem
can become worse as pile slumping on the edges (from faster decomposition) will cause further
preferential airflow in those locations, affecting the decomposition rate of the entire pile. Heat
Figure 13: Aeration Floor Spacing at UNH Compost Facility
132
losses will not only occur from reduced decomposition in the middle of the pile, but will also
occur from cold air being pulled into the aeration system from the edges of the pile.
In addition to preventing the Coanda Effect from the side walls, the length of aeration
floor along the back push wall should also have a 3-4’ aeration dead zone to prevent preferential
air channeling. At the UNH facility, the 3’ section of each aeration lines closest to the back push
wall, did not have aeration holes and had a layer of concrete overtop instead of a cover plate
(specifics will be discussed in detail later) (Figure 14).
Figure 14: Floor Spacing for the Aeration Lines at the UNH Compost Facility
Cost Saving Tip # 1: When considering cost-saving strategies, ensuring the facility is not
overbuilt is an obvious one. This is especially true for the aeration network, which needs to be
carefully planned and sized to meet the aeration demand of the various materials being
composted. When deciding on pipe diameter for the aeration floor, it is important to consider the
total airflow requirements in relation to the pile height and length. Increasing from a 4’’ to 6’’
diameter PVC aeration channel has substantial cost ramifications in the thousands of dollars.
133
The general aeration setup within the mechanical room of these systems involves at least one
size increase in PVC diameter beyond what was cast in the aeration floor. While the cost
difference between 6’’ and 8’’ PVC pipe and fittings is not too significant, the cost difference
between 8’’ and 10’’ PVC is enormous, costing several hundred dollars more per fitting and
section of pipe. In most cases, 10’’ PVC pipe is not necessary, and careful planning should verify
this point. The original aeration system at the UNH facility was constructed in a manner
requiring 10’’ PVC pipe in the back mechanical room. The extra cost of using 10’’ vs. 8’’ PVC
pipe and fittings was $7,500 for just the materials and did not include:
Extra cost in shipping weight from the heavier components
Extra labor in installing heavier and more bulky materials
Extra sealant required
Extra support structures (clevis hangers, pipe riser clamps, threaded rods, etc.)
Contractor markup (usually 25%)
A more realistic figure was around $12,000, when all of the other costs were included.
Cost Saving Tip # 2: A second cost-saving strategy is to purchase all the PVC pipe, PVC
fittings, sealant, aeration fan(s), flexible couplings, and support structures and have them
installed vs. purchased by a contractor. An important point to make is that this type of cost-
saving strategy often comes with the knowledge that the contractor will not warranty the
purchased products (standard practice if owner purchases the materials). Additionally, if there is
an insufficient supply of materials during working hours, you will end up paying for workers to
sit around and wait for your mistake. When purchasing materials to save cost, it is crucial to
ensure that all the supplies (and in some cases, extra) are available to truly save cost. It is also
important to sign a contract indicating that you will be purchasing materials and that a markup
134
will not be part of the deal, as some contractors will still charge a markup for materials
purchased by the owner.
Facility Location
Specific information on the steps involved in siting a facility were omitted from this
report, as each farm/compost operation will have tremendous variability with regard to proper
location. For reference, detailed information on this topic can be found in Epstein (2011) and
EPA (1994). However some basic guidelines are provided below:
Avoid close proximity to neighbors unless a powerful air filtration system and
biofilter are to be used. Single greatest cause of compost facility closures is due to
nuisance claims from odor (Epstein 2011).
Cost Saving Tip # 3: Site facility as close to feedstocks as possible and try to have
straight line transport of feedstocks to the composting bays. Minimizing feedstock
handling time by siting and orienting the facility properly can save a tremendous
amount of money (time, labor, fuel, etc.) with regard to material handling.
Ensure facility has adequate fire lanes on all sides and room for feedstock to be pulled
out and piled should an internal smoldering fire occur and require breakup (code
requirement for UNH facility).
In addition to the above recommendations, some specific location considerations for a heat-
recovery facility using Agrilab’s Isobar System are:
Minimize distance from hot water production to hot water use. However, if using an
underground insulated PEX pipe to transfer the hot water from source to sink, losses
are only 2-3ºF per 100’ length if buried properly (OWFB 2013). Siting to reduce
material handling should take precedence.
135
If planning on attaching a high tunnel or greenhouse, proper facility orientation is
needed to ensure shading does not become a problem.
For reference, UNH sited the compost facility in a location that was closest to the
feedstocks being composted. The reduction in time for material handling was determined to be
the greatest factor in locating the facility (Figure 15).
Figure 15: Aerial View of UNH Organic Dairy Research Farm
BUILDING A HEAT RECOVERY COMPOSTING FACILITY
The following sections outline the step-by-step process of building the UNH heat-
recovery composting facility, with recommendations to operators on design and the various cost-
136
saving strategies that can be used at their sites. The reader is encouraged to reference the
appendices for additional diagrams/specifications and cost structure.
Site Preparation
Due to the high variability in soils and site conditions, ground preparation should be
assessed by the contractor hired for that particular job. One important consideration that may be
slightly different than standard practices is that composting facilities require more attention with
regard to drainage. Because there is potential for pollution of waterways from nutrients
originating from the feedstocks, compost, and compost leachate (primarily nitrogen and
phosphorus), all drainage from the site should pass through some form of filtration, whether a
lagoon, engineered wetland, or a small portion of an agricultural field. At the UNH facility,
drainage is directed into a portion of an agricultural field, which eventually travels into a swale,
leading into the nearby farm woodlot. As with odor, careful planning of drainage needs to be
assessed before a problem actually arises. Failure to do so could result in significant fees or
facility closure.
Underground Slab and Concrete Wall Preparation
Underground cold and hot water lines [(1’’ PEX) Cresline HD-160] were installed prior
to concrete forming. Both lines were set in a 5’ trench between the milk house (location of water
supply) and the future mechanical room (280 linear feet). The lines were 8’’ apart, and had 6’’ of
sand surrounding them in all directions. Compacted backfill was put overtop. The 1’’ PEX cold
water line was connected to a ¾’’ PEX line at the entrance of where the mechanical room would
be located and led to a freeze-proof post hydrant (Campbell CYH-5 Frost Proof Yard Hydrant) in
the location of the main composting floor. A second ¾’’ cold water line was also installed off the
first line to a freeze-proof post hydrant at the mid-point of the mechanical room. Both hydrant
137
lines were buried below the frost line (5’) and were marked and taped to prevent soil from
entering the pipe until future hookup.
The 1’’ hot water supply and return lines were contained within a heavily insulated pipe
(Uponor Pre-Insulated Pipe Systems ASTM Ecoflex Thermal Twin), which are often used for
outside wood furnaces (Figure 16).
Figure 16: Insulated Underground PEX Pipe
As with the cold water lines, the ends were taped until future hook-up to the mechanical room.
The primary 1500 gallon precast concrete leachate tank (Phoenix Precast Products) and small
section of 4’’ PVC connecting ductwork were also installed at this time (Figure 17).
Figure 17: Compost Leachate Tank at UNH Facility
138
Cost Saving Tip # 4: While smaller leachate tanks are less capital initially, they also reduce the
ability to pump large volumes of leachate at a time for field application. Compost leachate is
high in nutrients and serves as a great fertilizer/irrigation source, especially on farms. However,
1500 gallons is typically not enough volume to justify bringing out a tanker truck that usually
has a capacity > 4000 gallons. As a consequence, the tank needs to be pumped more often, and if
this leachate cannot be spread on a nearby agricultural field, more pumping may cost the
operator more money in the long run.
After installing the water lines and leachate tank, forms for the side walls and back push
wall were constructed. The forms for the back push wall had 16 sleeves for the 4’’ PVC aeration
lines (4 lines per bay). The forms for the back mechanical room also had a sleeve installed for
the main electrical line.
Cost Saving Tip # 5: During this stage of construction, it is important to identify and plan for all
possible sleeve locations, as drilling holes after pouring concrete is much more expensive.
Additionally, the larger sleeves for the aeration lines through the back push wall should be a
tight fit and not oversized.
Pouring Concrete Walls
The first pour at the UNH compost facility was the push wall, two side walls, and five
concrete piers for the front supports of the building. The back push wall was the thickest and had
the largest footings to accommodate a front end loader pushing material against it. The
dimensions were 96’L * 12’’ W * 8’H. The footings were 96’L * 6’ 6’’ W * 1’ H (Figure 18).
139
Figure 18: Concrete Dimensions at UNH Compost Facility
The two side walls had the dimensions of: 40’ L * 8’’ W * 8’ H. The footings were 40’ L
* 2’ W * 1’ H. The two side concrete piers in the front of the building had the dimensions of
2’6’’ L * 10’’W * 8’ H, with footings of 2’6’’ L * 4’6’’ W * 1’H (Figure 19). The three internal
piers had dimension of 3’8’’ L * 10’’ W * 8’ H, with footings of 3’8’’ L *4’6’’ W * 1’ H. After
the walls and piers cured, they were backfilled and brought to grade with compacted fill.
Figure 19: Concrete Piers at the UNH Compost Facility
140
The second pour at the UNH facility was the wall and piers for the back mechancal room.
The wall had dimensions of 32’3’’ L * 8’’ W * 6’ H, with footings of 32’3’’ L * 2’ W * 1’ H. In
addition to this wall, eight concrete piers were cast to continue the structureal support for the
back of the building. The eight concrete piers were 12’’ L * 8’’ W * 4’ H with footings of 2’ *
1’ (Figure 20and Figure 21). After the wall and piers cured, they were backfilled and brought to
grade with compacted fill.
Figure 20: Back Mechanical Room after first Concrete Pour
141
Figure 21: Back Mechanical Room Concrete Dimensions
Cost Saving Tip # 6: The amount of concrete used to construct the UNH facility was substantial
(≈ 225 yd3 total) and was one of the larger expenses in the project (Appendix 4). However, cost
was reduced by using 5 ½’ of wood for the upper push wall and side walls, reducing the concrete
requirement by 30 yd3 (Figure 22). When substituting wood for concrete it is important to note
the decreased longevity of the wall and need to replace the wood at some point in the future.
Figure 22: Back Push Wall at UNH Composting Facility
142
If using wood, it is important to note that the material may warp, due to high heat and
moisture from the composting process. Warpage is especially problematic if the mechanical
room is on the other side of the wall and has a ventilation system. This is because the ventilation
system will draw compost vapor and dust through the cracks between the boards and into the
mechanical room. This poses a potential health concern, and needs to be addressed with some
form of vapor barrier. To address this issue, we used 4’ * 8’ * 3/8’’ plywood and attached plastic
sheeting (FrostKing 10’ * 25’ rolls) for the vapor barrier, and then used rough pine lumber (2’’ *
10’’* 16’) for the compost-to-wall interface (Figure 23).
Figure 23: Vapor Barrier on Back Push Wall at UNH Composting Facility
Cost Saving Tip # 7: A second major cost saving strategy, which could be utilized for those
installing a high-tension fabric structure, would be to use interlocking concrete waste blocks for
the side walls (Figure 24).
143
Figure 24: High-Tension Fabric Structure with Waste Block Walls (ClearSpan 2013)
Waste blocks come in various sizes, with the most common for this purpose being 6’ L *
2’ W * 2’ H and weighting 3600 lbs per block. If buying a trailer load, cost per block is often
under $75 per delivered block. Cost savings of using blocks are realized through a reduction in
ground preparation associated with the walls and footings, along with a reduction in labor cost
associated with forming and pouring the walls. This is especially true if the site has ledge. Had
UNH built a similarly-sized fabric structure with waste block side walls, the material cost for the
blocks would have been roughly $4,225 ($65/ block * 65 blocks).
If waste blocks are used, it is recommended to use them for only the side walls and not
the back push wall that contains the aeration channels and isobar unit behind it. The primary
reason why the push wall should be poured is that you want a structurally sound wall with
footings, as any movement could break seals in the aeration network and in the worst case
scenario, damage the heat exchange unit. The seams between blocks would also become
problematic with regard to compost leachate and vapor entering the back mechanical room. As
with the wooden walls, negative aeration from the air filtration system would pull air into the
mechanical room.
144
Insulating the Concrete Slab and Setting up the Aeration Ductwork
One of the most important steps in building a heat recovery composting facility is
ensuring enough insulation is put underneath the concrete slab, as this cannot be remedied
afterward. The goal of insulating the concrete slab is to prevent cold soil temperatures from
robbing heat from the slab and aeration ductwork. Insulating also reduces condensation from
forming on the bottom of the slab, due to varying temperatures. While heat loss is impossible to
avoid (1st law of thermodynamics), reducing losses pre-heat exchanger through proper insulation,
will increase the heat recovery rate. Proper insulation of the pad will also ensure the base
material is capable of reaching thermophilic temperatures. This is especially true during the
winter months in cooler regions. A good way to think of the concrete slab is to consider it as a
thermal battery for the compost – it has to be insulated to reduce heat from escaping. To reduce
this problem from occurring, two layers of 2’’ ridged extruded polystyrene foam (Foamular 250)
were used at the UNH facility (Figure 25). This 4’’ layer of foam had a total R-vale of 20. When
installing these boards, it is important to overlap the top boards with the bottom, preventing any
continuous vertical seams where thermal loss can occur.
Figure 25: Insulation below Main Composting Floor at UNH Composting Facility
145
Cost Saving Tip # 8: An alternative and possibly cost-saving strategy if labor costs are high
would be to use Insul-Tarp®. This product has an R-value of 5.9 and can be rolled out like a
tarp, saving labor costs. At the UNH facility, it would have required 15 rolls of the 12’*50’ tarp
at a cost of roughly $9,000 to match the same R-vale of the rigid foam insulation, which cost
$5,500 for the material (Appendix 1). Again, comparing R-values, local distributer prices, and
the cost of labor will determine which insulation is the most economical.
The degree to which the slab is insulated depends on ambient ground temperatures during
the winter season. Because the UNH facility is in New Hampshire, and experiences cold winters,
two layers of insulation were required to separate the system from the ground. This
recommendation originated out of lessons-learned from the first heat recovery facility built in the
neighboring state of Vermont at Diamond Hill. The first composting bay they built only had 2’’
of foam insulation, which was found to be inadequate during the winter months. Insulation of the
concrete slab is the last place money should be cut if one is planning to recover heat from this
type of system. A minimum of R-10 should be used in all geographic locations.
After installing the rigid form insulation, the side walls also have to have a thermal
break/expansion joint where the concrete pad meets the concrete walls. The function of this
break is to allow for expansion and contraction of the pad, but to also prevent the back and side
walls from robbing heat from the much warmer compost floor. To create this joint, two layers of
½’’ polyethylene foam were used (A.H. Harris ½’’ Polyethylene Expansion Joint Filler). The
double layer provided a 1’’ joint and an insulation value of R-6 (Figure 26).
146
Figure 26: Thermal Break Installation against Internal Walls
Structural Support (Joints and Pad Reinforcement)
The next step in the process was to set up the forms to connect the main composting pad
to the external concrete apron. This connection was made with 7/8’’ diameter * 16’’ L greased
dowels, 12’’ on center from one another (Figure 27). The dowels were greased to prevent cement
from bonding to them. If cement were to bond to the dowels, it would reduce their functionality
and prevent the two slabs from flexing, resulting in cracking.
Figure 27: Concrete Slab-Connecting Dowels at UNH Composting Facility
147
As with the side wall expansion joint/thermal break, the expansion joint between the two
slabs is crucial for both structural reasons and to prevent heat loss. Without this joint, the cooler
concrete apron without active compost over it, would start robbing heat from the warmer
composting floor. This would reduce heat recovery and effect the composting material nearest
the joint. Ideally, this expansion joint should be two feet beyond the end of the aeration
ductwork, with the idea of having the compost extend at least two feet beyond the joint. The 4’
of compost beyond the end of the aeration channels with the expansion joint in the middle is to
prevent the aeration ductwork from pulling cold air into the system from the tapered portion of
the pile, and to insulate the aeration ductwork and compost pad at the end of the aeration line
(Figure 28).
Figure 28: Expansion Joint between Compost Floor and External Apron
With insulation and forms in place, the next step was to install the welded wire mesh.
Galvanized steel continuous high chair upper supports (4’’ high) were placed on top of the
ridged foam insulation to hold the wire mesh at a pre-set level. Wire mesh was then placed on
148
top of the supports to a height of 4’’with sheets being lapped a minimum of 12’’ and connected
to maintain a continuous structure (Figure 29). The mesh used at the UNH facility was 6*6
W2.1*W2.1, meaning smooth (W) wire with 6’’ longitudinal and transverse spacing and a cross
sectional area of 2.1 hundredths of a square inch. This material serves to reinforce the concrete
pad by increasing the tensile strength. By increasing the tensile strength (up to 30%), the tensile
force caused by expansion/contraction and/or shifts in the sub-base is reduced (Aberdeen 1957).
An important note is that cracks can still form, but will be less severe, as the mesh spreads the
force across a much larger area.
Ensuring the above step is done correctly is very important, as the temperature profile
across the concrete pad can be quite variable depending on how the various compost batches are
loaded into the facility. Cracks in the concrete floor are of particular concern because of the
amount of leachate that drains from the compost.
Installing the Aeration Channels
When installing the 4’’ PVC (Sch. 40) aeration channels, two feet of pipe was extended
beyond the back push wall through the sleeves and into the mechanical room for future hook up
to the aeration network. Expanding joint filler foam was used to fill the gaps between the PVC
and sleeves (Figure 30). All PVC was connected using solvent cemented joints, as the
temperature within the aeration channels (> 110ºF) exceeds the max recommended temperature
for threaded joint connections in Sch. 40 pipe (GF Harvel 2013). In sum, each aeration line had
33’ of PVC (2’ extending in mechanical room, 1’ through the push wall sleeve, 3’ unperforated,
and 27’ perforated to the expansion joint), with two 4’’ couplings and one 4’’ end cap.
Each aeration line was held up by six 4’’ pipe risers (only half of the raiser used) and six
pairs of ½’’ threaded rods (18’’ long), hammered through the rigid insulation and down into the
149
sub-base (Figure 29). The pipe risers were 5’ apart and were used to easily establish a 1% grade
over the 30’ compost floor from the end of the pipe down to the back push wall. This allows for
any leachate from the pile to drain through the aeration ductwork down to the primary leachate
system in the mechanical room. Each pair of pipe raisers was accompanied by a pair of 18’’rebar
rods hammered through the insulation and down into the sub-base in the shape of an X (Figure
29). The six pairs of rebar rods were used to prevent the pipe from moving during the concrete
pour.
