Measuring and Modeling Greenhouse Gas Emissions from Agroecosystems

Raymond L. DesjardinsAgriculture and Agri-Food Canada

Ottawa, ON Canada K1A 0C6Ray.desjardins@agr.gc.ca

Presented at the China Ecological Forum, June 12, 2009

Beijing, China

2

Outline

•Greenhouse gas emissions from agroecosystems

•Flux meauring techniques

•Carbon dioxide

•Methane

•Nitrous oxide

.Modeling techniques

•Carbon dioxide exchange

•Methane emissions

•Nitrous oxide emissions

• Progress towards a more sustainable agriculture

•Greenhouse gas emissions from agroecosystems

•Flux meauring techniques

•Carbon dioxide

•Methane

•Nitrous oxide

.Modeling techniques

•Carbon dioxide exchange

•Methane emissions

•Nitrous oxide emissions

• Progress towards a more sustainable agriculture

3

Atmospheric gases

Our dry atmosphere is made up of a mixture of gases, consisting of:

•78.1% N2;Nitrogen

•20.9% O2; Oxygen

•0.9% Ar; Argon

•0.1% all other gases including the greenhouse gases CO2 (carbon dioxide), CH4 (methane), N2O (nitrous oxide) and O3 (ozone)

Our dry atmosphere is made up of a mixture of gases, consisting of:

•78.1% N2;Nitrogen

•20.9% O2; Oxygen

•0.9% Ar; Argon

•0.1% all other gases including the greenhouse gases CO2 (carbon dioxide), CH4 (methane), N2O (nitrous oxide) and O3 (ozone)

On a wet basis, our atmosphere contains ≈ 1-4% H2O

On a wet basis, our atmosphere contains ≈ 1-4% H2O

4

Atmospheric greenhouse gases (GHGs)

Atmospheric GHGs, including H2O, CO2, CH4, N2O and O3trap longwave radiation and maintain our climate at a temperature that can support life.

Atmospheric GHGs, including H2O, CO2, CH4, N2O and O3trap longwave radiation and maintain our climate at a temperature that can support life.

In recent years, human activities have led to an accumulation of GHGs in the atmosphere (mainly CO2, CH4 and N2O), resulting in an increased temperature on Earth through increased trapping of longwave radiation.

In recent years, human activities have led to an accumulation of GHGs in the atmosphere (mainly CO2, CH4 and N2O), resulting in an increased temperature on Earth through increased trapping of longwave radiation.

This is known as the enhanced greenhouse effect.This is known as the enhanced greenhouse effect.

5

Recent trends in atmospheric greenhouse gas concentrations

Source: IPCC (2007)

SoilSoil

AtmosphereAtmosphere

CH4 CH4 N2O CO2

Livestock ManureCrop

Management Plants

Agricultural GHG Emissions

7

Global Warming Potential (GWP100)

•Greenhouse gases are not equal in their ability to trap radiation and on a mass basis over a 100 year time horizon, are indexed relative to CO2

•CH4 is 21 times more powerful than CO2

•N2O is 310 times more powerful than CO2

•Using the global warming potential (GWP) of each gas, GHG emissions are often expressed as CO2e, or ‘carbon dioxide equivalents’

•Greenhouse gases are not equal in their ability to trap radiation and on a mass basis over a 100 year time horizon, are indexed relative to CO2

•CH4 is 21 times more powerful than CO2

•N2O is 310 times more powerful than CO2

•Using the global warming potential (GWP) of each gas, GHG emissions are often expressed as CO2e, or ‘carbon dioxide equivalents’

310211 2422 ×+×+×= ONCHCOeCO SAR (1996)

298251 2422 ×+×+×= ONCHCOeCO AR4 (2007)

8

Global sources of anthropogenic greenhouse gas emissions: Carbon Dioxide

GlobalCarbon Dioxide emissions from fossil fuel combustion

and cement amount to 7.2 Pg C as CO2 per year (1Pg = 1 billion tonnes).

Carbon Dioxide emissions from fossil fuel combustion and cement amount to 7.2 Pg C as CO2 per year

(1Pg = 1 billion tonnes).

Carbon Dioxide emissions from land use change (e.g. deforestation) amount to 1.6 Pg C as CO2 per year

Carbon Dioxide emissions from land use change (e.g. deforestation) amount to 1.6 Pg C as CO2 per year

Source: Denman et al. (2007)

9

Global sources of anthropogenic greenhouse gas emissions: Methane and Nitrous Oxide

Global

Source: Denman et al. (2007)

Methane emissions from the energy, waste and agriculture sectors amount to about 350 Tg CH4 per

year (1Tg = 1 million tonnes).

Methane emissions from the energy, waste and agriculture sectors amount to about 350 Tg CH4 per

year (1Tg = 1 million tonnes).

Nitrous oxide emissions from all sectors amount to 6.7-8.1 Tg N2O-N per year

Nitrous oxide emissions from all sectors amount to 6.7-8.1 Tg N2O-N per year

10

Agriculture’s contribution to global methane and nitrous oxide emissions

GlobalAgriculture is responsible for approximately 40-50% of

global methane emissions.Agriculture is responsible for approximately 40-50% of

global methane emissions.

Agriculture is responsible for approximately 50-70% of global nitrous oxide emissions.

Agriculture is responsible for approximately 50-70% of global nitrous oxide emissions.

Source: Denman et al. (2007)

11

Agricultural Sources of Methane

Enteric fermentation (digestion) by ruminant animals 86 Tg CH4 per year

China: 8.9 Tg Ch4 per year

Enteric fermentation (digestion) by ruminant animals 86 Tg CH4 per year

China: 8.9 Tg Ch4 per year

Management of animal manures 18 Tg CH4 per year

China: 3.8 Tg CH4 per year

Management of animal manures 18 Tg CH4 per year

China: 3.8 Tg CH4 per year

Rice cultivation 60 Tg CH4 per year

China: 6.0 Tg CH4 per year

Rice cultivation 60 Tg CH4 per year

China: 6.0 Tg CH4 per year

Source: Denman et al. (2007); FAO (2006); Huang et al. (2006)

12

Agricultural Sources of Nitrous Oxide

Manure Management – Direct emissions from manure storage 0.5 Tg N2O-N.

China: 0.15 Tg N2O-N

Manure Management – Direct emissions from manure storage 0.5 Tg N2O-N.

China: 0.15 Tg N2O-N

Source: US EPA (2006)

Agricultural soils – Direct and indirect emissions from application of synthetic/manure fertilizers, crop residue decomposition, waste deposition

by grazing animals and cultivation of organic soils 4.7 Tg N2O-N.

China: 1.2 Tg N2O-N

Agricultural soils – Direct and indirect emissions from application of synthetic/manure fertilizers, crop residue decomposition, waste deposition

by grazing animals and cultivation of organic soils 4.7 Tg N2O-N.

