23

United StatesDepartment ofAgriculture

NaturalResourcesConservationService

ConservationBuffers to ReducePesticide Losses

March 2000

24

The United States Department of Agriculture (USDA) prohibits discrimination in all its programs and activities on thebasis of race, color, national origin, gender, religion, age, disability, political beliefs, sexual orientation, and marital orfamily status. (Not all prohibited bases apply to all programs.) Persons with disabilities who require alternate meansfor communication of program information (Braille, large print, audiotape, etc.) should contact the USDA’s TARGETCenter at (202) 720-2600 (voice and TDD).

To file a complaint of discrimination, write USDA, Director, Office of Civil Rights, Room 326W, Whitten Building, 14thand Independence Avenue, SW, Washington, DC 20250-9410, or call (202) 720-5964 (voice or TDD). USDA is an equalopportunity employer.

Issued March 2000

Conservation Buffers to Reduce Pesticide Losses is a joint product of the United StatesDepartment of Agriculture Natural Resources Conservation Service (USDA-NRCS)National Water and Climate Center and the United States Environmental ProtectionAgency (USEPA) Office of Pesticide Programs. It was conceived by an ad hoc jointcommittee formed in 1994. Participants were Joe Bagdon, USDA-NRCS National Waterand Climate Center, Ron Parker, USEPA Office of Pesticide Programs, Tom Gilding,American Crop Protection Association, Nick Poletika, Dow AgroSciences LLC, andDick Fawcett, Fawcett Consulting. The committee would like to express its thanks toreviewers from the following organizations for comments on and additions to thepublication:

Iowa State University - Department of Agricultural Engineering

Michigan Agricultural Cooperative Marketing Association

New York State Department of Environmental Conservation

Rachel Carson Council, Inc.

United States Geological Survey

USDA Agricultural Research Service

USDA-NRCS Science and Technology Consortium and National Headquarters

USEPA Regional Offices

USEPA Office of Water

Virginia Polytechnic Institute - Dept. of Biological Systems Engineering

The committee hopes that the production of this document will demonstrate theopportunities for intergovernmental as well as public - private cooperation in furtheringgoals of common interest. While all of the above mentioned organizations share thecommon goal of reducing pesticide losses, not all items in the publication necessarilyrepresent the official policy of each organization. The cooperation of the Federalagencies does not imply endorsement of any commercial product or service mentionedin this publication.

25

Executive Summary

Conservation buffers are areas or strips of land maintained inpermanent vegetation. These buffers can be used in a systemsapproach to manage soil, water, nutrients, and pesticides for sus-

tainable agricultural production, while minimizing environmental impact.This publication summarizes recent research investigating the ability ofconservation buffers to trap and degrade pesticides carried in field runoff.It is designed to be used by change agents, such as NRCS field staff,Extension agents, certified crop advisors, land use planners, and agrichemicalsales and support personnel. When used in conjunction with local andregional research and information sources, such as the NRCS Field OfficeTechnical Guide (FOTG), this publication should help planners design,install, and maintain buffers for optimal pesticide trapping, while gainingother buffer benefits, such as erosion control, nutrient and sediment trap-ping, wildlife habitat, and streambank protection.

Some pesticides are highly adsorbed to soil particles. Because properlydesigned buffers are effective in trapping eroded sediment, runoff losses ofthis kind of pesticide have been consistently reduced by buffers. However,a number of modern pesticides are only moderately adsorbed to soilparticles and are carried with runoff from fields primarily in the dissolvedphase. To be effective in trapping this type of pesticide, buffers must slowrunoff and increase infiltration so that pesticides can be trapped anddegraded in buffer soil and vegetation. Many studies have demonstratedpesticide trapping efficiencies of 50 percent or more for this type ofpesticide, provided that sheet flow, not concentrated flow, occurs. Concen-trated flow is the nemesis of pesticide trapping by buffers. Techniques toencourage shallow sheetflow across buffers are described. Maintenance ofbuffers is critical. Trapped sediment can change the profile of buffers,increasing concentrated flow, and will need to be removed or leveled tomaintain buffer function. Soil conservation practices used above bufferswill prolong their effectiveness. Buffer location, size, and vegetation speciesselection are described.

Conservation buffers are not a substitute for careful pesticide selectionand use. They are a tool to further improve water quality and produceadditional environmental benefits when used along with other practicesdescribed in the text.

26

1

Conservation Buffers to Reduce Pesticide Losses

these pesticides in the same way they trap sediment.However, some pesticides are only moderately adsorbedto soil particles and are carried off fields primarily dis-solved in water. For buffers to be effective in trappingthis type of pesticide, either water must infiltrate into thebuffer, carrying the chemical into the soil, or chemicalsmust be removed from solution flowing over the soilsurface by contact with soil or vegetation. Most studiesindicate that an increase in water infiltration is the mostimportant factor in trapping these pesticides.

This booklet examines current knowledge of how con-servation buffers can be most effectively used to reducepesticide losses to water. Studies specifically measuringpesticide trapping by buffers will be reported, as well asrelevant studies on effectiveness of buffers in trappingsediment and increasing water infiltration. The effective-ness of buffers in reducing pesticide losses depends onthe properties of the specific pesticide, the design andmaintenance of the buffer, and local climate, weather,and soil conditions. When combined with specific localinput from such sources as the Natural Resources Con-servation Service (NRCS), Soil and Water ConservationDistricts, and the Extension Service, this booklet willprovide guidance to those advising and assisting farmersand landowners installing conservation buffers. OtherBest Management Practices (BMPs) which improve man-agement and reduce losses of pesticides and can be usedin combination with buffers, will also be described. Thispublication does not comprehensively examine bufferimpacts on nutrient losses to water. However, bufferimpacts on nitrate loss are briefly described and con-trasted to impacts on pesticide loss.

Literature Reviews

Gilliam, J.W., et al. 1997. Selected agricultural best man-agement practices to control nitrogen in the Neuse RiverBasin, North Carolina State University Technical Bulle-tin 311, Raleigh, NC.

Lowrance, R., et al. 1995. Water quality functions ofriparian forest buffer systems in the Chesapeake BayWatershed. U.S. EPA Publication EPA 903-R-95-004.

Bibliographies

Vegetated stream riparian zones: their effect on streamnutrients, sediments, and toxic substances. SmithsonianEnvironmental Research Center, Edgewater, MA. http://www.serc.si.edu/documents/ripzone.html

Introduction

Conservation buffers are areas or strips of land main-tained in permanent vegetation to help control pollut-ants and manage other environmental problems. Thesebuffers can be used in a systems approach to helpmanage soils, water, nutrients, and pesticides for sus-tainable agricultural production while minimizing envi-ronmental impact. Buffers have long been a staple inconservation systems designed to prevent erosion andtrap sediment and nutrients from field runoff. They alsoprovide other benefits, such as wildlife habitat improve-ment, streambank protection, and farming safety. Manystudies have been conducted to document these benefitsand to provide guidance in designing buffers for thesepurposes. But do buffers work to reduce pesticide losses?

Rainfall or irrigation can cause pesticides to run off thesurface of treated fields. Edge-of-field losses can rangefrom less than 1 percent of the amount applied to asmuch as 10 percent (Wauchope, 1978; Baker, 1983).Losses are greatest when severe rainstorms occur soonafter pesticide application. Edge-of-field concentrationsof pesticides in surface runoff can range from less than1 part per billion (ppb) or 1 microgram per liter (µ/L) to1 part per million (ppm) or 1 milligram per liter (mg/L) ormore.

Soil erosion runoff

Until recently, few studies had been conducted to mea-sure the effectiveness of buffers in trapping pesticides inrunoff. Physical and chemical properties of pesticidesaffect pesticide behavior and transport. Some pesticidesare highly adsorbed to soil particles and are carriedprimarily adsorbed to eroded sediment. Buffers trap

2

Function and design of vegetation filter strips: an anno-tated bibliography. Texas State Soil and Water Conserva-tion Board, Temple, TX. http://waterhome.tamu.edu/tsswcb/Projects/bibliography/index.html

NRCS Websites

National Handbook of Conservation Practices and BufferJob Sheets:

http://www.ncg.nrcs.usda.gov/nhcp_2.html

Natural Resources Conservation Service Homepage:http://www.nrcs.usda.gov

Types of Buffers

Water buffers within fields

Grassed waterway—a natural or constructed vegetatedchannel that is shaped and graded to carry surface waterat a nonerosive velocity to a stable outlet. Because ofconcentrated flow that normally occurs in waterways,sediment trapping and water infiltration can be minimalwith large runoff events, but substantial with smallerevents. Waterways are most effective in trapping sedi-ment and dissolved chemicals when designed to spreadconcentrated waterflow over a vegetated filter adjacentto streams.

Contour buffer strips— strips of perennial vegetationalternated with wider cultivated strips that are farmedon the contour. Buffers are most effective in trappingpesticides when runoff enters uniformly as sheetflow.Contour buffer strips are one of the most effectivebuffers to trap pesticides. There is less chance forconcentrated flow and smaller areas of cultivated fielddeliver runoff directly to each strip within a relativelyshort distance compared to some edge-of-field buffers.

