Arsenic in the Bangladesh soils related to physiographic region, paddymanagement, and micro- and macro- element status

Chowdhury , M. T. A., Deacon, C. M., Jones , G. D., Huq, S. M. I., Williams, P. N., Hoque , A. F. M. M., Winkel,L. H. E., Price, A. H., Norton , G. J., & Meharg, A. A. (2017). Arsenic in the Bangladesh soils related tophysiographic region, paddy management, and micro- and macro- element status. Science of the TotalEnvironment, 590-591, 406-415. https://doi.org/10.1016/j.scitotenv.2016.11.191

Published in:Science of the Total Environment

Document Version:Peer reviewed version

Queen's University Belfast - Research Portal:Link to publication record in Queen's University Belfast Research Portal

Publisher rights© 2017 Elsevier Ltd. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/ which permits distribution and reproduction for non-commercial purposes, provided the author and source are cited.

General rightsCopyright for the publications made accessible via the Queen's University Belfast Research Portal is retained by the author(s) and / or othercopyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associatedwith these rights.

Take down policyThe Research Portal is Queen's institutional repository that provides access to Queen's research output. Every effort has been made toensure that content in the Research Portal does not infringe any person's rights, or applicable UK laws. If you discover content in theResearch Portal that you believe breaches copyright or violates any law, please contact openaccess@qub.ac.uk.

Download date:06. Aug. 2020

1

ArsenicintheBangladeshsoilsrelatedtophysiographicregion,paddy1

management,andmirco-andmacro-elementalstatus2

3

M.TanvirA.Chowdhury1,2,ClaireM.Deacon1,GerradD.Jones3,S.M.ImamulHuq2,PaulN.4

Williams4,A.F.M.ManzurulHoque5,LennyH.E.Winkel6,AdamH.Price1,GarethJ.Norton1*,5

andAndrewA.Meharg4*6781InstituteofBiologicalandEnvironmentalSciences,UniversityofAberdeen,Cruickshank9Building,St.MacharDrive,Aberdeen,AB243UU,U.K.102DepartmentofSoil,WaterandEnvironment,UniversityofDhaka,Dhaka-1000,Bangladesh113Eawag:SwissFederalInstituteofAquaticScienceandTechnology,Ueberlandstrasse133,12P.O.Box611,CH-8600Duebendorf,Switzerland134InstituteforGlobalFoodSecurity,Queen’sUniversityBelfast,DavidKeirBuilding,Malone14Road,BelfastBT95BN,NorthernIreland,U.K.155SoilResourceDevelopmentInstitute(SRDI),KhamarBariRoad,Dhaka1215,Bangladesh166SwissFederalInstituteofTechnology(ETH),InstituteofBiogeochemistryandPollutant17Dynamics,ETHZurich,CH-8092Zurich,Switzerland18

19

Correspondingauthors20

*AndrewA.Meharg21Tel:+44(0)2890975413;fax:+44(0)2890976513;e-mail:aa.meharg@qub.ac.uk2223*GarethJ.Norton24Tel:44(0)1224272409;fax:+44(0)1224272396;e-mail:g.norton@abdn.ac.uk25 26

2

Abstract27

28

Whiletheimpactofarsenicinirrigatedagriculturehasbecomeamajorenvironmental29

concerninBangladesh,todatethereisstillalimitedunderstandingofarsenicin30

Bangladeshipaddysoilsatalandscapescale.Asoilsurveywasconductedacrossten31

differentphysiographicregionsofBangladesh,whichencompassedsixtypesof32

geomorphology(Bil,Brahmaputrafloodplain,Gangesfloodplain,Meghnafloodplain,33

Karatoya-BangalifloodplainandPleistoceneterrace).Atotalof1209paddysoilsand23534

matchednon-paddysoilswerecollected.Thesourceofirrigationwater(groundwaterand35

surfacewater)wasalsorecorded.Theconcentrationsofarsenicandsixteenotherelements36

weredeterminedinthesoilsamples.Theconcentrationofarsenicwashigherinpaddysoils37

comparedtonon-paddysoils,withsoilsirrigatedwithgroundwaterbeinghigherinarsenic38

thanthoseirrigatedwithsurfacewater.TherewasacleardifferencebetweentheHolocene39

floodplainsandthePleistoceneterrace,withHolocenefloodplainsbeinghigherinarsenic40

andotherelements.Theresultssuggestthatarsenicismostlikelyassociatedwithlesswell41

weathered/leachedsoils,suggestingitiseither due to the geological newness of Holocene 42

sediments or differences between the sources of sediments, which gives rise to the arsenic 43

problems in Bangladeshi soils.44

45Introduction46

47

Riceiselevatedininorganicarseniccomparedtoallotherdietarystaples(Mehargetal.,48

2009).Floodingofsoils,asinpaddycultivation,leadstothemobilizationofnaturaland49

anthropogenicinorganicarsenicstoresinironoxyhydroxidephases,causedbyboththe50

3

reductionofarsenicandironundernegativesoilredoxpotentials(MehargandZhao,2012).51

Paddysoilsaremanagedthroughtilling,fertilization,andsurfacewaterandgroundwater52

irrigation,withthelatteroftenelevatedininorganicarsenicthroughoutlargeareasof53

Bangladesh(Huqetal.,2003;MehargandRahman,2003;Robertsetal.,2007;Luetal.,54

2009).Furthermore,arseniccanundergoanumberofprocesseswithinpaddysoilsthat55

leadstoitssubsequentlosssuchaspartitioningtomonsoonalfloodwaters(Dittmaretal.,56

2007;SahaandAli,2007;Dittmaretal.,2010;Robertsetal.,2010),leachingtosub-surfaces57

(McLarenetal.,2006;Khanetal.,2009;Heikensetal.,2007),andbiovolatilizationtoarsines58

(Mestrotetal.,2011).Thus,thearsenicloadingofanyparticularpaddysoilwillbedueto59

geologicaloriginandthesubsequentweatheringofconstituentminerals,andthe60

agronomicmanagementofthatsediment(Luetal.,2009).61

62

Bangladeshhasthreemajorgeomorphologicalunits(Brammer,1996;HuqandShoaib,63

2013).Thesearehill,terrace,andfloodplainareas.Thehillsoccupytwelvepercentofthe64

country’slandarea.TheupliftedterraceareasareofPleistoceneageandoccupyeight65

percentofthecountry.ThefloodplainsareofHoloceneageandoccupyeightypercentof66

thecountry.TheHolocenefloodplainsincludethepiedmontplains,riverfloodplains,tidal67

floodplains,andestuarinefloodplains.Thesegeomorphologicalunitsarerelatedtothe68

parentgeologicalformations,however,theyarealsocharacterizedbylandtopographyand69

ageofthesoilformationthroughsedimentdepositionovertime(Brammer,1996).70

71

Tounderstandandcharacterisethephysiographyofthegeomorphologicalareas,72

Bangladeshisdividedintotwentymainphysiographicregions(FAO/UNDP,1988).This73

physiographicclassificationwasbasedontheparentmaterialinwhichindividualsoiltypes74