Figure 29: Aeration Line Form Setup Prior to Concrete Pour
After setting up the pipe risers and supports, the next step was to fill the pipe with water
to check for any leaks in the aeration line joints and to increase the weight of the pipe to prevent
it from floating during the concrete pour. To fill the aeration lines with water, the 2’ section of
pipe on the other side of the back push wall was furnished with a temporary 4’’ flexible rubber
end cap (Fernco) and hose bibb (Figure 30). This temporary part can either be made by
150
purchasing a flexible endcap and inserting a hose bibb with washers (method used at UNH), or
by purchasing it premade like those from Fernco (HBC-4), and Band-Seal® (0704510). The
latter option may end up being less expensive should that single part be on sale.
Figure 30: Aeration Lines through Back Push Wall with Hose Bibbs
After filling the pipes with water and ensuring no leaks were present in the aeration lines,
the forms for the aeration channel cover plates were installed. When creating these forms, the
goal is to create a lip for a cover plate to sit, and have the plate recessed ¼ - ½’’ below grade. It
is important to have a lip for the cover plate to rest, otherwise the PVC pipe can be crushed if a
loader drives over one of the cover plates. It is also important to have the cover plates recessed at
least ¼’’ as it reduces the possibility of a loader catching one when loading/unloading the
compost. The recession also allows for some warpage of the wood without worrying about
catching the plate with a loader. The wooden cover plate forms used at UNH were ½’’ and ¾’’
thick * 6’’ wide plywood stacked on top of each other for a total thickness of 1¼’’ (Figure 31).
151
Figure 31: Aeration Line Forms for Cover Plates
After concrete pouring, these forms were removed, allowing for a ¾’’ cover plate to be
recessed ½’’ below grade. As described previously, the first three feet of each aeration line did
not have aeration holes, and did not require a cover plate, as solid concrete was poured over that
section of pipe (reducing cold air intrusion).
Cost Saving Tip # 9: An alternative to cut plywood cover plate forms is to use dimensional
lumber (2’’ * 6’’ * 10’ or 1’’ * 6’’ * 10’). This will save labor cost in cutting individual cover
plates. Below are pictures from another ASP heat recovery composting facility (Sunset View
Farm in Schaghticoke, NY) illustrating how to set up the cover plates with dimensional lumber
(Figure 32, Figure 33 and Figure 34).
152
Figure 32: Alternate Cover Plate Form Setup (Jerose 2013)
Figure 33: Alternate Cover Plate Form Setup (Jerose 2013)
153
Figure 34: Alternate Cover Plate Form Setup (Jerose 2013)
When looking at the three previous figures from Sunset View Farm, it is important to
note that they placed the pipe directly on the foam insulation, and did not have it raised with pipe
raisers and rebar. They also omitted the welded wire mesh. While this saves cost initially, both of
these omissions are not recommended as both reduce the structural integrity of the concrete
floor. Placing the aeration pipe directly on the insulation could also pose a significant leachate
problem, should one of the aeration pipes (also the floor drain), crack. However, the three
previous figures illustrate cover plate design and how to use dimensional lumber to achieve the
desired cover plate floor recession.
Pouring the Slab and Finishing the Composting Floor
The main composting floor (94’ L *32’ W) received 88 yd3 of concrete (9 truckloads) to
a thickness of 9.5’’. When the concrete was being poured, it was first placed on either side of the
aeration each pipe, to ensure they were held in place during the rest of the pour (Figure 35).
154
Figure 35: Concrete Pour for Main Composting Floor at UNH Facility
After all 16 aeration lines had concrete on either side, the rest of the concrete was poured. When
pouring, the welded wire mesh was held up with a metal rake in the few areas that were
slumping, ensuring an even level of mesh across the whole surface of the slab. Additionally, the
3’ portion of concrete closest to the push that did not receive a cover plate was given a 1% slope
over 3’ from the back push wall. This allows leachate from that portion of the pile to drain into
the aeration/drainage pipe, preventing leachate accumulation against the wall, which could
potentially enter into the mechanical room should a crack form along the wall. The removal of
leachate also reduces the possibility of compost becoming saturated at the base of the pile. If this
were to occur, an anaerobic spot would develop, producing methane, and also reducing the heat
value from that portion of biomass. Figure 36 illustrates the profile of the main composting floor.
155
Figure 36: Profile of Compost Pile and Floor at UNH Facility
After curing, water was released from the aeration lines, and the wooden cover plate
forms were removed. Each aeration line had ½’’ diameter holes drilled 6’’ on center at the apex
of the pipe (Figure 37). On some of the aeration lines, concrete had to be gently chipped away to
be able to access the pipe to drill a hole. After drilling, the holes were taped to prevent
construction material from entering during the rest of the building process.
Figure 37: Drilling of Aeration Holes
156
Cost Saving Tip # 10: Drill holes in the aeration lines at the end of the construction process, as
this will reduce the labor involved in taping and untaping the holes. It also prevents debris from
entering the aeration network.
In addition to aeration holes, smaller leachate holes may be necessary toward the section
of line nearest the back push wall. Because aeration holes are usually drilled at the apex of the
pipe, leachate can accumulate and pool before draining into the lowest aeration hole by the back
push wall (Figure 38). This can be resolved by drilling 1/8’’ diameter holes in each aeration line
at the lowest point of the pipe closest to the back push wall. Additional 1/8’’ diameter holes may
also be necessary at other low spots along each aeration line to allow for drainage. It is important
to prevent this pooling, as it will reduce the longevity of the cover plates. A simple method to
assess the floor drainage and where additional leachate holes are needed is to fill each aeration
channel with water and drill where pooling occurs.
Figure 38: Drilling Location for Leachate Holes
After drilling the aeration holes, wooden cover plates made of marine-grade plywood
(10’L * 6’’W * ¾’’H) were fabricated on site. An arch was sawn lengthwise across each cover
157
plate to ensure that pressure from the wheels of a loader would not press down on the PVC
aeration lines (Figure 39). Each cover plate had ½’’ holes drilled 6’’ from one another. Unlike
the aeration holes in the PVC lines, the aeration holes in the cover plates were drilled slightly off
center from one another, to reduce the possibility of the boards splitting down the middle.
Additionally, the holes in the cover plates were not directly over the holes in the pipe, reducing a
direct path for fines to be sucked into the aeration system.
Figure 39: Profile of a Cover Plate over an Aeration Line
Marine-grade plywood was used instead of pressure treated, as the farm is organic and
there were concerns about the chemicals in the pressure treated wood leaching into the compost.
Ideally, black locust would have been used, as it is naturally rot resistant and is accepted under
organic practices. This wood will likely be used when the cover plates need replacing in the
future. Figure 40 illustrates the profile and dimensions of the aeration floor at the UNH facility.
158
Figure 40: Profile of the UNH Aeration Floor and Subfloor
An important point to mention is that the aeration holes in the pipe and cover plates need
to be drilled cleanly, without pieces of the material inhibiting the orifice (Figure 41). Prior to
loading the composting facility, all holes should be checked and smoothed with a rasp.
Figure 41: Cover Plate Aeration Hole with Chipping
In addition to ensuring the aeration holes are clean, the aeration channels themselves should be
checked for any welded wire mesh that may impede the ability of the cover plates to rest at the
appropriate height (Figure 42).
159
Figure 42: Aeration Line with Welded Wire Mesh Impeding Proper Cover Plate Fit
Prepping and Pouring the Internal Concrete Apron
The first step in prepping the internal apron (96’ L * 8’ W * 9.5’’ H) was to remove the
wooden forms encasing the dowels. After removal, two layers of ½’’ polyethylene foam (A.H.
Harris Polyethylene Expansion Joint Filler) were used for the expansion joint, providing a 1’’
joint and an R-6 insulation value. Plain 4’’ concrete dobies were then placed on the compacted
fill to hold up the 6*6 W2.1*W2.1 welded wire mesh (Figure 43).
160
Figure 43: Specifications when Pouring the Internal Concrete Pad
Because this section of concrete pad was for a walkway, and a thermal break was
installed between pads, insulation was not needed. A second line of 7/8’’ diameter * 16’’ long
greased dowels, 12’’ on center from one another were also installed in the wooden forms to
connect the internal apron to the external concrete apron for a future pour.
Prepping and Pouring the External Concrete Apron
The external concrete apron (96’L*4’W*4’’H) was prepared in a similar fashion to the
internal apron - two layers of ½’’ polyethylene foam around the slab-connecting dowels, with
welded wire mesh being held up by plain concrete dobies, with no ground-level insulation
(Figure 44). However, the thickness of the slab was reduced by 5 ½,’’ reducing the concrete
requirement by 6.5 yd3.
Cost Saving Tip # 11: The concrete slabs for the walkway and external apron can be less thick
than that of the main composting floor, due to a lack of aeration lines. This decision has to be
161
made early on in the process, as the placement of the dowels connecting the slabs will have to be
adjusted. Alternatively, depending on facility design, the walkway and external apron could also
be stone dust, forgoing concrete completely.
Figure 44: Specifications when Pouring the External Concrete Pad
Installing the Leachate Network and Pouring the Mechanical Room Floor
The mechanical room floor (96’L * 9’ 3 ½’’W * 4’’H) was poured the same day as the
external apron and was prepared with the same welded wire mesh and plain concrete dobies for
risers. However, before the floor was poured, the primary 80’ long 4’’ PVC leachate line was
installed against the back push wall with 4’’ riser clamps 4’ on center (21 riser clamps in total),
and connected to the 1500 gallon leachate tank. Along the 80’ leachate line were eight 4’’ – 2’’
PVC wye reducers, located directly under every aeration header. At a later step, these wye
reducers were used to connect the primary leachate line to the 2’’ leachate lines coming from
each pair of aeration lines forming a single header (Figure 45).
162
Figure 45: Specifications when Pouring Back Mechanical Floor
Before the pour, the mechanical room also had forms for a 24’’ L * 24’’ W * 12’’ H sump pit.
This pit was later covered with a welded grate cover (McNICHOLS GW-100A). The floor on
either side of the sump pit was also sloped toward the pit to allow for drainage.
Raising the Building
Following the concrete pours, the final carpentry for the pole barn began. Specific step-
by-step details about raising a pole barn are not included in this portion of the report, as a
detailed document on the topic already exists (Carson and Dougherty 1997). However, the
engineering/architectural diagrams used to construct the pole barn are included in Appendix 5.
Cost Saving Tip # 12: When considering what type of structure to build, one should understand
that the primary purpose of the structure is to enclose the mechanical room and keep the
elements off the compost, as wind, rain, and snow negatively affect the composting process and
heat recovery. That being said, any structure capable of protecting the compost and mechanical
room would be suitable for achieving the end goal of compost stabilization and heat recovery.
For this reason, tension fabric structures like those from ClearSpan will often offer the best
163
economies, unless the farmer/compost operator plans to build the pole barn themselves and can
do so more cheaply. A fabric structure of similar size to the pole barn described in this report
would have cost $62,400 (total cost for material, delivery, installation, etc.).
What is important to note is that the steps prior to raising the facility in this report, and
those that follow, will likely be the same regardless of building type. For reference, a pole barn
was constructed at UNH due to the research nature of the project, and the requirement for the
building to last for decades. Though the facility was designed for research, the aeration floor and
mechanical room setup would have been the same had a fabric structure been used.
Although step-by-step details on raising a pole barn are not provided, there are several
operational recommendations worth mentioning. Regardless of structure type, the operator
should consider what equipment will be used in the facility and scale the height of the building
accordingly. At the UNH facility, the height was based on a silage truck dumping material onto
the composting floor, requiring a building height of 22’. A second important consideration is
ventilation, and the need to allow these types of compost buildings to breathe. If the building is
too tight, it will result in the accumulation of bioaerosols, which are unhealthy to workers.
Buildings without adequate ventilation will also cause an over accumulation of moisture, which
will corrode and eat away at the building itself. If building a pole barn, an exaggerated ridge vent
will allow for ample ventilation, preventing the accumulation of compost bioaerosols and
moisture. When constructing the ridge vent, ensure a mesh is installed to prevent birds and wind-
driven snow from entering the facility (UNH had to install Cobra mesh following construction
for these purposes). If using a fabric structure, a mesh upper end wall can be used to allow the
building to breathe. More expensive mechanical ventilation systems can also be added, but are
not likely necessary for small-scale operations.
164
Setting up the Mechanical Room
The setup for the mechanical room will vary based on whether aeration lines have
individual blowers with timers and fan speed controllers, or the whole system has one large
blower with damper controls. The UNH facility was originally designed with individual blowers
and fan speed controllers. This decision was made to have greater manipulation of each bay for
research purposes. However, having individual blowers and timers significantly increased labor
when adjusting the aeration schedules. More importantly, the company selling the blowers
switched materials and blowers that lasted more than 3 years at other facilities, lasted no more
than 3 months at the UNH facility, due to corrosion from the compost vapor stream. After 1.5
years of operation, the mechanical room at the UNH facility was completely overhauled and
converted to a single fan system with pneumatic damper controls. The following sections
describe the updated mechanical room setup.
All PVC components installed in the mechanical room were Schedule 40 and were
connected using solvent cemented joints and flexible couplings. Threaded components were not
used as they are not capable of maintaining a seal in high heat conditions like those found in a
compost aeration system. To save cost, the mechanical room was built within the facility and
was not a separate entity requiring a second roof. Instead, the ceiling was made of clear
corrugated polycarbonate sheets (SUNTUF ®) and was arched away from the back wall to allow
compost material to slide off, should it make it that high during loadings (Figure 46).
165
Figure 46: Specifications for Mechanical Room Ceiling
Aeration Lines
Aeration lines were set up in pairs, forming a single aeration header (2 headers per bay).
Each 4’’ diameter PVC aeration line was extended 6’’ into the mechanical room (originally 2’
but cut to 6’’) and was connected to a 4’’ manual butterfly valve (Hayward 4’’) (Figure 47). If
using butterfly valves, it is important that the interior disc is corrosion resistant.
Figure 47: Butterfly Valve Connecting Aeration Line to Aeration Header
166
From each butterfly valve, an 8’’ 90° elbow and 12’’ section of PVC pipe was connected
into an 8’’ PVC tee, forming a single aeration header (Figure 48). Following the 8’’ PVC tee,
each aeration header was reduced to 4’’ diameter pipe.
Figure 48: Setup for Aeration Headers
Cost Saving Tip # 13: Reducing the pipe diameter at the header was to save cost in the
pneumatic gate valves, which increase in cost exponentially as pipe diameter increases
(4’’diameter valves ≈ $80/ valve vs. 8’’diameter valves ≈ $900/valve). While the reduction in
pipe does cause more resistance in the aeration system, Agrilab engineers confirmed that the fan
could handle the extra resistance. The valves used at the UNH facility were 4’’ Valterra
pneumatic gate valves (6201P) and were selected due to low cost and the fact that all the
internal components were corrosion-resistant.
167
Cost Saving Tip # 14: Each aeration header was equipped with a 4’’ diameter PVC sanitary tee
with a flexible end cap (Figure 49). Tees were installed at each header to be able to recirculate
exhaust vapor into newly loaded bays during the winter. This procedure can reduce composting
time by several weeks, by bringing compost batches up to temperature in climates with harsh
winters. This not only increases the volume of material that can be composted, but also reduces
the amount of cold vapor being sent into the heat exchanger as piles warm up. The recirculation
process is accomplished by connecting a 4’’ flexible drain pipe into the 4’’ tee at the header and
into an 8’’ * 8’’ * 4’’ PVC wye located in the exhaust line. Vapor is then forced into a bay by
closing a valve in the primary exhaust line and at the bay header. Since the heated vapor is from
the exhaust line, one bay can be heated without a loss of heat recovery from Agrilab’s system.
Figure 49: Exhaust Vapor Recirculation Connection Points
Connected to the bottom of each aeration heater was a 2’’ S-trap with waste, which
connected into the primary 4’’ leachate system through a 4’’ – 2’’ wye reducer (Figure 50).
168
Figure 50: Leachate Hookup Specifications for Each Pair of Aeration Lines
A P-trap with waste can also be used, but would not fit in our situation. The function of
the S-tap (or P-trap) is to capture any condensate, and to prevent the aeration system from
pulling vapor from the leachate line or tank. Having a waste valve is essential, especially for
facilities built in cooler regions where freezing may occur. Winter freezing can occur if: 1) a
compost batch is too old and not heating up enough, 2) a bay is empty, 3) a bay is unloaded
during the winter and not loaded fast enough to prevent a freeze and 4) if a power outage occurs
during the winter and the system is not hooked up to a generator. Having a waste valve on the
trap will allow drainage of the water until that bay is up and running again.
Primary Aeration Supply and Exhaust Lines
Each aeration header was connected to the primary 8’’ diameter aeration supply line
through an 8’’ * 8’’ * 4’’ PVC wye (Figure 48). Flexible couplings (8’’ to 8’’) were used to
connect lengths of pipe in between bays. Few solvent cement seals were made in the aeration
169
network, due to the research nature of the facility and the need to modify components for
continual experimentation. While this design is suited for research, there are advantages in using
flexible couplings over standard PVC solvent cemented couplings when connecting joints in this
type of system. Flexible couplings are leak proof and allow the aeration network to flex.
Additionally, if any upgrades or alterations are required, the flexible couplings allow for entire
sections pipe to be easily removed and reused without having to saw pieces off.
The primary vapor line coming off the fan was also made of 8’’diameter PVC pipe. As
with the primary supply line, 8’’ to 8’’ flexible couplings were used to connect sections of pipe.
In the original set up of the facility, the vapor line left the Isobar unit and went directly outside to
vent the compost vapor to the ambient air. While this method reduced material cost and labor, it
did not scrub the contaminants out of the vapor stream, which contributed to odors around the
farm. Removing the vapor immediately also reduced the ability for further heat extraction and
usage. During the renovation of the mechanical room in 2015, the primary exhaust line was
upgraded and extended along the entire length of the facility and out to a biofilter (Figure 51).
PVC wyes (8’’ * 8’’ * 4’’) were also installed two feet to the left of every aeration header to
connect into the 4’’ PVC tees located at every header for reutilization of exhaust vapor (Figure
49). Two sections of 8’’ diameter * 6’ long aluminum pipe were also installed to serve as space
heaters for the back mechanical room (Figure 49).
