Environmental factors influencing survival of threespine ...faculty.ucr.edu › ~walton › Walton...

90 Journal of Vector Ecology June2007

Environmental factors influencing survival of threespine stickleback (Gasterosteus aculeatus) in a multipurpose constructed treatment wetland in

southern California

WilliamE.Walton1,MargaretC.Wirth1,andParkerD.Workman1,2

1Department of Entomology, University of California, Riverside, CA 92521, U.S.A.2Present address: Department of Orofacial Sciences, University of California, San Francisco,CA 94143, U.S.A.

Received 14 August 2006; Accepted 16 November 2006

ABSTRACT: Survivalofthethreespinestickleback,Gasterosteus aculeatus,differedamongmarshesinademonstration9.9-hamultipurposeconstructedtreatmentwetlanddesignedtoimprovethequalityofsecondary-treatedmunicipalwastewaterinsouthernCalifornia.Atameanloadingrateof3.3kgNH4-Nha

-1d-1(6kgtotalNha-1d-1),thesuitabilityofthewetlandtosupportapopulationofsticklebackswasestimatedtobelow.Thedevelopmentofpotentiallytoxiclevelsofun-ionizedammonia,particularlyduringperiodswhenpHincreasedconcomitantlywithoxygengenerationbyphytoplanktonbiomass> 300 mg chlorophyll a liter-1, and disinfection by-products were associated with lowered survivorship of sentinel fish.Moreover,thehighoxygendemandfromnitrificationofNH4-Ncreateddailyperiodsoflowdissolvedoxygenconcentration(6-16hat<2mgliter-1)intheopenwaterareasoftheshallowmarshes.Lowdissolvedoxygenconcentrationinopenwaterzonesofthesevenmarshesduringapartofeachdayandpersistentanaerobicconditionsintheemergentvegetationrenderedthemajorityofthewetland’ssubstratesurfaceunavailableforsuccessfulreproductionbysticklebacks.ThepotentialsitesforGasterosteustoreplacemosquitofish,Gambusia affinis andG. holbrooki,asabiologicalcontrolagentagainstmosquitoesareprobablylimitedtocomparativelycool-waterhabitatswithhighwaterquality,suchasriverinewetlands.Journal of Vector Ecology 32 (1): 90-105. 2007.

Keyword Index:Constructedtreatmentwetlands,Gasterosteus,mosquitoabatement,un-ionizedammonia,disinfectionby-products.

INTRODUCTION

The use of constructed treatment wetlands as analternative to conventional wastewater treatment isincreasing worldwide (Kadlec and Knight 1996, Cole1998). In addition to tertiary treatment of municipalwastewater, man-made wetlands are used to improve thequality of agricultural wastewater (Karpiscak et al. 1999),mine runoff (Kadlec and Knight 1996), storm water(Metzgeretal.2002)andotherwatersources(Pries2002).Constructed treatment wetlands can provide additionalbenefitstowaterresourcesmanagementprogramssuchaswildlifeconservation,wetlandhabitatenhancement,publiceducation, and recreational sites. When wetland systemstreatwatercarryingcomparativelyhighlevelsofnutrientsandpollutants,itcanbeaconsiderablechallengetoprovidehighqualityhabitatfornativefishes,amphibians,waterfowl,andwaterbirds(Nelsonetal.2000).

A drawback of using wetlands for treatment ofwastewater is the potential for production of pestiferousand pathogen-vectoring mosquitoes (Mortensen 1983,Carlsonetal.1986,Waltonetal.1998,Russell1999).Thisconcern is particularly acute in regions where humandevelopment encroaches on historically isolated wetlandsandwhereconditions,suchasdenseemergentvegetation,large expanses of shallow water, and high nutrient loads,

are especially favorable for mosquito production (Walton2002).Surveillanceofmosquitoabundanceandpathogenactivityinthevectorandhost/reservoirspecies(Rose2001)is important and intervention to control mosquitoes maybenecessary.

Larvivorous fish can be an important componentof a mosquito abatement strategy for wetlands (Meisch1985,Krameretal.1988,WaltonandMulla1991,Walton2007).Integratedpestmanagementapproachescombiningmosquito-specific larvicides (e.g., Bacillus thuringiensis subsp. israelensis de Barjac and B. sphaericus Neide) andplanktivorousfishreducetheneedforchemicalpesticides,diminish the potential for evolution of resistance inmosquitoes to mosquito control agents, potentially lowerthecostofmosquitoabatement,minimizenegativeeffectsofpesticideapplicationonnontargetspecies,andlessenthelikelihood of pathogen transmission to humans, wildlife,anddomesticatedanimals.

Mosquitofish (Gambusia affinis (Baird&Girard)andG. holbrooki Girard) have been introduced worldwide asbiological control agents for mosquitoes since the early1900s(Meisch1985).Theireffectivenessasmosquitocontrolagentsandforreducingtheincidenceofdiseasecausedbymosquito-bornepathogens(Gratzetal.1996,Rupp1996),as well as their potential negative impact on native fishesandotherfauna(CourtenayandMeffe1989,Gamradtand

Vol.32,no.1 Journal of Vector Ecology 91

Kats1996,butseeLawleretal.1999),arestilldebated.Inlocalitieswheremosquitofisharenotnative,introductionsare discouraged or prohibited (Swanson et al. 1996) andalternative biological control agents for mosquitoes toGambusiaspp.areneeded.

The threespine stickleback (Gasterosteus aculeatusL.) may be an alternative biological control agent formosquitoes to Gambusia in many regions along bothcoasts of North America. The stickleback is predaceous,feeding on small animals throughout the water column(Coykendall1980,SchooleyandPage1984,HangelinandVuorinen1988,Sanchez-Gonzalesetal.2001)andpreyingreadily on immature mosquitoes (Bay 1985). The foodpreferences, foragingbehavior(Wootton1976,Whoriskeyetal.1985,FitzGeraldandWootton1993),shortgenerationtime (Wootton et al. 2005), adaptability (McKinnon andRundle2002),andwidegeographicdistribution(PageandBurr 1991, Swift et al. 1993) suggest that the threespinesticklebackmaybeaneffectivepredatorofmosquitoes inawidevarietyofhabitats (Hubbs1919;but seeOffillandWalton1999).

Here we report on the survival and growth of thethreespine stickleback in a multipurpose constructedtreatment wetland receiving secondary-treated municipaleffluent in southern California. Survival of fish and thepotentialforthewetlandtosupportapopulationofnativesticklebacksformosquitocontrolarerelatedtophysicalandchemicalfactorsmeasuredinthewetland.

MATERIALSANDMETHODS

This study was carried out in a 9.9-ha multipurposedemonstration constructed treatment wetland at EasternMunicipal Water District’s (EMWD) Hemet-San JacintoRegional Wastewater Reclamation Facility (HSJRWRF)in San Jacinto, CA (33°48’N, 117°1’W; 454 m ASL). Thewetlandreceiveddaily3800-7600m3ofsecondary-treatedmunicipaleffluentfromaconventionalwastewatertreatmentfacility.Secondary-treatedmunicipaleffluentwasblendedwith approximately 50% of the wetland effluent, dividedamong five inlet marshes, flowed to a central pond, andthen was divided between two outlet (polishing) marshes(Figure1)beforebeingincorporatedintoareclaimedwatersupply.Thenominaloperatingdepthoftheinletandoutletmarshes was 0.55 m; the central pond was 1.9 m. Dailytotal inflow volume varied according to the operationalneedsoftheHSJRWRF;hydraulicresidencetimewas9-14days(Sartorisetal.2000).Thewetlanddidnotcontainfishbecauseofconcernsthatfishmightclogcomponentsofthereclaimedwaterconveyancesystem.