China: 1.2 Tg N2O-N

13

Agricultural greenhouse gas emissions

Global

Methane Emissions

Nitrous oxide Emissions

China

120-180 Tg CH4 1.7-4.8 Tg N2O-N

Source: Denman et al. (2007); People’s Republic of China (1994)

18 Tg CH4 0.5 Tg N2O-N

14

Space and Time Scale of Measurement Techniques

Balloon

1 hour

1 Day

1 Month

1 Year

Aircraft

Laser

1 m2 1 Hectare 1 km2

Representative Area of Measurement

10 km2

Chamber

Rep

rese

ntat

ive

Tim

e of

Mea

sure

men

t

Tower

15

Non-Flow Through Non-Steady State (NFT-NSS) chambers: principles of operation

Time since deployment

Gas

co

ncen

tratio

n in

he

ad s

pace

•Insert collar into soil, affix chamber to collar

•Gas accumulates in head space, no replacement of air

•Gas concentration in the chamber rises continually

•Sample periodically, typically at intervals of a few minutes andfor periods of 15-30 minutes

•Gas samples returned to the lab and analyzed with e.g. gas chromatography

•Most common type of chamber because large concentration change is possible, mechanically simple, no need for power and no need for gas analysis on-line

•Insert collar into soil, affix chamber to collar

•Gas accumulates in head space, no replacement of air

•Gas concentration in the chamber rises continually

•Sample periodically, typically at intervals of a few minutes andfor periods of 15-30 minutes

•Gas samples returned to the lab and analyzed with e.g. gas chromatography

•Most common type of chamber because large concentration change is possible, mechanically simple, no need for power and no need for gas analysis on-line

collar

chamber

16

Flux calculation:

Fg = flux density of gasV = volume of head spaceA = area of land enclosedC = gas concentration in head spacet = time

Flux calculation:

Fg = flux density of gasV = volume of head spaceA = area of land enclosedC = gas concentration in head spacet = time

⎟⎠⎞

⎜⎝⎛×⎟

⎠⎞

⎜⎝⎛=

dtdC

AVFg

Non-Flow Through Non-Steady State (NFT-NSS) chambers: Closed Chambers

Because the head space is small and confined, the resolution for these measurements are much higher than using any other technique.

Because the head space is small and confined, the resolution for these measurements are much higher than using any other technique.

17

y = 0.2024x + 0.844R2 = 0.9608

y = -0.006x2 + 0.3465x + 0.4868R2 = 0.9959

0

1

2

3

4

5

6

0 5 10 15 20 25

Deployment time (min)

N2O

con

cent

ratio

n (祄

ol m

ol-1

)

y = 0.2024x + 0.844R2 = 0.9608

y = -0.006x2 + 0.3465x + 0.4868R2 = 0.9959

0

1

2

3

4

5

6

0 5 10 15 20 25

Deployment time (min)

N2O

con

cent

ratio

n (祄

ol m

ol-1

)dc/dt in non flow-throughNSS chamber

- 8%

Concentration Change Over Time in ClosedChambers

18

y = 0.2024x + 0.844R2 = 0.9608

y = -0.006x2 + 0.3465x + 0.4868R2 = 0.9959

0

1

2

3

4

5

6

0 5 10 15 20 25

Deployment time (min)

N2O

con

cent

ratio

n (祄

ol m

ol-1

)

y = 0.2024x + 0.844R2 = 0.9608

y = -0.006x2 + 0.3465x + 0.4868R2 = 0.9959

0

1

2

3

4

5

6

0 5 10 15 20 25

Deployment time (min)

N2O

con

cent

ratio

n (祄

ol m

ol-1

)

- 25%

dc/dt in non flow-throughNSS chamber

Concentration Change Over Time in ClosedChambers

19

y = 0.2024x + 0.844R2 = 0.9608

y = -0.006x2 + 0.3465x + 0.4868R2 = 0.9959

0

1

2

3

4

5

6

0 5 10 15 20 25

Deployment time (min)

N2O

con

cent

ratio

n (祄

ol m

ol-1

)

y = 0.2024x + 0.844R2 = 0.9608

y = -0.006x2 + 0.3465x + 0.4868R2 = 0.9959

0

1

2

3

4

5

6

0 5 10 15 20 25

Deployment time (min)

N2O

con

cent

ratio

n (祄

ol m

ol-1

)

- 38%

dc/dt in non flow-throughNSS chamber

Concentration Change Over Time in ClosedChambers

20

y = 0.6153x + 396.62R2 = 0.9964

y = -0.0005x2 + 0.7369x + 391.61R2 = 0.999

350

370

390

410

430

450

470

490

510

530

550

0 50 100 150 200 250

Deployment time (s)

Hea

dspa

ce C

O2 c

once

ntra

tion

(祄ol

mol

-1)

y = 0.6153x + 396.62R2 = 0.9964

y = -0.0005x2 + 0.7369x + 391.61R2 = 0.999

350

370

390

410

430

450

470

490

510

530

550

0 50 100 150 200 250

Deployment time (s)

Hea

dspa

ce C

O2 c

once

ntra

tion

(祄ol

mol

-1)

dc/dt in flow-through NSS chamber

- 17%

dC/dt

Concentration Change Over Time in Open Chambers

21

Criteria for reliable soil flux measurements

Rochette and Eriksen-Hamel (2007) evaluated the quality of soil N2O emissions that have been collected using closed chambers and has suggested the confidence level in 50% of recent (2005-2007) N2O flux measurements is low or very low owing to poor methodologies or incomplete reporting. They proposed the following series of requirements to ensure a minimum standard and confidence in chamber measurements:

Rochette and Eriksen-Hamel (2007) evaluated the quality of soil N2O emissions that have been collected using closed chambers and has suggested the confidence level in 50% of recent (2005-2007) N2O flux measurements is low or very low owing to poor methodologies or incomplete reporting. They proposed the following series of requirements to ensure a minimum standard and confidence in chamber measurements:

1. Use insulated and vented base and chamber design2. Avoid chamber heights lower than 10 cm3. Have a minimum collar insertion depth of 5 cm4. Avoid plastic syringes for sample storage5. Take a minimum of 3 samples, including 1 at time t=06. Test non-linearity of changes in headspace gas concentration

1. Use insulated and vented base and chamber design2. Avoid chamber heights lower than 10 cm3. Have a minimum collar insertion depth of 5 cm4. Avoid plastic syringes for sample storage5. Take a minimum of 3 samples, including 1 at time t=06. Test non-linearity of changes in headspace gas concentration

22

Mass Balance Technique – Micrometeorological Mass Difference approach

Flux from within the enclosed source area can be calculated as the difference between the total gas fluxes across the upwind and downwind boundaries of the space, as follows:

Where, = the gas concentration across each boundary= wind vectors perpendicular to the boundary

Flux from within the enclosed source area can be calculated as the difference between the total gas fluxes across the upwind and downwind boundaries of the space, as follows:

Where, = the gas concentration across each boundary= wind vectors perpendicular to the boundary

( ) ( )dxdzvuF zgzgzzgzg

z x

zMMD .1.3.2.40 0

ρρρρ −+−= ∫ ∫

ρvu,

However, the MMD technique is limited by the fact that it can only be applied at a scale of perhaps 10’s of meters, it computationally demanding and it requires a significant amount of set up.