4

wereformedandthelandscapeonwhichthesoilsweredeveloped(FAO/UNDP,1988).75

Therefore,thephysiographicregionshavedifferencesingeology,relief,drainage,ageof76

landformationandpatternofsedimentarydeposition.Thesedifferencesultimately77

influencethenatureandpropertiesofthesoilsinthedifferentphysiographicregions.78

79

Thebiogeochemicalcyclingofarsenicinsoilsisstronglyaffectedbyotherelements.Ironis80

centralduetothestrongassociationbetweeninsolublearsenateandiron(III)oxyhydroxides81

underaerobicconditionsandwiththemobilizationofiron(II)andarseniteunderreducing82

(thatis,paddy)conditions(BGS/DPHE,2001;SmedleyandKinniburgh,2002;McArthuret83

al.,2004;Polizzottoetal.,2005).Manganeseoxidesalsohaveasimilarredoxchemistryto84

ironandarestronglyimplicatedinarsenicimmobilization/mobilizationduringoxic/anoxic85

cyclingofpaddysediments(SmedleyandKinniburgh,2002;Hasanetal.,2007).Arsenateis86

aphosphateanalogueand,thus,keytocompetitionforbindingsiteswithinthesoilsolid87

phase,aswellashavingsimilarbiogeochemicalcyclingunderoxicconditions(Adriano,88

2001;MehargandHartley-Whitaker,2002;Smithetal.,2002;LambkinandAlloway,2003;89

Stachowiczetal.,2008).Calciumandmagnesiumimmobilizearsenateunderoxic90

conditions,andcouldalsohavearoleinthebiogeochemicalcyclingofarsenicata91

landscapelevel(Smithetal.,2002;Stachowiczetal.,2008;Fakhreddineetal.,2015).92

93

Here,wewantedtounderstandtherelationshipbetweensoilarsenicandpaddy94

managementpracticewithrespecttoarsenicloadingsinBangladeshisoils.Cultivationzones95

ofpaddysoils(n=1209)acrosstenphysiographicregionsofBangladesh,fromlatitude96

22°06'to24°53',andlongitude88°20'to90°59'weresampledandanalysedforarsenicand97

asuiteofsixteenotherelements(aluminium,calcium,cadmium,cobalt,chromium,copper,98

5

iron,lead,magnesium,manganese,molybdenum,nickel,phosphorus,potassium,sodium99

andzinc).Forasubsetofsoils(n=235),pairedpaddyandadjacentnon-paddysoilswere100

alsocollectedandcharacterised.Thedatawereusedtoaddressfourspecificobjectives:to101

assesstheimpactthatgeomorphologicaldifferenceshaveonsoilarsenicatalandscape102

level;tounderstandtherelationshipbetweentheconcentrationofarsenicinthepaddy103

soilswiththeconcertationofarsenicwithintheunderlyinggroundwater;todetermineif104

thesourceofirrigationwaterimpactsonsoilarsenicconcentrations;andbyexaminingthe105

concentrationsofarsenicandotherelementsinpaddyandnon-paddysoils,weaimedto106

understandtheimpactsthatpaddymanagementhasonsoilelementalconcentrations. 107

6

MaterialsandMethods108

CollectionofSoilSamples109

Atotalof1444soilsamples(topsoil,0-15cmfromthesurface)frompaddyfields(n=1209)110

andneighboringnon-paddyareas(n=235)werecollectedfrom10differentphysiographic111

regionswithin57sub-districts(upazilas)from17districtsofBangladesh(TableS1).Non-112

paddysoilsweredefinedasthesoilswherepaddycultivationandgroundwaterirrigation113

hadnotbeenpracticedwithinknownmemoryofthefarmers.Thephysiographicregions114

fromwherethesoilsampleswerecollectedincludedArialBil(n=42paddyand10non-115

paddysoils),BrahmaputraFloodplain(n=207paddyand64non-paddysoils),GangesRiver116

Floodplain(n=261paddyand58non-paddysoils),GangesTidalFloodplain(n=47paddy117

and11non-paddysoils),Gopalganj-KhulnaBils(n=63paddyand8non-paddysoils),118

Karatoya-Bangalifloodplain(n=15paddysoilsonly),MeghnaEstuarineFloodplain(n=204119

paddyand28non-paddysoils),andMeghnaRiverFloodplain(n=184paddyand26non-120

paddysoils)fromHolocenefloodplains,andBarindTract(n=68paddyand15non-paddy121

soils)andMadhupurTract(n=118paddyand15non-paddysoils)fromPleistocene122

terraces.Thesourceofirrigationwaterforthepaddysoilswasrecorded(groundwater,n=123

904;surfacewater,n=281;both,n=24).Onlythesoilsthathadanon-mixedirrigation124

sourcewereusedforanalyzingtheimpactofirrigationtypeonsoilarsenic.125

126

SampleProcessingandPreparationforAnalysis127

Thesoilsampleswereair-driedand,priortoanalysis,thesampleswereovendried(80°C±128

5°Cfor48h),andfinelygroundusingaball-mill.Thesoildigestionprocedurefollowedwas129

describedbyAdomakoetal.(2009).Briefly,0.1gofsoilwasplacedinaglassdigesttube130

and2.5mlofconcentratednitricacidwasaddedtothetubeandleftovernightforpre-131

7

digestion.Then,2.5mlofhydrogenperoxidewasaddedtothesamplejustbeforedigesting132

andthesamplewasheatedontheblockdigesterfor1hat80°C,for1hat100°C,for1hat133

120°Cand,finally,at140°Cfor3huntilthesolutionwasclear.Oncecooled,thedigested134

soilsamplesweretransferredinto15mlpolypropylenetubesandeachglasstubewas135

thoroughlyrinsed3timeswithultrapuredeionizedwater(Milli-Q18.2MΩ).Thevolumes136

weremadeupto15mlmarkusingthesamewater.Toobtaintheappropriatedilutionfor137

analysisbyinductivelycoupledplasma-massspectrometer(ICP-MS)andmicrowaveplasma-138

atomicemissionspectrometer(MP-AES),thesampleswerefurtherdilutedto1in10.139

Calibrationstandardswerepreparedfrom1000mg/lmulti-elementstocksolutions(SPEX140

CertiPrepReferenceMaterial).141

142

ChemicalAnalysis143

ThepHofthesoilsamplesweremeasuredatasoil:water(deionizedwater)ratioof1:2.5144

(HuqandAlam,2005).TheICP-MS(AgilentTechnologies7500c,Japan)wasusedto145

determinethetotalconcentrationsofarsenic,cadmium,cobalt,copper,chromium,lead,146

manganese,molybdenum,nickel,phosphorus,andzincinthesoildigestsandtheMP-AES147

(AgilentTechnologies4100Series,USA)wasusedtodeterminethetotalconcentrationsof148

aluminum,calcium,iron,magnesium,potassium,andsodiuminthesoildigests.Ineach149

batchofdigestion,tenpercentofthetotalnumberofsampleswereselectedrandomlyfor150

duplicateanalysis(n=172).Everybatchofsamplesconsistedof33randomlyselectedsoil151

samples,4duplicates,1blank,and1soilCRM(certifiedreferencematerial)(NCSZC73007,152

