I8211EN/1/11.17

ISBN 978-92-5-130050-3

9 7 8 9 2 5 1 3 0 0 5 0 3

Useof137Csforsoilerosionassessment

EmilFULAJTAR,LionelMABIT,ChrisS.RENSCHLER,AmeliaLEEZHIYI

SoilandWaterManagement&CropNutritionSection,JointFAO/IAEADivisionofNuclearTechniquesinFoodandAgriculture,InternationalAtomicEnergyAgency,Vienna,Austria

FoodandAgricultureOrganizationoftheUnitedNationsInternationalAtomicEnergyAgency

Rome,2017

RecommendedcitationFAO/IAEA.2017.Useof137Csforsoilerosionassessment.Fulajtar,E.,Mabit,L.,Renschler,C.S.,LeeZhiYi,A.,FoodandAgricultureOrganizationoftheUnitedNations,Rome,Italy.64p.

The designations employed and the presentation of material in this information product do not imply the expression of any opinion whatsoever on the part of the Food and Agriculture Organization of the United Nations (FAO) concerning the legal or development status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. The mention of specific companies or products of manufacturers, whether or not these have been patented, does not imply that these have been endorsed or recommended by FAO in preference to others of a similar nature that are not mentioned. The views expressed in this information product are those of the author(s) and do not necessarily reflect the views or policies of FAO. ISBN 978-92-5-130050-3 © FAO, 2017 FAO encourages the use, reproduction and dissemination of material in this information product. Except where otherwise indicated, material may be copied, downloaded and printed for private study, research and teaching purposes, or for use in non-commercial products or services, provided that appropriate acknowledgement of FAO as the source and copyright holder is given and that FAO’s endorsement of users’ views, products or services is not implied in any way. All requests for translation and adaptation rights and for resale and other commercial use rights should be made via www.fao.org/contact-us/licence-request or addressed to copyright@fao.org. FAO information products are available on the FAO website (www.fao.org/publications) and can be purchased through publications-sales@fao.org. Photo cover: Mountainous agriculture in Uganda (Kigwa, Kabale District) © FAO-IAEA/Emil Fulajtar

Contents

Foreword vExecutivesummary vii1. Introduction` 12. Purposeandmethodsmeasuringsoilerosion 33. Principlesof137Csmethod 94. Siteselectionandsamplingstrategy 175. Analysing137Csdata:gammaspectroscopy 276. Conversionof137Csdatatosoillossvalues 317. Dataanalysis,interpretationandpresentation 358. 137Csmethod–validationanduseofsoilerosionmodels 419. Advantagesof137Csmethod 5110. ActivitiesperformedbytheJointFAO/IAEADivisionandfurtherperspectives 53Glossary 54References 57Listoffull-pagephotos 63

v

Foreword

Soilisthefundamentalnaturalresourceforhumankindandmanylivingorganisms.Itisthebasis for foodproductionandhasseveralenvironmental functions. Itplaysakeyrole inwater,nutrientandcarboncyclesandservesasanenvironment for floraandfaunainandabovethesoil.

Soil is a fragile resource and its development from weathered rocks and sedimentsunder the effect ofwater, air and living organisms requires time. The soil formationrate usually does not exceed 0.1 mm of soil layer per year and to form fertile soilsuitable for agriculture often requires several hundred or even thousand years. Incontrast the lossofsoilcanbeaquickprocess,especially ifbarrensoil isexposedtoerosionbywaterorwind.

Combatingsoilerosionrequiresinvestmentand,duetothelimitedresources,itshouldbe targeted to critical areas and time period during the most vulnerable season.Comprehensive knowledgeof spatial and temporal variability of erosionprocesses isurgently needed. Gaining reliable information is, however, challenging. Conventionalmethodsformeasuringsoilerosionarelabourintensiveandtimeconsuming,anddataneed to cover several decades to get a good representation ofmean erosion rates.Furthermore,mostconventionalmethods(exceptforgeodeticmethod)donotprovideinformationonthespatialdistributionoferosion.