Theshallowmarshescomprised80%of thewetland’ssurfaceareaandcontainedtwobulrushspecies.Californiabulrush[Schoenoplectus (=Scirpus)californicus(C.A.Meyer)]andhardstembulrush[S. acutusG.H.E.MuhlenbergexJ.Bigelow)]werefoundinabout75%and25%oftheshallowmarshes,respectively(Sartorisetal.2000).Atsiteswherefishsurvivalwasmonitored,CaliforniabulrushwasprevalentinInletMarsh3andOutletMarshA,andhardstembulrush

wasprevalentneartheinflowofInletMarsh1(Figure1).Thevegetationhadbeenplantedon1.2-or2.4-mcentersinautumn1994(Sartorisetal.2000).Betweentwoandfour12.2-mwideopenwaterzoneswerepresentinitiallyineachofthesevenshallowmarshes(Figure1).

Survival and growth of sentinel fishGasterosteus aculeatusL.sticklebackswerecollectedby

dip net from the San Jacinto River near Cranston Stationin theSanBernardinoNationalForest (RiversideCounty,CA).Collectionpermitrestrictionslimitedthenumberoffish (n = 55) used in this study. Fish were transported inwaterfromthecollectionsiteunderaerationinacoolerandacclimatedfor30mininone-halfwaterfromthecollectionsiteandfromthewetland.Standardlengthofeachfishwasmeasured to thenearestmillimeterbeforeplacement intocagesinthewetland.

Survivorshipofsticklebackswasmonitoredbyplacingeight to eleven sentinel fish in duplicate cages at threelocationsinthewetland:InletMarsh1,InletMarsh3,andOutlet Marsh A (Figure 1). To reduce the likelihood thatfish were inadvertently introduced into the wetland, eachinternalcagewassurroundedbyanexternalcage(length×width×height:1.30m×1.20m×0.94m)thatwaspushedinto the substrate at the vegetation-open water interface.Externalcagesconsistedofawoodenframecoveredonfivesides with fiberglass window screen. Internal cages were1.25m3(1.22m×1.13m×0.91m)andwereconstructedoffiberglasswindowscreenaffixedonfivesidestoapolyvinylchloride(PVC)frame.Tolimitthehandlingoffishduringcensuses,theinnercagewasliftedoutofthewaterandthenumber of surviving fish in each cage was counted. Eachcage was covered with chicken wire or plastic netting(BirdBlock™:EasyGardner,Waco,TX;meshopening:0.016m)topreventpredationofthesentinelfishesbypiscivorousbirds.

Survivorship of sticklebacks was monitored approxi-mately biweekly from June 14 through October 3, 1996.Dead fish which were not significantly decomposed werecollected at each census and preserved in 10% formalin.Standardlengthofeachdeadfishwasmeasuredtothenear-estmillimeter.

Environmental factorsWaterqualityparametersweremeasuredcontinuously

forfour-tofive-dayintervalsusingtworecordingelectronicwaterqualitysensorarrays(ICMWaterAnalyzers,PerstopAnalytical,Wilsonville,OR)positionedintheopenwaterandinthevegetationnearthesentinelcages.Dissolvedoxygenconcentration, pH, conductivity, and water temperaturewere recordedat30min intervals.Oneprobewasplacedintheopenwaternearoneofthesentinelfishcagesandthesecondprobewasplacedat2or10mintothevegetationontheoutlet sideof thecages.Thesensorarraywassituatedinthemiddleofthewatercolumnatapproximately0.25mbelowthewatersurface.Thesensorsweremovedamongthethreemarshescontainingthesentinelfishduringthestudy;physicochemicalvariablesweremeasuredcontinuouslyfor


twotimeperiodsineachmarsh.Afterfourtofivedaysinthewetland,thesensorswerecleanedandcheckedagainststandards.Thesensorswerealsodeployedattwositesalongthe west and southwest sides of the central pond (Figure1)for7daysinSeptember.Additionalmeasurementsweretakenat5mintothevegetationandinothermarshesofthewetlandforaone-yearperiodthatencompassedthestudydiscussedhere.

Nutrientconcentrations,physicochemicalparameters,andphytoplanktonbiomassweremeasuredapproximatelybi-weekly across the wetland during the summer. Grabsamples(2.2liters)weretakenwithaVanDornbottle(BetaPlus Bottle: Wildlife Supply Co., Saginaw, MI) from themiddle of the water column at 15 stations in the wetlandduringJulyandat17stationsfromAugustthroughOctober(Figure 1). Additional samples were taken in December1996, and February, April, and June 1997. Samples werestored in acid-washed dark polyethylene bottles, placedimmediatelyoniceaftercollection,andanalyzedwithin2-6hofcollection.

Ammonia-nitrogen, nitrate-nitrogen, and nitrite-nitrogen were measured using ion-specific electrodes(Orion Research Inc., Beverly, MA: #9512, 9307, 9346,respectively).Thenitrogensamplesweremeasuredwithin2hofcollection.Un-ionizedammoniaconcentrationwascalculatedas

A=C/[1+10(0.09018+(2729.92/T)-pH)]

whereA is theaqueousNH3concentration (mg liter-1),C

is the ammoniacal nitrogen concentration in water (mgliter-1), pH is pH of the water, and T is temperature (°K)(Denmeadetal.1982).Un-ionizedammoniaconcentrationwascalculatedat30minintervalsduringtheperiodthatthesensorarraysweredeployed.

Total phosphorus, total dissolved phosphorus, andparticulate phosphorus were measured using the ascorbicacidmethod(APHA1995).Totalphosphorussampleswereoxidized using persulfate prior to colorimetric analyses.Particulate phosphorus concentration was calculated asthe difference between total phosphorus concentrationin unfiltered water and total dissolved phosphorusconcentration in water filtered through a glass fiber filter(WhatmanGF/F;WhatmanInternationalLtd.,Maidstone,England,U.K.).

Total inorganic carbon in the water column wasestimated using the known relationship between totalalkalinity, pH, and water temperature (Wetzel and Likens1991). Total alkalinity was measured by Gran titration(Wetzel and Likens 1991). Physico-chemical parameterswere measured at each sampling station using one of theelectronicsensorsdescribedpreviously.

Phytoplankton biomass was measured spectrophoto-metrically as chlorophyll a (Wetzel and Likens 1991) induplicatesamplesateachsamplingsite.Phytoplanktonwascollected on 0.45µm pore membrane filters (Supor 450:GelmanSciencesInc.,AnnArbor,MI)andchlorophyllouspigmentswereextractedin90%alkalineacetone.Totalphy-

toplanktonbiomassincludesphaeopigments.Waterqualityvariableswereanalyzedweeklyininflow

andoutflowwatersamplesbyEMWDstafffollowingAPHA(1995) methods. Sartoris et al. (2000) and Thullen et al.(2002)describethevariablesandanalyticalprocedures.Inaddition to nutrient concentrations, five-day biochemicaloxygen demand (BOD5), total suspended solids, residualchlorine,andcoliformbacteriaabundanceweremeasured.Residual chlorine in influent water to the wetland wasmeasured using the idiometric method (APHA 1995) oneightdatesduringtheperiodthatsentinelfishwerepresentin the wetland. Coliform bacteria density in the influentwaterwasmeasuredusing the fermentation tube test andcalculatedastheMostProbableNumber(MPN)usingthenumberofpositivereactionsinthedilutionseries(APHA1995).