However, the MMD technique is limited by the fact that it can only be applied at a scale of perhaps 10’s of meters, it computationally demanding and it requires a significant amount of set up.

23

1

23

4

u vU

Mass Balance Technique – Micrometeorological Mass Difference approach

Anemometer tower

Enclosed source of

emissions

PumpIntegrator

Site selector valve

Trace gas analyzer

z

x

24

Instrumentation

Open-path lasers and retroreflectorOpen-path lasers and retroreflector

• Gas detector (NH3, CO2, CH4 and H2S) for area sources• On board calibration cell and datalogger• Narrow wavelength minimizes interference with CO2 and H2O• Path length up to 1000 m (depends on target gas)• Detection limit for CH4 of 2.0 ppm-m• e.g. Boreal Laser GasFinder, PKL Spectra-1

• Gas detector (NH3, CO2, CH4 and H2S) for area sources• On board calibration cell and datalogger• Narrow wavelength minimizes interference with CO2 and H2O• Path length up to 1000 m (depends on target gas)• Detection limit for CH4 of 2.0 ppm-m• e.g. Boreal Laser GasFinder, PKL Spectra-1

Boreal GasFinder PKL Spectra-1

25

Mass Balance Technique – Modified MicrometeorologicalMass Difference approach

•Simplified approach, does not require air sampling on all four sides, only upwind and downwind

•Width of laser line should be at least 6 times the width the of the source area

•Limited by wind direction, which should not be more than 45º to the laser lines

•Simplified approach, does not require air sampling on all four sides, only upwind and downwind

•Width of laser line should be at least 6 times the width the of the source area

•Limited by wind direction, which should not be more than 45º to the laser lines

Laser Tower Reflector Tower

26

Mass Balance Technique – Modified MicrometeorologicalMass Difference approach

U

Source of emissions

Upwind laser reflectors

Downwind laser reflectors

Downwind open-path lasers

Upwind open-path lasers

27

If a laser based measurement technique is used, such that the width of the laser line is much greater than the width of the plume of gas being measured and that the wind speed at any given height for all points along the laser line is uniform, then

Where,

= the length of the laser path

= the wind speed normal to the boundaries

= the scalar concentration and the subscripts d and b denote the downwind and background, respectively

If a laser based measurement technique is used, such that the width of the laser line is much greater than the width of the plume of gas being measured and that the wind speed at any given height for all points along the laser line is uniform, then

Where,

= the length of the laser path

= the wind speed normal to the boundaries

= the scalar concentration and the subscripts d and b denote the downwind and background, respectively

( )dzUXFZ

ggng bzdz∫ −=0

,,ρρ

X

nU

gρ

Mass Balance Technique: Integratedhorizontal flux method

28

Recovery from Synthetic Gas Release

0

1

2

3

4

5

6

7

0 20 40 60

Recovered methane (mg CH4 m-1 s-1)

Heig

ht (m

)

0

1

2

3

4

5

6

7

0 20 40 60

Recovered methane (mg CH4 m-1 s-1)

Heig

ht (m

)

Measured flux of CH4 and standard errors as a function

of height based on three releases of 79 mg CH4 s-1

97% recovery

29

The MMD technique was tested using a synthetic tracer (CH4) release to simulate on farm release by cattle.

The MMD technique was tested using a synthetic tracer (CH4) release to simulate on farm release by cattle.

Testing the MMD technique with a synthetictracer release

Methane recovery slightly exceeded methane release and was foundthat it could identify 10% changes in the rate of emission, provided the rate of emission was greater than about 40 mg CH4 s-1, equivalent to roughly 10 dairy cattle.

Methane recovery slightly exceeded methane release and was foundthat it could identify 10% changes in the rate of emission, provided the rate of emission was greater than about 40 mg CH4 s-1, equivalent to roughly 10 dairy cattle.

Source: Desjardins et al. (2004)

30

CQ

• Calculate trajectories upwind from C, using e.g. the Windtrax model*• Efficient & simple to find the rate of emission, Q from an area source• C-Q relationship given by “touchdowns”

• Calculate trajectories upwind from C, using e.g. the Windtrax model*• Efficient & simple to find the rate of emission, Q from an area source• C-Q relationship given by “touchdowns”

Trajectories affected by average wind and turbulence, i.e., different touchdown pattern during day (unstable) and night (stable).

Trajectories affected by average wind and turbulence, i.e., different touchdown pattern during day (unstable) and night (stable).

windwind

touchdowntouchdown

sourcesource

Micrometeorological tools: bLS modelling

*See: http://www.thunderbeachscientific.com/windtrax.html for more details

31

+ simple measurements + no operation disruption+ total emissions + remote measurement+ simple measurements + simple measurements + no operation disruption+ no operation disruption+ total emissions + total emissions + remote measurement+ remote measurement

Inverse Dispersion Modeling

• Dispersion model relates concentration C to emission rate Q (“C-Q relationship” ) for prevailing winds

• Measure C then infers Q

• Dispersion model relates concentration C to emission rate Q (“C-Q relationship” ) for prevailing winds

• Measure C then infers Q

Advantages

Wind CQ (g s-1)

32

Open path lasers and the backwards Lagrangian Stochasticmodelling technique to estimate CH4 emissions

Advantages:

•Non-invasive measurement technique

•Simple measurement set-up (one open path laser, one sonic anemometer)

•Effective for a range of emission sources (e.g. manure tank, beef feedlot, whole farm)

•Non-invasive measurement technique

•Simple measurement set-up (one open path laser, one sonic anemometer)

•Effective for a range of emission sources (e.g. manure tank, beef feedlot, whole farm)

33

Open path lasers and the backwards Lagrangian Stochasticmodelling technique to estimate CH4 emissions

Disadvantages:

•Ineffective during light wind conditions and periods of extreme stability

•Obstructions to wind flow must be accounted for by placing the lasers downwind about 10-20 times the height of the wind obstruction

•Open path lasers are not available for all gas species of interest (e.g. N2O) therefore this technique cannot be used for such gases

•Ineffective during light wind conditions and periods of extreme stability

•Obstructions to wind flow must be accounted for by placing the lasers downwind about 10-20 times the height of the wind obstruction

•Open path lasers are not available for all gas species of interest (e.g. N2O) therefore this technique cannot be used for such gases

34

Micrometeorological tools: bLS modeling

simL

bLbLS QC

CCQ

)/()( −

= ⎟⎟⎠

⎞⎜⎜⎝

⎛= ∑∑

= 01

211)/(wNP

QCP

isimL

Using a line measurement:

where,QbLS = Estimated emission rate

Q = Uniform but unknown emission rate

CL = Line concentration

Cb = Background concentration

P = Point Concentrations evenly spaced along the path

N = Total number of particles released at each point

wo = Vertical touchdown velocities

* The inner summation refers only to touchdowns within the source

In practice, C is measured along a line with open path lasers

In practice, C is measured along a line with open path lasers

Wind speed and direction is measured using a sonic anemometer and friction velocity, u*, surface roughness, z0 and Obukhov stability length, L are calculated, from which the vertical touchdown velocity, w0 can be estimated.