ChinaNationalAnalysisCenterforIronandSteel),whichwererandomizedpriorto153

analyticalanalysis.154

155

8

SoilMapping156

ThedatausedtoperformthemappingofarsenicinpaddysoilsacrossBangladeshincluded157

the1209paddysoilsanalyzedinthisstudyaswellas395soilarsenicconcentrationsfrom158

previousstudies(Williamsetal.,2011;Luetal.,2009;Islametal.,2012).ArcGISv.10.2(Esri)159

wasusedtocreateandanalyzegroundwaterandsoilarsenicmap.Thegroundwaterarsenic160

datawereobtainedfromBGS/DPHE(2001).161

162

StatisticalAnalysis163

AllstatisticalanalyseswereperformedusingthestatisticalsoftwareMinitabv.16(State164

CollegePA)andSigmaPlotv.13(SystatSoftwareInc.,CA).Thedatawerecheckedfor165

normalityandweretransformedpriortostatisticalanalysiswhereappropriate.166

167

9

ResultsandDiscussion168

Inordertoverifytheaccuracyoftheanalyticalmethodsaswellasthequalityofthedata,169

percentrecoveriesoftheelementsinCRMandrelationshipsbetweentheelement170

concentrationsinthesamplesandintheduplicates(tenpercentofthetotalnumberof171

samples)werecalculatedandtheaveragerecoveries(inpercentages)oftheelementsinthe172

CRMsandtheresultsoftheduplicateanalysisarepresentedinTableS2andFig.S1,173

respectively.174

175

Todevelopasoilarsenicmapofthesampledsoils,allpaddysoilsamplinglocationswithina176

10km2gridwereaveraged(Fig.1).Individuallocationsandsamplingdensitiesareshownin177

Fig.S2.Thereisaclearnorth/southdivideinpaddyarsenicconcentrationswithmuchhigher178

concentrations,ingeneral,inthesouth.Thepaddysoilarseniclevelsreportedhere(1-88179

mg/kg,average=8mg/kg)arewithintherangesreportedforpreviousBangladeshpaddy180

soilsurveys(Huqetal.,2003;MehargandRahman,2003;Luetal.,2009;Williamsetal.,181

2011;HuqandShoaib,2013).Thepatternofpaddysoilconcentrationsrelatewellto182

groundwatermeasurements(BGS-DPHE,2001),againwithgroundwaterelevatedinthe183

south,excludingthecoastalzone.Theexceptionistheclusterofsamplingpointsinthe184

extremesouth-eastthathavealowsoilarsenicconcentrationandthehighestgroundwater185

arsenicconcentration.Thisisprobablyduetothesourceofirrigationwaterusedinthis186

south-eastregion,wherethemainirrigationmethodisfromsurfacewaterratherthan187

groundwater(Fig.S3).Whencomparingthearsenicconcentrationsinthepaddiesthathave188

beenirrigatedwithgroundwaterandsurfacewateracrossBangladesh,therewasa189

significantdifference(ANOVAF=26.23,p<0.001)inthesoilarsenicconcentration(Fig.2).190

Soilsirrigatedwithgroundwaterhadonaverageanarsenicconcentrationof8.5mg/kg191

10

whichwassignificantlyhigherthanthesoilsirrigatedwithsurfacewater,whichhadan192

averagearsenicconcentrationof5.7mg/kg.Fortheindividualphysiographicregions,seven193

oftheregionshadenoughgroundwaterandsurfacewaterirrigatedsoils(>10)todo194

comparisonsbetweenirrigationmethodandsoilarsenic.Therewasnosignificantdifference195

insoilarsenicbetweenthegroundwaterirrigationandsurfacewaterirrigationforfourof196

thesevenphysiographicregions.Forthethreeotherphysiographicregions,significant197

differencesinarsenicconcentrationswereobservedbetweenthesoilsirrigatedwith198

groundwater(GWI)andsurfacewater(SWI),withhigherarsenicconcentrationsinthe199

groundwaterirrigatedsoilsthaninthesurfacewaterirrigatedsoils(forGangesTidal200

Floodplain,ANOVAF=5.97,p<0.05,n=28(GWI)and20(SWI),mean=14.6mg/kg(GWI)and201

8.6mg/kg(SWI);forMeghnaEstuarineFloodplain,ANOVAF=14.84,p<0.001,n=69(GWI)202

and111(SWI),mean=8mg/kg(GWI)and3.9mg/kg(SWI);forMeghnaRiverFloodplain,203

ANOVAF=62.06,p<0.001,n=130(GWI)and54(SWI),mean=9.4mg/kg(GWI)and4.7204

mg/kg(SWI)).Asthesampleswerecollectedfromdifferentgeomorphicregions,these205

resultscouldbeconfoundedbytheunderlyinggeomorphology.However,differenceinsoil206

arsenicduetodifferentirrigationtechniquesappearstobeatrendacrossthecountry.207

208

SoilarsenicconcentrationsacrosstendifferentphysiographicregionsofBangladeshwere209

comparedtoseehowtheconcentrationsvariedbetweenthedifferentregions.Highly210

significantvariations(ANOVAF=75.28,p<0.001andANOVAF=6.33,p<0.001,respectivelyfor211

paddyandnon-paddysoils)wereobservedinsoilarsenicconcentrationsamongtheten212

physiographicregions(Fig.3forpaddysoils,fig.S4fornon-paddysoils).TheMadhupur213

TractandtheBarindTractwerefoundtohavethelowestarsenicconcentrations(0.6-10.3,214

mean=3.4mg/kg,and0.8-23.4,mean=2.8mg/kg,respectively)inthepaddysoils,whereas215

11

theGangesRiverFloodplain(1.6-68,mean=11mg/kg)andtheGangesTidalFloodplain(3-216

42.5,mean=13.1mg/kg)hadthehighestsoilarsenicconcentrations.Martinetal.(2014,217

2015)reportedhigherconcentrationsandmobilizationofarsenicintheGangesfloodplain218

soils,duetoenhancedinfluenceofthepedoenvironmentalpropertiesintheregion,219

comparedtothatintheMeghnafloodplainsoilssuggestingacomplexinteractionbetween220

soilproperties,climateandagriculturalmanagementpracticesinthepaddysoil221

environmentinBangladesh.InthepresentstudytheGangesfloodplainsoilswereclassified222

asGangesRiverFloodplainandGangesTidalFloodplain,andtheMeghnafloodplainsoils223

wereclassifiedasMeghnaRiverFloodplainandMeghnaEstuarineFloodplain,whileno224

significantdifferenceinarsenicconcentrationswasobservedbetweenGangesRiver225

FloodplainandMeghnaRiverFloodplainsoils(Fig.3).Similarobservationswerealso226

reportedforgroundwaterarsenicconcentrationsacrossthedifferentgeomorphological227

unitsofthecountry(BGS/DPHE,2001;Ravenscroft,2001).228

229

Atagrosslevel,highandlowgroundwaterarsenicconcentrationregionsareknowntobe230

basedonphysiographicunits,withlowconcentrationsofarsenicingroundwatersinthe231

higheraltitudePleistoceneterraces,andathighconcentrationsinHolocenefloodplains232