3

Vegetative barriers—narrow, permanent strips of stiffstemmed, erect, tall, dense, perennial vegetation estab-lished in parallel rows and perpendicular to the domi-nant slope of the field. These barriers function similar tocontour buffer strips and may be especially effective indispersing concentrated flow, thus increasing sedimenttrapping and water infiltration.

Edge-of-field

Field borders—a band or strip of perennial vegetationestablished on the edge of a cropland field. This bufferreduces pesticide runoff only when runoff flows over thestrip. Even when no water flows over the strip, somewater quality benefit may be gained because sprayingoperations are physically separated from adjacent areas,reducing drift and direct application to riparian areas.

Filter strips—areas of grass or other permanentvegetation used to reduce sediment, organics, nutrients,pesticides, and other contaminants in runoff and tomaintain or improve water quality. Filter strips arelocated between crop fields and waterbodies. Morepesticides can be removed by encouraging as muchsheetflow as possible across the strip and minimizingconcentrated flow. This may be accomplished bycombining filter strips with other conservation practicesthat control concentrated flow, such as vegetativebarriers, level spreaders, or water bars.

Terrace tile inlet buffer—setbacks surrounding inletsto tile-outlet terrace systems. Some herbicide labelsdescribe leaving untreated setbacks around these inletswhen tiles draining terraces outlet directly into streams.These terraces are designed to cause water to pond overareas adjacent to inlets following runoff. Therefore,buffers may not increase sediment trapping or waterinfiltration compared to cropped areas, but only reducetotal areas treated with pesticide.

4

Set back

Application area

Setbacks—untreated areas where surface runoff entersstreams. Some herbicide labels describe leaving theseareas untreated. Seeding these areas to perennial grassimproves herbicide trapping compared to trapping withuntreated row crop.

Constructed wetlands

Riparian forest buffer—an area of trees and shrubslocated adjacent to streams, lakes, ponds, and wetlands.Forest buffers are often combined with perennial grassbuffers. Woody vegetation provides food and cover forwildlife, helps lower water temperatures by shading thewaterbody, contributes energy sources to aquatic com-munities, protects streambanks, and slows out-of-bankflood flows. Deep tree roots may intercept nitrate enter-ing streams in shallow subsurface flow and provide soilcarbon for microbial energy. Microbes can degradepesticides and denitrify nitrate.

While pesticides are usually carried to streams in sur-face runoff, most nitrate is carried to streams in subsur-face flow. Subsurface flow may also carry low concen-trations of pesticides. Riparian buffers can interceptshallow subsurface flow and cause either uptake ofnitrate and utilization by plants or encourage denitrifi-cation. Drainage tiles bypass buffers and deliver subsur-face drainage directly to streams. Wetlands constructedat tile outlets or as part of a riparian buffer system caneffectively degrade pesticides and denitrify nitrate.

Wind buffers

Windbreak/shelterbelts—plantings of single or mul-tiple rows of trees established for environmental pur-poses. The primary purpose of such buffers is to protectleeward areas from troublesome winds. They may alsoseparate spraying operations from adjacent areas, re-duce drift resulting from lowered wind speed, and inter-cept spray drift. Taller plantings provide the most driftprotection.

5

Soil water

Soil particle

Kd = Conc. pesticide on soilConc. pesticide in H2O

PP

PP

P

PP

P P

Cross Wind Trap Strips—areas of herbaceous vegeta-tion, resistant to wind erosion, and grown in stripsperpendicular to the prevailing wind direction. Thesestrips trap wind-borne sediment and nutrients and pes-ticides carried by sediment.

Pesticide Trapping

Pesticides may be held on the surface of material, suchas soil particles. This phenomenon is called adsorption.Pesticides may also be taken inside a material, such asbeing taken up by a plant. This process is called absorp-tion. The term sorption is sometimes used to refer toboth processes. Pesticide interaction with soil is prima-rily an adsorption process.

Pesticides vary in how tightly they are adsorbed to soilparticles. Degree of soil binding is measured by bindingcoefficients, or K values. Koc

(K of organic carbon) is a

measure of adsorption to the organic matter or carboncontent of soil, with higher values indicating more bind-ing. While pesticides are also bound to clay particles,binding to organic matter is a useful predictor of pesti-cide behavior and movement in soil. Koc values can beused to predict whether a specific pesticide will becarried primarily in the sediment or dissolved phase offield runoff. Example Koc values for specific pesticidesrange from 2 for dicamba (which is held loosely in thesoil) to 1 million for paraquat (which is bound tightly tosoil). Koc values greater than 1,000 indicate that pesti-cides are highly adsorbed to soil. These pesticides tendto be carried off fields on eroded soil particles. Thus, ifconservation buffers are effective in trapping the sedi-ment particle sizes that transport the pesticides, theyeffectively trap this type of pesticide. Pesticides withlower Koc values (generally less than 500) tend to movemore in water than on sediment. Concentrations carriedon sediment are higher than concentrations in water,but because water quantities running off fields are somuch greater than eroded soil quantities, water ac-counts for the majority of chemicals leaving fields. Totrap low Koc pesticides effectively, buffers need to in-crease water infiltration and maximize runoff contactwith soil and vegetation that may adsorb pesticides.

Herbaceous Wind Barriers—tall grasses, up to 5 feet,and other non-woody plants established in 1- to 2-row,narrow strips spaced across the field perpendicular tothe normal wind direction. These barriers reduce windspeed and wind erosion and intercept wind-borne soilparticles that may carry pesticides and nutrients.

Other Barriers—other types of perennial vegetationon the landscape can serve as a buffer. These barriersinclude CRP fields, wood lots, terrace back slopes,ditchbanks, and wildlife plantings.

6

Percen

t o

f p

esti

cid

e t

rap

ped

Koc

In contrast to most pesticides, nitrate is water-solubleand not readily adsorbed by soil particles. Usually ni-trate is not in runoff because in enters the soil quickly.Rather, nitrate that is not taken up by plants may leachto ground water and be carried to streams by subsurfaceflow. (Significant losses of nitrate in surface runoff canoccur in certain situations, such as heavy rainfall aftersurface application of nitrogen fertilizer or manure.) Totrap nitrate effectively, roots of conservation bufferplants need to intercept this subsurface flow. Condi-tions for denitrification present in this biologically ac-tive zone also reduce nitrate reaching streams. Simi-larly, some weakly adsorbed pesticides may leach toshallow ground water in small amounts. Although sub-surface flow may carry small quantities of pesticides tostreams, quantities present in surface runoff are usuallymuch greater. The NRCS maintains a current PesticideProperty data base and can provide Koc values for spe-cific pesticides.

Study Results

One of the earliest studies of the impact of buffers onpesticide runoff investigated pesticide retention bygrassed waterways (Asmussen, et al., 1977). When run-off from a small plot was directed into an 80-foot-longgrassed waterway, 70 percent of the weakly adsorbedherbicide, 2,4-D, was trapped. In a similar study (Rohde,et al., 1980), 96 percent of the strongly adsorbed triflura-lin was trapped when the waterway was dry beforerunoff, and 86 percent was trapped when the waterwaywas wet before runoff occurred.

Hall, et al. (1983) studied the impact of strip cropping onatrazine runoff in Pennsylvania. Seventy-two-foot-longplots were constructed up and down a 14 percent slopewith a 19.7-foot-wide area at the base of the slope seededto oats (Avena sativa L.). Runoff of atrazine applied tocorn was measured throughout the season, which in-cluded a severe, once-in-100-year frequency storm inJune. The oats strip trapped 91 percent of atrazine inrunoff when the herbicide was applied at a rate of 2pounds per acre.

Results of these early studies surprised some scientistswho assumed that buffers would have minimal impacton runoff of moderately adsorbed pesticides like atra-zine and 2,4-D. However, significant infiltration of run-off water into buffers was identified in these studies asthe primary mechanism of pesticide removal.

In Mississippi, Webster and Shaw (1996) measured run-off of metolachlor and metribuzin from soybean plotswith and without tall fescue (Festuca arundinacea

Schreb.) filter strips. Plots were 13 by 72 feet with 3percent slope. Filter strips were either 13 or 6.5 feetwide. Width of the filter strip did not affect herbicidetrapping efficiency. Over 3 years, metolachlor loss wasreduced 55 to 74 percent by the filter strips, comparedto plots without filter strips, while metribuzin loss wasreduced by 50 to 76 percent. Much of the herbicidetrapping could be attributed to infiltration of runoff intothe filter strips. Using similar techniques in a later study(Rankins, et al., 1998), tall fescue buffers reduced runoffof fluormeturon and norfluorazon applied to cotton byat least 60 and 65 percent, respectively.

