Research YCnt Can Use

Best Management Practicesto Reduce Pesticide Runofffrom TurfA common-sense approach can greatlyreduce the risk of water contamination.BY B. E. BRANHAM, F. Z. KANDIL, AND J. MUELLER

Golf turf management has madehuge strides over the past 40years that have allowed golf

course superintendents to achieveexcellent turf quality. However, achiev-ing these very high levels of turf qualityrequires numerous inputs, includingfertilizers, irrigation, topdressing, culti-vation, wetting agents, biostimulants,and pesticides. While practices such astopdressing, cultivation, and wettingagents are considered environmentallybenign, fertilizers and pesticides havereceived much scrutiny since some ofthese products can move off the turfand into ground and surface water.

Pesticide leaching from turf has beenstudied intensively,1,5,6,9and while pesti-cide leaching is a major problem in rowcrops, leaching of pesticides from turfpresents much less risk than previouslysuspected. Pesticide leaching in turf is amuch smaller problem than in rowcrops for two primary reasons.

First, the acreage treated with pesti-cides on all the golf courses in theUnited States is a drop in the proverbialbucket compared to row crop agricul-ture. The National Golf Foundationreported that at the end of 2002, there-was the equivalent of 14,725 18-holegDlf facilities in the United States. If weassume that each golf course contains,on average, 3 acres of putting greens, 5acres of tees, and 30 acres of fairways,then the total number of golf course

26 G R E ENS E C T ION R E COR 0

acres in the United States receivingpesticide applications (roughs typicallyreceive little in the way of pesticideapplications, although weed.controlmay be practiced) would total 559,550acres. This is less total acreage than theamount of corn and soybeans plantedin a typical county in central Illinois.Nationally, in 2001, approximately75.752 million acres of land wereplanted to corn, while 74.105 millionacres were planted to soybeans. Most ofthese receive some kind of pesticideapplication. All the intensively managedgolf course acres in the United Statesrepresent less than 0.4% of the totalacreage planted to the two largest cropsgrown in the U.S.

A second reason why turf presentsless of a risk for pesticide leaching is theturf itself. A previous USGA-fundedresearch project examined the effectof turf on pesticide movement anddegradation.2,3,4 We found that whenpesticides are applied to turf, leachingis reduced and degradation rates areincreased when compared to the samepesticides applied to bare soil (acommon practice in row crops).

These two differences have led manyto conclude that the risk of ground-water contamination from turf grasspesticides is low, but not non-existent.Proper Inanagement is still key, and oncertain sites, particularly those withsandy soils, shallow groundwater, and

proximity to water bodies, turfmanagers need to pick the pesticidesthey do use with care.

Pesticide runoff, however, is a com-pletely different issue. What is runoff?Runoff is a natural event that occurswhen a rain or irrigation event pro-duces more water than the soil/turf canaccept. This is a fairly common occur-rence, and depending upon soil types,slopes, ete., it may occur often or rarelyon a particular site. Runoff per se is nota bad thing, but when the runoff carriespesticides, nutrients, or other pollutants,problems may arise.

Whereas pesticide leaching is mostlya threat to groundwater (although theuse of tile drains also can threaten sur-face waters with pesticide leachate),pesticide runoff is a threat to surfacewater. Most golf courses have somewater features associated with them, andoften streams, rivers, or storm drains areused to accept runoff from golf courses.Some initial research has shown thatpesticide runoff can be significant, withsome researchers reporting as much as10% of the applied pesticide transportedin runoff.7

INVESTIGATING RUNOFFWith this background in mind, weexamined some management practicesthat might reduce the concentration ofpesticides when runoff does occur froma golf course. We first constructed a site

Runoff plots with a 5% slope were constructed at the University of Illinois to study the effects ofpost-application irrigation and clipping management on runoff of pesticides of varying solubilities.

to conduct runoff research. This sitewas sloped, but it did require somemodification to suit our needs. Thatmodification was provided by MunieOutdoor Services, a St. Louis-basedcompany that donated time and equip-ment to produce a plot area with auniform 5% slope that was approxi-mately 150 ft. X 35 ft.