(BGS/DPHE,2001;SmedleyandKinniburgh,2002;Ahmedetal.,2004;Ravenscroftetal.,233

2005).TheexplanationforthisisthatPleistocenesedimentsaremorehighlyweatheredand234

leachedofarsenic(Ravenscroft,2001;Ravenscroftetal.,2005).Arecentstudyonthe235

sourceofarsenicintheHolocene/PleistocenesedimentsfromtheTeraiplainofNepal(that236

stratigraphicallyresembleBangladeshisediments)proposedanumberofcomplexprocesses237

whichcanexplainthedifferencesinarsenicconcentrationbetweenHoloceneand238

Pleistocenesediments(Guillotetal.,2015).However,theriversystemsofBangladesh239

12

activelyreworkthelandscape,givinglensesofsoilremobilizedandre-deposited,240

interlayeringHoloceneandPleistocenesoils(BGS/DPHE,2001;Polizzottoetal.,2005;241

Mehargetal.,2006;Guillotetal.,2015).Itisalsoknownthatdifferentiallossofarsenic242

occursfromgroundwaterirrigatedpaddysoilsduringthesubsequentmonsoonalfloods243

throughpartitioningofsoilarsenicintooverlayingfloodwaters(Dittmaretal.,2007;Saha244

andAli,2007;Dittmaretal.,2010;Robertsetal.,2010).245

246

Todeterminethecontributionofbothnaturalsoilarsenicconcentrationsandhowpaddy247

managementpracticeshavecontributedtowardsthecurrentsoilarsenicconcentration,248

pairednon-paddyandpaddysoilsfrommajorphysiographicunitsofBangladeshwere249

analysed(Fig.4).Therewasasignificantrelationshipforsoilarsenicbetweenthepaddyand250

non-paddysoils(linearregressionR2=0.26,p<0.001,n=235)(TableS3).Theslopeoftheoverall251

regression(thatis,forallsoils)is1.6:1forpaddy:non-paddy,thatis,ageneralincreasein252

arsenicof60%inpaddycultivatedsoils.Thesoilsformthefloodplainsandbils(low-lying253

floodplain)followedthesamepatternastheoverallregressionregardlessiftheyarefrom254

theBrahmaputra,GangesorMeghnafloodplains.Pleistoceneterracesoilsstandapartand255

donotfollowtheoverallregression,beingbothonaveragelowerinarsenic,andhavingless256

arsenicaccumulationinpaddysoilscomparedtoHolocenefloodplainsoils.Apairedt-testof257

thematchingpaddyandnon-paddysoilsforarsenicconcentrationswithinthePleistocene258

terracesoilsindicatedthatthesesoilsweresignificantlydifferent(p<0.05),withthenon-259

paddysoilshavingelevatedarsenicconcentrationsincomparisontothepaddysoils,onan260

averagethenon-paddysoilshad19percenthigherarsenic.Thisindicatesthatpaddy261

managementisnotincreasingarsenicconcentrationsintheseterracesoils.Pleistocene262

terracegroundwatersarelowinarsenic(Nicksonetal.,2000;BGS/DPHE,2001;Ahmedet263

13

al.,2004;Ravenscroftetal.,2005),andthus,irrigationofPleistoceneterracesoilswith264

groundwatersshouldnotleadtoelevationinarsenic.AsHolocenefloodplaingroundwaters265

usedinpaddyirrigationareelevatedinarsenic(Alietal.,2003;Huqetal.,2003;Meharg266

andRahman,2003;SahaandAli,2007;Luetal.,2009;HuqandShoaib,2013),irrigationof267

paddieswitharsenicelevatedgroundwatershasthepotentialtoleadtobuild-upinsoil268

arsenic.Arsenicinthenon-paddyfloodplainsoilsrangedfrom1.8-24.3mg/kg(mean±sd=269

5.6±2.9,coefficientofvariation=0.52,n=205),showingthatarsenicisnaturallyvariable270

inBangladeshifloodplainsoils,withthisrangebeing2-11mg/kg(mean±sd=3.9±1.8,271

coefficientofvariation=0.46,n=30)forthePleistoceneterracesoils.Thisemphasisesthe272

inherentvariabilityinnaturalsoilarsenic,butthatvariabilityislessonPleistoceneterrace273

soils.ItistheHolocenesoils/sedimentsthatareexposedtotheactivereworkingthat274

typifiesadynamicestuarinedepositionalenvironment(SullivanandAller,1996;BGS/DPHE,275

2001;Polizzottoetal.,2005;Mehargetal.,2006;Luetal.,2009;Guillotetal.,2015),and276

thismayexplainthevariability.Theinherentdifferencesinthesedimentsofthefloodplain277

basinsdepositedfromdifferentsourcesovertime,differencesinarsenicaccumulation/278

releaseequilibriarelatedtotheindigenoussoilchemistry,residencetime,depthand279

durationofmonsoonfloodwater,rateofparticledispersion,rateofleachingtosubsurface,280

andbiovolatilizationtotheatmospherecanalsocontributetoexplainthevariabilityof281

arsenicinthefloodplainsoilsofBangladesh(McLarenetal.,2006;Huqetal.2008;Khanet282

al.,2009;Robertsetal.,2010;Mestrotetal.,2011;Brammer,2012;Martinetal.,2015).In283

addition,thediversityandcomplexityofsoilsinthefloodplainsofBangladeshare284

influencedbyvariationsinfloodingdepthwithintheinundationlandtypes(Brammer1997;285

Huqetal.,2008),andhence,theaccumulationandreleaseofarsenicinsoilsvarywithinthe286