Statistical analysesSurvivor functions for sentinel sticklebacks were

calculatedusingKaplan-Meierestimation,anonparametricproduct limit estimator (SURVIVAL module: SYSTAT1999b).Survivalquantileswerecalculatedforeachsurvivorfunction.Location(marsh)wasastratificationvariable inthe analysis and the statistical significance of differencesof stickleback survival among the three marshes werecomparedbyalog-rank(Mantel-Haenszel)test.

Backward stepwise regression was used to assess theimportance of selected physico-chemical variables on theconditional probabilities of stickleback mortality betweensamplingdatesfortheenvironmentalfactors.Conditionalprobabilities of stickleback mortality within each intervalbetweenthemeasurementofenvironmentalvariableswereestimated from the survival function for sentinel fish ineachmarsh.Ifonlyonefishremainedinacageandthefishdiedintheintervalbeingconsidered,thentheconditionalprobability of mortality (p=1.0) for that interval was notincluded in the regression. The full regression modelconsidered the concentrations of ammonium-nitrogen,nitrate-nitrogen, nitrite-nitrogen, un-ionized ammonia,residualchlorineintheinfluentwater,andlog-transformedchlorophyll a biomass as independent variables. Thehigher-orderinteractionsofun-ionizedammonia,residualchlorine, and log-transformed phytoplankton biomasswere also included in the analysis. Un-ionized ammoniaconcentrationwasestimatedastheproductofammonium-nitrogenconcentrationintheopenwaterzoneadjacenttothe cages holding sentinel fish and the mean across datesfor themaximumproportionofNH4-Nconverted toun-ionized ammonia during the two periods that electronicwater quality sensors were present in each marsh (InletMarsh 1=0.090, Inlet Marsh 3=0.083, Outlet MarshA=0.009).Weeklyresidualchlorinesampleswereaveragedacrossdateswithineachofthefivetimeintervals.Becausewater residence time was 9-14 days, water entering theoutletmarshwas>7daysold; therefore, residualchlorinewasassumedtobebelowdetectionlimits(


(SYSTAT 1999a) using the mean standard length of fishcollected on three to four dates in each cage during thestudy. The slopes for the relationship between standardlength versus time differed significantly among the cagesinthethreemarshessoANCOVAwasinappropriate.Leastsquaresregressionwasusedtoestimategrowthrate.

Thedurationofpotentiallystressfuldissolvedoxygenconcentration (2mgliter-1(FeldmethandBaskin1976).

RESULTS

Survival and growth of sentinel fishSurvival of sentinel sticklebacks differed among the

marshesinthewetland.TherelativesurvivalofGasterosteusin Outlet Marsh A was > Inlet Marsh 3 > Inlet Marsh 1(Figure2).ThefishinbothcagesinInletMarsh1succumbedcomparativelyquicklyandallindividualsexpiredbyday58afterstocking.SentinelfishinthetwocagesinInletMarsh3 exhibited excellent survival through days 42 and 58;survivaldeclinedmarkedlythereafter.Approximately78%ofthesentinelfishinonecageinOutletMarshAsurvivedfor thedurationof the study (126days);however, allfishinthesecondcagesurvivedfornearlytwomonthsbeforedyingabruptlybyday64afterstocking.

One cage containing Gasterosteus in each marshexhibited catastrophic mortality associated with bloomsof the filamentous chlorophyte, Hydrodictyon. The algaerapidly filled a cage during the period between censuses.Deadsticklebackswerefoundthroughoutthewatercolumn,were entangled in the net-like Hydrodictyon, and nearlyallof the individuals inaparticularcageweredead: InletMarsh1cage1,26June;InletMarsh3cage1,11August;Outlet Marsh A cage 2, 17 August. Following the suddenappearance of Hydrodictyon, the percent mortality of fishextantinacageatthepreviouscensuswas100%,87%,and100%,respectively.

Meansurvivaltimesforsentinelfishincagesnotaffectedby Hydrodictyondiffered significantly among themarshesin the San Jacinto demonstration constructed treatmentwetland (log-rank test:χ2=24.74, df=2, p


Variable Mean ± SE n

Nitrogenloadingrates(kgha-1d-1)a

organicnitrogen 1.08totalammonianitrogen 3.30nitritenitrogen 0.99nitratenitrogen 0.61

Totalphosphorus(mgliter-1)Inletmarshes 1.98±0.36b 5Outletmarshes 1.78±0.18 5

Dissolvedphosphorus(mgliter-1)Inletmarshes 1.55±0.21 5Outletmarshes 1.43±0.17 5

Particulatephosphorus(mgliter-1)Inletmarshes 0.43±0.19 5Outletmarshes 0.35±0.06 5

Alkalinity(µeqliter-1)Inletmarshes 3316±118 4Outletmarshes 3209±108 5

Conductivity(mSm-1)Inletmarshes 926.6±9.9 5Outletmarshes 908.0±15.2 5

Dissolvedinorganiccarbon(mgliter-1)Inletmarshes 0.090±0.004 4Outletmarshes 0.087±0.004 2

Biologicaloxygendemand(BOD5:mgliter-1)Inflow ~10-30c

Outflow 4.82±0.33 11

Totalsuspendedsolids(mgliter-1)Inflow 11.27±5.66 11Outflow 7.27±2.69 11

a Data from Sartoris et al. (2000): 30 May – 19 Sept. 1996.b SE was calculated across five sampling dates for the mean concentration in the five inlet marshes or the two outlet marshes on each sampling date. c EMWD (unpublished data) and Barber et al. (2001).

Table1.Meannitrogenloadingrates,nutrientconcentrationsandphysicochemicalfactorsintheHSJRWRFwetlandfromJulythroughSeptember1996.


inorganicnitrogen(TIN) in thewatercolumn(Figure5).MeanTINintheopenwaterzoneclosesttotheinflowtoInletMarsh1was19.1mgNliter-1 (n=6,SD=±3.7)andtothe inflowtoInletMarsh3was19.9mgNliter-1 (n=6,SD=±4.3).MeanTINintheopenwaterzoneclosesttothecages holding sentinel fish was 23-26% lower than in theopenwaterzonesneartheinflow.MeanTINnearthecagesinInletMarsh1was11.2mgNliter-1(n=6,SD=±2.9)andin Inlet Marsh 3 was 13.5 mg N liter-1 (n=6, SD= ± 4.3).Ammoniumnitrogenwas78%to89%ofTIN.

TIN and the percentage of TIN as NH4-N in OutletMarsh A decreased from the inlet marshes (Figure 5).Mean TIN in the two zones of the outlet marsh did notdiffer(station14:5.1±3.3mgNliter-1;station16:5.8±3.8mgNliter-1).Ammonium-nitrogenwas66-67%ofTINinOutletMarshA.ThepercentageofTINasnitrate-nitrogen(NO3-N)intheoutletmarsh(31-32%)increasedfromthatobservedintheinletmarshes(7-10%).Nitrite-nitrogenwasacomparativelysmallcomponent(2-3%)ofTIN.

Chlorophyll a concentration fluctuated more thantwoordersofmagnitudeacrosstimeinthethreemarshesin the wetland (Figure 6a). Phytoplankton biomass waslowest(300mgchlorophyllaliter-1inInletMarsh1 in Julyand in InletMarsh3 in lateAugustwhenappreciablemortalityofsentinelfishwasobserved.

Water column transparency was inversely related tothetrendsinphytoplanktonbiomass.SecchidepthinInletMarsh1wasveryshallowinJuly(Figure6b)andindicated

theuppermost limitofadense layerof suspendedmatter(i.e., phytoplankton and bacteria). During late July andearly August, the Secchi disk was visible on the substrateofInletMarsh1andOutletMarshA.WatertransparencyinInletMarsh3declinedduringAugustandSecchidepthwas comparatively shallow in all of the marshes in earlySeptember.