170
Figure 51: Biofilter Processing Exhaust Vapor from UNH Composting System
The central 80’L * 8’’ diameter PVC exhaust and supply lines were held up with custom
made pipe stands (Figure 52). The pipe stands were made of 4’’*4’’ beams, ½’’ threaded rod
couplings, 7’’ threaded rods, and half of an 8’’ clevis pipe hanger. The stands were designed in
this manner to allow for 6’’ of vertical play, making it very easy to adjust the slope of the
aeration system for proper drainage. Free-floating stands were also used instead of fixed
supports, due to the research nature of the facility and the requirement to test various pipe and
aeration layouts. While the decision between fixed and free-floating pipe supports is up to the
operator, the flexibility of free-floating supports has been incredibly useful at the UNH facility,
especially when making alterations to increase system performance. Fewer holes and anchors in
the concrete also reduces the risk of cracking. However, regardless of support structure, all metal
components touching the aeration network should have a layer of insulation between the metal
and PVC. This will reduce conductive heat losses or points of condensation within the aeration
network.
171
Figure 52: Pipe Stands for Aeration System at UNH Facility
Installing Agrilab Technologies Isobar Unit
Agrilab’s heat recovery system came to UNH on a flatbed truck halfway through the
mechanical room setup. The unit was 30’L * 34.5’’W * 30’’H with six isobars. Because of its
narrow size, it was able to be brought in through one of the side doors of the mechanical room
with a telehandler (Figure 53).
Figure 53: Delivery of Agrilab Technologies Heat Exchange Unit
172
When delivered, the heat-exchange unit was 1300 lbs, (total weight of 3800 lbs after filling the
bulk tank with water and loading the Isobars with refrigerant). The system also came with
supports for the vapor portion of the unit. Supports included two adjustable steel cradle floor
stands and three 24’’ C-shaped steel brackets with a top loop for a ½’’ threaded rod to be
connected to the ceiling studs. Due to differences in facility layout, supports for the much
heavier bulk water tank were not included with the isobar unit. Instead, a support structure was
built ahead of time and was made of 6’’L * 6’’W * 25’’H wooden beams, 2’’ * 10’’ lumber, and
a layer of ½’’ cement board (Figure 54).
Figure 54: Support Structure for Bulk Storage Tank of Water
The layer of cement board was placed in between the water tank and wooden frame of the stand
due to a requirement from the state fire inspector, who was concerned about having water
temperatures in excess of 120°F being in contact with wood. Across the 30’ span of the heat
recovery system was a 4’’ drop, allowing for drainage of condensate and more efficient
circulation of the heat exchange refrigerant.
173
The isobar unit was connected to the primary 8’’ PVC aeration network with a 12’’ to 8’’
flexible Fernco coupling, and a section of SDR-35 8’’ pipe and elbow (Figure 55). SDR-35 was
used for this section as it was leftover material from another project. While the setup between the
Isobar unit and primary aeration network will vary between facilities, it is important that all
systems have at least one flexible coupling in this connection. The flexible coupling(s) at this
joint will allow the system to expand and contract, as it heats up and cools down between
composting batches.
Figure 55: Aeration Line Hookup with Heat Exchange Unit
The Isobar unit was also connected to the primary 4’’ leachate drain through ¾’’ nylon
tubing (Figure 55). The heat exchange system was then attached to the fan (NY Blower 126
CGI). The blower was deliberately placed post heat exchanger, as vapor temperatures and
moisture content are reduced following heat exchange. The blower was attached to the isobar
unit with a 6’’ to 8’’ flexible coupling and a few sections of 6’’ PVC pipe (Figure 56). From the
blower, an 8’’ to 8’’ flexible coupling was used to connect into the exhaust line.
174
Figure 56: Connections to and from Primary Blower
The final hook up to Agrilab’s heat recovery system was connecting the 1’’ copper pipe
(Type L) to the water lines in the bulk tank. From the heat exchange unit, copper pipe was
connected to a network of hose bibs, which were added for various research needs (Figure 57).
From the network of hose bibbs, the hot water supply and return lines were attached to the
underground PEX pipe leading to the milk house (Figure 58).
Figure 57: Hot Water Supply and Return Lines from Isobar Unit
175
Figure 58: Underground PEX Hot Water Supply and Return Lines
Once in the milk house, the underground PEX lines were connected to 1’’ copper pipe
(Type L) that lead to another batch of copper hose bibs to allow for direct hot water removal
from the closed loop system (Figure 59).
Figure 59: Water Lines between Composting Facility and Milk Room
176
The copper pipe was then connected to a plate heater (GEA #FG10X20-20 with 1 ½
threads), where heat is transferred from the hot water line to a separate cold water line coming
from the farm’s well (Figure 60).
Figure 60: Plate Exchanger in Milk House
During this transfer, the 50ºF well water is warmed to 90ºF, before entering the farm’s water
heater where it is heated up another 50-70ºF to serve all the farm’s hot water demands (Figure
61). The range in expected temperature is dependent on how many compost bays are loaded and
their age.
Figure 61: Hot Water Heater Receiving Compost-Heated Water
177
After heat transfer, the cooler water from the closed loop is sent back to the water tank
connected to the isobar unit by a circulation pump (GRUNDFOS UPS26-150SF) (Figure 59 and
Figure 62).
Figure 62: Isobars within the Bulk Storage Tank (unfilled)
Cost Saving Tip # 15: If possible, use PEX pipe to transfer the heated water over copper pipe.
PEX is much less expensive in both material and installation cost. Additionally, the thermal
conductivity of PEX is 0.51 w/mK vs. copper, which is 401 w/mK (Patterson and Miers 2015).
The lower thermal conductivity of PEX means less heat will be lost from the pipe transferring the
heated water.
Cost Saving Tip # 16: If underground insulated PEX pipe is used to transfer the heated water
between buildings, leftover material should be used within the building to connect to the heat
exchange unit. This not only saves material cost, but also reduces labor costs associated with
insulating piping.
178
Installing the Aeration Control System
The aeration control system at the UNH facility was designed and installed by Agrilab
Technologies. The system is run by a Do-More Programmable Logic Computer (PLC), which is
housed in one of two control boxes (Figure 63).
Figure 63: Agrilab's Aeration Control System
This PLC controls an air compressor and Valterra pneumatic gate valves that open and close the
various composting bays (Figure 48 and Figure 64). The system operates by running the fan
24/7, with the PLC controlling when bays open and close.
179
Figure 64: Compressor Running Pneumatic Lines in the Aeration System
Setting up the Aeration Schedule for Heat Recovery
The aeration on/off schedule for each composting bay varies by batch, season, and age of
compost since loading. If a composting batch has a high initial moisture content (>65%), the
aeration schedule for those bays will be on for a longer period of time than if the batch were in
the more optimal 60-65% range. In contrast, the aeration schedule is on for less time for batches
loaded during the winter months, due to cold air intrusion. If too much cold air is pulled into the
pile during the winter, it can freeze the top layer of compost. The length of aeration also depends
on how recently the compost was loaded. New batches require significantly more oxygenation
and a longer aeration on time than piles post thermophilic stage. With those considerations in
mind, the aeration schedule at the UNH facility is controlled by vapor temperatures from each
composting bay. Thermocouples inside each bay header transmit compost vapor temperatures to
a Web Energy Logger (WEL), which graphs bay temperatures every minute. These temperature
readings are used to determine the aeration period based on the temperature trajectory. If the
daily trend is increasing vapor temperatures prior to the off-period for that bay, then the bay is
180
aerated longer. Likewise, if the aeration schedule shows vapor temperatures decreasing within
the same day, the aeration schedule is likely too aggressive and is changed to reduce the on time.
The optimal aeration goal for the UNH operation is to maintain pile temperatures
between 135 – 150°F for as long as possible. During the startup phase, this usually means
aerating more intensively to get the pile oxygenated and prevent temperatures from exceeding
160°F. During the later phases, aeration on time is decreased to ensure that pile temperatures do
not decrease too quickly. Although one may think that having the aeration system on for longer
periods would increase the heat recovery process, this can actually have the opposite effect,
especially during the winter when the incoming air is cold. A second problem with pulling too
much air through the piles is excessive drying, which is a known problem with ASP composting
systems. Likewise, too little aeration results in anaerobic conditions that produce less heat and
increases the composting time. The most effective method for aeration control is careful
monitoring of compost pile and vapor temperatures.
In addition to the length of aeration on time, the specific timing of aeration between bays
is of great importance. What separates a typical aeration schedule in an ASP system to one that is
utilizing Agrilab’s heat recovery unit is that the heat recovery system requires a schedule with
designated aeration zones to pull air through the piles in a cascading effect across the aeration
floor (1/4 bays on at a time for UNH). This method allows the heat exchange unit to have hot
compost vapor blowing against it at all times, instead of having intermittent high-heat loads (all
bays open) with periods without heat (all bays closed). Extended periods of down time within the
aeration schedule causes the aeration system to cool down. This reduces heat recovery from
conductive losses, as the system has to be brought up to temperature again. Heat recovery from
latent heat is also reduced, as compost vapor will condense on the inside surfaces of the aeration
181
network leading to the heat exchange unit. As mentioned earlier, condensation on any surface
other than the isobars represents a loss of thermal energy that could be captured.
Cost Saving Tip # 17: If using a cascading effect for the aeration schedule, the actual aeration
demand and horse power of the fan does not have to be as large compared to aerating several
bays at once. This is advantageous in reducing the initial capital expense of the fan (can
purchase a smaller fan), and also from an operational standpoint (lower HP saves electricity).
Additionally, some of the larger fans suitable for composting operations require three-phase
electric power, which is costly to install if not already on the site.
Testing and Insulating the System
After all the connections were made between the heat exchange unit and the rest of the
mechanical room, two smoke tests were conducted to ensure all the aeration joints were sealed.
The first test involved smoke bombs, where smoke was forced out of unsealed joints by turning
on the fan system and capping the exhaust pipe with a smoke bomb within. The second smoke
test involved incense and was conducted by turning on the aeration system with all the bay
valves closed and searching for joints where smoke from the incense was sucked into the
aeration system. Each test found one bad joint, which were sealed with marine-grade silicone
seal following the test. Ensuring that all joints are sealed properly is crucial, as leaks will reduce
heat recovery and also pose a health concern to workers from the NH3 and bioaerosols in the
compost vapor. In addition to testing the aeration system, the water lines were also tested for
leaks. This test was conducted by filling the water tank and loading the isobars with refrigerant
through their Schrader valves. If performing this test during the winter and the room with the
heat exchange unit is not heated, it is imperative that the aeration system works and compost
182
batches have been loaded and are up to temperature. This will reduce the risk of freezing and
bursting pipes.
After all the leaks were sealed, the entire aeration network (Isobar unit included) was
double-wrapped with a radiant heat barrier (Reflectix® Insulation). This material reflects 97% of
radiant heat, has a combined R-vale of 8.4, and is antimicrobial. Joints were also sealed with foil
tape (Reflectix® foil tape) to ensure a proper seal (Figure 65). The copper pipes were also
insulated with a double layer of standard closed-cell polyethylene foam pipe insulation (R-12).
Following the insulation of individual components (aeration and leachate pipe), the entire
aeration system was enclosed in an insulated box made of dimensional lumber and 4’ * 8’ sheets
of foam board insulation (Lowes 2.5’’ R-10 Insulation). Because of the research and sampling
demands of the system, the insulated box was constructed in a way to allow for entire 4’ * 8’
panels to be removed easily. Small ports for data sampling were also installed.
Figure 65: UNH Aeration System following Insulation
183
COST OF THE UNH HEAT RECOVERY COMPOSTING FACILITY
Before providing the cost of the UNH Heat Recovery composting facility, several
important points need to be made to make sense of the amount paid.
1. This was a university building and was engineered to follow the University’s stringent
codes and long term building requirements.
2. The facility was built for research purposes and required a building that would last for
decades. As a consequence, a pole barn was constructed. Because the barn was contracted
out through the university bid system, it was much more expensive than if a farmer were
to build the barn themselves.
3. Because the technology is relatively new, a tremendous amount of time and money
($21,500) was spent designing the system. Future facilities pulling designs and ideas
from this report will not likely have to go through such a lengthy process.
4. An additional concrete pad at a cost of $54,668 was added to the budget to serve as a
staging area for mixing the feedstocks. Not a requirement for other farms.
5. Typical cost saving strategies utilized at previous sites (owner helping with construction
to reduce labor cost, owner purchasing materials, etc.) were not allowed in our situation.
With the above statements in mind, the total cost for the UNH project was $538,000. A
specific breakdown of cost is provided in Appendix 6. If a facility of similar-size were built
outside of a university/research setting, with a fabric structure, waste blocks for the side walls,
owners doing some of the labor, and purchasing the aeration components themselves, the total
cost could be well under $300,000. A specific breakdown of this estimated cost, along with a
summary of all the previously recommended cost-saving strategies throughout the report, can be
found in (Appendix 1 and Appendix 7).
184
CONCLUDING THOUGHTS
When looking at the economics of building a heat recovery composing facility, it is
important to realize that managing for optimal composting also results in optimal heat recovery.
As stated at the beginning of this report, a specific combination of feedstocks is required to
generate the kind of heat needed for heat recovery. Compost operators that are not capable of
acquiring the right combination of feedstocks in a large enough quantity throughout the entire
year should probably not consider an advanced heat recovery system like the one described in
this report. However, less efficient and costly systems could be used and are described in Smith
and Aber (2016).
A second important consideration when looking at the economics of this type of
operation is that heat capture is just one of several value-added products that will make this type
of operation profitable. Other important economic factors to consider for both farmers and
compost operators include:
Potential Revenue Streams
Sale of compost
Sale of compost leachate
Tipping fees for accepting municipal yard and food waste
Carbon credits (very possible if utilizing a greenhouse to scrub CO2 from waste
vapor)
Sale of crops produced in a greenhouse or high tunnel
Cost Reductions
Reduction in fossil fuels previously used to heat water
185
Reduction in time, fuel, and labor spent spreading more bulky manure on fields
(farmer benefit). Also reduces the problem of smothering grass with wet manure,
which can cause temporary dead spots reducing hay or pasture yields.
Reduction in fly control costs, as the composting process destroys larva and removes
the material the flies can breed in (farmer benefit). Biting flies are known to impact
herd health on dairy farms and reduce milk yields (Dyck et al. 2009).
Non-market Benefits
Reduction in compost bioaerosols and overall farm odor through biofilter usage.
Reduction in greenhouse gas emission through aerobic composting, biofilter use, and
possible scrubbing of CO2 through a high tunnel or greenhouse.
Reduction in nutrient leaching associated with producing a more stabilized product
(compost).
Increased awareness and management of the farm’s waste streams through the
composting operation (have to know the quantities of the various waste feedstocks for
compost recipe building).
The above only represent a few of the economic factors relating to a heat recovery
composting facility. As previously stated, a payback of 4-8 years (not including grants or cost
sharing) has been recognized for this type of system (Agrilab Technologies 2013). For those
already composting with the ASP method, the payback period is even shorter, as the heat
exchange system can simply be attached to the current aeration system. An important point to
note regarding the payback period is that it is dependent on a number of factors. What greatly
impacts the time at which these systems pay for themselves is whether new infrastructure
(compost building, compost storage facility, etc.) and machinery (loaders, screeners, mixers,
186
conveyors, etc.) are required. The quantity of available compost for sale, combined with
proximity to markets is also a major factor that will determine the payback period. In assessing
economic feasibility, all of these considerations need to be made.
It is important to note that the primary goal of this report is to provide food for thought
and to save those interested in the technology capital in engineering and consulting costs, which
are often high for new technologies. The case study in this report represents one possible design
for a heat recovery composting facility. The three other facilities utilizing this technology have
slight variations within the mechanical room and the type of building used to cover the aeration
floor. A summary table outlining the differences (and commonalities) between the four facilities,
along with references related to their construction/building procedures, can be found in
Appendix 2 of this document.
In deciding whether to go forward with a heat recovery composting facility, the authors
encourage the reader to reference portions or even the entire report to policy makers and/or
investors, as it should answer many questions/concerns individuals have regarding this type of
technology. We also encourage all users of this technology to share their ideas with one another
and join the network of composters extracting heat from waste biomass.
187
REFERENCES
Acrolab. 2013. Isobar heat pipe. [Internet]. Available from:
<http://www.acrolab.com/products/isobars-heat-pipes.php>.
Adams, Z. 2005. Understanding biothermal energy. [Master’s Thesis]. The University of
Vermont. 128 p.
Agrilab Technologies. 2013. Heat capture and aerobic composting: Renewable thermal energy
technology. [Internet]. Available from: <http://www.agrilabtech.com/>.
Allain, C. 2007. Energy recovery at biosolids composting facility. BioCycle 48(10): 50-53.
Aquatias, P. 1913. Intensive cultivation of vegetables on the French system. London:
Upcott Gill. 192 p.
Brown, G. 2014. The compost-powered water heater. Woodstock: The Countryman Press. 162 p.
Engineering Toolbox. 2013. Energy content in common energy sources. [Internet]. Available
from: <http://www.engineeringtoolbox.com/energy-content-d_868.html>.
Epstein, E. 2011. Industrial composting: Environmental engineering and facilities management.
Boca Raton, FL: CRC Press. 334 p.
Campbell, J.B., Catangui, M.A., Thomas, G.D., Boxler, D.J., and Davis, R. 1993. Effects of
stable flies (Diptera: Muscidae) and heat stress on weight gain and feed conversion of
feeder cattle. J. Agric. Entomol, 10(3), 155-161.
Carlson, N., and Rapport, J.L. 2011. Economic feasibility of dairy manure digester and co-
digester facilities in the Central Valley of California. Environmental Science Associates
Report 209481.
Carson, J.M., and Dougherty, M. 1997. Post-frame building handbook: Materials, design
considerations, construction procedures. NRAES-1, 78 pgs.
188
Chatterjee, N., Flury, M., Hinman, C., and Cogger, C.G. 2013. Chemical and physical
characteristics of compost leachates: A review. Washington State University Extension,
WA-RD 819.1.
Chen, P., Overholt, A., Rutledge, B., and Tomic, J. 2010. Economic assessment of biogas and
biomethane production from manure. CALSTART white paper, 22p.
Chiumenti, A., Chiumenti, R., Diaz, L., Savage, G.M., Eggerth, L.L., and Goldstein, N. 2005.
Modern composting technologies. Emmaus, PA: JG Press. 111 p.