Isotopetracerscanhelp tomeet thesedeficienciesassomeradionuclidesandstableisotopesoccurring in theenvironmentcanserveasenvironmental tracersandhencefacilitatetheinvestigationoftheselandscapeprocesses.Thesoilerosionratescanbeestimated using 137Cs, a human-induced radionuclide of caesium released into theatmosphere during nuclear weapon tests more than half a century ago. The 137Csmethod for soil erosion assessment effectively provides long-term mean soilredistributionrates,representingtheperiodsinceitsrelease(mid-1950s)untilthetimeofthe137Cssampling.Suchinformationonspatialandtemporaldistributionoferosionis crucial also for soil erosion modelling and soil conservation programmes andprovides essential information towards several strategic objectives of the Food andAgricultureOrganizationoftheUnitedNations(FAO).

This publication provides a brief overview of the 137Cs method and explains theprinciples and strengths of its applications in research towards climate-smartagriculture. It addresses a wide audience encompassing scientists and students,environmental specialists, agricultural managers and decision makers, farmers andindividualswhoareinterestedinsoilconservationandsustainablelandmanagement.ItsummarizestheexperienceofresearchactivitiescarriedoutbytheJointFAO/IAEADivisionofNuclearTechniquesinFoodandAgriculture.

vi

vii

ExecutivesummarySoil erosion bywater,wind and tillage are among themost common and importantlanddegradationprocesses,withbothon-siteandoff-site impacts. Theyaffectmoreland than all other degradation processes put together. A total of 75 billion tons offertilesoilisremovedeveryyearfromglobalsoilscapebyerosion.Asaresult,precioussoil resources, which should be preserved for next generations, are continuouslyreduced. Every year approximately 12 million ha of land is lost. It is thereforeimportant that erosion research is conducted to assess the soil redistribution rates,theirspatialdistributionandtemporaldynamicsinordertocounteractthisprocess.

The measurement of soil redistribution is not an easy task because soil erosion iscaused by a number of processes running at different temporal and spatial scales.Conventional erosion measurement methods, such as erosion plots, volumetric,hydrological,andgeodeticmethods,areusedfordifferenterosionprocessesandtheycoverdifferent spatial and temporal scales.Most conventionalmethodshave severelimitations. They are associated with point data (measurement profiles) and do notprovide information on spatial distribution of erosion. However the majordisadvantage is that they are labour-intensive and require long monitoring periods.Thesedisadvantagescanbeovercomebyusing137Csasanerosiontracer.

Theprincipleofthe137CsmethodforsoilerosionassessmentisbasedonthechemicalcharacteristicsofCaesium.The137Cs isahuman-inducedenvironmental radionuclide,released into the atmosphere by nuclear weapons testing in the 1950s and 1960s,wherebytheradionuclidespreadtothestratosphereandgraduallydescendedtotheland surface. In addition, smaller regional contaminations were caused by nuclearpowerplantaccidents(suchastheChernobylandFukushima-Daiichiaccident).When137Csgetsintocontactwithsoilmaterial,itbindsfirmlytosoilcolloidsandisoftennottransferredbyprocessessuchas leachingorplantuptake. Itcanmoveonly togetherwith soil particles and this means that any change in 137Cs inventories indicate theoccurrenceofprocessesofsoilredistributionbyphysicalagents(e.g.soilerosion).

The concept of reference sites was established to express loss and accumulationprocessesquantitatively.Areferencesiteisanundisturbedsitewhereneithererosionnor sedimentation occurs, so that the 137Cs inventory represents the original falloutreducedonlybyradioactivedecay.Theprincipleofthemethodisbasedoncomparisonofstudiedandreferencesites.Ifthestudiedsitecontainsless137Csthanthereferencesite, it implies that the studied site is eroded. If its 137Cs inventory is greater, it isaffectedbysedimentation.Thissimplerelationisinterpretedbyconversionmodelsinorder to convert the differentiated 137Cs inventories into soil erosion andsedimentationrates.

Forsuccessfulinterpretationofthe137Csmethod,properselectionofthereferencesiteiscrucial.The137Cssamplingshouldstartwithdepthincrementalsamplingperformedto determine the depth distribution of 137Cs contamination; subsequent bulk coresamplesshouldthenbecollectedfollowinggridormultipletransectdesigns.

viii

Theanalysingof137Cscontentinthesoilisdonebygammaspectroscopy.137Cscanbeeasilymeasuredbygammaspectroscopybecauseitprovidesastrongpeak(at662keVenergy) that is well identifiable in the gamma ray energy spectrum. After datacollection(137Csinventoriesandcalculatedsoilerosionrates),dataprocessingandtheinterpretation of data follows. The obtained data can be used for various purposes,such as characterizing the erosion over a range of environments or land uses;estimatingtheimpactoflandmanagementandcroprotationonthesoilerosion;andassessing the efficiency of particular soil conservation measures. This publicationdescribes the above steps, including data interpretation. It also describes the use of137Cs incombinationwitherosionmodelling.The137Csderivederosiondatahasgreatpotentialasatoolforvalidationoferosionmodels.