14

toposequenceofalandscapeduetovariationsinreliefandsoilproperties,particularlyiron,287

clayandorganicmattercontents(Huqetal.,2008;Brammer2012b;Ahmedetal.2011).288

289

GiventhatdifferentgeomorphicregionswithintheHolocenefloodplainandPleistocene290

terraceregionsfollowthesamegeneraltrends,withthemaindifferencesbeingbetween291

floodplainandterrace,furtheranalysisconcentratedonfloodplainversusterrace292

comparisons.ForthePleistocenesoils,comparingpaddyandnon-paddyrelationshipswere293

seenforallelementstested(Fig.5andFig.S?).However,itwasonlyforarsenicthatpaddy294

soilsmovedawayfroma1:1relationship,andgroundwaterisspecificallyonlyelevatedin295

arsenictoanysignificantextent(BGS/DPHE,2001)withrespecttolevelsalreadyfoundin296

soil,thisisfurtherevidencethatitisgroundwaterirrigationperse,ratherthanother297

aspectsoffieldmanagement,suchasfertilizerandmanuringpractices,thatperturbpaddy298

soilarseniclevelscomparedtonon-paddysoils.Thedepletioninmacro-nutrientsin299

Pleistocenesediments,particularlythealkalineearthscalciumandmagnesium,ismost300

apparent.Arsenicisalsopositivelycorrelated(r=0.3,p<0.001)withsoilpH(Fig.S3?),with301

lowpHcausedbylowcalciumandmagnesiumconcentrations,crossconfirmingthe302

interplayofsoilsfactorscorrelatedwitharsenic.Ironandphosphorus,twoelements303

intimatelyassociatedwitharsenic’sbiogeochemicalcycles(FitzandWenzel,2002;Smithet304

al.,2002;Heikensetal.,2007)arealsohighlydepletedinPleistocenesoils.Non-essential305

aluminium,cadmium,andleadalsofollowthesametrend.Ithasbeendemonstratedthat306

pedogenicprocessesareresponsibleforthedepletionofnutrientswithinsoilsovertime307

(Peltzeretal.,2010).Additionally,nutrientscanbedepletedinsoilsovershorterperiodsof308

time(Chenetal.,2011).Soilsthathavebeenundercontinuouspaddycroppinghavebeen309

showntobedepletedinkeymacronutrientsinaveryshortperiodoftime,forexample,310

15

calcium,magnesium,andsodiumhavebeendemonstratedtoberapidlylostinpaddysoils311

within50yearsofricecultivation(Chenetal.,2011).312

313

ThewidercharacterizationofBangladeshipaddysoils,whereelementalconcentrationis314

plottedagainstcorrespondingarsenicconcentration(Fig.6),showsthesametrendasthe315

pairedpaddy–non-paddysampleswithPleistocenedepletedinallelementstestedas316

comparedtoHolocene,withthosebeinghighinarsenicalso,ingeneral,beinghighinthe317

correspondingelements(TableS4andFig.S?).ThisindicatesagainthatPleistocenesoilsare318

lesssustainablethanHolocenewithrespecttotheirelementalnutritionalqualities.Itisoff319

concernthatricegrainslowinarsenicmaybelowerinnutrientsaswell.Thissubjectareais320

notwellinvestigatedexceptwhereitwasshownthatinaBangladeshicontextthaton321

arsenicenrichedgroundwaterirrigatedpaddiesthatenhancedgrainarsenichad,ingeneral,322

suppressionofmicro-nutrientlevelsinricegrain(Williamsetal.,2009;Nortonetal.,2010).323

Unfortunately,thereappearstobetwoglobalprocessesthatregulatearsenicingrain,low324

nutrientsoilshavelowarsenic,andhigharsenicinhibitsgrainnutrientlevels.Thiswarrants325

furtherstudyinBangladesh,namelybywidesurveyofgrainversussoilassociationsforthe326

primarymineralnutrientsofhumanhealthimportance.327

328

Whatisalsoapparentfromtheplotsofelementalconcentrationagainstarsenicisthat329

Holocenesoilshaveamuchwiderrangeofarsenicconcentrationsathigherconcentrations330

oftheotherelementscomparedtoPleistocenesoils,thatis,thereismuchgreaterinherent331

variabilityinarseniccomparedtootherelements,specificallywhenotherelemental332

concentrationsarehigh(Fig.6).Thisisindicativeagainthatagriculturalmanagement333

practicesspecificallyaltersoilarsenicconcentrationsinBangladesh.Groundwaterfor334

16

irrigationistheprimarysourceofarsenictofloodplainpaddiesthatarecroppedduringthe335

dryseasonandiswellknowntoelevatearsenicinpaddysoils(Alietal.,2003;Huqetal.,336

2003;Mehargetal.,2003;Dittmaretal.,2007;SahaandAli,2007;Huq,2008;Luetal.,337

2009;Ahmedetal.,2011;HuqandShoaib,2013).Paddysoilsalsohavedifferential338

interactionwithmonsoonalfloodsfollowingdryseasonapplicationofarsenic,witharsenic339

capableofpartitioningfromsoilsintofloodwaters(Dittmaretal.,2007;SahaandAli,2007;340

Dittmaretal.,2010;Robertsetal.,2010).Asthisinteractionbetweenfloodwaterandsoil341

arsenicwillbedependentonsoilpropertiesandonthedynamicsoffloodwaterpatternsfor342

anyspecificpaddysoil,heterogeneityinarsenicremovalisexpected.Asthepaddysoils343

haveahigherarsenicconcentrationcomparedtothematchednon-paddysoils,itwould344

indicatethatthisprocessoflossofarsenicformthesoilsbymonsoonalfloodsisnot345

sufficienttoreducethatarsenicconcentrationinthepaddysoilsbacktothenon-paddysoil346

backgroundconcentration.347

348

WhenPrincipleComponentsAnalysis(PCA)isusedtolookattheinterrelationshipsbetween349

arsenicandotherelements,thesoilsclusterintoPleistoceneandHoloceneusingthefirst350

andsecondcomponents(Fig.7,TableS5).Thereissomeoverlapinthemiddlebutthisis351

expectedperhapsasthelargescalesatwhichphysiographicregionsaredrawnwillmissthe352

finedetailontheground.Thisisfurtherconfoundedbythelensingofoldsoilsovernewand353

withthesedimentdepositionalenvironmentalsobeinghighlyactive(Polizzottoetal.,2005;354

Mehargetal.,2006;Luetal.,2009;Guillotetal.,2015).Thedirectionoftheloadingsforthe355

componentsshowsthatarsenictrendswithmostelements,anditisonlycadmiumand356

molybdenumthatgenerallydiffer.ThePCAanalysisgivesfurtherstrengthtothehypothesis357

thatarsenicissimplyassociatedwithlesswellweathered/leachedsediments,again358

17

suggestingitiseither due to the geological newness of Holocene sediments or differences 359

between the sources of sediments thatgivesrisetothearsenicproblemsinBangladesh,and360

elsewhere(SmedleyandKinniburgh,2002;McArthuretal.,2004;Nicksonetal.,2005;361

Polyaetal.,2005;Bergetal.,2007;Mukherjeeetal.,2008;Rowlandetal.,2008;Winkelet362

al.,2008;Guillotetal.,2015).363

364

Conclusion365

….366

SupportingInformation367

Fouradditionaltablesandfourfiguresasnotedinthetext.Thismaterialisavailablefreeof368

chargeviatheInternet.369

370

Acknowledgement371

ThisworkwasdoneaspartofadoctoralfellowshipplanfundedbytheCommonwealth372

ScholarshipCommissionintheUK.WegratefullyacknowledgetheComputerandGISUnitof373

BangladeshAgriculturalResearchCouncil(BARC)forprovidingGISfilesforsoilmapping.The374

soilsampleswereimportedintotheUKunderimportlicenseIMP/SOIL/6/2013issuedby375

ScienceandAdviceforScottishAgriculture.L.H.E.W.andG.D.J.acknowledgefundingbythe376