Wind speed and direction is measured using a sonic anemometer and friction velocity, u*, surface roughness, z0 and Obukhov stability length, L are calculated, from which the vertical touchdown velocity, w0 can be estimated.

35

bLS Modeling Software: WindTrax

•WindTrax is a software tool for simulating short-range atmospheric dispersion•Operates in backward and forward Lagrangian Stochastic modes to estimate the concentration and downwind concentration for ground level sources•Operates with point or line averaged concentrations

•WindTrax is a software tool for simulating short-range atmospheric dispersion•Operates in backward and forward Lagrangian Stochastic modes to estimate the concentration and downwind concentration for ground level sources•Operates with point or line averaged concentrations

WindTrax can be downloaded from: http://www.thunderbeachscientific.com/windtrax.html

Graphical display to enter instrument and emissions source locations.

Graphical display to enter instrument and emissions source locations.

36

Wind Direction

Open path laser

Upwind Reflector

Background CH4

concentration (Cb)

Estimate CH4 concentration in air using open path lasers

Open path laser

Downwind Reflectors

Downwind CH4concentration

(CL)

Source of gas emissions: Pure compressed CH4 released through a grid

Experimental setup for pure CH4 release to test the bLS technique

37

Testing the bLS technique under complex conditions: Methane release in a barn

h = 6mGao et al. 2006. Evaluating the backward Lagrangian Stochastic (bLS) technique for estimating methane emissions from a barn. In preparation.

0

0.2

0.4

0.6

0.8

1

1.2

5h 10h 15h 20h 25h 30h

Fetch to building height ratio (D:h)

QbL

S/Q

Downwind distance from the barn at which measurements were made (m)

38

Measuring methane emissions from beef feedlot

Canada

Study site:

•Near to Vegreville, Canada

•Beef feedlot: 17,000 head

Study site:

•Near to Vegreville, Canada

•Beef feedlot: 17,000 head

Source: Van Haarlem et al. (2008)300 m

39

backwards Lagrangian Stochastic modelling: CH4emissions from whole farm feedlots

Diurnal Cycle in CH4 emissions, corresponding with feeding times

Diurnal Cycle in CH4 emissions, corresponding with feeding times

Source: Van Haarlem et al. (2008)

Average daily emission rate: 320 g CH4 animal-1 d-1

The CH4 emission factor estimated using the IPCC methodology is

approximately 240 g CH4 animal-1 d-1

Average daily emission rate: 320 g CH4 animal-1 d-1

The CH4 emission factor estimated using the IPCC methodology is

approximately 240 g CH4 animal-1 d-1

EFCH4 = GEI × YmEFCH4 = GEI × Ym

40

Reynolds:

Flux:

,xxx ′+= ,0=′x yxyxxy ′′+=.

qwqwwqF ′′+==

.

x

x′

x0=w

assuming stationarity and horizontal homogeneity!

The Eddy Covariance Method

0 10 20 30 40

Time (with tower) or distance (with aircraft)

verti

cal w

ind

velo

city

scal

ar

vertical wind velocityscalar

Measuring Changes in Soil Organic Carbon

CO2 - flux data have the potential to provide high resolution measurements of changes in C sequestration on the scale of hectares.

CO2 - flux data have the potential to provide high resolution measurements of changes in C sequestration on the scale of hectares.

e.g. ChinaFlux; Carbo Europe;

AmeriFlux; Fluxnet - Canada

42

CO2

Soil organic matter

product

energy

Ecosystemboundary

0.1

1.5

1.5

1.5

4

7

The carbon cycle in agroecosystems

43

Measurement of Soil C Gain

0

10

20

30

40

50

60

70

80

Initial

Increase with improved management

VariabilityAnalytical Spatial

Soil

C (M

g C

ha-

1 )

44

Flux Underestimation

From the basic energy balance equation,

Qn – QG – ΔQS = QE + QH

However, experimentally it has generally been found that

Qn – QG – ΔQS > QE + QH

45

Flux Underestimation – A persistent Problem

Therefore, the eddy covariance method generally underestimates the turbulent energy fluxes by 5-30 % as

compared to available energy

Dryer

Dz ~ 0.5 -1m

TDL resolution ~ 20 ppt

10 s/level - 10 Hz

Air sample under vacuum

pump

Tower-based N2O Flux

Small Dewar~12-16 h

=> vibration - free environment Sample cell 1.5m long

Tunable Diode Laser

48

The aerodynamic method, stable and unstableconditions

The aerodynamic method can be extended to stable and unstable atmospheric conditions as follows:

where, ψ1 and ψ2 are corrections for stability. In stable conditions,

In unstable conditions,

and

Where,

The aerodynamic method can be extended to stable and unstable atmospheric conditions as follows:

where, ψ1 and ψ2 are corrections for stability. In stable conditions,

In unstable conditions,

and

Where,

( )( )( ) ( )[ ] ( ) ( )[ ]⎥

⎦

⎤⎢⎣

⎡−−⎟⎟

⎠

⎞⎜⎜⎝

⎛⎥⎦

⎤⎢⎣

⎡−−⎟⎟

⎠

⎞⎜⎜⎝

⎛−−

=

12221

21121

1

2

12122

lnln zzzzzz

zz

CCuukFg

ψψψψ

Lz521 −==ψψ

( ) ( )2

tan22

1ln2

1ln2 12

1πψ +−⎥

⎦

⎤⎢⎣

⎡ ++⎥⎦

⎤⎢⎣⎡ +

= − xxx ( )⎥⎦

⎤⎢⎣

⎡ +=

21ln2

2

2xψ

41

161 ⎟⎠⎞

⎜⎝⎛ −=

Lzx

49Canada

-50

0

50

100

150

200

N 2O

Flu

x (n

g N 2

O m

-2 s

-1)

Feb 1 Mar 1 Apr 1 May 1

Nitrous oxide emissions measured in 1996 over a soybean field in Ottawa Canada, using a tower-based system