Ciborowski, P. 2001. Anaerobic digestion of livestock manure for pollution control and energy
production: A feasibility assessment. Minnesota Pollution Control Agency, 1-233.
City Soil. 2013. Bioenergy Projects. [Internet]. Available at:
<http://citysoil.org/projects- bioenergy.html>.
ClearSpan Fabric Structures. 2013. Picture of compost facility using ClearSpan structure.
[Internet]. Available at:
<http://www.clearspan.com/wcsstore/EngineeringServices/ClearSpan/design/images/opti
ons/waste_decontamination_foundations_002.jpg>.
Dyck, E., S. Flack, K. Hoffman, E. Maltby, K. Mendenhall, R. Perry, S. Richards, L. Tikofsky,
and B. Wallis. 2009. The organic dairy handbook: A comprehensive guide for the
transition and beyond. K. Mendenhall, ed. Northeast Organic Farming Association of
New York, Inc., New York, NY.
Fulford, B. 1986. The composting greenhouse at New Alchemy Institute: A report on two years
of operation and monitoring. Cape Code (MA): New Alchemy Institute; Research Report
No. 3.
Georg Fischer Harvel LLC. 2013. PVC Industrial Pipe Temperature Derating. [Internet].
Available at:
<http://www.harvel.com/piping-systems/gf-harvel-pvc-industrial-pipe/schedule-40-
80/temperature-derating>.
Goodrich, P.R. 2005. Anaerobic digester systems for mid-sized dairy farms. The Minnesota
Project.
189
Gorton, S. 2012. Compost Power! Cornell Small Farms Program. [Internet]. Available from:
<http://smallfarms.cornell.edu/2012/10/01/compost-power/>.
Haug, R.T. 1993. The practical handbook of compost engineering. Boca Raton, FL: Lewis
Publishers, 717 pgs.
H.L. Turner Group Inc. 2013. University of New Hampshire aerobic digestion facility Burley-
Demerit Farm Lee, NH. Project Number 3852.
Liebrand, C.B., and Ling, K.C. 2009. Cooperative approaches for implementation of dairy
manure digesters. United States Department of Agriculture, Rural Development Research
Report 217.
Loughberry Manufacturing. 2013. University of New Hampshire 30’ isobar unit. DWG. NO.
B40953.
Loughton, A. 1977. Straw-bale culture of greenhouse crops. In Jensen, M.H. (ed.), Controlled
Environment Agriculture; Proc. Intern. Symp., p 208-215. Tuscon, AZ, 7-8 April 1977.
Tuscon, AZ: University of Arizona.
Misra, R.V., Roy, R.N., and Hiraoka, H. 2003. On-Farm Composting Methods. Food and
Agriculture Organization of the United Nations, Land and Water Discussion Paper 2, 1-
37.
Outdoor Wood Furnace Boiler. 2013. Choosing the right underground insulated pex pipe for
your outdoor wood furnace. [Internet]. Available from:
<http://www.outdoor-wood-furnace-boiler.com/index.htm>.
Pain, I. & Pain, J. 1972. The methods of Jean Pain: Another kind of garden. Draguignan:
Ancienne Imprimerie NEGRO. 88p.
Patterson, J.E., and Miers, R.J. 2015. The thermal conductivity of common tubing materials
applied in a solar water heater collector. [Internet]. Available from:
<http://www.healthyheating.com/HH_Integrated_Design/Week%205/CPRT192002010.p
df>.
190
Reflectix, Inc. 2013. Technical Literature: Performance information for reflective insulations and
radiant barriers. [Internet]. Available from:
<http://www.reflectixinc.com/basepage.asp?PageName=Literature&PageIndex=734>.
Rynk, R., van de Kamp, M., Willson, G.B., Singley, M.E., Richard, T.L., Kolega, J.J., Couin,
F.R., Laliberty, L., Kay, D., Murphy, D.W., Hoitink, H.A.J., and Brinton, W.F. 1992. On-
Farm Composting Handbook. Ithaca, NY: Northeast Regional Agricultural Engineering
Service (NRAES-54). 186 p.
Schlesinger, W.H. 1997. Biogeochemistry: An analysis of global change. San Diego, CA:
Academic Press, 588pp.
Themelis, N. J. (2005). Control of Heat Generation during Composting. Biocycle, 46(1), 28-30.
Tucker, M.F. 2006. Extracting thermal energy from composting. BioCycle 47(8): 38.
Warrenstreet Architects. 2013. Aerobic Digestion Facility. Project Number 10171-0001.
United States Department of Agriculture. 2010. National Organic Program; Access to Pasture
(Livestock); Final Rule. 7 CFR Part 205.
United States Energy Information Administration. 2013. Heating Fuel Comparison Calculator.
[Internet]. Available from: <http://www.eia.gov/tools/faqs/faq.cfm?id=8&t=7>
United States Environmental Protection Agency. 2010. Aerobic digestion capital costs for dairy
farms. AgSTAR Fact Sheet.
United States Environmental Protection Agency. 1994. Composting yard trimmings and
municipal solid waste. Solid Waste and Emergency Response (5305), EPA530-R-94-
003.
United States Environmental Protection Agency. 2013. Greenhouse gas emissions: Methane
emissions. [Internet]. Available from:
http://epa.gov/climatechange/ghgemissions/gases/ch4.html
191
United States Environmental Protection Agency. 2006. Market opportunities for biogas recovery
systems: A guide to identifying candidates for on-farm and centralization systems. 2006.
AgSTAR Program: USEPA, USDA, and DOE Energy and Pollution Prevention, EPA-
430-8-06-004.
United States Environmental Protection Agency. 1994. Composting Yard Trimmings and
Municipal Solid Waste. Office of Solid Waste and Emergency Response, EPA530-R-94-
003.
Wright, P. 2001. Overview of anaerobic digestion systems for dairy farms. Natural Resource
Agriculture and Engineering Service, NRAES-143, 1-15.
Wright, P., Inglis, S., Ma, Jianguo., Gooch, C., Aldrich, B., Meister, A., and Scott, N. 2004.
Preliminary comparison of five anaerobic digestion systems on dairy farms in New York
State. ASAE Annual International Meeting, Paper No. 04-4032.
192
APPENDICES
Appendix 1: Materials List & Estimated Cost of a Similarly-Sized Facility
Item Description Quantity Cost/Unit Store Total Cost
Main Composting Floor
4'' PVC Pipe (10’ sections) 52 $ 23.77 US Plastics Corp $ 1,236.04
4'' PVC Couplings 48 $ 3.35 US Plastics Corp $ 160.80
4'' PVC End Caps 16 $ 4.76 US Plastics Corp $ 76.16
Band-Seal End Cap w/hose Bibb 16 $ 9.48 Amazon $ 151.68
4'' Galvanized Steel Pipe Risers 48 $ 15.90 Grainger $ 763.20
1/2'' Diameter Black Threaded Rod (10’
sections) 15 $ 8.55
Platinum Fire
Supply $ 128.25
1/2'' Diameter Rebar Rod (20’ sections) 29 $ 7.47 Home Depot $ 216.63
4'' Foamular 250 Foam (4''*4'*4' sheet) 88 $ 65.50 Home Depot $ 5,764.00
Cover Plate Plywood Forms
(4' * 8' * 1/2'' sheet) 7 $ 25.98 Lowes $ 181.86
Cover Plate Plywood Forms
(4' * 8' * 3/4'' sheet) 7 $ 32.20 Lowes $ 225.40
Marine Plywood Cover Plates
(4' * 8' * 3/4'' sheet) 7 $ 95.00 GooseBay $ 665.00
Mechanical Room & Aeration System
Bay Headers (Eight Total)
8'' x 4'' Flexible Rubber Coupling 8 $ 43.06 Zoro $ 344.48
4'' Sanitary Tee 8 $ 17.35 SupplyHouse.com $ 138.80
4'' Flexible Rubber End Cap 8 $ 2.72 SupplyHouse.com $ 21.76
4'' x 4'' Flexible Rubber Coupling 16 $ 4.35 SupplyHouse.com $ 69.60
4'' Long Sweep Street 90° Elbow 8 $ 13.45 SupplyHouse.com $ 107.60
4'' Valterra Pneumatic Gate Valve
(Model 6401P) 8 $ 153.26 Amazon $ 1,226.08
8'' x 8'' x 4'' PVC Wye 8 $ 61.75 SupplyHouse.com $ 494.00
Aeration Supply and Exhaust Lines
8'' x 8'' Flexible Rubber Coupling 20 $ 15.35 SupplyHouse.com $ 307.00
8'' Flexible Cap 3 $ 18.86 Drainage Solutions $ 56.58
4'' PVC Sch. 40 Pipe (10’ sections) 2 $ 23.77 Lowes $ 47.54
8'' PVC Sch. 40 Pipe (20’ sections) 10 $ 174.00 Eliminator Systems $ 1,740.00
Oatey 32 oz PVC Primer 3 $ 16.51 Home Depot $ 49.53
Oatey 32 oz Heavy Duty PVC Cement 3 $ 16.20 Home Depot $ 48.60
8'' 90° PVC Sch. 40 Elbow 2 $ 78.95 SupplyHouse.com $ 157.90
8'' Clevis Hanger 2 $ 25.80 Grainger $ 51.60
193
7/8'' Diameter Galvanized Steel
Threaded Rod (6’ sections) 2 $ 82.00 Grainger $ 164.00
Custom Made Pipe Stands for Aeration
Supply & Exhaust Line 18 $ 12.00 Home Depot $ 216.00
8 Gauge Solid Bare Copper
Grounding Wire (200 ft roll) 1 $ 160.00 Home Depot $ 160.00
Aeration Control System
Laptop 1 $ 500.00 Staples $ 500.00
Aeration Control System (Labor +
Materials) 1 $ 7,200.00
Agrilab
Technologies $ 7,200.00
California Air Tools 4610A Ultra
Quiet Air Compressor 1 $ 259.00 Home Depot $ 259.00
1/4'' NITRA Pneumatic Tubing
(1000 ft roll) 1 $ 182.00 Automation Direct $ 182.00
Connection to Isobar
8'' PVC Sch. 40 Vent Tee 1 $ 84.95 SupplyHouse.com $ 84.95
8'' x 8'' Flexible Rubber Coupling 1 $ 15.35 SupplyHouse.com $ 15.35
10'' x 8'' Flexible Rubber Coupling 1 $ 29.14 Zoro $ 29.14
Isobar to Exhaust Line
6'' x 6'' Flexible Rubber Coupling 3 $ 10.15 SupplyHouse.com $ 30.45
6'' PVC 45° Elbow 2 $ 21.95 SupplyHouse.com $ 43.90
6'' PVC Sch. 40 Pipe (10 ft sections) 1 $ 38.49 Lowes $ 38.49
8'' to 6'' Flexible Rubber Coupling 1 $ 16.77 Zoro $ 16.77
1 HP Blower with Speed Control 1 $ 1,775.00
NY Blower (126
CGI) $ 1,775.00
8'' Flexible Coupling 1 $ 15.35 SupplyHouse.com $ 15.35
Drain Line for Exhaust
4'' to 2'' Flexible Rubber Coupling 1 $ 5.99 SupplyHouse.com $ 5.99
2'' x 2'' Flexible Rubber Coupling 2 $ 5.02 Zoro $ 10.04
2'' PVC P-trap with Cleanout 1 $ 9.27 Home Depot $ 9.27
2'' PVC Sch. 40 Pipe (10 ft sections) 1 $ 8.96 Lowes $ 8.96
Biofilter
8'' 90° PVC Elbow 3 $ 55.00 Eliminator Systems $ 165.00
8'' PVC Tee 1 $ 75.00 Eliminator Systems $ 75.00
8'' x 6'' Flexible Rubber Coupling 1 $ 20.00 Eliminator Systems $ 20.00
6'' SDR-35 Cross Tee 1 $ 45.00 Eliminator Systems $ 45.00
6'' SDR-35 6'' Pipe (10 ft sections) 6 $ 35.00 Lowes $ 210.00
6'' x 6'' Flexible Rubber Coupling 6 $ 12.40 Eliminator Systems $ 74.40
6'' SDR-35 90° Elbow 2 $ 10.98 Lowes $ 21.96
6'' Flexible Rubber End Cap 2 $ 10.28 Zoro $ 20.56
Water Lines in Compost Facility
1'' Type L Copper Pipe (10 ft sections) 1 $ 64.15 Grainger $ 64.15
194
1'' Copper Tee 5 $ 4.45 Pex Supply $ 22.25
1'' 90° Copper Elbow 15 $ 2.93 Pex Supply $ 43.95
8 oz Oatey Paste Flux for Soldering 2 $ 4.50 Home Depot $ 9.00
Underground PEX Hot Water Supply
and Return Line (sold per ft) 300 $ 6.00 Outdoor Boilers $ 1,800.00
Campbell 5' Frost-Proof Yard Hydrant 2 $ 83.10 Grainger $ 166.20
Water Lines in Milk House
1'' Copper Pipe Type L (10 ft sections) 12 $ 64.15 Grainger $ 769.80
1'' Copper Tee 6 $ 4.45 Pex Supply $ 26.70
1'' Copper 90° Elbow 27 $ 2.93 Pex Supply $ 79.11
Grundfos UPS 26-150SF Circ Pump 1 $ 550.00 Plumber Surplus $ 550.00
GEA 100 GPM Plate Exchanger 1 $ 1,436.95 Pex Supply $ 1,436.95
Leachate Network
4'' PVC Pipe (10’ sections) 8 $ 23.77 Lowes $ 190.16
4'' - 2'' PVC Wye Reducer 8 $ 8.38 Home Depot $ 67.04
2'' PVC 90° Elbows 16 $ 1.20 US Plastic Corp $ 19.20
2'' PVC S-trap with Waste 8 $ 9.95 Home Depot $ 79.60
2'' PVC Pipe (10’ sections) 1 $ 8.96 Lowes $ 8.96
2'' PVC 45º Elbows 24 $ 1.31 US Plastic Corp $ 31.44
4'' Galvanized Steel Pipe Riser Clamp 21 $ 15.90 Grainger $ 333.90
5/8 '' Threaded Rod (6’ sections) 1 $ 28.25 Grainger $ 28.25
Pre-cast 1500 Gallon Leachate Tank 1 $ 1,100.00 Phoenix Precast $ 1,100.00
Leachate Tank Pump Alarm (Zoeller) 1 $ 74.33 Amazon $ 74.33
4'' PVC from Facility to Leachate Tank
(10’ sections) 2 $ 23.77 Lowes $ 47.54
Portable Semi-Trash Water Pump 1 $ 259.00 Home Depot $ 259.00
Major Capital Expenses
Agrilab Isobar Heat Exchange Unit
w/insulation and Technical Support 1 $ 55,245.00
Agrilab
Technologies $ 55,245.00
Fabric Structure + Installation
(65'W * 70'L) 1 $ 62,400.00 ClearSpan $ 62,400.00
Interlocking Waste Blocks (6' * 2' * 2')
for Two 65' L Walls 2' W and 6' H 65 $ 65.00 Coleman Concrete $ 4,225.00
Concrete Pad and Back Push Wall
+ Forming of Main Composting Floor 1 $ 40,000.00 $ 40,000.00
Subtotal Materials
Cost $ 194,904.20
Additional Costs to be Included
Land (variable)
Site work (variable)
195
Installation of aeration system in
mechanical room
Installation of electrical
Permits (variable by state)
Total Estimated Cost
196
Appendix 2: Summary Specs from other Heat Recovery Composting Sites
Diamond
Hill
Custom
Heifers
Sheldon, VT
Sunset View
Farm
(Schaghticoke, NY)
Jasper Hill Farm
Greensboro, VT
UNH Organic
Dairy
Research
Farm
Durham, NH
Owners Terry and
Joanne
Magnan
Sean and Sandy
Quinn
Andy and Mateo
Kehler
University of
New
Hampshire
Operation Type Raise 1800-
2100 calves
and heifers
Raise 1300 to 2000
calves and heifers +
100 cows
On-farm cheese-
maker with 45
Ayrshire cows, +
replacements
Organic Dairy
with 50 cows,
+ 50
replacements
Isobar® System
Installation
2006 2010 2012 2013
Building Size Two 52’ *
60’ bays +
mechanical
room
130’ * 55’ 120’ * 55’ 96’ * 50’
Aeration Floor
Size
52’ * 60’ +
apron
53’ * 60’ + apron 80’ * 30’ 96’ * 32’
Aeration Zones 4 4 8 8
Aeration Line
Length and
Diameter
60’ of 6’’
diameter pipe,
4 per zone, 16
total
60’ of 6’’ diameter
pipe, 4 per zone, 16
total
30’ of 4’’
diameter pipe, 2
per 6 zones, 4 per
2 zones, 16 total
30’ of 4’’
diameter pipe,
2 per 8 zones,
16 total
Monthly
Feedstock
Tonnage
180-200 per
windrow with
4 contiguous
batches
200-250
60
65
Feedstocks for
Composting
Cow manure
and bedding
from calves
Cow manure,
separated solids, and
bedding from calves
Cow manure,
separated solids,
and bedding
Manure,
bedding, waste
feed hay
Method of
Mixing/Loading
Mix with
vertical mixer
into pile,
followed by
loading with
telehandler
Mix with manure
side slinger into
pile, followed by
loading with front
loader
Mix by unloading
directly into
facility with a
rear-discharge
manure spreader,
followed by piling
higher with tractor
Mix by
unloading with
a rear-
discharge
manure
spreader,
followed by
piling higher
with tractor
Compost
Residence Time
8-26 weeks
(12
Typically)
8-20
(12 Typically)
10-16
(12 Typical)
12-17 weeks
197
Size of Isobar
Unit
6 Isobars 60’
long with a
800 gallon
bulk tank
12 Isobars 42’ long
with 600 gallon bulk
tank
6 Isobars 30’ long
with 300 gallon
bulk tank
6 Isobars 30’
long with 295
gallon bulk
tank
Hot Water Uses Radiant floor
heat for calf
barn (keeps
floor warmer
and more
dry), and heat
milk formula
for calves
Sanitization of
equipment, calf
hutches and
preparing feed
Used as a heater
for three
anaerobic digester
tanks (maintains
digester at 100°F)
Hot water
heating and
sanitization in
milk room at
farm
Average Bulk
Storage Tank
Temperature
120ºF 115ºF 101°F 100ºF
Peak
Temperature in
Bulk Tank
146ºF 129ºF 109°F 110ºF
Final Target
Water Temp
155ºF 110-115ºF 101ºF 170ºF
Average
BTU/hr
Recovery
200,000 150,000 NA NA
Peak BTU
Recovery
200,0000
Btu/hr
195,000 Btu/hr NA 33,800 Btu/hr
Average Farm
Savings from
Just Heat
Recovery
$10,000 $9,200 NA NA
Total Heat
Exchange
Components
Cost
$60,000
$80,000
$39,500
$38,415
Total Project
Cost
$480,000
(Includes
composting
barn with
concrete
aeration floor,
storage area,
mechanical
room,
plumbing
connections
and compost
curing shed.