Several studies in Iowa have investigated herbicidetrapping by smooth bromegrass (Bromus inermis

Leysser) filter strips using either simulated or naturalrunoff. In some studies field runoff was simulated byadding known concentrations of herbicide to waterreleased above filter strips while rainfall was simulated(Mickelson and Baker, 1993). Simulated runoff solutionrepresenting runoff from a 150-foot-long area was ap-plied to the top of 15- and 30-foot-long filter strips (thusrepresenting source area to buffer ratios of 10:1 and 5:1).The 15-foot strip reduced atrazine runoff by 35 percent,while the 30-foot strip reduced atrazine runoff by 59percent.

Other similar studies with smooth bromegrass buffers(Misra, et al., 1996) showed less difference betweenbuffer sizes. When comparing 15:1 and 30:1 source areato buffer ratios, atrazine removal was 31.2 and 26.4percent, respectively. Herbicide concentrations in simu-lated runoff were applied at either 0.1 ppm or 1.0 ppm.A higher percentage of herbicide was trapped by bufferswhen inflow had higher concentrations. When inflowhad 1.0 ppm herbicide, 15:1 area ratio buffers trapped50, 47, and 47 percent of atrazine, metolachlor, andcyanazine, respectively. When inflow concentrationswere 0.1 ppm, these buffers trapped 31, 32, and 30percent of atrazine, metolachlor, and cyanazine, respec-tively. Infiltration of runoff water into buffers accountedfor most herbicide trapping.

7

In other studies, natural runoff from a treated area wascollected and distributed to replicated smooth brome-grass buffers 66 feet long (Arora, et al., 1996). Runoffwas distributed to represent source area to buffer arearatios of 15:1 and 30:1. Efficiency of herbicide trappingwas determined for six runoff events over a 2-yearperiod. Trapping of atrazine, cyanazine, and metolachlorranged from a low of 8 percent to a high of 100 percent,depending largely on the timing and intensity of rainfalland antecedent soil moisture conditions. Herbicide trap-ping was least efficient when soil was saturated fromprevious rains. For most events there were only smalldifferences in herbicide trapping efficiency betweenarea ratios. Averaging results for all herbicides and arearatios over the six events, 62 percent of herbicidescontained in runoff was trapped by buffers. Infiltrationof runoff into buffers was determined to be the majormechanism of herbicide removal. While sediment reten-tion ranged from 40 to 100 percent, only about 5 percent oftotal herbicide retention resulted from sediment trapping.

Analysis of soil within the buffer strips confirmed thatherbicides were being trapped and held by soil within thestrips (Fawcett, et al., 1995). Concentrations declinedduring the growing season, presumably because of degra-dation. No phytotoxicity was observed on buffer grasses.

Bermudagrass (Coynodon dactylon L.) and wheat (Triti-

cum aestivum L.) contour buffers were studied in Texas(Hoffman, 1995). Three 30-foot-wide buffers were equallyspaced within a 435-foot-wide watershed planted tocorn. Hydrologic data showed that water runoff wasreduced 57 percent by bermudagrass and 50 percent bywheat. Total atrazine loss was reduced 30 percent bybermudagrass and 57 percent by wheat in 1 year, and by44 to 50 percent by all buffers in another year.

Patty, et al. (1997) studied ryegrass (Lolium perenne L.)buffers in France under natural rainfall conditions. Bufferwidths varied from 20 to 59 feet Runoff volume wasreduced by 43 to 99.9 percent, suspended solids by 87 to100 percent, lindane losses by 72 to 100 percent, atrazineand metabolite losses by 44 to 100 percent, isoproturonlosses by 99 percent, and diflufenican losses by 97percent.

The ability of bermudagrass buffers to trap runoff of turfpesticides was studied in Oklahoma (Cole, et al., 1997).Buffers 16 feet wide trapped from 90 to 100 percent ofdicamba, from 89 to 98 percent of 2,4-D, from 89 to 95percent of mecroprop, and from 62 to 99 percent ofchlorpyrifos in runoff. In most instances buffer mowingheight (3.3 or 9.7 inches), buffer length (7.9 or 16 feet),and tine aeration did not significantly affect pesticidetrapping efficiency.

The effectiveness of a three-zone riparian buffer intrapping herbicide runoff was studied in Georgia(Lowrance, et al., 1997). Total buffer width was 164 feet,with a 26-foot-wide grass buffer adjacent to the cropfield, a managed pine forest downslope from the grass,and a narrow hardwood forest containing the streamchannel. More than 90 percent of atrazine and alachlorin field runoff was trapped by the buffer. Most herbicidewas trapped by the grass strip, with 60 to 70 percent ofherbicide in runoff trapped. Grass was a mixture ofbahiagrass (Paspalum notatum Flugge), bermudagrass,and perennial ryegrass.

The impact of untreated setbacks around tile inlets intile-outlet terrace systems was studied in Iowa, Ne-braska, and Missouri (Mickelson, et al., 1998; Franti, etal., 1998). In all studies setbacks provided no reductionsin herbicide runoff into inlets beyond what would beexpected because of reduced area treated. This result isnot surprising since these terraces are designed to pondwater around inlets, causing sediment to settle andincreasing water infiltration. Much of the untreatedsetback area is under water during runoff events andcannot be expected to increase infiltration or sedimenttrapping beyond that caused by normal functioning ofthe terrace system. The studies’ authors concluded thatalternative BMPs, such as herbicide incorporation andno-till production, were more effective in reducing her-bicide runoff into inlets. Based on this research, USEPAallowed label changes on atrazine-and cyanazine-con-taining products. Three alternative BMPs are now de-scribed on these labels for use in tile-outlet terracesystems: (1) 66-foot untreated setback around inlet, or(2) incorporation of herbicide in areas draining to inlet,or (3) no-till production with high crop residue levels inareas draining to inlet.

The importance of water infiltration as a mechanism oftrapping moderately adsorbed pesticides is illustratedby some studies that have shown that buffers do little toreduce concentrations of moderately adsorbed pesti-cide in runoff. In Nebraska (Yonts, et al., 1996), smoothbromegrass or intermediate wheatgrass buffers did notsignificantly reduce concentrations of alachlor,cyanazine, 2,4-D, or atrazine present in furrow irrigationrunoff water, although concentrations of the stronglyadsorbed chlorpyrifos were reduced.

In contrast, some studies have shown significant reduc-tions in concentrations of moderately adsorbed pesti-cides caused by buffers. In Mississippi (Tingle, et al.,1998), tall fescue buffers as narrow as 1.6 feet at the baseof 72-foot plots reduced concentrations of metolachlorand metribuzin in runoff by almost 50 percent.

8

Biological and physical conditions, which develop inbuffers planted to grasses and/or trees, favor increasedwater infiltration and subsequent attenuation of nutri-ents and attenuation and degradation of pesticides.Bharati (1997) compared infiltration and soil propertiesunder a multispecies riparian buffer in Iowa to adjacentcultivated fields and a grazed pasture. Average cumula-tive water infiltration was five times greater under thebuffer than that under the cultivated field and pasturesites. Soil bulk density was also consistently lowerunder the buffer. Wood (1977) compared hydrologiccharacteristics of forested land to adjacent sugarcane,pineapple, or pastureland of the same soil series at 15sites in the Hawaiian Islands. Infiltration rates werehigher under forest cover at 14 of the 15 sites. Meanweight diameters of the surface soil aggregates werelarger for forested soils.

The extensive root growth in buffers and superior soilstructure most likely explain observed water infiltrationincreases. This root growth also increases biologicalactivity by supplying an organic carbon energy source tosoil micro-organisms. These micro-organisms in turnare responsible for degrading pesticides and denitrify-ing nitrate. Such untilled areas also attenuate atmo-spheric carbon dioxide through carbon sequestration intree and grass vegetation and soil organic matter. Tillageand crop production often deplete soil organic matter(Reicosky, et al., 1995), releasing carbon dioxide to theatmosphere. Increased soil organic matter in buffersserves to better adsorb pesticides in runoff. Nutrientsare taken up by vegetation and stored in living tissue.Periodic harvest may be desirable to prevent later re-lease. Pesticides are also taken up by roots and may bemetabolized in plants. In addition, vegetation at the soilsurface adsorbs pesticides during runoff events. In Iowa(Fawcett, et al., 1995), atrazine concentrations in plantresidue collected in buffers ranged from 80 ppb to 740ppb, depending on collection date, and were similar toconcentrations found in surface buffer soil.

Few studies have investigated pesticide trapping byconstructed wetlands. Some pesticides are relativelyshort-lived in water and will degrade while sequesteredin wetlands, therby reducing contaminants reachingstreams. Wetlands can also serve to attenuate pulses ofconcentrated runoff before it enters streams. Somepesticides are relatively persistent once they reach wa-ter. However, the high organic matter content of wet-land sediment binds these pesticides, removing themfrom water. Matter (1993) used intact freshwater wet-land sediment microcosms to study the behavior ofatrazine. Atrazine was removed from the overlying wa-ter column at a rate of 15.8 percent per day for 3 days

after introduction. After 10 months, 88 percent of theapplied atrazine was unextractable from sediment andnone was recovered from the overlying water.