50

Tower based N2O flux estimates over corn, 2001

-50

0

50

100

150

200

250

300

350

400

15-Mar 25-Mar 4-Apr 14-Apr 24-Apr 4-May 14-May 24-May 3-Jun 13-Jun

N2O

Em

issi

ons

(g N

2O-N

ha-1

d-1)

Tower - Corn

Canada

Source: Desjardins et al (2009)

51

Tower based N2O flux estimates over spring wheat, 2004

-50

0

50

100

150

200

250

15-Mar 25-Mar 4-Apr 14-Apr 24-Apr 4-May 14-May 24-May 3-Jun 13-Jun

N2O

Em

issi

ons

(g N

2O-N

ha-1

d-1)

Tower - Spring wheat

Canada

Source: Desjardins et al (2009)

52

Aircraft Instrumentation for Regional Flux Measurement

53

Comparing Tower and Aircraft Flux Measurements

QE QE QE

QH QH QH

SGP project 1997

SGP project 1997

USA

54

Relaxed Eddy Accumulation (REA)

• Alternate to eddy covariance technique to measure fluxes of trace gases for which there are no fast-response analyzers

• Air samples from updrafts and downdrafts are collected in two separate reservoirs for later analysis

• In EA, sample flow rate is proportional to w; this requirement is ‘relaxed’ in REA (i.e., full flow into up or down reservoir depending on the direction of the vertical wind)

• Alternate to eddy covariance technique to measure fluxes of trace gases for which there are no fast-response analyzers

• Air samples from updrafts and downdrafts are collected in two separate reservoirs for later analysis

• In EA, sample flow rate is proportional to w; this requirement is ‘relaxed’ in REA (i.e., full flow into up or down reservoir depending on the direction of the vertical wind)

( )F w wχ χ σ χ χ= −' ' = A Up Down

Vent (Dead band)

PTFESample Bag

DC Power supply

3-wayValve

Mass-FlowController

2-μmFilter

Reliefvalve

DiaphragmPump 12 l/min

Inlet

UP

DOWN

¼” PTFEtubing

55

Relaxed eddy accumulation technique

• Easier to apply than eddy accumulation• Fast-response anemometer measures w and controls a simple valving

system• Air is sampled at a constant rate and diverted into ‘up’ and ‘down’ bins

depending on the direction of w

• Easier to apply than eddy accumulation• Fast-response anemometer measures w and controls a simple valving

system• Air is sampled at a constant rate and diverted into ‘up’ and ‘down’ bins

depending on the direction of w

( )dnupwg CCbF −= σ

Where,b = empirical coefficient that varies with the dead bandσw and σC = standard deviation of vertical velocity (w) and concentration (C)r = correlation coefficient between w and C

Theory and experiments have suggested that b is constant and independent of stability, ≈ 0.6.

Where,b = empirical coefficient that varies with the dead bandσw and σC = standard deviation of vertical velocity (w) and concentration (C)r = correlation coefficient between w and C

Theory and experiments have suggested that b is constant and independent of stability, ≈ 0.6.

Cwg rF σσ=

56

-1.0

-0.8

-0.6

-0.4

-0.2

0.0-1.0-0.8-0.6-0.4-0.20.0

CO2 Flux (EC) (mg CO2 m-2 s-1)

CO

2 Flu

x (R

EA) (

mg

CO

2 m

-2 s

-1)

FREA = (0.84 +/- 0.10) FEC - (0.03 +/- 0.05)

R2 = 0.83

• Forest and agricultural fields

• Stainless steel canisters, with in-line magnesium percholateto remove water

• Forest and agricultural fields

• Stainless steel canisters, with in-line magnesium percholateto remove water

Comparison of EC and REA Flux Estimates of CO2

57

ΔN2O (pptv) σw = 0.3 m s-1 σw = 0.9 m s-1

10 1.6 5.5 g N2O-N ha-1 d-1

FN2O = ρN2O × 0.56 × σw × ΔN2O

Resolution of the REA system for N2O flux measurement

μ=-1.1 sc=17

-40

-30-20

-100

10

2030

40

ΔN

20 =

up

-dow

n (p

ptv)

58

Morewood

Casselman

N

0 5 km5 km

Morewood

Casselman

N

0 5 km5 km

N

0 5 km5 km

Crop types in the aircraft footprint

soy

cereals

pasture/grass

alfalfa

forest

corn

town

LEGEND

59

0

10

20

30

40N

o D

ata

Con

ifero

us F

ores

t

Mix

ed F

ores

t

Bro

adle

af F

ores

t

Low

Den

sity

For

est

Bar

e S

oil

Cor

n

Soy

bean

Cer

eals

Alfa

lfa

Ope

n Fi

eld

Wat

er

Bui

ltup

Are

as

Wet

land

% o

f Ave

rage

Foo

tprin

t

CasselmanMorewood

Land use information in 2001 within footprint of aircraft transects

60

N2O emissions during and right after snowmelt at the Eastern Canada study sites in 2001

Each data point represents the average of 3 samples, collected during two consecutive 10 km flight legs (total flight distance for one data point is ≈ 20 km)

Each data point represents the average of 3 samples, collected during two consecutive 10 km flight legs (total flight distance for one data point is ≈ 20 km)

-25

0

25

50

75

100

12515

-Mar

25-M

ar

4-A

pr

14-A

pr

24-A

pr

4-M

ay

14-M

ay

24-M

ay

3-Ju

n

13-J

un

N2O

Em

issi

ons

(g N

2O-N

ha-1

d-1

) CasselmanMorewood

61

N2O emissions right after snowmelt and after plantingat the Eastern Canada study sites in 2003

-25

0

25

50

75

100

125

15-M

ar

25-M

ar

4-A

pr

14-A

pr

24-A

pr

4-M

ay

14-M

ay

24-M

ay

3-Ju

n

13-J

un

N2O

Em

issi

ons

(g N

2O-N

ha-1

d-1

)

CasselmanMorewood

62

-25

25

75

125

175

15-M

ar

25-M

ar

4-A

pr

14-A

pr

24-A

pr

4-M

ay

14-M

ay

24-M

ay

3-Ju

n

13-J

un

TowerMorewoodCasselman

-25

25

75

125

175

15-M

ar

25-M

ar

04-A

pr

14-A

pr

24-A

pr

04-M

ay

14-M

ay

24-M

ay

03-J

un

13-J

un

-25

25

75

125

175

15-M

ar

25-M

ar

04-A

pr

14-A

pr

24-A

pr

04-M

ay

14-M

ay

24-M

ay

03-J

un

13-J

un

-25

25

75

125

175

15-M

ar

25-M

ar

04-A

pr

14-A

pr

24-A

pr

04-M

ay

14-M

ay

24-M

ay

03-J

un

13-J

un

N2O

Flu

x (g

N2O

-N h

a-1

d-1 )