SwissNationalScienceFoundation(SNFPP00P2_133619;PP00P2_163747).377

378

References379

Adomako,E.E.,Solaiman,A.R.M.,Williams,P.N.,Deacon,C.,Rahman,G.K.M.M.,Meharg,380A.A.2009.EnhancedtransferofarsenictograinforBangladeshgrownricecomparedtoUS381andEU.Environ.Int.35,476–479.382

Adriano,D.C.,2001.TraceElementsintheTerrestrialEnvironment.Springer,NewYork.383

18

Ahmed,K.M.,Bhattacharya,P.,Hasana,M.A.,Akhter,S.H.,Alam,S.M.M.,Bhuyian,M.A.H.,384Imam,M.B.,Khan,A.A.,Sracek,O.2004.Arsenicenrichmentingroundwaterofthealluvial385aquifersinBangladesh:anoverview.Appl.Geochem.19,181–200.386

Ahmed,Z.U.,Panaullah,G.M.,DeGloria,S.D.,Duxbury,J.M.2011.Factorsaffectingpaddy387soilarsenicconcentrationinBangladesh:Predictionanduncertaintyofgeostatisticalrisk388mapping.Sci.Tot.Environ.412–413,324–335.389

Ali,M.A.,Badruzzaman,A.B.M.,Jalil,M.A.,Hossain,M.D.,Ahmed,M.F.,Masud,A.A.,390Kamruzzaman,M.,Rahman,M.A.2003.Arsenicinplant-soilenvironmentinBangladesh,in:391Ahmed,M.F.,Ali,M.A.,Adeel,Z.(Eds.),FateofArsenicintheEnvironment.ITNCentre:392BangladeshUniversityofEngineeringandTechnology,Dhaka,pp.85–112.393

Berg,M.,Stengel,C.,Trang,P.T.K.,Viet,P.H.,Sampson,M.L.,Leng,M.,Samreth,S.,394Fredericks,D.2007.MagnitudeofarsenicpollutionintheMekongandRedRiverDeltas—395CambodiaandVietnam.Sci.Tot.Environ.372,413–425.396

BGS/DPHE(BritishGeologicalSurvey/BangladeshDepartmentforPublicHealth397Engineering),2001.ArseniccontaminationofgroundwaterinBangladesh,FinalReport.398Kinniburgh,D.G.andSmedley,P.L.(Eds.),BritishGeologicalSurveyReportWC/00/19.BGS,399Keyworth,Vol.2.400

Brammer,H.,1996.TheGeographyoftheSoilsofBangladesh.TheUniversityPressLtd.,401Dhaka.402

Brammer,H.,2012.ThePhysicalGeographyofBangladesh.TheUniversityPressLtd.,Dhaka.403

Chen,L.-M.,Zhang,G.-L.,Effland,W.R.2011.Soilcharacteristicresponsetimesand404pedogenicthresholdsduringthe1000-Yearevolutionofapaddysoilchronosequence.Soil405Sci.Soc.Am.J.75,1807–1820.406

Dittmar,J.,Voegelin,A.,Roberts,L.,Hug,S.J.,Saha,G.C.,Ali,M.A.,Badruzzaman,A.B.M.,407Kretzschmar,R.2007.Spatialdistributionandtemporalvariabilityofarsenicinirrigatedrice408fieldsinBangladesh.2.Paddysoil.Environ.Sci.Technol.41,5967–5972.409

Dittmar,J.,Voegelin,A.,Roberts,L.C.,Hug,S.J.,Saha,G.C.,Ali,M.A.,Badruzzaman,A.B.M.,410Kretzschmar,R.2010.ArsenicaccumulationinapaddyfieldinBangladesh:Seasonal411dynamicsandtrendsoverathree-yearmonitoringperiod.Environ.Sci.Technol.44(8),4122925–2931.413

FAO/UNDP(FoodandAgricultureOrganization/UnitedNationsDevelopmentProgramme),4141988.LandresourcesappraisalofBangladeshforagriculturaldevelopment,Report2:415AgroecologicalRegionsofBangladesh.ReportpreparedfortheGovernmentofthePeople’s416RepublicofBangladeshbytheFoodandAgricultureOrganizationofUnitedNations,United417NationsDevelopmentProgramme,BGD/81/035,Technicalreport2.418

Fakhreddine,S.,Dittmar,J.,Phipps,D.,Dadakis,J.,Fendorf,S.,2015.Geochemicaltriggers419ofarsenicmobilizationduringmanagedaquiferrecharge.Environ.Sci.Technol.49(13),4207802–7809.421

19

Fitz,W.J.,Wenzel,W.W.2002.Arsenictransformationsinthesoil-rhizosphereplantsystem:422fundamentalsandpotentialapplicationtophytoremediation.J.Biotechnol.99,259–278.423

Guillot,S.,Garçon,M.,Weinman,B.,Gajurel,A.,Tisserand,D.,France-Lanord,C.,vanGeen,424A.,Chakraborty,S.,Huyghe,P.,Upreti,B.N.,Charlet,L.2015.OriginofarsenicinLate425PleistocenetoHolocenesedimentsintheNawalparasidistrict(Terai,Nepal).Environ.Earth426Sci.74,2571–2593.427

Hasan,M.A.,Ahmed,K.M.,Sracek,O.,Bhattacharya,P.,vonBrömssen,M.,Broms,S.,428Fogelström,J.,Mazumder,M.L.,Jacks,G.2007.Arsenicinshallowgroundwaterof429Bangladesh:investigationsfromthreedifferentphysiographicsettings.Hydrogeol.J.15,4301507–1522.431

Heikens,A.,Panaullah,G.M.,Meharg,A.A.2007.Arsenicbehaviourfromgroundwaterand432soiltocrops:Impactsonagricultureandfoodsafety.Rev.Environ.Contam.Toxicol.189,43343–87.434

Huq,S.M.I.,2008.Fateofarsenicinirrigationwateranditspotentialimpactsonfoodchain,435in:Ahuja,S.(Ed.),ArsenicContaminationofGroundwater:Mechanism,Analysis,and436Remediation.Wiley,Hoboken,pp.23–49.437