Considering buffer research to date, buffers have beeneffective under controlled conditions, in trapping highlyadsorbed and moderately adsorbed pesticides. Table 1summaries buffer studies showing trapping efficiencyfor specific pesticides and pesticide Koc values. Highlyadsorbed pesticides were trapped at rates of from 62 to100 percent. Trapping of moderately adsorbed pesti-cides was more variable and ranged from 8 to 100percent. Buffers retained the lowest percent of pesti-cide when buffer soil was saturated from previous rains.Many studies found pesticide trapping efficiencies of 50percent or more.

Do results of these controlled studies predict what willhappen in the real world? Nearly all of these studies (withthe exception of early grassed waterway studies) weredesigned to encourage sheetflow across buffers. There-fore, they represent the maximum trapping that can beexpected. In the real world, concentrated flow often oc-curs across buffers, reducing their effectiveness. To maxi-mize trapping of sediment-adsorbed and dissolved pesti-cides, sheetflow needs to be encouraged through properbuffer design and maintenance.

Dillaha, et al. (1989) analyzed 33 existing buffers inVirginia for sediment trapping efficiency. They foundsediment trapping was often poor because of eitherconcentrated flow where topography was hilly or sedi-ment that accumulated in the buffer, causing runoff toflow parallel to the buffer until a low point was reachedwhere concentrated flow occurred.

Excessive sediment load in runoff may not only changeflow patterns caused by accumulation in buffers, butmay also reduce water infiltration, making buffers lesseffective in trapping dissolved pesticides. In an Iowasimulation study (Misra, 1994), runoff with and withoutsuspended sediment was introduced into buffers. Inabsence of sediment, buffers removed over 80 percentof atrazine, cyanazine, and metolachlor. When sedimentat 10,000 mg/L was included in runoff, trapping of thethree herbicides fell to about 50 percent. Accumulationof sediment apparently caused soil surface sealing, re-ducing total water infiltration from 83 percent, in theabsence of sediment, to 30 percent with sediment. It isthus critical that soil conservation methods be usedabove conservation buffers to reduce the amount ofsediment entering buffers.

9

Pesticide Koc Study reference Percent pesticide trapped

Highly adsorbed pesticides

Chlorpyrifos 6,070 Boyd, et al., 1999 57–79Cole, et al., 1997 62–99

Diflufenican 1,990 Patty, et al., 1997 97Lindane 1,100 Patty, et al., 1997 72–100Trifluralin 8,000 Rhode, et al., 1980 86–96

Moderately adsorbed pesticides

Acetochlor 150 Boyd, et al., 1999 56–67Alachlor 170 Lowrance, et al., 1997 91Atrazine 100 Arora, et al., 1996 11–100

Boyd, et al., 1999 52–69Hall, et al., 1983 91Hoffman 1995 30–57Lowrance, et al., 1997 97Mickelson and Baker 1993 35–60Misra, et al., 1996 26–50Patty, et al., 1997 44–100

Cyanazine 190 Arora, et al., 1996 80–100Misra, et al., 1996 30–47

2,4-D 20 Asmussen, et al., 1977 70Cole, et al., 1997 89–98

Dicamba 2 Cole, et al., 1997 90–100Fluormeturon 100 Rankins, et al., 1998 60Isoproturon 120 Patty, et al., 1997 99Mecoprop 20 Cole, et al., 1997 89–95Metolachlor 200 Arora, et al., 1996 16–100

Misra, et al., 1996 32–47Webster and Shaw 1996 55–74Tingle, et al., 1998 67–97

Metribuzin 60 Webster and Shaw 1996 50–76Tingle, et al., 1998 73–97

Norflurazon 600 Rankins, et al., 1998 65

Table 1 Summary of buffer studies measuring trapping efficiencies for specific pesticides. Koc values listed for each pesticideare from the NRCS Field Office Technical Guide, Section II Pesticide Property data base.

10

Adehesion

Ponded flow

Sediment deposition

Infiltration

Sorption

12

3

3

33

3

33

44

4

44

5

5

55

2

2

2

2 22

2

1

1

1

1

11

11

11

Designing Buffers forMaximum Pesticide TrappingEfficiency

LocationAll buffers can provide some protection of waterbodiesif they are sited between pesticide-treated fields andwater. Physical separation of spraying operations andwater reduces the chances for direct application towater where spray booms overhang water when turningat field ends. It can also reduce spray drift into water.However, to trap the pesticides in runoff and drift,buffers must be sited so that water runs over, or windpasses through, the buffer area. Concentrated flow isoften prevalent by the time field runoff reachesstreambanks. Natural berms may develop along banks,preventing overland flow into streams. This phenom-enon was illustrated by a study in Nebraska, wherewater runoff patterns were characterized between fieldedges of watersheds and streams (Eisenhaurer, et al.,1997). In one watershed, about 51 percent of the areahad runoff pathways that experienced sheetflow, butonly 22 percent of the area had sheetflow distances ofmore than 10 feet. Thus, buffers adjacent to water maybe limited in their ability to trap pesticides unless landcan be shaped to encourage sheetflow or spreader de-vices are incorporated into the design.

intercept both sheet and concentrated flow from fieldsand also can intercept pesticides close to the source.Wider grass strips encourage more sheetflow and infil-tration as runoff enters the edges of waterways.

Buffers are most effective in trapping pesticides whenlocated as close to treated fields as possible. Contourbuffer strips and vegetative barriers are most effectivebecause they are located within fields and are on thecontour, thus maximizing sheetflow across the buffer.Herbaceous wind barriers and cross wind trap stripslocated close to treated fields trap wind eroded particlescontaining adsorbed pesticides. Typically, the ratio ofrunoff source areas to buffers is smaller for this type ofbuffer than most edge-of-field buffers, which alsoincreases trapping efficiency. Grassed waterways

Numbering system of stream orders

Stream networks are designated by using stream orders.First order streams have no tributaries. A second-orderstream starts at the confluence of two first-order streams.The confluence of two second-order streams is a third-order stream, and so on. Most experts conclude thatstreamside conservation buffers are most effective onfirst- and second-order streams at the “top” of water-sheds. The greatest volume of runoff water, and there-fore pollutant volume, enters most stream systems fromthese small streams. Thus, intermittent as well as first-and second-order perennial streams require more veg-etative buffer protection. Little “new” water enters third-and fourth-order streams over banks. Conservation buff-ers along these larger streams provide other benefits,such as wildlife habitat and streambank protection, buthave less opportunity to intercept pesticides and im-prove water quality. In watershed planning, likely sourcesof pesticides can be identified based on cropping pat-terns. This information can be used to prioritize place-ment of conservation buffers.

11

Fields

Level spreader

Diversionberm

Level spreader

Forested Riparian Buffer

Gravel Erosionfabric

Spreader ditch

Gravel

Design of level spreader used for dispersing agricultural runoff through

a forested riparian buffer

Concentrated flow through a grass buffer

Concentrated flow is the nemesis of pesticide trappingby buffers. Natural berms often develop along fieldedges from deposition of sediment. Such berms becomebarriers to sheet flow off fields and should be removedby leveling when possible. Land can also be shaped toencourage broad, shallow sheet flow. New techniqueshave been developed to disperse concentrated flow. Forexample, level spreaders are constructed to laterallydisperse runoff uniformly across a slope. They consistof a long, narrow trench with an outlet lip of uniformelevation constructed in stable, undisturbed soil. Theoutlet area should have uniform slope and be well-vegetated. Small berms, or “water bars” may be con-structed to break up concentrated flow and redirect it assheet flow across buffers. Strategically located vegetativebarriers perpendicular to the flow can serve the samepurpose to slow runoff velocity and redirect runoff acrossan associated grass buffer as shallow sheetflow.

Detailed cross-section of level spreader trench for dispersing runoff

along the contour

How wide is wide enough?Appropriate widths for conservation buffers are debat-able. Widths are defined here as flow length across thebuffer. Buffer per unit area is affected by runoff flow rateand depth as well as by conditions within the buffer, suchas soil type and antecedent moisture, that affect waterinfiltration. Amount of runoff is affected by source areasize and properties as well as rainfall intensity and quantity.Selecting an appropriate buffer size often involves consid-eration of several desired functions, site conditions, andwhat is economically or politically practical.

Many studies have investigated sediment trapping effi-ciency of grass buffers. Dillaha, et al. (1989) found that30- and 15-foot strips of orchardgrass trapped 84 and 70percent of incoming solids, respectively. The source areaof runoff was 60 feet, or 4 times as wide as the15-foot buffers. Magette, et al. (1989) found that 30- and 15-foot strips of fescue trapped 75 and 52 percent of incomingsolids, respectively. The source area was 72 feet deep, or4.8 times as wide as the 15-foot buffers. Castelle, et al.,(1994) reviewed literature on buffer size requirements andconcluded that a range of buffer widths from 10 to 650 feetwas effective, depending on site-specific conditions. Abuffer width of at least 50 feet was necessary to protectwetlands and streams under most conditions.