They also installed a mist irrigationsystem that could provide two intensi-ties of simulated rain events. Rain dropshave much different energy than theoutput from a mist head, which is veryimportant on bare soil, but we believethe energy difference is less importantwhen a turf cover is in place. Mter theplots were constructed in the fall, theywere allowed to settle over the winterand were sodded the next spring withcreeping bentgrass. The rest of thesummer was spent installing the runoffcollection equipment and testing thesystem, and by the end of the SUffill1erwe conducted a test run.

In the summer of 2003, we had thepersonnel and equipment in place toconduct the experiments. We evaluatedthree possible strategies to reduce pesti-cide runoff. First, can irrigation applieda short time after pesticide application

significantly reduce pesticide runoff?By washing the pesticide off the leafsurface and deeper into thatch and soil,can the concentration and total quantityof pesticide in runoff be reduced?

The second experin1ent examinedthe length of time between pesticideapplication and runoff event. Some turfmanagers and n1any homeowners usenatural rainfall in place of irrigation. Ifrain is forecast, an application of pesti-cide or fertilizer may be applied and therain is used to water-in the product. Ofcourse it the rain produces runoff,pesticide loss could be quite high. Canrunoff potential be reduced by applyinga small amount of irrigation prior tothe runoff event and thus reducepesticide runoff?

The third experiment centered onclipping management. Turf is a uniquecrop in that each pesticide application ismade directly onto the foliage. Evenwhen a pesticide is primarily root-absorbed, a significant quantity of thepesticide will adhere to leaf tissue. Idon't believe that we have consideredclippings to be a source of pesticidecontamination, but the first mowingfollowing a pesticide application effec-tively frees up a significant portion of

the pesticide application. If a rain eventmoves these clippings, a significantamount of pesticide will be transportedwith the clippings.

An even thornier issue results whenclippings are collected. If the clippingsare composted, rapid degradation of thepesticide residues will result, but caremust be taken to prevent rainfall fromleaching pesticides from the clippings. Ifthe clippings are simply scattered in therough, turf managers may be uninten-tionally producing areas with high con-centrations of pesticides that may besusceptible to leaching or runoff.

EXPERIMENTAL PROCEDURESIn each experiment, pesticides wereapplied as a three-way tank mix. Weselected pesticides based upon theirwater solubility and ease of analysis byhigh-performance liquid chromatog-raphy (HPLC). Each tank mix containeda pesticide we classified as having high,medium, or low water solubility. Watersolubility plays a donunant role in theavailability of the pesticide for runoff.Pesticides with higher water solubilitiesare more readily moved with flowingwater. Pesticides with very low watersolubilities will move in lower concen-

) A N U A R Y - F E B R U A R Y 2 () () 5 27

Following pesticide application. irrigation was applieduntil all plots produced at least 40 liters of runoff.

trations in water. Best managementpractices may need to be modifiedbased upon water solubility. In otherwords, what works best to reduce run-off of a highly water-soluble pesticidemay not be as effective with a water-insoluble pesticide.

Following pesticide application, themist irrigation system was turned on atthe appropriate time for each experi-ment to produce runoff. Irrigation wasapplied until all plots produced at least40 liters of runoff. In each experiment,approximately 2 hours of irrigation wasapplied. From each 40-liter runoffsample, a 4-liter subsample was collectedinto amber glass jugs. The samples wereanalyzed by HPLC to determine theamount of each pesticide present in thewater samples.

The first experiment examined theeffectiveness of post-application irriga-tion in reducing pesticide runoff. Threepesticides - chlorothalonil (DaconilUltrex™), paclobutrazol (Trimmit™),and mefanoxam (Subdue Maxx™) -were applied and 0.2 inch of post-application irrigation was hand applied

28 G R E ENS E C T ION R E COR D

at 0.25, 1,4,8, or 24 hours after pesti-cide application. The simulated runoff-producing rain event was initiated at 25hours after pesticide application (i.e.,simulated rainfall began 1 hour after thelast pesticide washoff treatment wasapplied).