2000 2001

2003 2004

Peak, April 14, 360 g N2O-N ha-1 d-1

Multi-year comparison of aircraft and tower basedestimates of N2O emissions

63

Multi-year comparison of aircraft and model estimates of N2O emissions

-25

25

75

125

175

15-M

ar

25-M

ar

4-A

pr

14-A

pr

24-A

pr

4-M

ay

14-M

ay

24-M

ay

3-Ju

n

13-J

un

-25

25

75

125

175

15-M

ar

25-M

ar

4-A

pr

14-A

pr

24-A

pr

4-M

ay

14-M

ay

24-M

ay

3-Ju

n

13-J

un

-25

25

75

125

175

15-M

ar

25-M

ar

4-A

pr

14-A

pr

24-A

pr

4-M

ay

14-M

ay

24-M

ay

3-Ju

n

13-J

un

DNDCCasselman

-25

25

75

125

175

15-M

ar

25-M

ar

4-A

pr

14-A

pr

24-A

pr

4-M

ay

14-M

ay

24-M

ay

3-Ju

n

13-J

un

N2O

Flu

x (g

N2O

-N h

a-1

d-1 )

2000 2001

2003 2004

Total emissions (kg N2O-N ha-1)DNDC: 0.34Aircraft: 0.53

Total emissions (kg N2O-N ha-1)DNDC: 0.76Aircraft: 0.55

Total emissions (kg N2O-N ha-1)DNDC: 1.44Aircraft: 1.87

Total emissions (kg N2O-N ha-1)DNDC: 1.11Aircraft: 1.77

6464

ΔCH4(ppbv) σw = 0.3 m s-1 σw = 0.9 m s-1

0.2 2 6 mg m-2 d-1

0.2 700 2100 kg km-2 y-1

FCH4 = 0.56 × σw × ΔCH4

Resolution of the REA system for CH4 flux measurement

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

dCH

4= U

P - D

OW

N (p

pbv)

65

Estimated Agricultural CH4 Emissions

Canada

66

6 284 kg km-2 yr-1

7 252 kg km-2 yr-118 119 kg km-2 yr-1

9 209 kg km-2 yr-1

7 114 kg km-2 yr-1

10 373 kg km-2 yr-1

6 992 kg km-2 yr-1

6 631 kg km-2 yr-1

Regional CH4 Emissions, Sept. 3 and 18, 2003

Canada

67

Develop ‘Model’

Develop ‘Model’

MeasureGHG’s

MeasureGHG’s

Verified Model

Verified Verified Model Model

timetime

. . .

Process of Model Development

Learn Learn BuildBuild ApplyApply

MeasureMeasure

68

Nutrient and energy cycles in agricultural ecosystems

OrganicOrganicmattermatter

nutrients

NN22OO

CHCH44 COCO22export

import

energy

Ecosystem boundary

69

Influence of time on net GHG emissions

0

Time

Cha

nge

in n

et G

HG

E

mis

sion

s

+

- BMP 1

BMP 2

BMP 3The relative benefit of adopting any one beneficial management practice (BMP) may change over time

The relative benefit of adopting any one beneficial management practice (BMP) may change over time

Other factors may change over time that can affect GHG emissions. For example:

•Temperature

•Atmospheric CO2

•Crop types

Other factors may change over time that can affect GHG emissions. For example:

•Temperature

•Atmospheric CO2

•Crop types

•Technology

•Economics

•Land use

•Government policy

•Population

•Social value of farms

70

Interactions between management decisionsand GHG emissions

Management decision

Grow leguminous forage crops

Grow leguminous forage crops

Implications

Residue return to soil

Residue return to soil

Forages fed to cattle

Forages fed to cattle

Increased manure production

Increased manure production

Decreased synthetic fertilizer

Decreased synthetic fertilizer

Effect on GHG emissions

Decreased CO2Decreased CO2

Increased CH4Increased CH4

Increased CH4and N2O

Increased CH4and N2O

Decreased N2O and CO2

Decreased N2O and CO2

What is the cumulative effect of this management decision on net greenhouse gas emissions?

What is the cumulative effect of this management decision on net greenhouse gas emissions?

Models are necessary to answer these types of questionsModels are necessary to answer these types of questions

71

CO2

Soil organic matter

product

energy

Ecosystemboundary

The carbon cycle in agroecosystems

72

Soil

carb

on

Initialcultivation

ΔC<0 ΔC≈ 0 ΔC>0

Effect onatmospheric CO2 ~

Managementchange

agriculture

Management-Induced Carbon Change on Agricultural Land

73

Estimating C factors

• Factors from empirical data and/or Century • Century used extensively because need to obtain

time effects (rate, duration) of C change

• Factors from empirical data and/or Century • Century used extensively because need to obtain

time effects (rate, duration) of C change

74

5000

5200

5400

5600

5800

6000

6200

6400

6600

1990 2000 2010 2020 2030 2040 2050 2060 2070 2080

Time (yr)

Tota

l Soi

l C (g

m-2

)

Land management change(switch to perennial crops)in 2000

CENTURY simulation withoutland management change

CENTURY simulation withland management change

Estimating soil C factors

75

-200

0

200

400

600

800

1000

1200

1990 2000 2010 2020 2030 2040 2050 2060 2070 2080

Time (yr)

Cha

nge

in S

oil C

(g m

-2) Curve fit to C change

ΔC = 1336 x [1 – exp(-0.024 x t)]

Estimating soil C factors

76

Zone LMC

K (/ year)

ΔCLMCm

ax(Mg/ha)

Final Yearof Effect

after LMC

Mean annual Linear Coefficient over Duration of Effect of LMC (Mg/ha

per year)

Mean Annual Linear Coefficient over First 20 years

after LMC(Mg/ha per year)

IT to NT 0.0261 4.9 63 0.06 0.10

IT to RT 0.0188 2.3 30 0.03 0.04

RT to NT 0.0222 2.5 37 0.04 0.05

Decrease fallow 0.0305 13.1 91 0.14 0.30

Increase perennial 0.0281 26.1 120 0.21 0.56

Semi-Arid Prairies

Soil Carbon Change Factors in Canada

Reducing tillage, decreasing fallow and converting to permanent cover all increase soil carbon

Canada

77

Soil organic matter

N2

fertilizer

manure

NO3,NH4

legumesN2,N2O

The nitrogen cycle in agroecosystems

78

DNDC – DeNitrification and DeCompositionmodel for estimating soil N2O emissions

Thermal-hydraulic Sub-model

DenitrificationSub-model

Decomposition Sub-model

Plant Sub-model

•Hourly N2O •Soil microbial respiration•Hourly N2O•C decomposition•NH3 volatilization