12

Flow

Contributing area

Filter strip length

Filter stripwidth

Width and Length Used in Conservation Buffers

The width is measured in the direction of flow. Sinceconservation buffers are placed along the contour orperpendicular to the prevailing wind direction as fea-sible as possible, their direction at the narrow point iscalled width. This is analogous to the width of a culti-vated area in stripcropping or width of a contour strip.The flow of water moves parallel with the width. (Thesame is the case of other conservation practices, such ascross wind trap strips, which have movement of thewind across the width.) The length of a conservationbuffer is the longitudinal distance across the landscapethat the strip occupies perpendicular to the direction offlow. Other terms, such as flow length, may be used todepict the direction of flow (see insert figure).

The slope of the buffer and soil in the buffer area impactthe overall buffer performance. Steeper slopes in theconservation buffer strips increase flow velocity andshorten the time the contaminant material carried in therunoff water, both particulate and soluble, has to inter-act with the vegetation and soil in the buffer area. Thesoil is an important parameter in judging buffer effec-tiveness. Hydrologic soil groups (A, B, C, and D) areindicative of the infiltration and runoff potential of thesoil. Soil groups A and B have higher infiltration poten-tial; therefore, less runoff than groups C and D. Bufferstrips located on hydrologic soil groups C and D are lesseffective in treating run-on than buffer areas on A and Bsoils. The soil drainage class also determines the extentof soil moisture conditions and water storage availablein a soil.

With most conservation buffers, the greater flow length(width) of buffer area provides the greater entrapmentand removal of contaminants. However, an optimallength or area is soon reached where further distancedoes not result in proportionally greater efficiency. Abuffer strip that achieves 100 percent removal of con-taminants or completely reduces the water discharge tozero would be difficult and impractical to design andmaintain. Most practical designs are based on contami-nant removals of at least 50 to 60 percent (up to 80percent for sediment) and at the same time allow somedischarge at the end of the buffer.

Type and density of vegetation also influence buffereffectiveness. A large number of vegetative stems (usu-ally greater than 50 per square foot for most grasses) is

required to retard water and wind flows and provideenough surface area for attachment of contaminatematerial passing through the buffer strip. Stems shouldstand upright during runoff and wind events.

The width of a conservation buffer strip depends on anumber of factors. The purpose of the buffer strip mustbe defined. Buffers to entrap and deposit sediment arenot required to be as wide (only at least 20 feet) as buffersused to remove soluble compounds, such as nitratenitrogen or pesticides (as wide as 100 feet or longer). Ittakes more surface area and longer flow paths to adsorband infiltrate soluble material than to entrap solidmaterial. Climate conditions and storm eventsanticipated during the expected runoff events influencethe effectiveness of the buffer to retard flow and removepollutants. At times, conditions such as frozen and snowcovered soil, saturated soil, and crusted soil surfaces,severely reduce the function and effectiveness of bufferstrips.

13

Adequate buffer widths depend on field slopes andsource areas. A draft NRCS Conservation Practice Stan-dard for Filter Strips requires a minimum flow length of30 feet for the purpose of reducing sediment and sedi-ment-adsorbed contaminant loadings. It also sets ratiosof filter strip area to field area based on Universal SoilLoss Equation R factor values (rainfall amount andintensity) of regions: “The ratio of the field or disturbedarea to the filter strip area shall be less than 70:1 inregions with USLE R factor values 0 to 35, less than 60:1in regions with USLE R factor values 35 to 175, and lessthan 50:1 in regions with USLE R factor values of morethan 175.” Consult the local NRCS Field Office Techni-cal Guide for filter strip standards because these criteriavary depending on local conditions. Additional criteriamay apply to reduce dissolved contaminants in runoff.The draft national standard states: “Filter strip flowlength required to reduce dissolved contaminants inrunoff shall be based on management objectives, con-taminants of concern, and the volume of runoff from thefilter strip drainage area compared with the filter strip’sarea and infiltration capacity.”

Several site characteristics may dictate wider buffers,especially when trying to maximize water infiltrationand trapping of dissolved pesticides. For example, fine-textured soils generally have lower water infiltrationrates; or a high water-table underlying buffers may limitinfiltration. Iowa studies found that water infiltrationand trapping of dissolved herbicides by buffers wasleast effective when previous rains saturated soils.Vegetation within the buffer improves surface soil condi-tions, improving infiltration rates and internal soildrainage.

Narrow buffers have sometimes trapped pesticides ef-fectively. The specific pesticide studies cited in thispublication found that buffers as narrow as 1.6 feetcould be effective in trapping significant quantities ofpesticides. Increasing buffer width did not always sig-nificantly improve pesticide trapping. Tingle, et al. (1998)compared tall fescue buffers measuring 1.6, 3.2, 6.6, 9.8,and 13.1 feet wide placed below 72-foot-long soybeanplots. No significant differences in pesticide trappingefficiencies were found between buffer widths. Runoffloss of metribuzin was reduced by at least 73 percent, andrunoff loss of metolachlor was reduced at least 67 percentby all buffer widths.

While site characteristics, such as large source areas orslow permeability soils, may dictate larger buffers forhigh pesticide trapping efficiency, relatively smallbuffers should provide significant water quality benefits.Typical buffer widths of about 50 feet can be effective inreducing pesticide runoff by at least 50 percent if sheet

flow is maintained, depending on a number of factors asdescribed previously. Wider buffers may providegreater protection than narrow buffers in many settings,but where space or cost considerations limit bufferwidths, a narrow buffer is better than no buffer at all.

For more information

Section IV of the NRCS Field Office Technical Guide ineach state contains Conservation Practice Standardsdeveloped for that state. These state standards are basedon national standards in the NRCS National Handbook ofConservation Practices. National standards establishminimum requirements for state standards, which arespecifically tailored to each state's local conditions.Conservation Practice Standards include a practicedefinition, purposes of the practice, conditions wherethe practice applies, criteria for applying the practice,special considerations in applying the practice, practiceplans and specifications, and practice operation andmaintenance requirements. All applicable conservationbuffer practices have standards in the local Field OfficeTechnical Guide which include required buffer widths.

Information on selecting and sizing buffer practices forthe conservation buffer initiative is available at:

http://www.ftw.nrcs.usda.gov/tpham/buffer/akey.htm

Species selectionConservation buffers can be planted to perennialgrasses, legumes and forbs, woody plants, or acombination of the three. When available and able toperform the desired functions of specific buffer types,native plant species are preferred. Some annuallyharvested crops, such as small grains or legume-grassforages, can serve the purpose of buffers, either whenplanted adjacent to watercourses or in stripcroppingsystems–alternating strips of row crop and denselyplanted crops. In Texas (Hoffman, 1995), wheat wasmore effective in trapping herbicides than bermudagrasswhen planted in contour strips below a corn field.

Perennial grasses—Many buffer studies have usedcommon forage grass species, such as bromegrass,orchardgrass, fescue, and bermudagrass. While thesespecies have performed satisfactorily, researchers areinvestigating other species including native warm-season grasses. To date, few studies have compared theeffectiveness of grass species in trapping pesticides.Rankins, et al. (1998) compared giant reed (Arundo donax

L.), eastern gammagrass (Tripsacum dactyloides L.), bigbluestem (Andropogon gerardii Vitman), Alamo switch-grass (Panicum virgatum L.), and tall fescue planted infilter strips below cotton treated with fluormeturon andnorflurazon. All species were similar in effectiveness.

14

Ground water

Subsurface flow

Surface runoff

ZONE 2Managed Forest

ZONE 3Runoff Control

ZONE 1Undisturbed Forest

Hyporheic Zone

The native warm-seasongrass (switchgrass), wascompared to cool-seasongrasses (bromegrass), timo-thy (Phleum pratense L.),and fescue in ability to trapsediment and nutrients inIowa (Lee, 1997). Switch-grass filter strips removedsignificantly more sedi-ment, total N, nitrate-N, to-tal P, and PO4-P than cool-season grass filter strips.

Ideally, buffer grasses shouldproduce dense vegetationwith stiff, upright stems nearground level. Species thatform sods rather than clumpsprovide more uniform cov-erage. Because increased in-filtration and percolation ofwater into buffers is an important pesticide removal mecha-nism, species with deeper rooting patterns may also bemore effective. Upright growth and stiff stems can slowrunoff velocity and increase sediment deposition and infil-tration. Weak-stemmed species may be pushed over byrunoff, mat on the soil surface, and decrease infiltration.Because buffers trap considerable quantities of sediment,buffer species should be able to tolerate deposition ofsediment over crowns.

Stiff-stemmed grass species have received recent atten-tion for use as narrow hedges. Meyer, et al., (1995) foundthat a 19-inch-wide hedge of switchgrass or vetiver[Vetiveria zizanioides (L.) Nash.] ponded runoff to adepth of 10 inches and trapped more than 90 percent ofsediment coarser than 125 mm (fine sands and coarser).Such hedges can be used in contour strips. Over time,trapped sediment forms natural terraces. Short hedgescan also be integrated with other buffers to break upconcentrated flow and direct it across buffers. Vetiver isnot winter-hardy, but switchgrass is adapted to northernand southern climates and is widely used for conserva-tion buffer and grass hedge applications. Warm-seasongrasses, such as switchgrass and big bluestem, are alsotolerant to triazine herbicides that may be present in fieldrunoff.