RESULTSThe results of the first experimentwere disappointing. No matter howwe examined the data, there were fewmeaningful differences. The largestpoint from the trial was that post-appli-cation irrigation was not effective inreducing the amount of pesticide avail-able for runoff. Closer inspection of thedata Yielded one significant finding.Chlorothalonil runoff was reduced bypost-application irrigation at 15 minutesafter pesticide application. This maymake sense from a pesticide chemistryviewpoint. Chlorothalonil is very waterinsoluble, with a commonly acceptedwater solubility of 0.6 PPM.8 Productswith water solubilities this low areusually applied as an emulsion in waterin order to get the product into a spray-

able form. Once the spray dries on theleaf surface, the emulsifying characteris-tics are lost and the pesticide behavesaccording to its natural water solubility.

A pesticide, or any organic chemical,with water solubility below 1 PPM willbe very strongly sorbed to the wax andother non-polar compounds of the leafsurface. Once these pesticides dry onthe leaf surface, they're literally stuckthere. By applying irrigation soon afterapplication, some of this drying will beprevented and a larger mass of thepesticide can be moved deeper into theturf profile. Once a water-insolublepesticide has dried on the leaf surface,post-application irrigation will not beeffective in moving the pesticideoff the leaf.

With the fungicide chlorothalonil,post-application irrigation immediatelyafter application would not be a goodpractice since the product needs to beon the leaf surface to exert its fungi-cidal activity. However, if the intendedsite of action is the soil or thatch sur-face, as, for example, preemergenceherbicides, these products shouldreceive post-application irrigation assoon as the application is completed.This not only reduces the amount ofpesticide available for runoff; it alsoincreases the amount of pesticidereaching the soil or thatch surface.

The second experiment examinedthe impact of the interval betweenpesticide application and runoff event.While no one can control when itrains, it is still instructive to understandthe importance of the interval betweenpesticide application and runoff. In thisexperiment, pesticides were applied at12,24, 48, or 72 hours prior to therunoff event. The pesticides appliedwere pendimethalin (PreM™), propi-conazole (Banner Maxx™), andmefanoxam (Subdue Maxx™).

In this experiment, the results weredramatic. Regardless of water solubility,the longer the time between pesticideapplication and runoff, the less pesticidewas detected in runoff. And while thiswould be expected, what was interest-

amounts of active ingredient. On amass basis, more chlorothalonil was lostthan either of the other two pesticides.However, on a percent-of-applied basis,chlorothalonillost much less than theother two pesticides (Table 2). Chloro-thalonil is an older product that requireshigher use rates than many newer pesti-cides, thus chlorothalonil was applied ata rate of 11.2 lbs. ail A, while newerchemistries are usually applied at ratesof lIb. ailA or less. Even though chloro-thalonil is very water insoluble and lesslikely to run off (as shown by thepercentage data), more chlorothalonilwas recovered in runoff because it wasapplied at rates of 16 to 44 times higherthan the other two pesticides. Second,pesticide mass is the product ofpesticide concentration in runoff andthe total volume of runoff collected.The plots we used in this trial weredeveloped to be as uniform as possible,and yet there were still large differencesin runoff volumes between plots. Thisdirectly affects the runoff mass and canmake the data difficult to interpret.

Clipping management can have a bigimpact on pesticide runoff. Pesticiderunoff was reduced by 34% to 57% byremoving clippings. We doubt that thehigher mass of pesticide runoff whereclippings were returned can be attrib-uted to clippings in the runoff. Whilewe did observe some clippings in therunoff water, we removed the clippingsby filtration prior to analysis. The massof pesticide found on the sediment(clippings and other particles) was asmall fraction of the amounts recoveredfrom the runoff water. Thus, the reduc-

,Clipping, Treat~tmt

Removed::Returned

RemovedReturned

Removed.Returned, I, .

11.2

0.25

ApplicationRate (Ibs. ai/A) ..

0.7mefanoxam

paclobutrazol

chlorothalonil

, i, 1

: Pesticide

Our third experiment evaluated theeffects of removing clippings on pesti-cide runoff. On golf course greens, tees,and fairways, pesticides are applied asoften as once every two weeks duringthe summer. A significant portion ofthe pesticide application is deposited onthe leaf tissue, and much of the appli-cation will remain sorbed (a term thatdescribes substances that can be bothadsorbed and absorbed) to the leaftissue. This study was simplified so thatwe compared only two treatments,clippings removed versus clippings re-turned. In this experiment, pesticideswere applied at 9 a.m. on July 15,2003.The plots were mowed the followingday at 9 a.m. and the runoff event wasinitiated one hour later at 10 a.m. bysimulating runoff via irrigation.