Un-ionizedammoniaconcentrationwas>0.5mgliter-1forapartofeverydayinInletMarsh1andfor67%ofthedays in Inlet Marsh 3 (Figure 7a). Un-ionized ammoniaconcentrationexceeded1.25mgliter-1onfourdaysintheinletmarshesduringthe17daysthatthesensorsrecordedphysicochemical data, and fluctuated daily as pH in thewatercolumnchanged.Un-ionizedammoniaconcentrationwas much lower in Outlet Marsh A, fluctuating between0.02and0.05mg liter-1during late Julyandbetween0.07and0.14mgliter-1duringmid-August(Figure7a).

Residual chlorine in the influent water was abovedetection limits (0.2 mg liter-1) during three time periodswhilesentinelfishwerepresentinthewetland:11-31July,20August,and17September(Figure7b).Residualchlorineconcentrationwas>8mgliter-1inJuly,>3mgliter-1inAugustand near the limit of detection in September. Coliformbacteriaabundancewasrelatedinverselytoresidualchlorineconcentrationinthewetlandinfluentanddeclinedto≤100MPN(100ml-1)duringperiodsthatresidualchlorinewaspresentat>0.5mgliter-1(Figure7b).

The conditional probability of mortality of sentinelG. aculeatus increased directly with residual chlorineconcentrationinthe influentwaterandtheestimatedun-

Figure1.Schematicillustrationofthe9.9-haHSJRWRFwetlandinSanJacinto,CA.Emergentvegetation(Schoenoplectusspp.)isillustratedbycross-hatching.SentinelfishcagesareillustratedbyfilledsquaresinInletMarsh1,InletMarsh3andOutletMarshA.Waterchemistrysamplingstationsareindicatedbynumbers1through17.Theteardrop-shapedobjectsarenestingislandsforbirds.

In let M arsh 5In let M arsh 1

In let M arsh 2

In let M arsh 3

In let M arsh 4

C entra l P ond

O utlet M arsh A

O utlet M arsh B

312

45

6789

10

11

1213

17

15

14

16

N

Inlet M arsh 5In let M arsh 1

In let M arsh 2

In let M arsh 3

In let M arsh 4

C entra l P ond

O utlet M arsh A

O utlet M arsh B

312

45

6789

10

11

1213

17

15

14

16

N


M

arsh

Dat

esO

pen

wat

er

Ve

geta

tion

Inle

tMar

sh1

July

16-

197.

38±

0.3

525

.2±

1.6

186

7.06

±0

.19

25.2

±1

.514

12

Aug

ust5

-97.

28±

0.2

724

.5±

1.8

207

7.14

±0

.03

23.4

±1

.718

110

Inle

tMar

sh3

July

22-

277.

42±

0.3

425

.0±

1.0

191

7.42

±0

.27

24.8

±0

.919

12

Aug

ust1

2-15

7.33

±0

.26

26.7

±0

.919

77.

32±

0.0

726

.0±

0.9

197

10

Out

letM

arsh

AJu

ly2

9-A

ug.2

7.12

±0

.04

26.8

±0

.919

67.

23±

0.0

326

.6±

1.0

146

2

Aug

ust1

6-20

7.18

±0

.04

24.7

±1

.219

47.

37±

0.1

024

.7±

0.9

165

2

pH

°C

n

pH

°Cn

Dist

ance

of

sens

orin

ve

geta

tion

(m)

Tabl

e2.

The

mea

n(±

SD

)pH

and

tem

pera

ture

ofw

ater

ino

pen

wat

era

djac

entt

oca

gesh

oldi

ngse

ntin

elfi

sha

ndin

veg

etat

edz

ones

oft

hree

mar

shes

inth

eH

SJRW

RFw

etla

nd

duri

ngsu

mm

er1

996.

Dat

aw

ere

reco

rded

eve

ry0

.5h

and

at0

.25

md

epth

.


ionized ammonia concentration in the open water zoneadjacent to the cages holding fish. Stepwise regressionidentified the two main effects as significant predictorsof conditional mortality of the sticklebacks ([NH3]:(coefficient ± SE) 0.280 ± 0.102; t=2.733, df=10, p


Figu

re4

.Diss

olve

dox

ygen

con

cent

ratio

n,w

ater

tem

pera

ture

and

pH

ats

ites

ino

pen

wat

era

ndw

ithin

the

vege

tatio

nne

arc

ages

con

tain

ing

sent

inel

thre

espi

nest

ickl

ebac

kin

th

ree

mar

shes

with

inth

eH

SJRW

RFw

etla

ndin

San

Jaci

nto,

CA

,dur

ing

1996

.a:I

nlet

Mar

sh1

,ope

nw

ater

;b:I

nlet

Mar

sh3

,ope

nw

ater

;c:O

utle

tMar

shA

,ope

nw

ater

;d:I

nlet

M

arsh

1,v

eget

atio

n;e

:Inl

etM

arsh

3,v

eget

atio

n;f

:Out

letM

arsh

A,v

eget

atio

n.


Figure 5. Concentration of inorganic nitrogen compounds in three marshes of the HSJRWRF wetland, San Jacinto CA, during 1996. a: Inlet Marsh 1, b: Inlet Marsh 3, c: Outlet Marsh A. Station refers to the sampling sites illustrated in Figure 1.

1

10

1 00

10 00

4 -Jul 1 9-Jul 3 -Aug 18 -Aug 2-S ep 1 7-S ep

Chlor

ophy

ll a

(mg L

-1)

Inlet M arsh 1Inlet M arsh 3O utlet M arsh A

(a)

0

0.1

0.2

0.3

0.4

0.5

0.6

4-Jul 19 -Jul 3 -Aug 18-Aug 2-Sep 17-SepDate

Secc

hi de

pth (m

)

(b)

**

******

1

10

1 00

10 00


Chlor

ophy

ll a

(mg L

-1)


(a)

0

0.1

0.2

0.3

0.4

0.5

0.6


Secc

hi de

pth (m

)

(b)

1

10

1 00

10 00


Chlor

ophy

ll a

(mg L

-1)


(a)

0

0.1

0.2

0.3

0.4

0.5

0.6


Secc

hi de

pth (m

)

(b)

**

******

Figure6.Phytoplanktonbiomass(a:totalchlorophylla)andwater column transparency (b: Secchi depth) in the openwater zone adjacent to cages holding sentinel fish in theHSJRWRFwetlandduring1996.*indicatesthattheSecchidiskwasvisibleonthebottomofthemarsh.

themajorityof thewetland’ssubstratesurfaceunavailableforsuccessfulnest-buildingactivitiesbysticklebacks.

Survival of sentinel G. aculeatus in the inlet marsheswas lower than in the polishing marsh, Outlet Marsh A,andwasrelatedtochangesinwaterqualityfromtheinflowto the outflow of the treatment wetland. The mean NH4-N concentration (16mg liter-1), suspended matter (shallow Secchi depth andalgalbiomass>300mgchlorophylla liter-1),andresidualchlorine.