Conservation buffer grass species and varieties shouldbe adapted to local conditions. Check local informationsources, such as NRCS and Extension, before makingselections. These sources can also recommend seedingrates and procedures. Some cost sharing programs mayalso have specific seeding requirements.

Woody species—Trees and shrubs can also trap sedi-ment, nutrients, and pesticides, as well as provide wildlifehabitat and streambank protection. The deep roots of treesalso help to intercept subsurface water-flow containingnitrate and introduce organic matter deep into the soil,facilitating denitrification and acting as a carbon source forpesticide-degrading micro-organisms.

Trees and shrubs are often used in combination with agrass buffer located adjacent to crop fields. Schultz, et al.(1995) describe a three-zone buffer with a 23-foot-widestrip of perennial grass (switchgrass preferred) adjacentto the crop field, two rows of shrubs next downgradient,and four or five rows of trees adjacent to the stream, fora total width of 66 feet. Gilliam, et al. (1997) describe a 50-foot-wide buffer with half in perennial grass and halfforest species. Welsch (1991) describes a three-zoneriparian buffer, where zone 1 is permanent woody veg-etation immediately adjacent to the streambank, zone 2is managed forest occupying a strip upslope from zone 1,and zone 3 is an herbaceous filter strip upslope from zone2. (See figure above.)

Selection of appropriate shrubs and trees for three-zoneriparian buffers depends highly on the following factors:• climate• site conditions, including soil type and depth to water

table• intended uses (possible harvest)• species of wildlife desired• tolerance of the vegetation species to the pesticides

contained in the runoff

A three-zone riparian forest buffer

15

For more information contact the local Soil and

Water Conservation District office. Additional

information:

• Banks and Buffers, A Guide to Selecting NativePlants for Streambanks and Shorelines. Availableon CD-ROM from Tennessee Valley Authority.Call 423-751-7338.

• Plants for Conservation Buffers. USDA, NRCS.

Economic considerationsWhen planning installation of a buffer, consider the costof installation and maintenance and the economic conse-quences of taking land out of crop production. Properdesign and installation can increase the expected lifespan of a buffer and reduce maintenance costs. Pay-ments from Federal, State, and private sources can helpto offset reductions in income and cover installationcosts. Some states allow a real estate tax exemption forthe area of a farm planted to a permanent conservationbuffer. Some income may be generated from buffers ifhaying or tree harvest is allowed, helping to offset loss ofincome from the traditional crop.

Cost-sharing programs often have specific design crite-ria that vary with locality. Local input is essential toensure that buffers are properly designed and installedso that they qualify for payments. Installation costs vary,depending on species and design. A typical riparianforest buffer with mixed hardwood seedlings and aswitchgrass strip can cost about $400 per acre to install(USDA-NRCS, 1999). Cost-share funds contributing 50 to75 percent or more of the cost of practice installation arewidely available.

Maintenance

Sediment removal—Conservation buffers must beintensively managed to maintain pesticide-trapping effi-ciency. Sediment trapped by buffers changes land shapeand may cause runoff to flow parallel to buffers, ratherthan across them. Similarly, sediment trapped in thecenter of grassed waterways may cause runoff to flowalong the edge of waterways, eroding gullies and in-creasing concentrated flow. Sediment must be removedperiodically from these areas, and vegetation reestab-lished when necessary. It is critical that sediment loadsflowing across buffers be limited as much as possible bysoil conservation practices applied to source fields. Thedraft NRCS Conservation Practice Standard for FilterStrips requires that average sheet and rill erosion abovethe filter strip be less than 10 tons per acre per year.

Mowing—Buffers may require mowing for weed con-trol or aesthetic reasons. Mowing can both positivelyand negatively affect pesticide trapping efficiency. Itcan encourage some grass species to tiller and producedenser vegetation at the soil surface. Mowing too short,especially with stiff-stemmed species, may reduce theflow retardance of the vegetation and injure the grass.Actively growing vegetation is more biologically active,taking up and degrading pesticides, and supplying car-bon for microbial degradation.

Harvest of grass or trees—One function of conserva-tion buffers is to trap nutrients, such as nitrogen andphosphorus. Periodic harvest of buffer vegetation re-moves trapped nutrients from the system, preventingeventual release to the soil and potential movement towater.

Impact of trapped herbicides—Herbicides trappedby buffers are degraded in the soil by microbial andchemical processes. Some herbicide may be taken up bybuffer plants either by roots or through foliage andmetabolized. However, excessive loads of certain

16

herbicides can injure buffer vegetation. Some preemer-gence herbicides have little impact on established plants.However, a few, such as the triazines, can injure grassesif present in high enough concentrations. Most bufferstudies have reported either no injury to buffer grasses,or only slight injury. In Iowa, bromegrass in buffers grewmost vigorously nearest source areas, apparently be-cause of nutrients trapped by the buffer (Arora, et al.,1996). Atrazine, cyanazine, and metolachlor trapped bythe buffers did not harm the grass. Grass seedlings aremost sensitive to herbicides in runoff. Thus, the greatestchance for harmful impact of herbicides in runoff wouldoccur during buffer establishment. Warm-season grasses,such as switchgrass, big bluestem, and bermudagrass,tolerate triazine herbicides well and would not be in-jured by runoff.

Avoid overspray—While buffers usually tolerate her-bicide concentrations in runoff, direct application ordrift of some herbicides can harm grasses and woodyplants. Nonselective herbicides used as burndown treat-ments in no-till production systems and on some herbi-cide-tolerant crop varieties can be especially damagingto buffer vegetation. Care should be taken to turn offspray booms and use control drip nozzles when drivingover buffers, or if booms extend over buffers whenturning. Squaring up cropland areas by varying bufferwidths along irregular streams or field borders makesapplication of herbicides easier, with fewer “point rows”and chances for overspray over buffers.

Turning on buffers—While buffers at the edge of fieldsmake convenient turning areas, driving heavy equip-ment on buffers can cause damage, compacting soil andreducing water infiltration, and causing ruts when soil iswet. Ruts may then encourage concentrated flow thatbypasses filtering ability of the buffer. Avoid driving onbuffers as much as possible, especially under wet soilconditions.

Livestock grazing—Grazing reduces buffer efficiencyby compacting soil and reducing grass heights. Woodyspecies may also be injured. Livestock can cause signifi-cant streambank degradation and directly contaminatewater. Some livestock producers would like to grazelivestock in fields adjacent to buffers for limited timeswithout having to fence buffers, for example, to allowgleaning of waste grain following harvest. Plan a grazingsystem to allow quick, intensive foraging under goodsoil moisture conditions. Remove livestock when soilsare wet to reduce potential damage to buffers.

Weed control—Buffers may harbor weeds requiringcontrol. Vigorous grass growth prevents growth of manyannual weeds, but some perennial weeds may requireeither mowing or spot treatment with herbicides. Nox-ious weeds must be controlled in the buffer.

Insect concerns—Buffers may harbor insect pests thatmove into crop fields. In some cases, such grassy areasare sprayed with insecticides to prevent damage toadjacent crops. If necessary, such treatments should beselected considering potential risks to adjacent aquaticecosystems. Buffers also can be a safe harbor for benefi-cial insects. Populations of these insects can build upwithin buffer areas and stay outside the cropland areatreated with insecticides. Tailor vegetation and buffermaintenance to promote beneficial insect populationswithin buffer areas.

Regional considerations in bufferdesign and maintenanceAppropriate buffer design depends on many local factorsincluding climate, soils, hydrology, and farmingpractices. Buffer species need to be adapted to the regionand appropriate for other functions, such as wildlifehabitat. Climate can greatly affect buffer function. Forexample, winter rains on frozen soil in northern areasproduce runoff that cannot be processed by buffers.Rainfed agriculture produces different runoff patternsthan agriculture dominated by irrigation. For thesereasons, planners must get local input to benefit fromlocal experience and research. In addition, specificbuffer design specifications are often required to qualifyfor cost-sharing programs.

Local NRCS and soil and water conservation districtoffices can provide design specifications for your area.Extension is also a source of expertise and information.Many State Extension Services have developed publica-tions on buffer design and maintenance.

17

Example of Extension publications:

• Buffer Strip Design, Establishment, and Maintenance.Publication Pm-1626b, Iowa State University Extension.

• Landowner’s Guide to Managing Streams in the East-ern United States. Publication 420-141, Virginia Co-operative Extension.

• Vegetative Filter Strips: Application, Installation, andMaintenance. Publication AEX-467-94, Ohio StateUniversity Extension.