As might be expected, removingclippings reduced pesticide runoff(Table 2). When examining the data ona mass basis, i.e., the total quantity ofpesticide removed, the data must beconsidered in view of several importantfactors. First, an important factor inreducing pesticide runoff (as well asother forms of off-site transport) is touse pesticides that require smaller

Table IPesticides used in runoff studies at the University of Illinois

, Common Name Trade Name . Water Solubility (mg/L)

mefanoxam SubdueMaxx 26,000

propiconazole BannerMaxx 110

paclobutrazol Trimmit , 35

chlorothalonil Daconil 0.6pendimethalin Pendulum 0.3

ing was that, in general, the differencesin runoff were significant between run-off at 12 hours following applicationversus 24,48, or 72 hours after appli-cation. In other words, if runoff occurs1, 2, or 3 days following application,there is not a great difference in theamount of pesticide that runs off. But ifthe runoff event occurs at 12 hours orless after application, there will be asubstantial increase in the amount ofpesticide runoff that occurs. Forexample, on a mass basis, we recovered8.9 mg of pendimethalin in runoffwater when runoff occurred at 12hours after application, but only 1.5,1.6, or 1.2 mg if runoff occurred at 72,48, or 24 hours following application,respectively. Similar results wereobtained for the other two pesticidesin this study.

One surprising result of this trial wasthat, on a mass basis, there was morepropiconazole in the runoff thanmefanoxam. This result was counter toour hypothesis that the more watersoluble a pesticide, the more susceptibleit is to runoff. In general, the initialconcentration of mefanoxam in therunoff was higher than propiconazole,but as more runoff came off, the con-centration of mefanoxam decreasedwhile that of propiconazole did notdecrease appreciably. Perhaps sincemefanoxam is much more water soluble(see Table 1), some of it may move intothe soil and thatch more readily withthe onset of precipitation, whereas pro-piconazole, which is less water soluble,may remain in the upper canopy whereit can continue to partition into waterflowing across the turf surface.

JANUARY-FEBRUARY 2005 29

tion in pesticide runoff where clippingswere removed is most likely a directresult of the decrease in the amount ofpesticide available when the runoffoccurs. However, while the reductionin pesticide in the runoff was substan-tial, it begs the question of what hap-pens to the clippings. If the clippingsare simply deposited elsewhere on thegolf course, then the runoff problemhasn't necessarily been reduced; it's justredistributed.

LESSONS LEARNEDThe purpose of this research was todevelop best management practices toreduce pesticide runoff. The mosteffective practice was to remove clip-pings, but the clippings themselves con-tain a significant amount of pesticide,and these must be dealt with respon-sibly.The turf in the field representswhat is termed a non-point sourcepollution problem; that is, the potentialpollutants are distributed across a largearea at low concentrations. Collectingclippings and putting them in a pilewould essentially create a point sourcepollution problem. However, creating acompost pile of clippings should permitrelatively rapid degradation of the pesti-cides in the pile, and if drainage is con-trolled, this would be a particularlygood option.

Regardless of whether or not youremove clippings as part of a best man-agement program to reduce pesticiderunoff, this research illustrates that clip-pings can be an important source ofpesticides. Whether you return clip-pings or collect them, be aware thatclippings harvested immediately follow-ing a pesticide application will containa significant quantity of pesticide.Returning those clippings to the turfwould be valuable particularly in thecase of soil-active pesticides such aspreemergence annual grass controlherbicides and root-absorbed productssuch as the plant growth regulatorspaclobutrazol or flurprimidol.

Pesticide application within 12 hoursof an expected rain event should be

30 G R E ENS E C T ION R E COR 0

avoided. Runoff events occurring at24- 72 hours after pesticide applicationwill contain reduced pesticide concen-trations versus runoff that occurswithin 12 hours of a pesticideapplication.