The high dissolved oxygen concentration andconcomitant increase of pH associated with sustainedprimary production in the inlet marshes shifted thechemical equilibrium from ammonium to un-ionizedammoniacreatingpotentiallytoxiclevelsofammonia(>0.5mgNH3-Nliter

-1)forthesticklebacks.Theratesofoxygenproductionanddemandinthewatercolumndidnotdiffersignificantlyamongthethreemarshesstudied(unpublisheddata), but autotroph biomass in the water column variedover100-foldineachmarsh.Thedailymaximumdissolvedoxygenconcentrationinopenwaterzones(


salinity,andconcentrationsofothertoxicants(e.g.,metalssuchasAl,Cu,NiandZn)(Hazeletal.1971,Thurstonetal.1981,Nimmoetal.1989).The96-hLC50forguppy(Poecilia reticulataPeters)juvenileswas1.24mgNH3liter

-1(Ruffieretal.1981).The96-hLC50forfatheadminnow,Pimephales promelas Rafinesque, in well water (1.12 mg NH3 liter

-1)was similar to P. reticulata but was reduced by 50% inmunicipal wastewater (0.56 mg NH3 liter

-1: Nimmo et al.1989).Thechronicvalueforthisspecieswas0.13mgNH3liter-1.Swansonetal.(1996)recommendedthatun-ionizedammonianitrogenabove0.1mgliter-1shouldbeavoidedinculturesystemsofG. affinis.The96-hTLm(mediantolerancelimit)forG. aculeatusjuvenilesandadultsinfreshwaterat15°and23°Cwere2.1and1.8mgundissociatedNH4OHliter-1(Hazeletal.1971);usingtheformulaofDenmeadetal.(1982)andsubstitutingthevaluesinTable6ofHazeletal.,theselevelsareequivalenttoLC50sof0.58and0.38mgNH3liter

-1,respectively.Residual chlorine in the influent water and the

interaction of disinfection by-products with ammoniacalnitrogen and wetland-derived organic matter can createtoxic conditions for fish and other aquatic organisms inconstructed treatment wetlands where NH4-N levels areelevated.Othercompoundscommonlyfoundinwastewater,suchasendocrinedisruptorchemicals,canbedetrimentalto fish (Bell 2001, 2004, Pottinger et al. 2002, Wibe et al.2002).Theresultsof regressionstudiescarriedout inourstudyshowthat the interactionbetweenresidualchlorineandun-ionizedammoniawasstronglyrelatedtosticklebackmortalitywithinsamplingperiods.Ifdechlorinationoftheinfluentwater toa treatmentwetlandisnot fullyeffectiveortheinfluentwatersupplyiscontaminatedbychlorinatedwater, then chlorine and halogenated compounds (e.g.,chloramines, trihalomethane, or other disinfection by-products;Nimmoetal.1989,MESC,NBS,andCSU20001,Rostad et al. 2000) can enter a constructed treatmentwetland.Eventhoughtheconcentrationsofpotentiallytoxicdisinfection by-products released from water treatmentplants decline as water moves through constructedtreatment wetlands, enhanced concentrations of wetland-derived organic carbon (Sartoris et al. 2000, Barber et al.2001) are thought to increase dramatically (~30-fold) thepotential for toxic effects from disinfection by-products(Rostadetal.2000).

Differences in the natural cycles of bulrush diebackmay have contributed to the differences in survival ofsticklebacksinthetwoinletmarshes.DespitebeingsituatedatagreaterdistancefromtheinfluentweirofInletMarsh1 where the effects of chlorine contamination might belessened as compared to being positioned closer to theinfluentweirofInletMarsh3,G. aculeatusinInletMarsh11MESC, NBS, and CSU (Midcontinent Ecological ScienceCenter, National Biological Service and ColoradoState University). 2000. Hemet/San Jacinto RegionalWater Reclamation Facility: Demo Wetlands ToxicityCharacterization.Phase1Report,May1996.U.S.GeologicalSurvey,U.S.Dept.Interior,Denver,CO.11pp.

exhibitedmortalityhigherthandidthefishinInletMarsh3.Highmortalityofsentinelsticklebacksmighthavebeenrelatedtothe interactionofdisinfectionby-productswithhigher levels of wetland-derived organic matter beingreleased from a natural mid-winter (1995-1996) diebackofS. acutus thatdidnotoccurinS. californicus-dominatedInletMarsh3.

Reducedsurvivalininletmarshwateralsowasobservedinotherfishspecies.DuringtoxicitystudiesattheHSJRWRFwetland in mid-May 1996, 100% of P. promelas juvenilessuccumbedinwatercollectedneartheweirsofthefiveinletmarshes and all fish survived in water collected from theoutlet marshes (MESC, NBS and CSU 20001). Organic oroxidizablecompoundssuchaschloramineswerethoughttohavecontributedtothetoxicityofinletwater(MESC,NBSandCSU20001).SubsequentbioassaysinMarch1999,whenresidual chlorine was not detected in the influent water,failedtoshowtoxicityofinletwatertofatheadminnowsatNH4-Nconcentrationsbetween9and14mgliter

-1(Castle19992). Survival of sentinel mosquitofish also was lowestinInletMarsh1;however,GambusiasurvivorshipinInletMarsh3andOutletMarshAwas100%(Waltonetal.1997).SticklebacksurvivalwaslowerthanthatofmosquitofishineachofthethreemarshesoftheHSJRWRFwetland.

Water temperatures in the HSJRWRF wetland werebetween 21° C and 29° C during the period that sentinelfishweremonitoredanddidnotexceedtheupperthermaltolerance of the sticklebacks. The thermal tolerance of G. aculeatus varies among populations and is related to theambient temperatures found in the native habitat and tothe acclimation temperature (Jordan and Garside 1972,BlahmandSnyder1975).Theupperlethaltemperatureforwarm-adaptedsticklebacks collectedfromtheSantaClaraRiver in southern California was 34.6° C (Feldmeth andBaskin1976).OffillandWalton(1999)foundthatmortalityof G. aculeatus collected from the Mojave River (SanBernardino County, CA) increased when the maximumwater temperature of earthen ponds exceeded 33° C. Thesticklebacksusedinthepresentstudywerecollectedfromthe San Jacinto River (Riverside County, CA) and wereexpected to have an upper thermal tolerance (~34° C)thatwassimilartothoseobservedintheothersticklebackpopulations collected from shallow lotic environments ofsouthernCalifornia.

The catastrophic mortality of sticklebacks associatedwithbloomsof thenet-likegreenalgae,Hydrodictyon sp.,is attributed to an unforeseen effect of the cages ratherthan a natural phenomenon in the open water zones ofthe constructed treatment wetland. Blooms of water netoccurred in one replicate cage in each marsh and rapidlyfilled the water column of each cage. The natural mid-wateractivitiesofthestickleback(SchooleyandPage1984)wereseverelyrestrictedbydensebloomsofHydrodictyon.2Castle, C. J. 1999. Hemet/San Jacinto Regional WaterReclamationFacility:DemoWetlandsToxicityAssessment.AcuteBaselineConditionsofConstructedWetlands-Phase1Report,March1999.U.S.GeologicalSurvey,U.S.Dept.Interior,Denver,CO.7pp.


0

2

4

6

8

10

24-Jun 24-Ju l 23-Aug 22-Sep

Resid

ual c

hlorin

e (mg

L-1

)

0

1

2

3

4

5

6

7 Bacterial density (log MPN 100ml-1)

res idua l ch lo rinebacteria

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0 1 2 3 4 5 6N H 3 * res idua l chlo rine

Cond

itiona

l mor

tality

0

0 .5

1

1.5

2

2.5

15- Ju l 29- Ju l 12 -A ug

Un-io

nized

ammo

nia (m

g L-1

) In le t M arsh 1In le t

M arsh 3In le t

M arsh 1In le t

M arsh 3

O utle t M arsh A

O utle t M arsh A

(a)

(b)

(c)

0

2

4

6

8

10

24-Jun 24-Ju l 23-Aug 22-Sep

Resid

ual c

hlorin

e (mg

L-1

)

0

1

2

3

4

5

6

7 Bacterial density (log MPN 100ml-1)

res idua l ch lo rinebacteria

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0 1 2 3 4 5 6N H 3 * res idua l chlo rine

Cond

itiona

l mor

tality

0

0 .5

1

1.5

2

2.5

15- Ju l 29- Ju l 12 -A ug

Un-io

nized

ammo

nia (m

g L-1

) In le t M arsh 1In le t

M arsh 3In le t

M arsh 1In le t

M arsh 3

O utle t M arsh A

O utle t M arsh A

(a)

(b)

(c)

Figure 7. Factors associated with mortality of sentinelsticklebacksintheHSJRWRFwetlandduringthesummer,1996. a: estimated un-ionized ammonia concentration inthe open water regions of marshes, b: residual chlorineconcentration and coliform bacteria abundance inthe influent water, c: conditional mortality of sentinelsticklebacksvs.theinteractionofun-ionizedammoniaandresidualchlorineconcentration.