$819,000
(Includes
composting barn
with concrete
aeration floor,
storage area,
mechanical room,
plumbing
connections and
dewatering/separator
equipment, pumps
and building. Costs
Estimated
$750,000
(Includes
composting barn
with concrete
aeration floor,
storage area,
mechanical room,
plumbing
connections,
greenhouse,
digestion tanks,
liquid biofiltration
$538,000
(Includes
composting
barn with
concrete
aeration floor,
compost and
feedstock
mixing area.
Costs include
design, labor
and materials.)
198
Costs include
design, labor
and
materials.)
include design, labor
and materials.)
cells, vapor
biofilter bed and
manure
dewatering/separa
tor equipment,
pumps and
building. Costs
include design,
labor and
materials.)
199
Appendix 3: UNH Facility Layout
200
Appendix 4: Quantity of Concrete used at UNH Facility
Concrete Pads
o Main composting floor 32’ * 94.33’ * 9.5’’ = 88 yd3
o Internal apron 8’ * 94.33’ * 9.5’’ = 22 yd3
o External apron 4’ * 96’ * 4’’ = 5 yd3
o Mechanical room 10’ * 94.33’ * 4’’ = 12 yd3
Total Concrete for Pads = 127 yd3
Push Wall and Side Frost Walls
o Back push wall footing 6.5’ * 96’ * 12’’ = 23 yd3
o Back push wall 8’ * 94.33’ * 12’’ = 28 yd3
o Side frost wall footing 2’ * 40’ * 12’’= 3 yd3 * 2 = 6yd3
o Side frost wall 8’ * 40’ * 8’’ = 8 yd3 = 16yd3
Total Concrete for Main Composting Room = 73 yd3
Mechanical Room Walls
o Mechanical room wall footing 1 2’ * 32.25’ * 12’’ = 2.4 yd3
o Mechanical room wall 1 6’ * 32.25 * 8’’= 4.8 yd3
o Mechanical room wall footing 2 2’ * 10’ * 12’’ = 0.75 yd3
o Mechanical room wall 2 6’ * 10’ * 8’’ = 1.5 yd3
Total Concrete for Mechanical Room = 9.5 yd3
Concrete Piers
o Front concrete pier footings (2) 2.5’ * 4.5’ * 12’’ = 0.42 * 2 = 0.84 yd3
o Front concrete pier (2) 8’ * 2.5’ * 10’’ = 0.62 * 2 = 1.24 yd3
o Front concrete pier footings (3) 3.67’ * 4.5’ * 12’’ = 0.62 * 3 = 1.86 yd3
o Front concrete pier (3) 3.67’ * 8’ * 10’’ = 0.91 * 3 = 1.82 yd3
o Mechanical room pier footing 2’ * 1’ * 12’’ = 0.07 * 8 = 0.56 yd3
o Mechanical room piers (8) 4’ * 1’ * 8’’ = 1.09 * 8 = 8.72 yd3
Total for Piers = 15 yd3
Total Concrete for Facility = 225 yd3
201
Appendix 5: Diagrams for UNH Pole Barn
(H.L. Turner Group Inc. 2013)
202
(H.L. Turner Group Inc. 2013)
203
(H.L. Turner Group Inc. 2013)
204
(Warrenstreet Architects 2013)
205
(H.L. Turner Group Inc. 2013)
206
Appendix 6: Breakdown of the Cost for the UNH Facility
Item Cost
Preliminary Design Services $ 7,179.00
Detailed Design Submission $ 14,358.00
General Requirements (site costs that
include temp toilets and office trailer
rental, temp telephone and fax
connection/usage, temp electrical,
temp water, travel, office supplies,
cost of superintendent, etc. $ 74,020.00
Sitework $ 68,284.00
Demolition $ 1,730.00
Concrete $ 64,924.00
Metals $ 150.00
Rough Carpentry $ 66,421.00
Finish Carpentry $ 15,316.00
Moisture Protection (roof, insulation
of all water and aeration lines, water
proofing foundations, etc.) $ 35,580.00
Doors, Frames and Hardware
(includes 4 standard doors and 4
20'*20' compost bay doors) $ 21,685.00
Painting $ 150.00
Specialties $ 100.00
Agrilab Technologies Isobar Unit +
Support $ 52,400.00
Mechanical $ 65,550.00
Electrical $ 19,595.00
Overhead and Profit $ 29,214.00
Other $ 1,675.00
Total $ 538,131.00
207
Appendix 7: Recommended Cost-Saving Strategies Found Throughout Report
1. Carefully assess the aeration demands of the system, as the cost of PVC pipe increases
substantially as pipe diameter increases. This is especially true for the difference between
8’’ and 10’’ diameter PVC pipe and fittings.
2. Purchase materials yourself [PVC pipe and fittings, fan(s), flexible couplings, insulation
(rigid foam for concrete sub-base, thermal joints, pipe insulation, etc.), support structures
(clevis hangers, piper riser clamps, threaded rod, etc.), plate exchanger, Isobar heat
exchanger, etc.]. As indicated earlier, ensure a contract is established where you are still
not paying a contractor for these purchases.
3. Ensure the facility is sited properly to reduce the travel distance between the feedstocks
and the composting floor. This is one of the more important cost-saving strategies, as
extra time maneuvering around objects will add substantial cost over the long run.
4. Consider installing a larger leachate tank (≥ 3500 gallons), capable of filling a tanker
truck in one load. This will reduce the number of pumpings and allow for full loads when
transporting and spreading leachate on fields for fertilizer/irrigation.
5. Ensure all possible holes through the concrete are planned ahead of time and receive
sleeves during forming to prevent the need to drill through the concrete.
6. For the primary back push wall, wood can be used for the upper portion of the wall,
reducing the cost of concrete. This cost-saving strategy needs to be looked at carefully
though, as it will require a vapor barrier, and the need to replace the wood in the future.
7. Instead of pouring side walls, investigate the cost of concrete waste blocks. These blocks
are commonly used for side walls for fabric structures. UNH received quotes in the $65 -
$75 range per 2’ * 2’ * 6’ delivered interlocking block (when purchasing a trailer load).
This cost-saving strategy will not only reduce concrete costs, but could significantly
reduce the ground preparation costs as well.
8. If labor costs are high in the area, a cost comparison between rigid foam insulation and
Insul-Tarp is warranted when deciding which insulation to place under the concrete slab.
9. Make the cover plates out of dimensional lumber to save labor cost and increase the
recession on the aeration floor.
10. Do not drill the aeration holes in the PVC pipe until all construction is complete – this
reduces the labor involved in cleaning should the holes get plugged with construction
material.
11. The concrete aprons (internal and external) and mechanical room slab do not have to
have the same thickness as the primary aeration floor, saving concrete costs.
12. Use a high tension fabric structure if possible. This represents one of the greatest cost-
saving strategies, especially if the operator was not planning on building the structure
themselves.
13. If using pneumatic valves to control the aeration system, valves increase exponentially
with an increase in pipe diameter size. Reducing the pipe diameter prior to the valve can
save thousands of dollars in cost, provided the fan is capable of handling the increase in
resistance.
14. Installing PVC tees or wyes at every header and in the adjacent exhaust line allows for
compost vapor recirculation. This method can be used to start bays, especially in the
winter, reducing the time to reach thermophilic conditions. This not only increases the
208
volume that can move through the composting facility, but also reduces cooler air being
sent into the heat exchange system as a pile warms up.
15. PEX pipe should be used over copper whenever possible. Copper pipe has a higher
thermal conductivity, meaning more heat will be lost from the heated water as it travels
from source to sink. PEX pipe is also less expensive to purchase and install.
16. If using an underground PEX pipe, use as much of it as possible and bring it as close to
the isobar unit as possible, reducing the need for insulating additional PEX or copper
pipe.
17. Setting up the aeration schedule in a cascading effect will allow for a smaller horse power
fan. This saves capital initially and operating expenses perpetually from reduced
electricity costs. Many of the higher powered composting fans > 2 HP also require 3-
phase electricity, which is expensive to install if it is not already on site.
209
Appendix 8: Summary Steps for UNH Facility Construction
1. Size Facility - Size according to feedstock quantity, residence time, and available funds.
Remember to size facility with appropriate aeration dead zones, walkways/internal
concrete apron, mechanical room, etc.
2. Ground Preparation – Ensure any potential runoff from facility does not enter
neighboring waterways, as high nitrogen and phosphorus levels could cause
environmental problems/litigation from those downstream.
a. Install underground water lines that go under the concrete footings.
b. Install leachate line and tank.
3. Footing and Wall Forming – Install 4’’ sleeves for the aeration lines on the back push
wall. Install all other sleeves (electrical, thermocouples, etc.).
4. Pouring Walls and Piers – Following pours, bring ground up to grade with fill.
5. Aeration Floor Forming and Insulation - Lay down pad insulation, supports, welded wire
mesh, and pipe risers/rebar rods.
a. Install slab-connecting dowels between aeration floor and internal concrete apron.
b. Add in thermal break to all portions of the facility where the concrete pad touches
the side walls.
c. Lay down aeration lines with PVC going through back wall sleeves and extending
2’ into mechanical room. Cap PVC ends and fill aeration lines with water.
6. Pouring Aeration Floor - Pour around aeration lines first to hold them in place. Ensure
there is a 1% slope across the 3 ft. section of concrete without a cover plate, and at least a
1% slope from the end of the aeration line to the back push wall.
a. After cure, remove PVC flexible end caps from the aeration lines and release the
water. Also remove all the forms (slab-connecting and cover plate).
b. Fill any gaps around the aeration lines sleeves with Leakmaster LV-1 (or similar
product).
c. Install cover plates (do not drill yet, but saw arch across length).
d. Install leachate line against the back push wall in the mechanical room.
7. Slab Forming and Insulation for Internal Apron – Lay down concrete dobies and welded
wire mesh.
a. Ensure insulation is placed around the dowels from the aeration floor to the
internal apron for a thermal break/expansion joint between the two slabs.
b. Install dowels to connect internal apron with external apron.
8. Pouring Internal Apron – Ensure internal apron has 1% slope toward aeration floor for
drainage.
9. External Apron Forming and Insulation – Lay down concrete dobies and welded wire
mesh. Ensure insulation is installed around dowels between internal and external concrete
aprons.
10. Pouring External Apron – Pour concrete with a 1% slope away from the building
11. Forming Mechanical Room Floor – Install forms for the sump pit
12. Pouring Mechanical Room Floor – Ensure floor is sloped to back floor drain/sump pit.
13. Raising the Building - (pole barn, ClearSpan, etc.)
210
14. Aeration Line Holes - Drill ½’’ holes in aeration lines and cover plates once the building
is raised. Ensure all orifices are properly drilled and are clean.
15. Leachate holes – Fill aeration lines with water and assess where pooling occurs. If
aeration holes were drilled at the apex of the pipe, small diameter (1/8’’) will likely be
required by the back mechanical wall. Additional leachate holes may also be warranted if
low spots exist in the aeration lines.
16. Mechanical Room Aeration Setup - Install aeration ductwork to where the Isobar unit
will be installed.
a. Ensure appropriate insulation is placed around all metal components (stands, riser
clamps, clevis hangers, etc.) that come into contact with the aeration network. Do
not insulate the pipe itself yet.
b. Connect all aeration channels to primary leachate network.
17. Install aboveground waterlines (cold and hot) to where the Isobar unit will be housed (do
not insulate the copper pipe yet).
18. Install isobar unit
a. Hook Isobar unit to aeration, exhaust, leachate and water (cold and hot water
supply/return) networks
b. Install aeration control system
c. Smoke test aeration system for leaks
d. Insulate aeration ductwork
e. Fill bulk storage water tank
f. Test water lines for leaks
g. Insulate copper pipe
h. Insulate Isobar unit after system has run/tested for leaks
211
CHAPTER 6: RECOVERING HEAT FROM COMPOSTING AS AN
INNOVATIVE APPROACH TO REDUCE GREENHOUSE GAS
EMISSIONS AND FOSSIL FUEL CONSUMPTION IN THE
AGRICULTURAL SECTOR
ABSTRACT
As nations move forward in developing their mitigation plans to reduce greenhouse gas
emissions as part of the 2015 Paris accord, it will be necessary to develop innovative strategies
for the agricultural sector, which currently looks to anaerobic digestion as the primary mitigation
strategy. However, recovering the heat generated by aerobic microbes during the composting
process serves as an alternative method to mitigate greenhouse gas emissions. This study
describes a commercial-scale composting system capable of extracting the metabolically-
generated heat from microbes during the composting process to warm water for on-farm heating
needs. When the compost heat recovery system was managed for 90°F water with 121 – 133°F
compost vapor, a recovery rate of 293,909 - 811,332 BTU/day was achieved. With regard to
fossil fuel avoidance, this recovery rate was equivalent to an offset of 911 – 2515 gallons of oil
or 1382 – 3814 gallons of propane per year. From a GHG mitigation standpoint, roughly 10 – 28
TCO2E/yr can be avoided if switching from an oil boiler, or 9 – 24 TCO2E/yr can be avoided if
switching from a propane boiler. The composting system also resulted in an additional 278 tons
of carbon being stored in the soil annually, following compost land application.
INTRODUCTION
In late 2015 France hosted the 21st Conference of the Parties to the United Nations
Framework Convention on Climate Change (COP21). The international agreement that came
forth between the 195 member states was to mitigate greenhouse gas (GHG) emissions to
212
prevent global warming from exceeding 2°C above pre-industrial levels (UN COP21 2015). The
proposed mitigation strategy was to reduce GHG emissions, while also enhancing carbon sinks.
To achieve this comprehensive strategy requires participation from all sectors of the economy.
This is especially true for the agricultural sector, which contributes to global warming through
GHG emissions associated with growing crops, raising livestock and the management of their
waste streams. More specifically, the agricultural sector accounts for 19-29% of GHG emissions
globally (Vermeulen et al. 2012).
Of the various mitigation strategies that can be used in the agricultural sector, recovering
and utilizing the heat generated during the composting process may prove to be one of the more
universal and easily achievable strategies to reduce GHG emissions and increase carbon sinks in
both the industrialized and developing world. This is largely due to the low capital investment
and maintenance cost of compost heat recovery systems (CHRS), the ability to utilize any type of
organic waste, and the ease at which systems can be scaled (Smith and Aber 2016). CHRSs
extract and utilize the metabolic heat generated by aerobic microbes as they consume organic
feedstocks. During this process compost pile and vapor temperatures will often peak at 160°F,
while being maintained between 120 – 140°F for several months. This heat can be extracted and
used directly for crop growth and season extension, or can be used to warm water for radiant
floor heating within a greenhouse, farmhouse or adjacent building, maintain temperature of an
anaerobic digester, or serve as a preheater for a standard hot water boiler (Smith and Aber 2016).
Energy savings and GHG reductions are realized from a decline in fossil fuel used for on-farm
heating. The composting process is also considered a GHG mitigation strategy, due to the
avoidance of CH4 production, which has a global warming potential 21x that of CO2 (Brown
2008, Ayalon et al. 2001). The end product (compost) is also a long-term carbon sink when
213
applied to urban, agricultural, or forested lands (EPA 2016, DeLonge et al. 2013, Brown et al.
2012). This means composting systems with heat recovery are capable of global warming
mitigation in terms of emission reduction, emission avoidance, and carbon storage.
Over the years, three primary mechanisms have developed to extract heat from the
composting process. The least costly and least efficient method is via convection, where
composting materials are placed in a trench to heat crops growing in a layer of topsoil above. By
warming the root-zone of the crops, the growing season can be extended by several months in
the spring and fall. This method has its origins in China 2000 years ago (Brown 2014), and was
refined in the early 1600’s by farmers in France, who added glass enclosures to reduce heat loss
from the convecting compost heat below (Aquatias 1913).
The second type of CHRS is conductive-based, where heat from the compost pile itself is
used to heat water. Pain and Pain (1972) pioneered this method by using a 55 ton pile of
brushwood with several hundred feet of water-filled tubing within the pile to warm well water
from 50 – 140°F at a rate of one gallon per minute for 6 months. The heated water was used to
warm a farmhouse and a high tunnel using cast iron heaters. Heat recovery was 47,500 BTU/hr.
Schuchardt (1984) used a similar method to heat a 1184 ft2 greenhouse in Germany, maintaining
the heat exchange water at 86 – 104°F for 9 months, and recovering 11,584 BTU/hr. New
approaches to this method are described in detail in Brown (2014). Conductive systems made of
composting chambers with water jackets have also been used to heat water (Vemmelund and
Berthelsen 1979, Viel et al. 1987). However, conductive systems are not very efficient, as the
cellulosic substances in compost have a low heat conduction value (Verougstraete et al. 1985).
The third mechanism to extract heat from composting is through the use of a heat
exchanger and aeration system that pulls compost vapor from the pile to capture the latent heat.
214
The efficiency of these systems is higher than convective or conductive systems, as the majority
of the energy balance within a composting pile is contained in the water vapor. Jaccard et al.
(1993) found that 81% of the thermal balance from their CHRS was from latent heat, while only
12% and 8% were from the sensible heat in the vapor and sensible heat in the extract material,
respectively. Themelis (2005) found similar results, where 63% of the compost energy balance
was latent heat and 13.4% was in sensible heat. However, the mechanical components of these
systems (aeration fan, heat exchanger, aeration pipes, computer system to run the fan, etc.) make
them more expensive and therefore tend to be more practical for larger composting operations.
The first commercial-scale system of this type was installed at Diamond Hill Custom Heifers in
Vermont, USA in 2007. The CHRS uses Agrilab Technologies heat exchange system. A second
system of this design was also installed at the University of New Hampshire’s (UNH) Organic
Dairy Research Farm in 2012, and is where this research took place.