Huq,S.M.I.;Alam,M.D.AHandBookonAnalysesofSoil,Plant,andWater;Bangladesh438AustraliaCenterforEnvironmentalResearch:UniversityofDhaka,Dhaka,Bangladesh,2005.439

Huq,S.M.I.,Shoaib,J.U.M.2013.TheSoilsofBangladesh,worldsoilsbookseries.Springer,440Dordrecht.441

Huq,S.M.I.,Rahman,A.,Sultana,N.,Naidu,R.2003.Extentandseverityofarsenic442contaminationinsoilsofBangladesh,in:Ahmed,M.F.,Ali,M.A.,Adeel,Z.(Eds.),Fateof443ArsenicintheEnvironment.ITNCentre:BangladeshUniversityofEngineeringand444Technology,Dhaka,pp.69–84.445

Islam,M.R.,Brammer,H.,Rahman,G.K.M.M.,Raab,A.2012.Arsenicinricegrowninlow-446arsenicenvironmentsinBangladesh.WaterQual.Expo.Health4(4),197–208.447

Khan,M.A.,Islam,M.R.,Panaullah,G.M.,Duxbury,J.M.,Jahiruddin,M.,Loeppert,R.H.2009.448Fateofirrigation-waterarsenicinricesoilsofBangladesh.PlantandSoil322(1),263–277.449

Lambkin,D.C.;Alloway,B.J.2003.Arsenate-inducedphosphatereleasefromsoilsandits450effectonplantphosphorus.WaterAirSoilPollut.144,41–56.451

Lu,Y.;Adomako,E.E.;Solaiman,A.R.M.;Islam,M.R.;Deacon,C.;Williams,P.N.;Rahman,452G.K.M.M.;MehargA.A.2009.Baselinesoilvariationisamajorfactorinarsenic453accumulationinBengalDeltapaddyrice.Environ.Sci.Technol.43,1724–1729.454

Martin,M.,Bonifacio,E.,JakeerHossain,K.M.,ImamulHuq,S.M.,Barberis,E.,2014.Arsenic455fixationandmobilizationinthesoilsoftheGangesandMeghnafloodplains.Impactof456pedoenvironmentalproperties.Geoderma228–229,132–141.457

20

Martin,M.,Stanchi,S.,Hossain,K.M.J.,ImamulHuq,S.M.,Barberis,E.,2015.Potential458phosphorusandarsenicmobilizationfromBangladeshsoilsbyparticledispersion.Sci.Tot.459Environ.53629,973–980.460461McArthur,J.M.;Banerjee,D.M.;Hudson-Edwards,K.A.;Mishra,R.;Purohit,R.;462Ravenscroft,P.;Cronin,A.;Howarth,R.J.;Chatterjee,A.;Talukder,T.;Lowry,L.;Houghton,463S.;Chadha,D.K.2004.Naturalorganicmatterinsedimentarybasinsanditsrelationto464arsenicinanoxicgroundwater: theexampleofWestBengalanditsworldwideimplications.465Appl.Geochem.19,1255−1293.466

McLaren,R.G.;etal.Fateofarsenicinthesoilenvironment.InManagingArsenicinthe467Environment:FromSoiltoHumanHealth;Naidu,R.,Smith,E.,Qwens,G.,Bhattacharya,P.,468Nadebaum,P.Eds.;CSIROPublishing:Collingwood2006;pp157−182.469

Meharg,A.A.,Hartley-Whitaker,J.2002.Arsenicuptakeandmetabolisminarsenicresistant470andnon-resistantplantspecies.NewPhytol.154,29–43.471

Meharg,A.A.,Rahman,M.2003.ArseniccontaminationofBangladeshpaddyfieldsoils:472Implicationsforricecontributiontoarsenicconsumption.Environ.Sci.Technol.37(2),473229−234.474

Meharg,A.A.,Scrimgeour,C.,Hossain,S.A.,Fuller,K.,Cruickshank,K.,Williams,P.N.,475Kinniburgh,D.G.2006.Co-depositionoforganiccarbonandarsenicinBengalDeltaaquifers.476Environ.Sci.Technol.40,4928−4935.477

Meharg,A.A.,Zhao,F.J.2012.Arsenic&Rice.Springer,Netherlands.478

Meharg,A.A.,Williams,P.N.,Adomako,E.,Lawgali,Y.Y.,Deacon,C.,Villada,A.,Cambell,479R.C.J.,Sun,G.,Zhu,Y.-G.,Feldmann,J.,Raab,A.,Zhao,F.-J.,Islam,R.,Hossain,S.,Yanai,J.4802009.Geographicalvariationintotalandinorganicarseniccontentofpolished(white)rice.481Environ.Sci.Technol.43,1612-1617.482

Mestrot,A.,Feldmann,J.,Krupp,E.M.,Hossain,M.,Roman-Ross,G.,Meharg,A.A.2011.483Fieldfluxesandspeciationofarsinesemanatingfromsoils.Environ.Sci.Technol.45,4841798−1804.485

Mukherjee,A.,Bhattacharya,P.,Savage,K.,Foster,A.,Bunschuh,J.2008.Distributionof486geogenicarsenicinhydrologicsystems:controlsandchallenges.J.Contam.Hydrol.99,1–7.487

Nickson,R.T.,McArthur,J.M.,Ravenscroft,P.,Burgess,W.G.,Ahmed,K.M.2000.488Mechanismofarsenicreleasetogroundwater,BangladeshandWestBengal.Appl.489Geochem.15,403−413.490

Nickson,R.T.,McArthur,J.M.,Shrestha,B.,Kyaw-Myint,T.O.,Lowry,D.2005.Arsenicand491otherdrinkingwaterqualityissues,MuzaffargarhDistrict,Pakistan.Appl.Geochem.20,49255−68.493

Norton,G.J.,Dasgupta,T.,Islam,M.R.,Islam,S.,Zhao,F.-J.,Stroud,J.L.,McGrath,S.P.,494Feldmann,J.,Price,A.H.,Meharg,A.A.2010.Arsenicinfluenceoncultivarvariationingrain495

21

trace-elementnutrientcontentinBengalDeltagrownrice.Environ.Sci.Technol.44,4968284−8288.497

Peltzer,D.A.,Wardle,D.A.,Allison,V.J.,Baisden,W.T.,Bardgett,R.D.,Chadwick,O.A.,498Condron,L.M.,Parfitt,R.L.,Porder,S.,Richardson,S.J.,Turner,B.L.,Vitousek,P.M.,Walker,499J.,Walker,L.R.2010.Understandingecosystemretrogression.Ecol.Monogr.80(4),500509−529.501

Polizzotto,M.L.,Harvey,C.F.,Sutton,S.R.,Fendorf,S.2005.Processesconductivetothe502releaseandtransportofarsenicintoaquifersofBangladesh.PNAS102,18819−18823.503