•daily root respiration

•N utilization

Soil properties

Air temperature/ precipitationManagement

practices

79

DNDC-MFT Interface

Simulate default crop rotations, soil and climate

User can choose or modify management practices and soil

data as required

DNDC-MFT Interface

Simulate default crop rotations, soil and climate

User can choose or modify management practices and soil

data as required

DNDCSimulation processes

DNDCSimulation processes

OutPut

Estimated N2O, CO2and Net GHG

emission factors ecodistrict level

for each managementpractice and rotation

OutPut

Estimated N2O, CO2and Net GHG

emission factors ecodistrict level

for each managementpractice and rotation

EcodistrictDatabase

EcodistrictDatabase

DailyMeteorological

Data

SoilProperties

Crop Rotations

Validate and compare against Tier 2 IPCC

factors

Validate and compare against Tier 2 IPCC

factors

Canada

Integrating Process-based and Empirical Models

80

Estimated N2O and CO2 emissions as influenced by agricultural practices in Canada

Combined CO2 & N2O coefficients (Mg CO2 equiv. ha-1 y-1)

No-tillage ReducedFallow

150%Fertilizer

50%Fertilizer

FallFertilizer

PermanentGrassland

Brown Chern. -0.33 -0.43 0.04 0.01 0.03 -0.97

Dark Brown Chern. -0.64 -0.80 -0.03 0.12 0.14 -1.33

Black Chern. -0.72 0.19 0.01 0.28 -3.44

Dark Gray Luvisol -0.80 -0.61 0.26 -0.09 0.46 -4.24

Gray Brown Luvisol -0.54 -0.11 0.33 -2.56

Gray Luvisol -0.55 -0.27 0.39 -2.13

Gleysolic -0.40 0.21 0.12 -2.36

Minus sign represents a net reduction in GHG

• Century and DNDC

81

Methane emissions calculations, IPCC Methodology

I - Methodology - OverviewI - Methodology - Overview

Sources

Levels

Calculations

Tier 2 Tier 2Tier 1

Enteric Fermentation Manure management

Cattle Other All Animal Categories

GEI VS

(Bo x MCF x MS)(Ym)x x

CH4 (kg/head/day) CH4 (kg/head/day)

GEI = Gross Energy IntakeYm = methane conversion

rate (% of feed energyconverted to CH4)

GEI = Gross Energy IntakeYm = methane conversion

rate (% of feed energyconverted to CH4)

VS = daily Volatile Solidexcreted (kg of dry matter)

Bo = maximum methaneproducing capacity(m3CH4/kg of VS) for manure produced by animal

MCF = MethaneConversion Factors for each manuremanagement system

MS = fraction of animal type's manure handledusing a definedManure System

VS = daily Volatile Solidexcreted (kg of dry matter)

Bo = maximum methaneproducing capacity(m3CH4/kg of VS) for manure produced by animal

MCF = MethaneConversion Factors for each manuremanagement system

MS = fraction of animal type's manure handledusing a definedManure System

82

CH4 is a molecule containing energy

The amount of energy lost though CH4 emissions is : ym

So:

GEI

GEI = Gross Energy IntakeYm = methane conversion rate

(% of feed energy converted to CH4)

CH4

Calculation of the Emission Factor

(EF)

EF(Gg/hd/yr) = (GEI x ym) x 36555.65 (Mj/kgCH4)

Methane emissions from enteric fermentation

Depending on the type and quality of feed, Ym can vary from about 2 to 12% of GEI. Recent research from Canadian feedlots suggests Ym is about 4% compared to the IPCC default Ym value for feedlot cattle of 3% (6.5% for grazing cattle). About10% of Canadian beef cattle are housed in feedlots.

Depending on the type and quality of feed, Ym can vary from about 2 to 12% of GEI. Recent research from Canadian feedlots suggests Ym is about 4% compared to the IPCC default Ym value for feedlot cattle of 3% (6.5% for grazing cattle). About10% of Canadian beef cattle are housed in feedlots.

83

Other (heat, urine, CH4)

Digestible Energy

Net Energy

NEm(maintenance)

NEa(activity)

NEg(growth)

NEl(lactation)

NEp(pregnancy)

Liveweight

Weightgain

Matureweight

Liveweight

Fatcontent

Milkprod.

Feedingsituation

% femalethat give

birth

Category (female, castrate, bull)

+ +++

+

GEI

(NEm + NEa + NEl + NEp + NEg)GEI (MJ/day) =

x F

DE (%GEI)

Digestible Energy (DE MJ/day) = Net Energy x F

/ DE(%)

GEI

Fecal Energy

Manure

Enteric fermentation

84

Methane emissions from manure management

EF = VS x Bo x Σ(MCF x MS) x 0.67 x 365EF = VS x Bo x Σ(MCF x MS) x 0.67 x 365

EF = annual emission factor (kgCH4/hd/yr)

VS = daily Volatile Solid excreted (kg of dry matter)

Bo = maximum methane producing capacity (m3CH4/kg of VS) for manure produced by animal

MCF = Methane Conversion Factors for each manure management system

MS = fraction of animal type's manure handled using a defined Manure System

0.67 = conversion factor of m3 CH4 to kilograms CH4 (kg CH4 /m3 CH4)

EF = annual emission factor (kgCH4/hd/yr)

VS = daily Volatile Solid excreted (kg of dry matter)

Bo = maximum methane producing capacity (m3CH4/kg of VS) for manure produced by animal

MCF = Methane Conversion Factors for each manure management system

MS = fraction of animal type's manure handled using a defined Manure System

0.67 = conversion factor of m3 CH4 to kilograms CH4 (kg CH4 /m3 CH4)

85

Greenhouse Gas Emissions in Canada, 1990-2006

0

100

200

300

400

500

600

700

800

900

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

Gre

enho

use

Gas

Em

issi

ons

(Mt C

O2e

)

Agricultural Emissions

Kyoto Objective: -6% compared to 1990Kyoto Objective: -6% compared to 1990

+28%+28%

Agricultural emissions represent about 8.5% of national emissionsAgricultural emissions represent about 8.5% of national emissions

86

ModelModel

MethodMethod

Sources of Sources of emissionsemissions

Holistic Approach for Estimating GHG Emissions from the Dairy Sector

CH 4 N2O CO 2

Feed ration Population - Dairy cows

Synthetic Nitrogen Fertilizers

Farm Fieldwork

CH4 emissionsIPCC TII

Crop Complex For dairy cows

Nitrogen fertilizerFor the crop complex

N2O EmissionsIPCC TII

Manure

Enteric Fermentation

Manure Mgmnt

Fertilizer Manufacture

CO2 EmissionsFossil Energy F4E2 Model

: Inputs

: Calculation

87

PrinciplePrinciple

Yields kg / haYields kg / ha

Silage

Straw

Barley

Pasture

Corn

Others(wheat, soybean,

etc.)