Choosing pesticides that require lowactive ingredient application ratesdramatically reduces the amount ofpesticide runoff. Many newer pesticidechemistries have application rates of30-120 grams ai/A (~0.1-0.3Ibs. ai/A).The best way to reduce pesticide runoffor leaching is to not use a pesticide. Thesecond best way is to choose a pesticidewith good environmental properties,and one of the best is a low applicationrate.

Lastly, the use of buffer strips is a bestmanagement practice. A buffer strip is avegetated strip that is not treated withpesticide. In our runoff experiments,the pesticides were applied within 2feet of the runoff collection apparatus.Any increase in the length of untreatedturf or other landscape plantings be-tween the treated turf and the pointwhere runoff water would enter astream, drain, or other direct access towater will dramatically reduce pesticiderunoff. This occurs for two reasons.First, turf will remove some of thepesticide that is flowing across it; that is,some pesticide will absorb to the turf-grass plants. Second, as runoff contain-ing pesticide enters the buffer stripwhere no pesticide is present, simpledilution reduces the pesticide concen-tration that ultimately enters the waterbody.

Pesticide runoff is an important issuethat golf course superintendents mustbe aware of and recognize wherepotential problems exist. Bodies ofwater flowing through the golf courseneed to be protected. Even if your golfcourse does not have a surface waterfeature, care must still be exercised.Many golf course superintendents usesurface drains to remove excess waterfrom low-lying or poorly drained areas.Often these drains ultimately lead to asurface water body. As a result, pesti-

cides applied to a fairway may bereadily moved off the golf course ifsurface drains are used to removeexcess water.

LITERATURE CITED1. Cisar,]. L., and G. H. Synder. 1996. Mobilityand persistence of pesticides applied to a USGAgreen. III: Organophosphate recovery in clip-pings, thatch, soil, and percolate. Crop Sa.36: 1433-1438.

2. Gardner, 0. S., and B. E. Branham. 200l.Mobility and dissipation of ethofumesate andhalofenozide in turf grass and bare soil.] Agric.Food Chem. 49:2894-2898.3. Gardner, 0. S., and B. E. Branham. 2001.Effect of turf grass cover and irrigation on soilmobility and dissipation of mefanoxam andpropiconazole.] Environ. Qual. 30: 1612-1618.

4. Gardner, 0. S., B. E. Branham, and 0.WLickfeldt. 2000. Effect of turf grass on soilmobility and dissipation of cyproconazole.Crop Sci. 40:1333-1339.5. Gold, A.]., T. G. Morton, W M. Sullivan, and]. McClory. 1988. Leaching of 2, 4-D anddicamba from home lawns. Ttater,Air, and SoilPollution. 37:121-129.

6. Petrovic,A. M., W C. Barrett, 1. Larsson-Kovach, C. M. Reid, and D.]. Lisk. 1996. Theinfluence of a peat amendment and turf densityon downward migration of metalaxyl fungicidein creeping bentgrass sand lysimeters.Chemosphere.33(11):2335-2340.7. Smith,A. E., and 0. C. Bridges. 1996. Move-ment of certain herbicides following applicationto simulated golf course greens and fairways.Crop Sci. 36:1439-1445.8. Wauchope, R. D., T. M. Butler, A. G. Hornsby,P.WM. Augustijn-Beckers, and]. P. Burt. 1991.The SCSI ARSI CES pesticide database forenvironmental decision-making. Rev. Environ.Contam. TOxicol.123:1-155.9. Yates, M. v: 1995. The fate of pesticides andfertilizers in a turf grass environment. USGAGreen Section Record. 33(1):10-12.

Editor's Note: This article and manyothers reporting the results of projectsfunded by USGA's Turfgrass andEnvironmental Research Program canbe found in USGA's Tuifgrass andEnvironmental Research Online(http://usgatero.msu.edu).

B. E. BRANHAM, PH.D., AssociatePrifessor; F Z. KANDIL, PH.D., ResearchAssistant; and J. MUELLER, ResearchAssistant; Department if Natural Resourcesand Environmental Sciences)University ifflllnois} Urbana} fllinois.