While recovering heat from composting has been in practice for over two millennia, few
controlled scientific studies have actually been undertaken. Of the limited number of peer-
reviewed articles that exist, a majority of the CHRS have been small-scale experiments
(Vemmelund 1979, Viel et al. 1987, Jaccard et al. 1993, Seki and Komori 1992, Gnanaraj 2012,
Seki et al. 2014), or modeling of commercial-scale systems that are not operational (Di Maria et
al. 2008, Irvine et al. 2010, Rada et al. 2014). As a consequence, a significant gap exists in the
literature involving the feasibility of commercially-viable CHRS. The aim of this research is to
fill this gap, by presenting heat recovery and utilization values from an active commercial-sized
aerated static pile (ASP) heat recovery composting facility. Heat recovery values are then
compared to a fossil fuel and greenhouse gas emission equivalent for the purpose of carbon
pricing.
215
MATERIALS & METHODS
Research Site and Operational Considerations
Research was conducted at the UNH heat recovery composting facility located at the
university’s Organic Dairy Research Farm in Lee, NH. The ASP composting facility processes
cow manure, spent animal bedding (pine wood shavings), and waste feed hay. Monthly batches
containing 150 yd3 (75 wet tons or 30 dry tons) of feedstock are loaded into the facility with a
rear discharge manure spreader (Kuhn Knight 1230). Each batch consists of two bays (75
yd3/bay), with one bay serving as a treatment and one as a control. The compost residence time is
60 days, resulting in an annual composting volume of 3600 yd3 or 1800 wet tons.
Compost Recipe
The composting recipe for this study was kept constant and was a mixture of feedstocks
(40% manure, 40% bedded pack, 10% waste hay, and 10% woodchips by volume) achieving a
carbon-to-nitrogen (CN) ratio of 25-35:1, a moisture content (MC) of 60-65%, and a bulk density
of < 1200 lbs/yd3. A computer-based compost recipe builder from Agrilab Technologies was
used to generate the specific ratio of feedstocks (manure, spent bedding, and waste feed hay)
needed to achieve the desired chemical and physical characteristics that result in high-heat
composting. The input variables for the recipe builder were CN ratio, MC, and bulk density for
each feedstock, with the output being the number of tractor buckets per manure spreader load of
each feedstock to achieve the desired mix. The input values for each variable were based on farm
averages from previous sampling of the feedstocks (data not presented).
Compost Aeration and Vapor Temperature for BTU Estimation
Feedstocks are aerated by pulling oxygen through the piles with a 1 HP blower (NY
Blower 126 CGI), which is connected to a network of perforated pipes located below each
composting bay. There are two 4’’ diameter PVC aeration lines 30 ft in length below each bay
216
with ½’’ holes drilled 6’’ on center. All 16 aeration lines go through the back push wall of the
composting floor and are connected to a single 8’’ diameter PVC aeration line that runs into the
heat exchange unit. The aeration blower is located after the heat exchange unit, pulling air
through the piles, into the aeration channels, through the primary aeration line and into the heat
exchange unit (Figure 66). Exhaust vapor post heat exchanger is then sent through a biofilter
made of woodchips and finished compost.
Figure 66: CHRS Diagram for UNH Facility
The blower is set at a constant airflow (250 CFM), with the amount of aeration on time
per day being adjusted by incoming compost vapor temperatures. Initial aeration on-time varies
by batch, but is managed to allow compost vapor temperatures to reach 150°F. Upon reaching
this temperature, the aeration on time per batch is adjusted weekly to maintain compost
temperatures in the 140-150°F range for as long as possible. Bays are aerated in a cascading
effect, ensuring the heated compost vapor is always being blown against the CHRS.
The aeration control system was designed by Agrilab Technologies and is run by a Do-
More Programmable Logic Computer (PLC). The PLC controls an air compressor and pneumatic
217
gate valves (4’’ Valterra) located at the aeration header of every composting bay. While the
aeration blower runs constantly, the pneumatic valves at every bay open and close depending on
the aeration demands of each batch, as determined by the incoming vapor temperature. Vapor
temperatures are monitored with a Web Energy Logger (WEL), which tracks vapor temperature
on a one minute interval. Aeration on time is adjusted weekly, unless vapor temperatures
increase or decrease sharply (± 10°F) within the same day, indicating too little or too much
aeration.
Heat Exchange System
The CHRS was developed by Agrilabs Technologies and uses six super-thermal
conductor heat pipes (Isobars) that transfer heat from the extracted compost vapor to water
within a 295 gallon heat sink tank (HST). The Isobars are 30 ft in length, with 20 ft contained
within a vapor chamber and another ten feet contained within the HST. As compost vapor passes
over the 20 ft section within the vapor chamber, it condenses and transfers the latent heat of
condensation to a refrigerant within the Isobar. This energy vaporizes the refrigerant, which
travels up the Isobar to the cooler section contained within the HST. Upon reaching the cooler
section, the refrigerant condenses, transferring the latent heat of condensation to the water in the
HST. The condensed refrigerant travels back to the evaporator side through gravity, completing
the cycle.
Heat Recovery Rate
The rate at which heat is exchanged to the 295 gallon HST is dependent on incoming
compost vapor temperature, relative humidity, airflow, aeration on time, and the temperature of
the water in HST. During the research trials, airflow and humidity were kept constant, at 250
CFM and 100% relative humidity respectively. Prior to every heat recovery trial, the aeration
218
system was turned off and the HST drained and refilled with 55°F well water, ensuring the
starting water temperature of each trial was the same. Upon refilling the HST, the aeration
system was turned back on. Incoming vapor temperature and the temperature of the HST were
recorded on a five minute interval, using thermocouples from a HOBO Onset U12-008 4-
Channel External Data Logger. Each heat recovery trial ran for 24 hrs. Heat recovery was
calculated in BTUs, due to the ease at which recovery rates can be calculated:
1 BTU = the amount of energy to raise 1 lb of water 1°F.
System BTU Recovery Rate = ∆THST * LBS WATERHST * TIMEcS
Where: ∆THST = Change in water temperature in the HST since the cold start
LBS WATERHST = 295 gallons * 8.34 lbs/gallon = 2460 lbs
TIMECS = Aeration On-time Since Cold Start (cs)
Heat recovery in BTU/hr were compared to the incoming compost vapor temperature pre-heat
exchanger, to establish a range of heat recovery values by vapor temperature. Heat recovery by
average compost vapor temperature were compared three different ways, which included: 1) the
number of BTUs/hr during the first hour since the cold start, 2) the number of minutes required
to raise the HST to 90°F, which is the target water temperature desired at the farm for the heat
exchange system, and 3) daily BTU recovery if utilizing all the produced 90°F water.
Energy Equivalent to Fossil Fuel
Heat recovery values by compost vapor temperature were compared to energy values if
using fossil fuel to heat the water instead of the CHRS. No. 2 heating oil (138,500 BTU/gallon)
and propane (91,333 BTU/gallon) were used for comparison (EIA 2016). These heating values
were adjusted to account for boiler efficiency, assuming an annual fuel utilization efficiency
(AFUE) corresponding to an Energy Star boiler with 85% efficiency (EIA 2015). This resulted in
219
BTU equivalents of 117,725 BTU/gallon for No. 2 heating oil and 77,633 BTU/gallon for
propane. The number of gallons of oil or propane saved per day was calculated by dividing the
daily BTU recovery if utilizing all the produced 90°F water by the fossil fuel BTU/gallon
equivalent.
Greenhouse Gas Mitigation from Emission Reductions and Carbon Storage
The degree to which greenhouse gas emissions were offset was based on the gallon
equivalent of No. 2 heating oil or propane per day that were avoided by using the CHRS to heat
water to 90°F. Greenhouse gas emission values of 22.4 lbs CO2/gallon for No. 2 heating oil and
12.7 lbs CO2/gallon propane were used for comparison (EIA 2013).
Compost storage was based on EPA’s Waste Reduction Model (WARM), which predicts
a carbon storage of 0.154 tons of CO2 equivalent (TCO2E) per wet ton of organic materials
composted and input into soil (EPA 2015). An annual facility throughput of 1800 wet tons was
used for this calculation.
Statistical Analysis
Temperature readings from the various data loggers (WEL and HOBO) were compiled
and summarized using Microsoft Excel. Data analysis was conducted using JMP® software from
SAS. Linear regression analysis was used to explore the relationship between compost vapor
temperature and the heat recovery rates from the CHRS.
RESULTS & DISCUSSION
Average Compost Vapor and Pile Temperature at Facility
The compost vapor temperatures used in this study came from multiple composting
batches, which varied slightly in their heat production rate over time. The variability in vapor
temperatures over time for each composting batch was not of concern, as only average vapor
220
temperature, pre-heat exchange unit was necessary to establish the relationship between vapor
temperature and heat recovery rate over time. This held true because the relative humidity
(100%) and aeration rate (250 CFM) were constant between batches. The relationship between
compost vapor and pile temperature by compost age for a typical trial batch in this study is
depicted in Figure 67.
Figure 67: Typical Composting Temperature over Time at the UNH Facility
This relationship exhibited a pattern characteristic of the batch loading process, which occurred
at the UNH facility. Following loading, several days of increasing temperature occurred,
followed by a long period of stabilized temperatures near 130°F for the remainder of the trial
period. This pattern was typical for all the batch trials undertaken during this study (Appendix 1).
Heat Recovery Rate
The relationship between average vapor temperature entering the heat exchange system
and the heat recovery rate was significant (P = 0.04 and R2= 0.91) (Figure 68). When the CHRS
was managed for 90°F water with average vapor temperatures of 133°F, a recovery rate of
811,332 BTU/day (33,806 BTU/hr.) was achieved. On a unit weight basis, this was equivalent to
2704 BTU/day/wet ton or 6761 BTU/day/dry ton of compost feedstocks. When managed for
70
90
110
130
150
0 1 2 9 17 23 31 37 44 51 58
Ave
rage
Tem
per
atu
re (F
)
Compost Age (Days)
Average Compost Vapor and Pile Temperatures by Compost Age for Batch 9
Compost Pile Compost Vapor
221
90°F water with average vapor temperatures of 121°F, a recovery rate of 293,909 BTU/day
(12,246/hr) was achieved. This corresponded to a unit weight basis of 980 BTU/day/wet ton or
2449 BTU/day/dry ton (Figure 68).
Figure 68: Compost Heat Recovery Rate vs. Vapor Temperature
When looking at the heat recovery rate, it is important to note that it is based on incoming vapor
temperature and the target water temperature for utilization, assuming relative humidity and
aeration rate are constant. If the same 133°F vapor were used to heat water to 80°F instead of
90°F, the recovery rate would increase from 33,806 BTU/hr to 45,132 BTU/hr (33.5% increase).
Utilizing water at a lower temperature increases the heat recovery rate because the rate of heat
transfer from the CHRS decreases as the temperature of the HST increases. The significant
relationship (P < 0.001) in (Figure 69) illustrates this point, and is due to the smaller temperature
gradient (∆T) between the water in the HST and the Isobars that are heating the water.
y = 42378x - 5E+06R² = 0.9189
0
100000
200000
300000
400000
500000
600000
700000
800000
900000
120 122 124 126 128 130 132 134
BTU
/Day
Vapor Temperature (°F)
Compost Heat Recovery Rate per Day by Average Vapor Temperature when Managing for 90°F Water
222
Figure 69: Heat Recovery Rate by HST Temperature at the UNH Facility
This relationship was described in Delta (2016) and can be illustrated using the following heat
transfer equation: Q = UA ∆TM
Where Q = heat transfer rate (BTU/hr)
U = Overall heat transfer coefficient [BTU/(ft2 °F hr)]
A = heat transfer surface area (ft2)
∆TM = logarithmic mean temperature difference (°F)
The decreasing efficiency of the CHRS as the HST temperature increased is a crucial component
of how the system is managed. In the current system, managing for 90°F HST water with an
average compost vapor temperature of 133°F resulted in a recovery rate of 811,332 BTU/day or
2930 gallons of 90°F water/day, while managing for 80°F water resulted in a recovery rate of
1,083,177 BTU/day or 5310 gallons of 80°F water per day. Therefore, in any CHRS, a decision
needs to be made regarding water need, both in temperature and quantity.
When comparing the heat recovery rates obtained from our system to those reported in
the literature, tremendous variability exists. Tucker (2006) described a large-scale Agrilab
y = -63.497x + 6171.8R² = 0.8881
0
500
1000
1500
2000
2500
3000
50 55 60 65 70 75 80 85 90 95
BTU
Rec
ove
ry/5
Min
ute
Inte
rval
HST Temperature (oF)
Heat Recovery Rate By HST Temperature
223
system located at a 2000 heifer dairy facility in Vermont, capable of recovering 200,000 BTU/hr
from waste bedding and manure. A second large-scale Agrilab system at another 2000 heifer
dairy in New York reported a similar recovery rate of 195,000 BTU/hr (Brown 2014). A third
small-scale dairy farm in Vermont with 240 cows recovered 20,000-30,000 BTU/hr with a
homemade CHRS, using a large insulated box (48 ft long *12 ft wide *10 ft high) with an
aerated floor and heat exchange tubing on the ceiling (Brown 2014). A fourth farm also using an
aeration system and small insulated boxes was able to recover 3813 BTU/hr (Smith and Aber
2016). However, these recovery rates were in contrast to those from conductive-based systems
described in the literature. Pain and Pain (1972), described a system capable of recovering
47,500 BTU/hr when using a 50 ton pile of brushwood with recirculating heat exchange tubbing
within. Schuchardt (1978) used a similar conduction-based method to recover 11,500 BTU/hr.
Finally, Brown (2014) reported recover rates of 10,000 BTU/hr for a small-scale conductive-
based system.
The tremendous variability in heat recovery rates between composting systems was
described in Smith and Aber (2016). In their comprehensive review of 45 different CHRS,
recovery rates were 1796 BTU/hr (s = 1525 BTU/hr) for lab-scale systems, 18,990 BTU/hr (s =
15,644 BTU/hr) for mid-scale systems, and 194,214 BTU/hr (s = 112,295 BTU/hr) for
commercial-scale systems. The variability in heat recovery rates was due to system scale,
biodegradability of feedstocks being composted, type of heat exchange system, composting
method, continuous vs. batch loading, method of reporting energy recovery, model vs
operational data, geographic location and duration of heat recovery (Smith and Aber 2016). Due
to these variables, making system level comparisons is difficult.
224
In addition to the numerous variables that affect heat recovery, a majority of the systems
described in Smith and Aber (2016) reported peak recovery values, assuming that the highest
achievable composting temperature from the system could be maintained indefinitely. Of the 45
systems reviewed, only three (Brown 2014, Irvine et al. 2010, and Allain 2007) described a
CHRS with a range of recovery values based on changing compost parameters (compost age or
hot water utilization). While some systems will be able to maintain peak temperatures (150-
160°F) through the sheer volume of feedstock being processed or via continuous loading, many
composting systems will be sending cooler compost vapor into the heat exchange system from
the beginning of newly loaded batches, or from aerating older batches following the peak
thermophilic stage. At these stages of the composting process, heat recovery values would be
decreasing. Due to this method of reporting, a majority of the heat recovery values reported in
the literature should be viewed at as best case scenarios, unless the authors specifically provide a
range in recovery values based on changing compost temperatures, relative humidity, and how
much hot water can be produced at varying temperatures.
225
Fossil fuel Equivalent
The quantity of fossil fuel energy avoided by using the CHRS to warm water was driven
by incoming compost vapor temperature and the actual utilization of the heated water.
Figure 70 illustrates the gallons of propane and oil avoided if managing the CHRS for 90°F
water and utilizing all the heated water produced per day.
Figure 70: GHRS Energy Saving Equivalent by Average Vapor Temperature
0.00
2.00
4.00
6.00
8.00
10.00
12.00
120 122 124 126 128 130 132 134
Gal
lon
Eq
uiv
alen
t
Vapor Temperature (°F)
Energy Saving Equivalent in Oil and Propane by Average Compost Vapor Temperature if Managing for 90°F Water
Oil Propane
0.00
2.00
4.00
6.00
8.00
10.00
12.00
120 122 124 126 128 130 132 134
Gal
lon
Eq
uiv
alen
t
Vapor Temperature (°F)
Energy Saving Equivalent in Oil and Propane by Average Compost Vapor Temperature if Managing for 90°F Water
Oil Propane
226
At a vapor temperature of 133°F, the CHRS saved roughly 10 gallons of propane or 7 gallons of
oil in heating water per day (Figure 70). This assumes that the system is managed for 90°F water
and all 2930 gallons are utilized per day. Should the hot water demand be less than 2930 gallons
per day, the oil and propane equivalent would decrease. Likewise, if the incoming vapor
temperature was less than 133°F, the oil and propane equivalent would decrease. Additionally,
the final target temperature and quantity of water will alter how many gallons of fossil fuel are
avoided. For example, if the CHRS was on a large farm utilizing 5000 gallons of water per day
and the target temperature was 90°F, 2930 gallons could be heated for free, and another 2070
would have to be heated with another source. If heated with a propane boiler, 604,233 BTUs
(7.78 gallons propane) would be required. Alternatively, the 5000 gallons could be heated to
80°F with the CHRS and only pay for propane to raise the temperature another 10°F. If heated
with a propane boiler this would be require 417,000 BTU (5.37 gallons of propane). By
managing for 80°F water instead of 90°F and using a secondary boiler, the farm would save 2.41
gallons of propane per day. This example illustrates the importance of understanding the hot
water demand of the operation in both temperature and quantity of water to maximize energy
savings.
Of the 45 CHRS described in Smith and Aber (2016), only two reported energy savings
in terms of fossil fuel reduction. Schuchardt (1978) used a large conduction-based heat
exchanger within a mound of brushwood, saving three gallons of heating oil per day over a nine
month period. Alwell (2014) also reported a fossil fuel saving of one gallon per day when using a
CHRS on a small farm in Iowa, USA. The lack of reporting of actual energy savings in the form
of fossil fuel is largely driven by the fact that most of the composting systems described in the
literature are lab-based or are theoretical in nature. Additionally, many of the operating systems
227
that do report energy savings (Day 2014, Brown 2014, Tucker 2006, Beck et al. 1992) report
fossil fuel savings in monetary terms, making it difficult to assess the quantity of fuel avoided,
due to changing energy prices over time.
Greenhouse Gas Mitigation
As with the fossil fuel equivalent, the degree to which the CHRS mitigated GHG
emissions was dependent on incoming compost vapor temperature and the hot water demand of
the system (Figure 71).