Normalswimmingactivitiesofwarm-adaptedsticklebacksoccurred at dissolved oxygen concentrations >2 mg L-1(Feldmeth and Baskin 1976). Under hypoxic conditions,G. aculeatusgulpair/waterattheair-waterinterfacewhileeither swimming parallel to the water surface or bobbingwiththebodymaintainedatacomparativelysteepanglewithrespecttothewatersurface(Whoriskeyetal.1985,Waltonetal.1997);bothbehaviorswouldhavebeenrestrictedbythe dense blooms of water net. The moribund fish wereentangledinthenet-likealgaethroughoutthewatercolumnandappearedtobeveryrobustsuggestingthattheonsetof

mortalitywasrapidasphyxia.LargebloomsofHydrodictyonwereneverobservedoutsideofthecages.Watermovement,shadingbydenseplanktonicalgalpopulationsintheinletmarshes,orherbivorybywaterfowlmaylimitHydrodictyonfromproliferatingintheopenwateroutsidethecages.

Because low dissolved oxygen concentration wasobserved in each of the three marshes in the HSJRWRFwetland, differences of fish survival in cages whereHydrodictyon was not present were unlikely to have beencausedbyhypoxia.Dissolvedoxygenconcentrationinthewatercolumnofthethreemarsheswasseverelydepressedatnightandintheearlymorning.Dissolvedoxygenlevelswereextremelylow(~0mgO2liter

-1)atalltimesat>5mintothebandsofvegetationandascloseas2mfromtheopenwater-vegetation interface in the inlet marshes. Rose andCrumpton(1996)alsofoundpersistentoxygendepressionin stands of emergent vegetation in a prairie potholewetland. Unlike an Iowa wetland where open water areasadjacent to emergent vegetation were usually well-mixed(RoseandCrumpton1996),thehighoxygendemandfromnitrification in the HSJRWRF wetland created stressfulconditions for sticklebacks in open water zones adjacentto vegetation, particularly in the inlet marshes. Jones(1964) reported that sticklebacks can survive at dissolvedoxygen concentrations as low as 0.25-0.5 mg liter-1 butwhen fish were exposed to water with different dissolvedoxygenlevelsat13°Ctheymovedfromwaterwithoxygenconcentration


0

5

10

15

20

25

Diss

olved

oxyg

en (p

pm

)

0

5

10

15

20

25

30

35

Temperature (oC)

C en tra l Po nd

789

11-Sep9 :00 AM

13-Sep9 :00 AM

15-Sep9 :00 AM

17-Sep9 :00 AM

pH

0

5

10

15

20

25

2-O ct 1-D ec 30-Jan 31-M ar 30-M ay 29-Ju l

Ammo

niaca

l nitro

gen (

mg L

-1)

In le t M arsh 1 , s ta tion 2In le t M arsh 1 , s ta tion 3In le t M arsh 3 , s ta tion 6In le t M arsh 3 , s ta tion 7O utle t M arsh A, s ta tion 14O utle t M arsh A, s ta tion 16

(a )

(b)

0

5

10

15

20

25

Diss

olved

oxyg

en (p

pm

)

0

5

10

15

20

25

30

35

Temperature (oC)

C en tra l Po nd

789

11-Sep9 :00 AM

13-Sep9 :00 AM

15-Sep9 :00 AM

17-Sep9 :00 AM

pH

0

5

10

15

20

25

2-O ct 1-D ec 30-Jan 31-M ar 30-M ay 29-Ju l

Ammo

niaca

l nitro

gen (

mg L

-1)

In le t M arsh 1 , s ta tion 2In le t M arsh 1 , s ta tion 3In le t M arsh 3 , s ta tion 6In le t M arsh 3 , s ta tion 7O utle t M arsh A, s ta tion 14O utle t M arsh A, s ta tion 16

(a )

(b)

Figure8.Physico-chemicalfactorsalongtheperipheryofthecentralpondduringSeptember1996andinthreemarshesbetweenOctober1996andJune1997intheHSJRWRFwetland,SanJacinto,CA.a:dissolvedoxygenconcentration,watertemperatureandpH;b:ammoniacalnitrogenconcentration.

Figure9.SchematicillustrationofconditionsaffectingsitesforsticklebackreproductionintheHSJRWRFwetlandduringsummer1996.


inthearidsouthwesternUnitedStatesmustresidenearthewatersurfaceandareatagreaterriskofbirdpredation.Birdpredation is a significant mortality factor for sticklebackpopulations residing in natural wetlands and mortalityincreases appreciably where water depths are < 0.2 m(FitzGerald and Wootton 1993). The HSJRWRF wetlandsupportslargepopulationsofpiscivorousbirds(e.g.,heronsandegrets:Andersonet al. 2003). Inwetlandscontainingpiscivorousfish,hypoxiacanlimitthedistributionofyoungfish to regions with little protective cover and reduce thesurvivalofjuvenilefish(SuthersandGee1986).

Seasonal and successional changes in the wetlandcan alter the nutrient removal efficiency of constructedtreatment wetlands, exposing fish to potentially toxiclevels of municipal wastewater constituents that fluctuateseasonallyandincreasetemporally.Lowernitrogenremovalefficiency associated with internal nitrogen loading bydense emergent macrophyte populations increased NH4-Nconcentrationintheoutletmarshes(Sartorisetal.2000)and phytoplankton biomass in open water zones (Smithetal.2000).Ammoniacalnitrogenlevelsincreasedduringautumn(Figure8b)andremainedcomparativelyhighuntilspring. Warm-water aquatic communities were one andone-half to six timesmoresensitive to the toxiceffectsofammonia at cold versus warm temperatures (Nimmo etal. 1989). The high ammoniacal nitrogen levels observedduringtheautumnandwinter1996andinearly1997wouldpresumablybedetrimentaltooverwinteringsticklebacks.

Environmental factors associated with ammonia-rich municipal effluent entering a constructed treatmentwetland in southern California are predicted to limit thesuitability of the wetland as a habitat for the stickleback,an alternative to the mosquitofish as a biological controlagent of mosquitoes. Both external and internal loadingof potential toxins influenced the survival of sentinel G. aculeatus. If larvivorous/planktivorous fish species are tobeincorporatedintointegratedpestmanagementprogramsformosquitoesinhabitingconstructedtreatmentwetlands,then species with broad environmental tolerances willbe needed in wetlands that treat ammonia-dominatedmunicipaleffluent.Thefunctionalandnumericalresponsesof the fish populations are also important; mosquitoesshould be an important component of the fish’s diet andthe habitat should be conducive to rapid increases of fishabundance following stocking and natural reductions inpopulation size. The stickleback is commonly found instreamsandriversofsouthernCalifornia(Swiftetal.1993)andmaybeaneffectivelarvivoreofmosquitoesinhabitingthe comparatively cool and high quality water found inriverine wetlands. For constructed treatment wetlandstreatingammonia-richmunicipaleffluentthatareeffectivelyisolatedfromnaturalaquaticecosystems,themosquitofishis likely to be more effective than the stickleback as abiologicalcontrolagentformosquitoes.