Impact of Buffers onLeaching of Pesticides andNitrate

Because buffers increase water infiltration, concern hasbeen expressed that leaching of pesticides and nitratemight be increased, possibly to shallow ground water.When examining this possibility, consider the proper-ties of pollutants normally present in field runoff. Be-cause nitrate is water soluble and rarely adsorbed to soilparticles, it quickly moves into the soil with rainfall. Inmost settings surface runoff contains little nitrate. Asdescribed previously, nitrate is carried to surface waterprimarily by subsurface flow. Similarly, weakly adsorbedpesticides (which would have the greatest leaching risk)often are not detected at significant concentrations inrunoff, as they quickly move into the soil. Pesticidesdetected in runoff are primarily strongly adsorbed com-pounds attached to suspended sediment and moder-ately adsorbed compounds both adsorbed to sedimentand dissolved in water.

Strongly adsorbed pesticides have very low leachingpotential because of adsorption to soil. Moderatelyadsorbed pesticides can sometimes leach below theroot zone in small concentrations. However, quantities

leached are normally as much as 1,000 times smallerthan quantities carried off fields in surface runoff. Partsper million concentrations of some products can bedetected in runoff at the field edge, while concentra-tions detected in shallow ground water are often only afew parts per billion, if detected at all. Because manybuffers are located near streams, pesticides or nitrateleaching into buffers would most likely be carried bysubsurface flow to streams. Movement of runoff throughthe root zone soil of buffers before discharge to streamsby subsurface flow is much better than allowing surfacerunoff to directly enter streams. Pesticides can beadsorbed and degraded and nitrate taken up by plants ordenitrified within buffers.

Because of relatively low concentrations of pesticidetrapped in buffers, leaching risk from buffers should bemuch less than that from source fields. For example, inan Iowa study, atrazine concentrations in a source cornfield were 4,800 ppb in the surface 2 cm of soil after thefirst runoff event of the season (Fawcett, et al., 1995).Atrazine concentrations in the buffer strip were 750 ppb.Using BMPs to reduce pesticide runoff from sourcefields not only reduces pesticide loads ultimately reach-ing surface water, but also reduces loads trapped bybuffers. Conservation buffers have been shown to de-grade pesticides and to attenuate pesticide concentra-tions in subsurface water-flow. In Iowa (Schultz, et al.,1997), atrazine concentrations in soil water 2 feet belowa corn field were 13 ppb. Atrazine concentrations be-neath an adjacent grass and woody vegetation bufferwere only 0.2 ppb. In a Georgia study (Lowrance, et al.,1997), no atrazine was detected in shallow ground waterbeneath a 3-zone buffer for the first 2 years of the study.In the third year, a large rain event soon after herbicideapplication resulted in atrazine detection in monitoringwells 6.6 feet deep. A concentration of 6 ppb was de-tected at the field edge. At the downslope edge of a 26-foot-wide grass strip adjacent to the field, atrazine con-centrations declined to 2 ppb. At the downslope edge ofthe tree strip at the stream edge, atrazine was detectedat only 0.2 ppb.

Considering the relatively small load of pesticide inter-cepted by buffers compared to that applied to cropfields, and the adsorption and degradation of pesticidesby soil and vegetation in buffers, increased leaching ofpesticides does not appear to be a significant risk fromconservation buffers.

18

Integrating Buffers withBMPs

Conservation buffers can trap and degrade part of thepesticides that run off fields either adsorbed to sedi-ment or dissolved in water. However, buffers seldomtrap all pesticides in runoff. Buffers have been describedas “the last line of defense” or as acting to “polish” runoffafter it has been treated by other practices. Other BMPsare needed in a systems approach to adequately protectwater quality. Many practices can be used to reduceoffsite movement of pesticides, and can be selected andintegrated into cropping systems where appropriateand effective. Although not an exhaustive list, somecommon pesticide BMPs and descriptions are listedbelow.

Integrated Pest Management (IPM)—IPM systemsutilize pesticides in concert with nonchemical pest man-agement techniques. Pest populations are determinedand pesticides or other techniques are used only whenpest populations exceed economic thresholds. The low-est effective pesticide rate is used, and pesticide prod-ucts are selected to target specific pests and protectnontarget organisms.

Pesticide selection—Pesticides applied at low ratesreduce amounts available to run-off. Products that arestrongly adsorbed are less likely to move off fieldsdissolved in runoff.

Pesticide application timing—Risk of pesticide run-off is greatest when heavy rains closely follow pesticideapplication. Avoid applying pesticides if heavy rain isimminent. Sometimes long-term weather records canindicate application times when heavy rains are lesslikely. Post-emergence applications result in less runoffthan soil applications, as the crop and weeds behavesimilar to a buffer, increasing infiltration and pesticideadsorption by soil and foliage.

Banded application—Application of herbicides inbands over crop rows, combined with cultivation to con-trol weeds between rows, reduces total amounts of chemi-cal applied compared to broadcast applications.

Soil incorporation—Some herbicides and insecticidesare effective when mechanically incorporated into the soil(in the case of herbicides) or placed in crop furrows (in thecase of insecticides). Placing some of the applied chemicalbelow the soil surface protects it from surface runoff.Because tillage performed to incorporate pesticides buriessurface crop residue, it may increase erosion risk.

Conservation tillage—Surface crop residue reduces ero-sion and often increases water infiltration, reducing pesti-cide runoff (Fawcett, et al., 1994). No-till, especially afterhaving been practiced for several years, has sometimesdramatically reduced pesticide runoff, although pesticidesintercepted by surface crop residue may be subject torunoff if heavy rains follow application.

Nutrient management—Supplying the amount, select-ing the form, and determining the timing and placement ofcrop nutrients provide adequate soil fertility for food, fiber,and forage production while at the same time protectagainst any detrimental environmental risk.

Contour planting—Contour rows reduce erosion, slowrunoff, and increase infiltration. Orientation of rows adja-cent to buffers may need to be adjusted to direct runoff assheetflow across buffers.

Stripcropping—When strips of densely planted crops,such as forages or small grains, are alternated with stripsof row crops, the densely planted crop acts as a buffer.When the strips are planted on the contour, runoff anderosion are reduced and more runoff water enters the soil.

Crop rotation—Rotation of crops can disrupt life cyclesof insects, diseases, and weeds, and reduce the necessityfor pesticide treatments. Pesticides can be rotated as thecrop is rotated, thus reducing the amount of any onepesticide used on that field.

Terraces/detention ponds—These structures shortenslope length, trap sediment, and increase water infiltration,reducing pesticide runoff.

Irrigation timing—Irrigation after application of soil-applied herbicides moves the chemical into the soil profile,improving weed control and protecting the chemical fromlater rainfall-runoff events. Use of polyacrylamide (PAM),which reduces irrigation-induced erosion, also improveswater infiltration, thereby reducing pesticide runoff.

19

Irrigation water management—Improved manage-ment of the rate and amount of irrigation water canreduce deep percolation, tailwater, and erosion losses.

Compaction reduction—Correcting compaction prob-lems encourages water infiltration and reduces runoff.

Subsurface drainage—A high water table can result inexcessive runoff and pesticide loss. Improving drainageincreases water infiltration and reduces pesticide run-off. As subsurface drainage carries nitrate to streams,treatment of tile effluent in a constructed wetland orbuffer area may be desirable.

Summary

Conservation buffers are an effective tool to reducepesticides losses to water when used in conjunctionwith other BMPs. Pesticide trapping is most efficientwhen sheetflow, rather than concentrated flow, occursacross buffers. Sheetflow can be encouraged by properbuffer design, including such innovations as level spread-ers, water bars, and stiff-grass hedges. As sediment istrapped, waterflow patterns are changed. Thus, buffermaintenance is critical. Sediment must be periodicallyremoved and buffers reshaped to maintain effective-ness. Other soil conservation practices must be used inconjunction with buffers to prolong the effective bufferlife.

Conservation buffers provide many other benefits, in-cluding trapping sediment and nutrients, providing wild-life habitat, protecting streambanks, and increasing farm-ing safety. By varying buffer width along irregular streamsor field borders, the cropped areas can be “squared up,”reducing sprayer overlaps and making fields more com-patible with Global Positioning Systems controls usedin precision farming. Many buffer types can be selectedto match site conditions and desired benefits. Appropri-ate buffer plant species should be selected to matchlocal conditions. Research into buffer effectiveness inpesticide trapping is a relatively new field. As researchcontinues, buffer designs and maintenance procedureswill undoubtedly be refined to maximize their effective-ness.

Literature Cited

Arora, K., S.K. Mickelson, J.L. Baker, D.P. Tierney, andC.J. Peter. 1996. Herbicide retention by vegetativebuffer strips from runoff under natural rainfall.Trans of the ASAE 30(6):2155-2162.

Arora, K., S.K. Mickelson, and J.L. Baker. 1995. Evaluat-ing vegetative buffer strips for herbicide retention.Paper No. 95-2699. American Society of Agricul-tural Engineers, St. Joseph, MI.

Arora, K., J.L. Baker, S.K. Mickelson, and D.P. Tierney.1993. Evaluating herbicide removal by buffer stripsunder natural rainfall. Paper No. 93-2593. Ameri-can Society of Agricultural Engineers, St. Joseph,MI 49085.