Figure 71: Greenhouse Gas Emission Avoidance in Oil and Propane from the CHRS
If the facility managed water at 90°F and had average daily vapor temperatures of 133°F,
emission reductions of 154 lbs CO2 for oil or 133 lbs/CO2 for propane would be avoided per day,
assuming all the hot water was utilized. If these operating conditions were maintained throughout
the year, the system would mitigate 28 TCO2E if replacing oil or 24 TCO2E if replacing propane.
If these emission reductions were included in EPA’s WARM carbon model, an additional 0.014
– 0.016 TCO2E per wet ton of feedstock would have to be added to the 0.15 TCO2E mitigated
per wet ton of compost applied to the land. The addition of the CHRS would result in a 9 – 10%
0.00
20.00
40.00
60.00
80.00
100.00
120.00
140.00
160.00
180.00
120 122 124 126 128 130 132 134
Lbs
CO
2E/
Day
Vapor Temperature (°F)
Greenhouse Gas Emission Avoidance for Oil and Propane by Average Vapor Temperature if Managing for 90°F Water
Greenhouse Gas Emission Avoidance in Oil Greenhouse Gas Emission Avoidance in Propane
228
increase in greenhouse gas mitigation potential compared to traditional composting operations
without heat recovery. Additionally, the system could obtain carbon credits from CH4 avoidance.
A review of composting operations and their greenhouse gas impact by Brown et al. (2008),
found that the greatest effect composting has on GHG emissions is through CH4 avoidance.
However, this review did not contain any CHRS, which would further enhance the GHG
mitigation potential of the composting systems reported.
When comparing mitigation potential between our system and other CHRSs reported in
the literature, few inferences can be made. Of the 45 CHRS described in Smith and Aber (2016),
GHG mitigation was only addressed in one study. Winship et al. (2008) used a model to estimate
that 0.58 TCO2E could be avoided or sequestered per ton of biowaste composted if using their
in-vessel composting system. However, this study based their life-cycle analysis on a proposed
commercial-scale CHRS, modeling potential sequestration vs. using an actual operation utilizing
the heat. This is an important point, as actual heat utilization from a CHRS is quite different than
what the system is capable of producing via modeling.
It is also important to compare composting systems with heat recovery to anaerobic
digestion, which is currently the most common approach to recover energy and mitigate GHG
emissions from farm wastes. A study by Avalon et al. (2001) found that aerobic composting was
the most cost-effective waste management strategy to treat organic waste for GHG emission
reductions, when compared to landfilling with flare, landfilling with energy recovery,
incineration, or anaerobic digestion. This study also found that investment costs of aerobic
composting were 5x less than anaerobic digestion and the mitigation cost per ton of CO2
processed was 4x lower for aerobic composting when compared to anaerobic digestion.
Importantly, this study did not include the added benefit of heat recovery from the composting
229
material itself, which would further enhance this strategy over others. The high cost of anaerobic
digestion was also reported in Wright (2001) and EPA (2006). Both studies found that
economies of scale were required to have a positive return on a digester operation,
recommending individuals not pursue the technology unless they have waste from > 500 cows. A
study by Lansing and Klavon (2012) also looked at 18 small-scale anaerobic digesters utilizing
waste from 100-250 cattle, and found that of the 18 digesters, only two had a positive cash flow.
System Potential and Room for Improvement
It is important to note that the heat recovery rate, fossil fuel equivalent, and GHG
mitigation potential reported in this study represents the low range of the actual capability of the
CHRS. The primary reason for this is that the average vapor temperatures typically achieved
from the farm’s composting feedstocks are lower than other commercial composting systems
utilizing heat recovery units. Tucker (2006) reported average compost vapor temperatures in the
130-160°F range for an ASP composting facility using an Agrilab system in Vermont. Similarly,
Winship et al. (2008) and Irvine et al. (2010) reported average compost vapor temperatures in
excess of 140°F for commercial in-vessel composting operations with heat recovery units. There
are several explanations for why the current system has lower vapor and pile temperatures than
the other commercial CHRS reported in the literature. The first reason is that the current system
loads feedstocks in batches. When feedstocks are loaded in batches, the stored materials have
already begun to compost (Rynk et al. 1992), reducing the total amount of available heat that can
be captured from the material. The primary carbon source at the UNH facility is spent bedding
from a bedded pack barn that is cleaned out twice a year. Bedded pack barns operate by
topdressing the pack with bedding several times a week, allowing the material to accumulate for
a several month period. As bedding accumulates, the material in the lower layers begins to
compost, warming the pack for the herd. At the UNH facility, this material is unloaded in the
230
spring and fall, stored on a concrete pad and mixed with manure when enough has accumulated
to fill a composting bay. Because of this management system, a large fraction of the available
carbon, which is the energy source for the microbes and driver of the heat recovery system, has
already composted in the pack barn or on the concrete pad prior to entering composting facility.
Haug (1993) also reported reduced temperatures from batch loading systems.
A second reason for the lower temperatures found in this study relates to the
biodegradability of the feedstocks used in the composting mix. Because the biodegradability of
the feedstocks in the compost mix is directly related to the quantity of heat that can be produced
from that material (Naylor 1996, Haug 1993), poor degradability results in lower composting
temperatures and heat recovery potential. In the present study, the primary carbon source for the
compost mix is bedding from eastern white pine wood shavings. Pine is used as a bedding source
because the resin acids, terpenes, and phenolic components in the wood are antimicrobial
(Zehner et al. 1986). While ideal for maintaining herd health, the antimicrobial nature of the
material results in poor composting. The poor degradability of pine was also described in Haug
(1993), where a decomposition study of woods incubated in soil for 60 days showed that 9.5% of
the carbon in white pine was released as CO2 from decomposition, compared to 45.1% for nine
different hardwoods. Haug (1993) also described a case where a commercial biosolids
composting facility was having trouble maintaining thermophilic temperatures, due to the use of
pine sawdust in their composting mix. The solution to the problem was switching from the low
energy yielding pine sawdust to hardwood sawdust, which is more degradable and energy rich.
The material switch produced composting temperatures in excess of 176°F, requiring the facility
to increase aeration rates to prevent temperatures from being too excessive (Haug 1993).
231
A final reason for the lower than standard composting temperatures was due to the
elevated bulk density and CN ratio of the composting batches. Because the manure used in the
mixes was not covered, rain events saturate the material, requiring additional carbon in the form
of spent bedding and woodchips to be added to the mix to compensate for the higher bulk
density. A higher CN ratio results in slower composting and reduced pile and vapor temperatures
(Naylor 1996), while a higher bulk density restricts airflow to the pile, also reducing pile and
vapor temperatures.
As a consequence of the lower vapor temperatures, the UNH facility will begin importing
horse manure from another university farm to supplement the current composting recipe with
more energy rich manure and hay bedding. The more energy rich and more degradable material
should increase composting temperatures into the range more typical of other commercial ASP
composting operations. This in turn will increase the heat recovery rate, fossil fuel avoidance and
greenhouse mitigation from the system.
CONCLUSION
As nations look for innovative approaches to meet the goals of the Paris accord,
composting with heat recovery should be one of the strategies used to mitigate GHG emissions
and enhance carbon sinks. This study described a CHRS capable of recovering 293,909 -
811,332 BTU/day with compost operating temperatures in the 121 – 133°F range and managing
for 90°F water. With regard to fossil fuel avoidance, the system has the capability to reduce
fossil fuel consumption by 911 – 2515 gallons of oil or 1382 – 3814 gallons of propane per year.
The system also has the capability to reduce GHG emission by 10 – 28 TCO2E/yr if offsetting
heating costs from an oil boiler or 9 – 24 TCO2E/yr if offsetting heating costs from a propane
232
boiler. The operation also resulted in an additional 278 tons of carbon being stored in the soil
annually, following compost land application.
When looking at the feasibility of recovering heat from composting, it is important to
realize that systems are scalable and can be constructed inexpensively with locally sourced
materials to match the specific demands of the operation, regardless of the wastes being
composted or the geographic location of the system. This is in contrast to anaerobic digestion,
which is costly upfront, has a high maintenance cost, requires more mechanical components, and
requires large volumes of biomass. All these variables may prove to be barriers for smaller
farms, especially in the developing world. The scalability of CHRS allows operations of all sizes,
from small farms processing agricultural wastes, to industrial composting facilities processing a
city’s waste, to have heat recovery as an option. Finally, the end product following heat recovery
is compost, which has the dual benefit of improving soil quality and serving as a long-term
carbon sink.
233
REFERENCES
Aquatias, P. 1913. Intensive cultivation of vegetables on the French system. London: Upcott
Gill. 192 p.
Ayalon, O., Avnimelech, Y., and Shechter, M. 2001 Solid waste treatment as a high-priority and
low cost alternative for greenhouse gas mitigation. Environ. Manage. 27:697-704.
Brown, G. 2014. The compost-powered water heater. Woodstock: The Countryman Press. 162 p.
Brown, S., Miltner, E., and Cogger, C. 2012. Carbon sequestration potential in urban soils. In:
Lal, R., and Augustin, B (eds). Carbon sequestration in urban ecosystems. New York:
Springer. P 173–196.
DeLonge, M.S., Ryals, R., and Silver, W.L. 2013. A lifecycle model to evaluate carbon
sequestration potential and greenhouse gas dynamics of managed grasslands. Ecosystems.
16:962-979.
Di Maria, F., Benavoli, M., and Zoppitelli, M. 2008. Thermodynamic analysis of the energy
recovery from the aerobic bioconversion of solid urban waste organic fraction. Waste
Manage. 28:805-812.
Energy Information Administration (EIA). 2016. Energy units explained. U.S. Department of
Energy. [Internet]. Available from:
http://www.eia.gov/Energyexplained/?page=about_energy_units
234
Energy Information Administration (EIA). 2015. Updated buildings sector appliance and
equipment costs and efficiencies. U.S. Department of Energy, Washington, DC.
[Internet]. Available from:
https://www.eia.gov/analysis/studies/buildings/equipcosts/pdf/full.pdf.
Energy Information Administration (EIA). 2013. Carbon dioxide emissions coefficients. U.S.
Department of Energy, Washington, DC. [Internet]. Available from:
https://www.eia.gov/environment/emissions/co2_vol_mass.cfm
Environmental Protection Agency (EPA). 2015. Waste Reduction Model (WARM) Version 13:
Composting. [Internet]. Available from:
http://www3.epa.gov/warm/pdfs/Composting.pdf.
Fulford, B. 1986. The composting greenhouse at New Alchemy Institute: A report on two years
of operation and monitoring. Cape Code (MA): New Alchemy Institute; Research Report
No. 3.
Gnanaraj, R.A. 2012. Energy recovery from sugarcane press mud. Asian J. Sci. Res. 5(4):185-
195.
Haug, R.T. 1993. The practical handbook of compost engineering. Boca Raton, FL: CRC Press.
717p.
Irvine, G., Lamont, E.R., and Antizar-Ladislao, B. 2010. Energy from waste: Reuse of compost
heat as a source of renewable energy. Int. J. Chem. Eng. 2010:1–10.
Jaccard, L., Lehmann, P., Civilini, M., and de Bertoldi, M. 1993. Yard waste composting with
heat recovery. Compost Sci. Util. 1(3):10-14.
Kalra, Y. P. and D. G. Maynard (1991). Methods manual for forest soil and plant analysis.
Forestry Canada, Northwest Region, Northern Forestry Centre, Edmonton, Alberta.
Information Report NOR-X-319E. 116 p.
Lansing, S., and Klavon, K. 2012. Small-scale anaerobic digestion in the United States: Design
options and financial viability. In: Got manure? Enhancing environmental and economic
sustainability; Proc., p. 177-187. Liverpool, NY, 28-29 March 2012.
235
Naylor, L. M. 1996. Composting. In: Girovich, M.J (ed). Biosolids treatment and management:
Processes for beneficial use. New York: Marcel Dekker, Inc. p 193-269.
Pain, I. & Pain, J. 1972. The methods of Jean Pain: Another kind of garden. Draguignan:
Ancienne Imprimerie NEGRO. 88p.
Rada, E.C., Ragazzi, M., Villotti, S., and Torretta, V. 2014. Sewage sludge drying by energy
recovery from OFMSW composting: Preliminary feasibility evaluation. Waste Manage.
34:859-866.
Robarge, W. P. and I. J. Fernandez. 1987. Quality assurance methods manual for laboratory
analytical techniques. U.S. Environmental Protection Agency and U.S. Forest Service
Forest Response Program, Environmental Research Laboratory, Corvallis, OR. 204 pp.
Robin, P., Ablain, F., Yulipriyanto, H., Pourcher, A.M., Morvan, T., Cluzeau, D., and Morand,
P. 2008. Evolution of non-dissolved particulate organic matter during composting of
sludge with straw. Bioresour Technol. 99:7636–7643.
Schuchardt, F. 1984. Heat loss during composting of sawtimber. Landbauforschung Völkenrode
34(4):189-195.
Seki, H., Kiyose, S., and Sakida, S. 2014. An experimental system for the recovery,
accumulation, and utilization of heat generated by bamboo chip biodegradation using a
small scale apparatus. J. Agri. Met. 70(1): 1-11.
Seki, H. & Komori, T. 1992. Packed-column-type heating tower for recovery of heated
generated in compost. J. Agri. Met. 48(3): 237-246.
Tucker, M.F. 2006. Extracting thermal energy from composting. BioCycle 47(8):38.
United Nations. 2015. Adoption of the Paris Agreement. Conference of the Parties to the United
Nations Framework Convention on Climate Change (COP21). United Nations Office,
Geneva Switzerland. FCCC/CP/2015/L.9/Rev.1. [Internet]. Available from:
http://unfccc.int/resource/docs/2015/cop21/eng/l09r01.pdf.
236
United States Environmental Protection Agency. 2006. Market opportunities for biogas recovery
systems: A guide to identifying candidates for on-farm and centralized systems.
AgSTAR, EPA-430-8-06-004.
Vemmelund, N. & Berthelsen, L. 1979. A note on heat recovery from mechanically aerated
farm-yard manure. Agr. Wastes 1:157-160.
Vermeulen, S. J., Campbell, B. M., and Ingram, J. S. I. 2012. Climate change and food systems.
Annu. Rev. Environ. Resour. 37:195:222.
Verougstraete, A., Nyns, E.J., and Naveau, H.P. 1985. Heat recovery from composting and
comparison with energy from anaerobic digestion. In J.R.K. Gasser (ed.), Composting of
agricultural and other wastes; Proc., p. 135-146.Oxford, UK, 19-20 March 1984.
London: Elsevier Applied Science Publishers.
Viel, M., Sayag, D., Peyre, A., and Andre, L. 1987. Optimization of in-vessel co-composting
through heat recovery. Biological Wastes. 20:167-185.
Washington State University. Organic Farming Systems and Nutrient Management: Calculating
Compost Density. Department of Crop and Soil Sciences. [Internet]. Available from:
<http://puyallup.wsu.edu/soilmgmt/BulkDensity.html>.
Winship, E.A.N., Holmes, D., and Notion, D. 2008. Com- bined heat and composting. In
Rodic-Wiersma, L., Barth, J., Bidlingmaier, I.H.W., de Bertoldi, M., and Diaz, L.F. (eds),
Moving organic waste recycling toward resource management and biobased economy;
Orbit 6th International Conference., p 1451-1463. Wageningen, Netherlands, 13-15
October 2008. Wageningen, Netherlands: Orbit.
Wright, P. 2001. Overview of anaerobic digestion systems for dairy farms. Natural Resource
Agriculture and Engineering Service, RAES-143.
237
APPENDIX
Appendix 1: Average Compost Vapor and Pile Temperatures by Age for Additional Batches
used in the Study
40
60
80
100
120
140
0 1 5 9 16 30 37 41 44 48 54 61 68Ave
rage
Tem
per
atu
re (F
)
Compost Age (Days)
Average Compost Vapor and Pile Temperatures by Age for Bays 7 and 8 from Batch 2
Bay 7 Pile Bay 7 Vapor Bay 8 Pile Bay 8 Vapor
60
80
100
120
140
160
0 3 10 14 17 21 27 34 41Ave
rage
Tem
per
atu
re (F
)
Compost Age (Days)
Average Compost Vapor and Pile Temperatures by Age for Bays 5 and 6 from Batch 3
Bay 5 Pile Bay 5 Vapor Bay 6 Pile Bay 6 Vapor
238
CHAPTER 7: CONCLUDING THOUGHTS ON THE AGROECOSYSTEM
STUDY
This dissertation explored two new approaches to assist dairy farmers in reducing
externalities associated with waste accumulation on their farms. This was accomplished by
finding value in underutilized farm resources and waste streams and recycling them back through
the system or by converting them into revenue generating products that could be exported off the
farm. The two innovative approaches were producing animal bedding from low-grade eastern
white pine using a wood shaving machine, and building and operating a novel composting
system capable of processing farm waste streams, while also recovering the thermal energy for
on-farm hot water heating. An agroecosystem approach was used in this research, connecting the
various components of the system through the theme of waste management. The UNH ODRF
was used as a case study.
The following sections summarize the key findings from this study, as they pertain to the
original research goals and questions outlined in Chapter 1. The practical applications of this
research, limitations, and recommendations for future research are also presented.
Chapter 2: Animal bedding cost and somatic cell count across New England dairy farms:
Relationship with bedding material, housing type, herd size, and management system
The first step in this agroecosystem study was to conduct a survey of New England dairy
farmers, addressing what bedding materials they use, why, the cost, and whether they are
interested in the on-farm production of bedding using a wood shaving machine. Prior to this
study, no such information existed for the New England region, proving to be a knowledge gap
in both the extension and academic literature. The specific research questions asked in this study
were:
239
1. What bedding materials do New England dairy farmers use and why?
2. What percent of dairy farmers experienced increased bedding costs over the last decade,
and how were those costs managed?
3. What is the current annual bedding material cost per cow?
4. Does the bedding type, housing system, farm scale, or management system have an effect
on farmers’ self-reported SCC or bedding cost?
5. Are regional dairy farmers interested in producing animal bedding using a wood shaving
machine as a potential cost-saving and revenue-generating alternative?