Acknowledgments

WethankL.Randall,A.Luck,J.Krumm,andA.Laquifor assistance in the field and laboratory; C.J. Castle, S.Denison,J.Malcolm,D.Nimmo,R.Rodriquez,J.Sartoris,L.Smith,C.Swift,andJ.Thullenforbeneficialdiscussions,andtheEasternMunicipalWaterDistrictforaccesstothefieldsiteandassistancewithaspectsofthisstudy.J.Jiannino,G. Peck, D. Popko, M. Sanford, J. Sartoris, and J. Thullenprovidedhelpfulcommentsonthemanuscript.ThisresearchwassupportedbySpecialFundsforMosquitoResearchfromtheDivisionofAgricultureandNaturalSciencesUniversityofCalifornia,theAgriculturalExperimentStationatU.C.-Riversideandcooperativeagreement1445-CA09-0011withtheU.S.DepartmentoftheInterior.

REFERENCESCITED

American Public Health Association (APHA). 1995.Standard Methods for the Examination of Water and Wastewater.19thed.Am.Publ.Hlth.Assoc.,Am.WaterWorksAssoc.andWaterEnvt.Fed.,Washington,D.C.

Anderson,D.C.,J.J.Sartoris,J.S.Thullen,andP.G.Reusch.2003.Theeffectsofbirduseonnutrientremovalinaconstructed wastewater–treatment wetland. Wetlands23:423-435.

Barber,L.B.,J.A.Leenheer,T.I.Noyes,andE.A.Stiles.2001.Natureandtransformationofdissolvedorganicmatterintreatmentwetlands.Environ.Sci.Technol.35:4805-4816.

Bay, E.C. 1985. Other larvivorous fishes. In: H. Chapman(ed.)Biological Control of Mosquitoes.pp.18-24.Am.Mosq.Contr.Assoc.,Fresno,CA.

Bell, A.M. 2001. Effects of an endocrine disrupter oncourtship and aggressive behaviour of male three-spined stickleback, Gasterosteus aculeatus. AnimalBehaviour62:775-780.

Bell, A.M. 2004. An endocrine disrupter increasesgrowth and risky behavior in threespined stickleback(Gasterosteus aculeatus). Hormones and Behavior 45:108-114.

Blahm, T.H. and G.R. Snyder. 1975. Effect of increasedwater temperatures on survival of adult threespinesticklebackandjuvenileyellowperchintheColumbiaRiver.Northwest.Sci.49:267-270.

Carlson, D.B., R.R. Vigliano, and G.L. Wolfe. 1986.Distribution of mosquitoes in different wastewaterstages of secondarily treated domestic effluent anduntreated citrus washwater. J. Amer. Mosq. Contr.Assoc.2:516-521.

Cole, S. 1998. The emergence of treatment wetlands.Environ.Sci.Technol.32:218-223.

Courtenay, W.R., Jr. and G. K. Meffe. 1989. Small fishesin strange places: a review of introduced poeciliids.In:G.K.MeffeandF.F. Senlson Jr. (eds.) Ecology and Evolution of Livebearing Fishes (Poeciliidae). pp. 301-331.PrenticeHall,EnglewoodCliffs,NJ.

Coykendall,R.L. (ed.)1980.Fishes in California Mosquito


Control.Biol.ControlCommission.Calif.Mosq.VectorContr.Assoc.Press,Sacramento.

Denmead, O.T., J.R. Freney, and J.R. Simpson. 1982.Dynamics of ammonia volatilization during furrowirrigationofmaize.SoilSci.Soc.Am.J.46:149-155.

Feldmeth, C.R. and J.N. Baskin. 1976. Thermal andrespiratory studies with reference to temperatureand oxygen tolerance for the unarmored sticklebackGasterosteus aculeatus williamsoniHubbs.Bull.South.Calif.Acad.Sci.75:127-131.

FitzGerald, G.J. and R.J. Wootton. 1993. The behavioralecologyofsticklebacks.In:T.J.Pitcher(ed.)Behaviour of Teleost Fishes. 2nd ed. pp. 537-572. Chapman andHall,London.

Gamradt, S.C. and L.B. Kats. 1996. Effect of introducedcrayfishandmosquitofishonCalifornianewts.Conserv.Biol.10:1155-1162.

Gratz,N.S.,E.F.Legner,G.K.Meffe,E.C.Bay,M.W.Service,C.Swanson,J.J.Cech,Jr.andM.Laird.1996.Commentson“AdverseAssessmentsofGambusia affinis.”J.Am.Mosq.Contr.Assoc.12:160-166.

Hangelin, C. and I. Vuorinen. 1988. Food selection injuvenile three-spined sticklebacks studied in relationtosize,abundanceandbiomassofprey.Hydrobiologia157:169-177.

Hazel,C.R.,W.Thomsen,andS.J.Meith.1971.Sensitivityofstripedbassandsticklebacktoammoniainrelationtotemperatureandsalinity.Calif.FishGame57:138-153.

Hubbs, C.L. 1919. The stickleback: a fish eminently fittedbynatureasamosquitodestroyer.Calif.FishGame5:21-24.

Jones, J.R.E. 1964. Fish and River Pollution. Butterworths,London.203pp.

Jordan, C.M. and E.T. Garside. 1972. Upper lethaltemperatures of threespine stickleback, Gasterosteus aculeatus (L.), in relation to thermal and osmoticacclimation,ambientsalinity,andsize.Can.J.Zool.50:1405-1411.

Kadlec,R.H.andR.L.Knight.1996.TreatmentWetlands.CRCPress,BocaRaton.893pp.

Karpiscak,M.M.,R.J.Freitas,C.P.Gerba,L.R.Sanchez,andE. Shamir. 1999. Management of dairy waste in theSonoranDesertusingconstructedwetlandtechnology.Wat.Sci.Tech.40:57-65.

Kramer, V.L., R. Garcia, and A.E. Colwell. 1988. AnevaluationofGambusia affinisandBacillus thuringiensisvar.israelensisasmosquitocontrolagentsinCaliforniawildricefields.J.Am.Mosq.Contr.Assoc.4:470-478.

Lawler, S.P., D. Dritz, T. Strange, and M. Holyoak. 1999.Effects of introduced mosquitofish and bullfrogs onthe threatened California red-legged frog. Conserv.Biol.13:613-622.

McKinnon,J.S.andH.D.Rundle.2002.Speciationinnature:Thethreespinesticklebackmodelsystems.TrendsEcol.Evol.17:480-488.

Meisch,M.V.1985.Gambusia affinis affinis.In:H.Chapman(ed.) Biological Control of Mosquitoes. pp. 3-17. Am.

Mosq.Contr.Assoc.,Fresno,CA.Metzger, M.E., C.M. Myers, and V.L. Kramer. 2002. The

“darkside”ofstormwaterrunoffmanagement:vectorsassociated with BMPs. Proc. Mosq. Vector Contr.Assoc.Calif.70:2-10.

Mortensen,E.W.1983.Mosquitooccurrenceinwastewatermarshes: a potential new community problem. Proc.Calif.Mosq.VectorContr.Assoc.50:65-67.

Nelson,S.M.,R.A.Roline,J.S.Thullen,J.J.Sartoris,andJ.E.Boutwell. 2000. Invertebrate assemblages and traceelementbioaccumulationassociatedwithconstructedwetlands.Wetlands20:406-415.