Asmussen, L.E., A.W. White Jr., E.W. Hauser, and J.M.Sheridan. 1977. Reduction of 2,4-D load in surfacerunoff down a grassed waterway. J. Environ. Qual.6(2):159-162.

Baker, J.L. 1983. Agricultural areas as nonpoint sourcesof pollution. In M.R. Overcash and J.M. Davidson(eds.) Environmental Impacts of Nonpoint SourcePollution. Ann Arbor Sci. Publ., Inc., Ann Arbor,MI. pp. 275-310.

Bharati, Luna. 1997. Infiltration in a Coland clay loamunder a six-year old multi-species riparian bufferstrip, cultivated row crops and continually grazedpasture. M.S. Dissertation, Iowa State Univ., Ames,IA.

Boyd, P.M., L.W. Wulf, J.L. Baker, and S.K. Mickelson,1999. Pesticide transport over and through the soilprofile of a vegetative filter strip. American Soci-ety of Agricultural Engineers. ASAE Paper no.992077

Castelle, A.J., A.W. Johnson, and C.Conolly. 1994. Wet-land and stream buffer size requirements - a re-view. J. Environ. Qual. 23:878-882.

Cole, J.T., J.H. Baird, N.T. Basta, R.L. Huhnke, D.E.Storm, G.V. Johnson, M.E. Payton, M.D. Smolen,D.L. Martin, and J.C. Cole. 1997. Influence of buff-ers on pesticide and nutrient runoff frombermudagrass turf. J. Environ. Qual. 26:1589-1598.

Dillaha, T.A., R.B. Reneau, S. Mostaghimi, and D. Lee.1989. Vegetative filter strips for agriculturalnonpoint source pollution control. Trans. of theASAE 32(2):513-519.

Eisenhaurer, Dean, Thomas Franti, Mike Dosskey, andKyle Hoagland. 1997. Field assessment of surfacerunoff patterns into riparian areas. Proceedings Veg-etative Filter Strip and Riparian Buffer ResearchSymposium, Dec. 17, 1997. University of Nebraska,Lincoln.

20

Fawcett, Richard S., Brian R. Christensen, and Dennis P.Tierney. 1994. The impact of conservation tillage onpesticide runoff into surface water: a review andanalysis. J. Soil and Water Cons. 49(2):126-135.

Fawcett, R.S., D.P. Tierney, C.J. Peter, J.L. Baker, S.K.Mickelson, J.L. Hatfield, D.W. Hoffman, and T.G.Franti. 1995. Protecting aquatic ecosystems withvegetative filter strips and conservation tillage.Proceedings of the National Agricultural Ecosys-tem Management Conference. New Orleans, LA,December 13-15, 1995. Conservation TechnologyInformation Center, West Lafayette, IN.

Franti, T.G., C.J. Peter, D.P. Tierney, R.S. Fawcett, and S.A.Meyers. 1998. Reducing herbicide losses from tile-outlet terraces. J. Soil and Water Cons. 53(1):25-31.

Gilliam, J.W., D.L. Osmond, and R.O. Evans. 1997. Se-lected agricultural best management practices tocontrol nitrogen in the Neuse River Basin, NorthCarolina State University Technical Bulletin 311,Raleigh, NC.

Hall, J.K., N.L. Hartwig, and L.D. Hoffman. 1983. Appli-cation mode and alternate cropping effects onatrazine losses from a hillside. J. Environ. Qual.12(3):336-340.

Hoffman, Dennis W. 1995. Use of contour grass andwheat filter strips to reduce runoff losses of herbi-cides. Proc. Austin Water Quality Meeting, Texas A& M Univ., Temple, TX.

Lee, Kye-Han. 1997. Nutrient and sediment removal byswitchgrass and cool-season grass filter strips.Proceedings Vegetative Filter Strip and RiparianBuffer Research Symposium, Dec. 17, 1997, Uni-versity of Nebraska, Lincoln.

Lowrance, R., L.S. Altier, J.D. Newbold, R.R. Schnabel,P.M. Groffman, J.M. Denver, D.L. Correll, J.W.Gilliam, J.L. Robinson, R.S. Brinsfield, K.W. Staver,W. Lucas, and A.H. Todd. 1995. Water quality func-tions of riparian forest buffer systems in the Chesa-peake Bay Watershed. EPA 903-R-95-004.

Lowrance, R., G. Vellidis, R.D. Wauchope, P. Gay, andD.D. Bosch. 1997. Herbicide transport in a man-aged riparian forest buffer system. Trans. of theASAE 40(4):1047-1057.

Magette, William L., Russel B. Brinsfield, Robert E.Palmer, and James D. Wood. 1989. Nutrient andsediment removal by vegetated filter strips. Trans.of the ASAE 32(2):663-667.

Matter, Michael Patrick. 1993. Sorption kinetics of atrazineand hydroxyatrazine in freshwater wetlands. M.S.Dissertation, Iowa State Univ., Ames, IA.

Meyer, L.D., S.M. Dabney, and W.C. Harmon. 1995. Sedi-ment-trapping effectiveness of stiff-grass hedge.Trans. of the ASAE 38(3):809-815.

Mickelson, S.K., J.L. Baker, S.W. Melvin, R.S. Fawcett,D.P. Tierney, and C.J. Peter. 1998. Effects of soilincorporation and setbacks on herbicide runofffrom a tile-outlet terraced field. J. Soil and WaterCons. 53(1):18-25.

Mickelson, S.K. and J.L. Baker. 1993. Buffer strips forcontrolling herbicide runoff losses. Paper No.932084. American Society of Agricultural Engi-neers, St. Joseph, MI.

Misra, Akhilesh Kumar. 1994. Effectiveness of vegeta-tive buffer strips in reducing herbicide transportwith surface runoff under simulated rainfall. Ph.D.Dissertation, Iowa State Univ., Ames, IA.

Misra, A., J.L. Baker, S.K. Mickelson, and H. Shang. 1996.Contributing area and concentration effects onherbicide removal by vegetative buffer systems.Trans. of the ASAE 39(6):2105-2111.

Misra, A., J.L. Baker, S.K. Mickelson, and H. Shang. 1994.Effectiveness of vegetative buffer strips in reduc-ing herbicide transport with surface runoff undersimulated rainfall. Paper No. 942146. AmericanSociety of Agricultural Engineers, St. Joseph, MI.

Patty, L., B. Real, and J.J. Gril. 1997. The use of grassedbuffer strips to remove pesticides, nitrate andsoluble phosphorous compounds from runoff wa-ter. Pesticide Sci. 49:243-251.

Rankins, A., Jr., D.R. Shaw, M. Boyette, and S.M. Seifert.1998. Minimizing herbicide and sediment losses inrunoff with vegetative filter strip. Abstracts WeedSci. Soc. Am. 38:59.

Reicosky, D.C., W.D. Kemper, G.W. Langdale, C.L. Dou-glas, Jr., and P.E. Rasmussen. 1995. Soil organicmatter changes resulting from tillage and biomassproduction. J. Soil and Water Cons. 50:253-261.

Rohde, W.A., L.E. Asmussen, E.W. Hauser, R.D.Wauchope, and H.D. Allison. 1980. Trifluralin move-ment in runoff from a small agricultural water-shed. J. Environ. Qual. 9(1):37-42.

21

Schultz, R.C., J.P. Colletti, T.M. Isenhart, W.W. Simpkins,C.W. Mize, and M.L. Thompson. 1995. Design andplacement of a multi-species riparian buffer stripsystem. Agroforestry Syst. 29:201-226.

Schultz, Richard C., Amy Kuehl, Joe P. Colletti, PaulWray, Tom Isenhart, and Laura Miller. 1997. Ripar-ian buffer systems. Iowa State University Publica-tion Pm-1626a, Ames, IA.

Tingle, C.H., D.R. Shaw, M. Boyette, and G.P. Murphy.1998. Metolachlor and metribuzin losses in runoffas affected by width of vegetative filter strips.Weed Sci. 46:475-479.

USDA-NRCS. CORE4 Conservation Practices. The Com-mon Sense Approach to Natural Resource Conser-vation. Natural Resources Conservation Service.Fort Worth, Texas. 385 pp.

Wauchope, R.D. 1978. The pesticide content of surfacewater draining from agricultural fields – a review.J. Environ. Qual. 7:459-472.

Webster, E.P. and D.R. Shaw. 1996. Impact of vegetativefilter strips on herbicide loss in runoff from soy-bean (Glycine max). Weed Sci. 44:662-671.

Welsch, D.J. 1991. Riparian Forest Buffers. USDA-FSPub. No. NA-PR-07-91. USDA-FS, Radnor, PA.

Wood, H.B. 1977. Hydrologic differences between se-lected forested and agricultural soils in Hawaii.Soil Sci. Soc. Am. J. 41:132-136.

Yonts, C.D., R.G. Wilson, and G.L. Hein. 1996. Control ofpesticides and nitrates in surface irrigation runoffwater. Paper No. 96-1041. American Society ofAgricultural Engineers, St. Joseph, MI.