Results from this study of 129 organic and conventional New England dairy farmers
found that the primary bedding materials used were sawdust, sand, and wood shavings. From
2003 – 2013 bedding costs increased by 89% for the surveyed population. In relation to other
farming expenses, conventional and organic dairy farmers ranked bedding 4th and 5th most costly
respectively. The cost of sand ($115/cow/yr) and manure solids ($39/cow/yr) were significantly
less than other bedding selections. This was in contrast to wood shavings, which were
significantly more expensive ($250/cow/yr). In terms of housing type, farmers using freestalls
realized significant cost savings in bedding purchases ($99/cow/yr), while farmers using tie stalls
experienced higher bedding material costs ($229/cow/yr). No difference in bedding cost was
observed when comparing management system (organic vs. conventional). However, economies
of scale were observed between small and large dairies, with bedding cost decreasing as herd
size increased. On conventional operations, bedding cost decreased from $272/cow/yr on farms
with ≤ 49 cows to $111/cow/yr on farms with ≥ 200 cows. Organic operations had a similar trend
with cost decreasing from $160/cow/yr on farms with ≤ 49 cows to $101/cow/yr on farms with ≥
200 cows. When looking at producer-reported SCC, no relationship was found with housing type
240
(freestalls, tie stalls, and compost bedded packs), herd size or management system (organic and
conventional). However, farmers using sawdust reported elevated SCC (152,695 cells/mL) when
compared to producers using other bedding materials. When dairy farmers were asked about
their interest in participating in local cooperatives to produce their own bedding with a wood
shaving machine, the majority showing interest were small-scale dairy producers. Likewise,
operators of small-scale dairies were the most interested in purchasing bedding from those
cooperatives.
Prior to this study, research on animal bedding was focused primarily on the relationship
between bedding material and bacterial counts. While this study addressed this issue with the
self-reported SCC, it went one step further by addressing cost. The addition of cost data was a
crucial component missing from the academic and extension literature, as an overwhelming
majority of studies, including this one, found no significant relationship between bedding type
and SCC. This means dairy farmers have the ability to switch bedding types to more cost-
effective materials, if they choose to do so. Importantly, this study provided the necessary
economic data that will allow a dairy farmer to determine whether switching makes economic
sense. This study also verified regional farmer interest in the production of wood shavings using
a wood shaving machine, justifying the purchase of such a machine for the second phase of this
agroecosystem project.
One of the limitations of this study related to the cost of the various bedding materials,
which represented a snapshot of the current market. Had this study been conducted pre-2005,
before the collapse of the new home construction market and subsequent closing of many of the
region’s sawmills, the conclusions regarding woody bedding may have been different. A second
limitation was the use of producer-reported SCC. While the focus of this study was on cost, the
241
values for SCC could have been stronger if a subset of the sample population has BTSCC tested
by the research team for cross-reference. However, Wenz et al., (2007) conducted a
questionnaire in this manner, and found that a majority of producers across the 21 surveyed
states, did not underestimate BTSCC and that the producer-reported SCC was an accurate
representation.
When looking to the future, this study would serve as a great baseline to assess the trend
in the regional bedding market, with regard to cost and material selection. It would be interesting
to assess whether dairy farmer perception of various bedding materials changes over time, as
research continually suggests that proper manure management is more crucial than the specific
bedding material selected with regard to SCC. A follow-up questionnaire sent ten years after the
first would answer many of these interesting questions and assess whether dairy farmers find
economic relief in the future by switching to more cost-effective alternatives.
Chapter 3: Financial viability of producing animal bedding with a wood shaving machine
After confirming interest in producing animal bedding with a wood shaving machine
from regional dairy farmers, the next phase of this agroecosystem project was purchasing a wood
shaving machine. The UNH ODRF was used as a case study, where low quality eastern white
pine trees from the farm’s woodlot were converted into animal bedding. Information from two
years of operating the wood shaving machine were used to create a financial decision model that
allows for site and machine-specific parameters to be used. The specific research questions for
this study were:
1. Is this type of venture economically feasible for the average-sized organic and
conventional dairy farm?
2. What is the minimum size farm operation that could support such a venture?
242
3. How much money could an individual make when using the machine for 10, 20, 30,
or 40 hours per week for an entire year?
4. What are operational considerations that make or break this type of venture?
Results from this study found that the average-sized organic and conventional dairy farm
in New England could not support a wood shaving venture if only producing bedding for on-
farm consumption. In this scenario, operating and capital costs resulted in an annual loss of
$6,844 per year during a five year financing period, with a machine payback period of 11 years.
If using the same parameter values, costs were only covered during the 5 year loan period if the
farm utilized 1420 yd3 of bedding per year, costing $21,300 ($15/yd3). Post loan period, the farm
would save $13,497 per year and receive an implicit wage of $88/hr for one worker. Economies
of scale were found, as increased bedding output spread the capital cost of the machine over
more units of bedding. Model results suggested that larger farms with greater bedding demand
would be more capable of covering costs when compared to smaller farms. However, a
cooperative of locally clustered small farms using a trailer-mounted wood shaving machine
could pool resources to gain the benefits of economies of scale by bringing the machine from
farm to farm when bedding is needed. Model output also found that if the wood shaving venture
was profit-driven, where bedding was exported, annual revenue during the loan period would
range from $36,553 - $186,703 when operating 10 - 40 hours per week. The implicit hourly
wage for one worker would be $70 - $90 per hour. Outside the loan period, annual revenue
increased to $50,050 - $ 200,200, with an implicit wage of $96 per hour for one worker operating
10 - 40 hours per week. However, if operating for 40 hours per week throughout the year, supply
would likely outweigh demand, as enough bedding could be produced for 30 organic dairies or
25 conventional dairies of average size in the New England region.
243
When looking at operational considerations, tax incentives would likely increase the
economics of the venture and the profit margin in each of the scenarios. In contrast, the potential
need for a debarker or drying system, and the reliance of parameter values from machine
manufactures, would all increase the cost of the venture and the values presented in the study.
Prior to this study, the feasibility of using a wood shaving machine to produce animal
bedding was absent from the literature (academic and extension), other than unreliable
information from various manufacture’s websites selling the machines. The financial decision
model filled this gap, by allowing individuals to obtain operator-specific economic data on the
feasibility of the venture. Importantly, the model was created in a manner allowing use by both
farmers and non-farmers.
While robust, one of the primary limitations of the model was that fact that some of the
parameter values need to be obtained from the manufacturer. This could pose a potential problem
if the manufacturer provides best-case scenarios for machine output and cost, which would
overestimate any financial benefits of the venture. This limitation could be avoided if the
individual filling out the model verified parameter values with someone already using that
specific machine and model. A second limitation related to the bedding material itself. The
model calculated the value and cost of green shavings, which are not preferred by a majority of
dairy farmers. Future research should investigate various low-cost drying systems, which could
be incorporated into the model. This would not only strengthen the model output, but would
make the data more applicable to a greater number of stakeholders.
Chapter 4: Heat recovery from composting – A comprehensive literature review
With bedding being produced onsite, the next phase of this agroecosystem study was to
find an innovative approach to handle the large quantities of spent bedding and manure
244
accumulating at the farm. Prior to this study, waste materials were stockpiled in a back field,
where they were left to decompose anaerobically. The piles emitted foul odors and were leaching
nitrate into an adjacent waterway. The solution to the waste management problem was to invest
in a novel composting system capable of recovering heat from the metabolic activity of the
microbes decomposing the material. This approach was selected in an effort to simultaneously
solve the waste management problem on the farm, while reducing fossil fuel imports associated
with warming water. Chapter 4 explored various compost heat recovery systems (CHRS) in a
detailed review, dating from ancient China 2000 years ago to systems being developed in 2016.
The specific research questions were:
1. What is the difference in heat recovery rate by system type (convective, conductive
and latent) and scale (lab, pilot, and commercial)?
2. Are there economies of scale with regard to heat recovery?
3. Are some systems more suitable for commercial applications than others?
4. Is the technology beyond the prototype phase, and can it serve as a mainstream
process to reduce energy costs on farms?
Results from this study, which incorporated 45 different CHRSs from 16 different
countries, found no predictable trend of heat recovery rate by system type or scale. Recovery
rates averaged 1895 kJ/hr for lab-scale systems, 20,035 kJ/hr for pilot-scale systems, and
204,907 kJ/hr for commercial-scale systems. On an energy per weight basis, recovery rates were
1159 kJ/kg DM for lab-scale systems, 4302 kJ/kg DM for pilot-scale systems, and 7084 kJ/kg
DM for commercial-scale systems. Heat recovery rates varied significantly and were dependent
on a combination of the following factors: system scale, type of heat exchange system,
composting method, composting feedstocks, continuous vs. batch loading, model vs. operational
245
data, geographic location, duration of heat recovery, and method of reporting thermal energy
recovery. Preference for vapor exchange was found for active commercial sites covered in the
review, with 86% of the facilities using that form of heat exchange.
Results from this review were noteworthy, as the last detailed review on CHRSs was
conducted by Fulford (1983) 33 years ago. Over this period, new technologies have increased the
efficiency of heat recovery systems. Additionally, this review went beyond the academic
literature, and reported on systems from theses, technical reports, and practitioner and trade
journals, where a majority of compost heat recovery work is published. The review also
described system design for the purpose of replication, which is important for compost
practitioners. The breadth of the systems covered in the review will allow stakeholders to
develop systems that are suitable for their site-specific variables. Finally, this review illustrated
how compost heat recovery technologies are beyond the prototype phase, and can serve as an
innovate method to process wastes, while also recovering thermal energy.
While this review served as a valuable baseline for the current status of compost heat
recovery technologies, it exposed a very important flaw in how heat recovery values have been
reported in the literature. A majority of the literature covered in the review reported peak heat
recovery rates, and did not report a range of values based on compost temperature over time.
Most studies, especially those from modeled commercial sites, assumed thermophilic conditions
indefinitely. With the exception of a few large-scale operations with tremendous quantities of
biomass, this assumption is invalid, as compost temperatures decline over time, and new batches
need time to warm up. Consequently, the primary flaw of this study relates to the heat recovery
rates, which may be an overestimate of the true average value one may be able to attain.
246
A valuable study pertaining to future research would be to write a methods article on how
to calculate and report compost heat recovery values, as no such information exists. If this were
published in an academic journal and then through cooperative extension, there would be a
greater likelihood of more consistent reporting, which does not currently exist.
Chapter 5: Heat recovery from composting: A step-by-step guide on building an aerated static
pile heat recovery composting facility
With information on the various types of CHRS, the next phase of this agroecosystem
study was constructing an ASP heat recovery composting facility to process the farm’s spent
animal bedding, manure, and waste feed. A detailed report on how to construct this system was
created during the construction phase of the project (October 2012 – June 2013), and refined in
2016 to reflect upgrades and modifications to the system. Prior to this research, only one short
article from Tucker (2006) addressed facility design for a commercial-scale heat recovery
composting facility. Consequently, UNH spent over $50,000 in design and consulting fees to
build the facility. The high cost was due to a lack of publically available information on how to
construct this type of facility. The specific research questions for this observational study were:
1. What methods and materials were used to construct the UNH facility?
2. Are alternative methods and materials available and what are their cost?
3. What cost-saving strategies could be used at future composting sites?
The final outcome from this study was an 80 page report, detailing how to design,
construct and reduce costs when building a heat recovery ASP composting facility. The report
will assist compost practitioners in designing their own ASP composting system, reducing the
amount of time and money that would otherwise be spent on engineering and consulting costs.
The step-by-step instructions from the planning phase through project completion, along with the
247
materials and cost list, will provide guidance to operators on how to construct and purchase large
portions of the system themselves, leading to cost savings in the tens of thousands of dollars for
each project. This report can also be used to answer many technology and cost questions that are
pertinent to policy makers and investors who may be considering supporting this type of venture.
This will hopefully expedite the often time-consuming design and funding phases that exist for
these types of projects. Compost operators in the U.S., Canada, U.K and Japan are currently
using information from this report for potential systems in their countries.
The greatest limitation of this study was the site-specific nature of the case study farm.
While three other CHRSs of similar design were described in the appendix, future research
should focus on facility design at other commercial-scale ASP heat recovery composting
facilities. This would provide a more comprehensive view of how these types of facilities can be
designed and constructed. More importantly, this type of study would allow for comparison of
what has and has not worked at each facility. Having managed the UNH system for almost three
years, a large majority of time has been dedicated toward upgrades and troubleshooting the
system. A detailed manuscript on the successes and failures from other facilities would hopefully
reduce the troubleshooting phase at other composting sites, saving compost practitioners time
and money.
Chapter 6: Recovering heat from composting as an innovative approach to reduce greenhouse
gas emissions and fossil fuel consumption in the agricultural sector
The final phase of the agroecosystem study was composting the farm’s wastes (spent
animal bedding, manure and waste feed) in the newly constructed ASP heat recovery composting
facility, and determining the energy and GHG offset from the process. The specific research
questions were:
248
1. What is the heat extraction and utilization rate from a commercial-scale heat recovery
composting facility?
2. What is the fossil-fuel equivalent of the captured heat?
3. To what degree does the heat recovery system offset GHG emissions for the purpose
of carbon pricing?
Results from this study found that the UNH system was capable of recovering 293,909 -
811,332 BTU/day with compost vapor temperatures in the 121 – 133°F range and managing for
90°F water. With regard to fossil fuel avoidance, these compost temperatures would result in an
annual reduction of 911 – 2515 gallons of oil or 1382 – 3814 gallons of propane. From a GHG
mitigation standpoint, roughly 10 – 28 TCO2E/yr would be avoided if switching from an oil
boiler, or 9 – 24 TCO2E/yr would be avoided if switching from a propane boiler. Following
compost application, an additional 278 tons of carbon would be stored in the soil annually.
This study was unique in that it presented heat recovery rates from an active commercial-
scale facility and presented a range in heat recovery values based on incoming vapor temperature
and utilization efficiency. Prior to this research, only Allain (2007) presented compost heat
recovery rates in a range based on utilization. All previous studies on this subject, which are
described in detail in Chapter 4, presented a static heat recovery rate based on optimal
composting conditions existing in perpetuity. This assumption is not realistic for an active
composting site, meaning compost heat recovery rates are largely overestimated in the literature.
This study was also the first to assess the GHG mitigation potential of a CHRS. This is
important, as nations are looking for innovative approaches to meet the goals of the Paris accord
on climate change. Recovering heat from composting could serve as an important method in
achieving this goal, with this study being the first to report on this issue.
249
The primary limitation of this study was the small number of replicates used to calculate
the relationship between compost vapor temperatures with the heat recovery rate and the GHG
mitigation offset. This limitation was due to several renovations and system upgrades to the
CHRS, which made it impossible to compare previous system recovery rates with those of the
upgraded system. Continual sampling on the current system will better assess the relationship
between compost vapor temperatures with the heat recovery rate and GHG mitigation offset.
A second limitation of this study was the limited range in compost vapor temperatures
(118 – 133°F) used to calculate the heat recovery rate and GHG mitigation potential. Ideally, the
vapor range would extend to 160°F, representing the upper limit of compost vapor temperatures.
Importantly, this higher range should cover the optimum operating temperatures of commercial-
scale facilities, which are the most likely candidates for installing this type of system. Future
work at the UNH facility should focus on compost recipes capable of producing vapor
temperatures in the 150-160°F range. This means importing different feedstocks to the farm, as
the current feedstocks are not suited for maintaining temperatures in this range. Future research
should also explore the vapor in the exhaust gas pre and post biofilter, as additional GHG
mitigation potential could be included.
Agroecosystem Study as a Whole
When looking at this agroecosystem study, it was successful in using innovative methods
to bring value to the underutilized resources and waste products at the UNH ODRF. Value was
found in the underutilized farm woodlot through the production of animal bedding, resulting in
two harvests and the production of over 700 yd3 of animal bedding, which was provided to the
UNH equine facility. Research from this phase of the study proved that this type of venture could
be economic for a cooperative of small dairy farmers, or one large operation with a bedding
250
demand exceeding 1420 yd3. Importantly, the decision model was created in a manner not
limiting its use to just dairy operations. Farmers requiring bedding for other animals (swine,
poultry, equine, etc.) or even non-farmers, have the ability to assess whether an animal bedding
venture is profitable when using their site-specific variables.
Substantial improvements to the farm’s waste management system were also recognized
over the course of this study. The odorous and pollution-causing piles of manure and spent
animal bedding that were present during the onset of this study no longer exist, as all waste from
the farm is processed in the ASP heat recovery composting facility. The benefits from the facility
were immediate and included:
1. Reduction of: farm odors, biting fly habitat, CH4 emissions from anaerobic piles of
decomposing material, pasture weed pressure originating from the spread of
unmanaged piles of manure, fossil fuel associated with warming water, labor
associated with moving manure to the lower field, and labor associated with
spreading manure
2. Elimination of: nutrient runoff into neighboring streams, unsightly piles of manure,
and dragging the fields after spreading manure
3. Provided research on the novel system and created a more stabilized nitrogen source
As with the wood shaving venture, publications from this portion of the agroecosystem
study were presented in a manner not limiting the research to just dairy operations. This is
important as the composting facility is a feasible option for any individual, institution, or
municipality processing large volumes of waste. In fact, a majority of the individuals contacting
the research team regarding this type of composting system have been municipal composters
wanting to process food waste.
251
Finally, the overarching goal of becoming a more closed agroecosystem using methods
that can be replicated by stakeholders at other sites was accomplished. The approaches used to
achieve this goal, along with the supporting economic data, were the backbone of this
dissertation and associated publications. This was done in recognition that both the wood shaving
venture and composting facility are capital intensive projects that carry financial risk if
individuals are unaware of all the costs and requirements associated with having a successful
operation. Importantly, this dissertation focused on the real world applications of each of these
strategies in an effort to illustrate how the more sustainable approach can be the more practical
and profitable one as well.
252
REFERENCES
Allain, C. 2007. Energy recovery at biosolids composting facility. BioCycle 48(10): 50-53.
Fulford, B. 1983. Biothermal energy: Cogenerants of thermophylic composting and their
integration within food producing and waste recycling systems. In E. Stentiford (ed.),
Composting of solid wastes and slurries; Proc. Intern. Symp., Leeds, England, 28-30
September 1983. Leeds, England: University of Leeds.
Tucker, M.F. 2006. Extracting thermal energy from composting. BioCycle 47(8): 38.
Wenz, J. R., S. M. Jensen, J. E. Lombard, B. A. Wagner, and R. P. Dinsmore. 2007. Herd
management practices and their association with bulk tank somatic cell count on United
States dairy operations. J. Dairy Sci. 90:3652-3659.