Farm

NN22OO

CHCH44COCO22

NN22OOCHCH44

COCO22

“Crop Complex”

Feed Rations Feed Rations

Crops (kg/anl-yr)Wheat 50

Oats 30Barley 870

Corn 440Dry peas 5Soybean 90

Canola 125

Straw-Silage CensusPasture

Dairy cows (2001)

Defining the Crop Complex

Intensity IndicatorIntensity Indicator

Areas (ha)Areas (ha)

88

Dairy Population and Milk Production

0.0

0.5

1.0

1.5

2.0

1981 1986 1991 1996 2001 2006

Dai

ry C

ow P

opul

atio

n (M

hea

d)

0

10

20

30

40

Milk

Pro

duct

ion

(kg

hd-1

d-1)

PopulationMilk Production

Met

hane

Em

issi

on F

acto

r (k

g C

H4

head

-1ye

ar-1

)

0

25

50

75

100

125

150

175

200

1981 1986 1991 1996 2001 2006

Methane Emission Factor

Increasing milk production per head has resulted in an increase in the methane emission factor for dairy cattleIncreasing milk production per head has resulted in an increase in the methane emission factor for dairy cattle

89

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1981 1986 1991 1996 2001 2006GH

G E

mis

sion

s pe

r kilo

gram

of m

ilk p

rodu

ced

(kg

CO

2e)

GHG emissions intensity: Dairy Sector

1981-2006: 26% reduction in emissions intensity

1991-2006: 18% reduction in emissions intensity

1981-2006: 26% reduction in emissions intensity

1991-2006: 18% reduction in emissions intensity

Source: Vergé et al (2007); Dyer et al (2008)

90

GHG Emissions Intensity in Canada

0

2

4

6

8

10

12

14

16

18

Dairy Beef Pork Poultry -broiler meat

Poultry - eggs

GH

G e

mis

sion

s pe

r kilo

gram

of m

ilk o

r liv

e w

eigh

t pr

oduc

ed o

r doz

en e

ggs

1981 1986 1991

1996 2001 2006

-26 %

-56 %

-56 %-19 % -8 %

Improved breeds, adoption of BMPsfeeding of leguminous crops have led todairy,

such as no-tillage and increased a reduction in emissions intensity

beef, pork and poultry production in Canada.

Improved breeds, adoption of BMPsfeeding of leguminous crops have led todairy,

such as no-tillage and increased a reduction in emissions intensity

beef, pork and poultry production in Canada.

91

System Boundaries for Emissions Calculations

Dairy FarmInputs Milk ProcessingTransportation

WholesaleRetail

Transportation

Transportation

Consumer

Transportation

Inputs

InputsInputsDepending on how the system boundaries are defined, emissions per

unit of product can change significantly. Life Cycle Analysis must define the system boundaries which then ensures the consistency

and comparability of studies.

Depending on how the system boundaries are defined, emissions per unit of product can change significantly. Life Cycle Analysis must define the system boundaries which then ensures the consistency

and comparability of studies.

92

An estimate of the carbon footprint of milk production in Canada

Crop production

Milk production Processing Packaging

Transport/distribution Retail

GH

G E

mis

sion

s (M

t CO

2e)

0

1

2

3

4

5

6

7

8

All emissions, CO2 (estimated)Electricity, CO2

Manure management, CH4 + N2OEnteric fermentation, CH4

Fuel use, CO2

Fertilizer supply, CO2

Soil, N2OCrop and milk production represent

about 80% of total emissionsCrop and milk production represent

about 80% of total emissions

Assuming a similar proportion of off-farm emissions as the US:

Total CN emissions: 12.42 Mt CO2e per year, or about 1.24 kg CO2e per kg of milk

produced

Assuming a similar proportion of off-farm emissions as the US:

Total CN emissions: 12.42 Mt CO2e per year, or about 1.24 kg CO2e per kg of milk

produced

93

Modeling tools for producers

Knowledge gained during the Model Farm program has been synthesized in a user friendly computer program, Holos, which estimates whole farm GHG emissions.

Knowledge gained during the Model Farm program has been synthesized in a user friendly computer program, Holos, which estimates whole farm GHG emissions.

Holos – Greek, meaning whole or complete

Holos – Greek, meaning whole or complete

“Model Farm” was a Canadian agricultural research program that lasted from 2002-2006 and focused on improving estimates of GHG emissions from Canadian farms, and finding methods of mitigating those GHG emissions.

“Model Farm” was a Canadian agricultural research program that lasted from 2002-2006 and focused on improving estimates of GHG emissions from Canadian farms, and finding methods of mitigating those GHG emissions.

94

Holos: What is it?

1. A simple, user friendly ecosystem model that estimates net-greenhouse gas emissions from individual Canadian farms

2. Describes best understood biophysical processes and their interconnections

3. Mathematical equations combined with expert knowledge and findings from experiments for a variety of farming practices

4. Allows for the evaluation of mitigation practices

5. Allows farmers to experiment with ‘what if?’ scenarios to reduce on-farm GHG emissions.

1. A simple, user friendly ecosystem model that estimates net-greenhouse gas emissions from individual Canadian farms

2. Describes best understood biophysical processes and their interconnections

3. Mathematical equations combined with expert knowledge and findings from experiments for a variety of farming practices

4. Allows for the evaluation of mitigation practices

5. Allows farmers to experiment with ‘what if?’ scenarios to reduce on-farm GHG emissions.

95

Holos output

Summary

•Process-based models provide useful information on GHG emissions from agroecosystems

•Empirical models are sometimes used because our knowledge of processes is still fairly poor

•Management practices are changing rapidly in agriculture, hence we continuously need new measurements to improve the models

•Much progress has been achieved in reducing the GHG emission intensities from major animal production sectors

•Tools like Holos should help producers reduce GHG emissions on their farms

•Process-based models provide useful information on GHG emissions from agroecosystems

•Empirical models are sometimes used because our knowledge of processes is still fairly poor

•Management practices are changing rapidly in agriculture, hence we continuously need new measurements to improve the models

•Much progress has been achieved in reducing the GHG emission intensities from major animal production sectors

•Tools like Holos should help producers reduce GHG emissions on their farms

97

Conclusions

•Substantial progress has been made in measuring GHG emissions from agroecosystems

•Models have also been improved substantially

•Management practices are continuously changing, hence the need to measure emissions and modify our models accordingly

•Tools like Holos are increasing awareness among producers and are helping to reduce GHG emissions from farms

•Because of increasing demand for food, the greatest progress one can expect is a reduction in GHG emission intensities

•Substantial progress has been made in measuring GHG emissions from agroecosystems

•Models have also been improved substantially

•Management practices are continuously changing, hence the need to measure emissions and modify our models accordingly

•Tools like Holos are increasing awareness among producers and are helping to reduce GHG emissions from farms

•Because of increasing demand for food, the greatest progress one can expect is a reduction in GHG emission intensities