Polya,D.A.,Gault,A.G.,Diebe,N.,Feldman,P.,Rosenboom,J.W.,Gilligan,E.,Fredericks,D.,504Milton,A.H.,Sampson,M.,Roland,H.A.L.,Lythgoe,P.R.,Jones,J.C.,Middleton,C.,Cooke,D.505A.2005.ArsenichazardinshallowCambodiangroundwaters.Mineral.Mag.69,807−823.506

Ravenscroft,P.2001.DistributionofgroundwaterarsenicinBangladeshrelatedtogeology,507in:Jacks,G.,Bhattacharya,P.,Khan,A.A.(Eds.),GroundwaterArsenicContaminationinthe508Bengal Delta Plains of Bangladesh. Proceedings of the KTH-Dhaka University Seminar,509UniversityofDhaka,Bangladesh.KTHSpecialPublication,TRITA-AMIReport3084,pp.41–56.510

Ravenscroft,P.,Burgess,W.G.,Ahmed,K.M.,Burren,M.,Perrin,J.2005.Arsenicin511groundwateroftheBengalBasin,Bangladesh:Distribution,fieldrelations,and512hydrogeologicalsetting.Hydrogeol.J.13(5-6),727–751.513

Roberts,L.C.,Hug,S.J.,Dittmar,J.,Voegelin,A.,Kretzschmar,R.,Saha,G.C.,Ali,M.A.,514Badruzzaman,A.B.M.2007.Spatialdistributionandtemporalvariabilityofarsenicin515irrigatedricefieldsinBangladesh.1.Irrigationwater.Environ.Sci.Technol.41,5960-5966.516

Roberts,L.C.,Hug,S.J.,Dittmar,J.,Voegelin,A.,Kretzschmar,R.,Wehrli,B.,Cirpka,O.A.,517Saha,G.C.,Ali,M.A.,Badruzzaman,A.B.M.2010.Arsenicreleasefrompaddysoilsduring518monsoonflooding.Nat.Geosci.3(1),53–59.519

Rowland,H.A.L.,Gault,A.G.,Lythgoe,P.,Polya,D.A.2008.Geochemistryofaquifer520sedimentsandarsenic-richgroundwatersfromKandalProvince,Cambodia.Appl.Geochem.52123,3029–3046.522

Saha,G.C.,Ali,M.A.2007.Dynamicsofarsenicinagriculturalsoilsirrigatedwitharsenic523contaminatedgroundwaterinBangladesh.Sci.Tot.Environ.379(2-3),180–189.524

Smedley,P.L.,Kinniburgh,D.G.2002.Areviewofthesource,behaviouranddistributionof525arsenicinnaturalwaters.Appl.Geochem.17,517–568.526

Smith,E.,Naidu,R.,Alston,A.M.2002.Chemistryofinorganicarsenicinsoils:II.Effectof527phosphorus,sodium,andcalciumonarsenicsorption.J.Environ.Qual.31(2),557–563.528

Stachowicz,M.,Hiemstra,T.,vanRiemsdijk,W.H.2008.Multi-competitiveinteractionof529As(III)andAs(V)oxyanionswithCa2+,Mg2+,PO4

3-,andCO32-ionsongoethite.J.Colloid530

InterfaceSci.320(2),400–414.531

22

Sullivan,K.A.,Aller,R.C.1996.DiageneticcyclingofarsenicinAmazonshelfsediments.532Geochim.Cosmochim.Acta60,1465–1477.533

Williams,P.N.,Islam,S.,Islam,R.,Jahiruddin,M.,Adomako,E.,Soliaman,A.R.M.,Rahman,534G.K.M.M.,Lu,Y.,Deacon,C.,Zhu,Y.-G.,Meharg,A.A.2009.Arseniclimitstracemineral535nutrition(Selenium,Zinc,andNickel)inBangladeshricegrain.Environ.Sci.Technol.43,5368430–8436.537

Williams,P.N.,Zhang,H.,Davison,W.,Meharg,A.A.,Hossain,M.,Norton,G.J.,Brammer,H.,538Islam,M.R.2011.OrganicMatter—SolidPhaseInteractionsAreCriticalforPredicting539ArsenicReleaseandPlantUptakeinBangladeshPaddySoils.Environ.Sci.Technol.45,5406080–6087.541

Winkel,L.,Berg,M.,Amini,M.,Hug,S.J.,Johnson,C.A.2008.Predictinggroundwaterarsenic542contaminationinSoutheastAsiafromsurfaceparameters.Nat.Geosci.1,536–542.543

23

544

Fig.1.Samplinglocationsgroupedper10km2andsamplelocationmarkerscaledtosize545foraveragearseniccontentofthatlocationforsurfacesoils.Theunderlyingcontourmap546isforgroundwaterarsenicwithdatainputtedfromtheBGS/DPHE(2001)arsenicsurvey.547548

549

550551

24

552

Fig.2.Boxandwhiskerplotshowingconcentrationsofarsenicinpaddysoilsirrigatedwith553groundwaterandsurfacewater.Theboxplotsindicatethelowerandupperquartile(box),554themedian(solidline),themean(dashedline),the10thand90thpercentiles(whiskers)555andthe5thand95thpercentiles(circles).556557558

25

559

Fig.3:Arsenicconcentrationsinthepaddysoilsfromdifferentphysiographicregions.The560numbersofsamples(n)ateachofthephysiographicregionsaregivenwithinthe561parentheses.Tukey’sposthocanalysiswasperformedwithone-wayanalysisofvarianceto562comparepair-wisethemeansofarsenicconcentrationsateachofthephysiographicregions563toshowwhichregionshadsignificantdifferencesinsoilarsenic.Regionsthatsharethe564sameletter(A–E)arenotsignificantlydifferent.ThelettersindicateTukeygroupingsforthe565physiographicregionswithrespecttotheirmeansoilarsenicconcentrations.566567568569570571572573574575576577578579580

26

581Fig.4.Relationshipsbetweenarsenicinpaddyandnon-paddysoilsfromdifferent582physiographicregionsofBangladesh.Theregressionlineineachgraphistheregression583lineforallthedata.ThefitandlineequationsaregivenintableS2.584585

27

586

Fig.5.Paddyversusnon-paddyelementalrelationshipswithsoilsclassifiedasHolocene587andPleistocene.Thelineoneachofthegraphsistheregressionlineforeachofthe588elements.ThefitandlineequationsaregivenintableS2.589

28

590Fig.6.RelationshipsforarsenicversuselementsforpaddysoilsgroupedintoHolocene591andPleistocene.Thelineoneachofthegraphsistheregressionlineforthecorresponding592elements.ThefitandlineequationsaregivenintableS3. 593

29

594

595Fig.7.PCAofpaddysoilsclassifiedintoHolocenefloodplainandPleistoceneterracealong596withloadingplot.Thefirstandsecondcomponentcontributed54.7and10.3percent,597respectively,tothevariations.598