Nimmo, D.W.R., D. Link, L.P. Parrish, G.J. Rodriguez, W.Wuerthele,andP.H.Davies.1989.Comparisonofon-site and laboratory toxicity tests: derivation of site-specificcriteriaforun-ionizedammoniainaColoradotransitional stream.Environ.Toxicol.Chem.8:1177-1189.

Offill, Y.A. and W.E. Walton. 1999. Comparative efficacyof the threespine stickleback (Gasterosteus aculeatus)andthemosquitofish(Gambusia affinis)formosquitocontrol.J.Am.Mosq.Contr.Assoc.15:380-390.

Page,L.M.andB.M.Burr.1991.A Field Guide to Freshwater Fishes of North America North of Mexico. HoughtonMifflin,Boston.432pp.

Pottinger,T.G.,T.R.Carrick,andW.E.Yeomans.2002.Thethree-spinedsticklebackasanenvironmentalsentinel:Effectsofstressorsonwhole-bodyphysiologicalindices.J.FishBiol.61:207-229.

Pries, J. (ed.)2002. Treatment Wetlands for Water Quality Improvement: Quebec 2000 Conference Proceedings(selectedpapers).CH2MHillCanada,PandoraPress,Waterloo.228pp.

Rose, C. and W.G. Crumpton. 1996. Effects of emergentmacrophytesondissolvedoxygendynamicsinaprairiepotholewetland.Wetlands16:495-502.

Rose, R. I. 2001. Pesticides and public health: integratedmethodsofmosquitomanagement.Emerg.Infect.Dis.7:17-23.

Rostad, C.E., B.S. Martin, L.B. Barber, J.A. Leeheer, andS.R. Daniel. 2000. Effect of a constructed wetlandon disinfection byproducts: removal processes andproduction of precursors. Environ. Sci. Technol. 34:2703-2710.

Ruffier,P.J.,W.C.Boyle,andJ.Kleinschmidt.1981.Short-termacutebioassaystoevaluateammoniatoxicityandeffluentstandards.J.WaterPollut.Contr.Fed.53:367-377.

Rupp,H.R.1996.AdverseassessmentsofGambusia affinis:analternateviewformosquitocontrolpractitioners.J.Am.Mosq.Contr.Assoc.12:155-159.

Russell,R.C.1999.Constructedwetlandsandmosquitoes:healthhazardsandmanagementoptions–anAustralianperspective.Ecol.Eng.12:107-124.

Sanchez-Gonzales,S.,G.Ruiz-Campos,andS.Contreras-Balderas. 2001. Feeding ecology and habitat ofthe threespine stickleback, Gasterosteus aculeatus microcephalus,inaremnantpopulationofnorthwestern


BajaCalifornia,Mexico.EcologyofFreshwaterFish10:191-197.

Sartoris,J.J.,J.S.Thullen,L.R.Barber,andD.E.Salas.2000.InvestigationofnitrogentransformationsinasouthernCaliforniaconstructedwastewater treatmentwetland.Ecol.Eng.14:49-65.

Schooley, J.K. and L.M. Page. 1984. Distribution andabundance of two marsh fish: the mosquitofish(Gambusia affinis) and the threespine stickleback(Gasterosteus aculetus). Proc. Mosq. Vector Contr.Assoc.Calif.52:134-139.

Smith, L.K., J.J. Sartoris, J.S. Thullen, and D.C. Andersen.2000. Investigation of denitrification rates in anammonia-dominated constructed wastewater-treatmentwetland.Wetlands20:684-696.

Suthers,I.M.andJ.H.Gee.1986.Roleofhypoxiainlimitingspring and summer distribution of juvenile yellowperch(Perca flavescens)inaprairiemarsh.Can.J.Fish.Aquat.Sci.43:1562-1570.

Swanson, C., J.J. Cech, Jr., and R.H. Piedrahita. 1996.Mosquitofish: Biology, Culture, and Use in Mosquito Control. Mosq. Vector Contr. Assoc. Calif. and Univ.Calif. Mosq. Research Program. Sacramento, CA. 87pp.

Swift,C.C.,T.R.Haglund,M.Ruiz,andR.N.Fisher.1993.The status and distribution of freshwater fishes ofsouthern California. Bull. South. Calif. Acad. Sci. 92:101-167.

SYSTAT.1999a.SYSTAT®9StatisticsI.SPSSInc.,Chicago,IL.660pp.

SYSTAT.1999b.SYSTAT®9StatisticsII.SPSSInc.,Chicago,IL.552pp.

Thullen, J.S., J.J. Sartoris, and W.E. Walton. 2002. Effectsof vegetation management in constructed wetlandtreatment cells on water quality and mosquitoproduction.Ecol.Eng.18:441-457.

Thurston,R.V.,G.R.Phillips,R.C.Russo,andS.M.Hinkins.1981.Increasedtoxicityofammoniatorainbowtrout(Salmo gairdneri)resultingfromreducedconcentrationsofdissolvedoxygen.Can.J.Fish.Aquat.Sci.38:938-988.

Walton, W.E. 2002. Multipurpose constructed treatmentwetlands in the arid southwestern United States: arethebenefitsworththerisks?In:J.Pries(ed.)Treatment Wetlands for Water Quality Improvement: Quebec 2000 Conference Proceedings(selectedpapers).pp.115-123.CH2MHillCanada,PandoraPress,Waterloo.

Walton, W.E. 2007. Larvivorous fish including Gambusia.In: T. Floore (ed.) Biorational Control of Mosquitoes.Am.Mosq.Contr.Assoc.,MountLaurel,NJ.Inpress.

Walton, W.E. and M.S. Mulla. 1991. Integrated control ofCulex tarsalis larvae using Bacillus sphaericus andGambusia affinis:effectsonmosquitoesandnontargetorganismsinfieldmesocosms.Bull.Soc.VectorEcol.16:203-221.

Walton, W.E., M.C. Wirth, P.D. Workman, and L.A.Randall. 1997. Survival of two larvivorous fishes ina multipurpose constructed wetland in southernCalifornia.Proc.Mosq.VectorContr.Assoc.Calif.65:51-57.

Walton, W.E., P.D. Workman, L.A. Randall, J.A. Jiannino,andY.A.Offill.1998.EffectivenessofcontrolmeasuresagainstmosquitoesataconstructedwetlandinsouthernCalifornia.J.VectorEcol.23:149-160.

Wetzel,R.G.andG.E.Likens.1991.Limnological Analyses.2nded.Springer-Verlag,NewYork.391pp.

Whoriskey,F.G.,A.Gaudreault,N.Martel,S.Campeau,andG.S.FitzGerald.1985.Theactivitybudgetandbehaviorpatternsoffemalethreespinesticklebacks,Gasterosteus aculeatus(L.),inaQuébectidalsaltmarsh.Nat.Can.(Rev.Écol.Syst.)112:113-118.

Wibe, A.E., G. Rosenqvist, and B.M. Jenssen. 2002.DisruptionofmalereproductivebehaviorinthreespinesticklebackGasterosteus aculeatus exposed to 17beta-estradiol.Environ.Res.90:136-141.

Wootton, R.J. 1976. The Biology of Sticklebacks. AcademicPress,NewYork.387pp.

Wootton, R.J. 1984. A Functional Biology of Sticklebacks.Univ.CaliforniaPress,Berkeley,CA.265pp.

Wootton,R.J.,C.E.Adams,andM.J.Martin.2005.Empiricalmodelling of the population dynamics of a smallpopulationof the threespinestickleback,Gasterosteus aculeatus.Environ.Biol.Fishes74:151-161.

Environmental factors influencing survival of threespine ...faculty.ucr.edu › ~walton › Walton...

Documents

Transcript of Environmental factors influencing survival of threespine ...faculty.ucr.edu › ~walton › Walton...