The primary purpose of erosion assessment with 137Cs method is to provideinformation needed for identification of erosion hot spot areas and selection ofconservation measures. This helps to implement the soil conservation programmesandthus137Csmethodcontributestomaintainingfoodsecurity.

1

1 IntroductionThe suggestion to use caesium-137 (137Cs) as a tracer for soil erosion assessmentoriginatedinthe1960swhenenvironmentalscientistsinvestigatedlandcontaminationby radionuclides released during nuclear weapons tests. It was found that someradionuclidesreleasedbythesetestscouldbeusedastracerstoinvestigatelandscapeprocesses,suchaswatercirculationandtransportofsoluble/insolublecompounds.

Initial studies on erosion assessment focused on strontium-90 (90Sr). Menzel (1960)noticedthatthelossof90Sratslopingreliefpositionscanbeattributedtoerosion.As90Sr is challenging toanalyse, theattentionsoon turned to 137Cs.Graham (1963)andFrere andRoberts (1963) began the investigations of 90Sr and 137Cs redistribution bysoilerosion.RorowskiandTamura (1965,1970a,b) started touse 137Csasartificiallyadded tracers to investigate correlation between soil erosion or soil loss and 137Csinventories at experimental plots. The 137Cs method underwent significantdevelopment and became a well-established method for soil erosion quantitativeassessmentinthe1990sandearly2000s(WallingandQuine,1990,1991,1992,1995;ZupancandMabit,2010;IAEA,2014).

The Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture plays animportantroleinthedevelopmentofthe137Csmethod.Itwasestablishedin1964andrepresentsastrategicpartnershipbetweenFAOandIAEA,mobilizingtheresourcesofbothorganizations topromote theapplicationofnuclear scienceand technology forsupportingthesustainableagricultureandsecurefoodproduction.

In 1995 the Joint FAO/IAEA Division launched its first coordinated research project(CRP) on the ‘Assessment of soil erosion through the use of 137Cs and relatedtechniquesasabasisforsoilconservation,sustainableproduction,andenvironmentalprotection (D1.50.05)’. Lateractivities aimedat FRNmethods continuedwith severalother CRPs (‘Conservation Measures for Sustainable Watershed Management UsingFallout Radionuclides, D1.50.08’; ‘Integrated Isotopic Approaches for an Area-widePrecision Conservation to Control the Impacts of Agricultural Practices on LandDegradation and Soil Erosion, D1.20.11’; and, recently, ‘Nuclear Techniques for aBetterUnderstandingoftheImpactofClimateChangeonSoilErosioninUplandAgro-ecosystems, D1.50.17’). The Joint FAO/IAEA Division through the Soil and WaterManagementandCropNutrition (SWMCN)Subprogrammecanbe consideredas theleading international institution inworkinganddevelopingthe137Csmethod.Theuseof137Csfortheassessmentoferosionisbecomingwell-established.Itiswidelyusedforerosion studies to provide key information on sheet and rill erosion rates andsoil/sedimentdynamicsandredistribution.Thisworkissupportedbytheinternationalresearchcommunity.The137Csmethodwasdescribedseveraltimesindetailinseveralhandbooks (WallingandQuine,1993; Zapata,2002;Mabitetal., 2014), the last twobeingbasedontheresultsofthementionedCRPs.

Later investigations identified few other radionuclides for soil erosion assessment.Lead-210 (210Pb) (Walling and He, 1999) and beryllium-7 (7Be) (Walling et al., 2000)broughtgoodresults,andmorerecentlyinnovativestudiestestingplutonium-239+240

2

(239+240Pu)werepublished(Alewelletal.,2014).Alltheseradionuclidesarecommonlytermedasfalloutradionuclides(FRNs)astheyarefallenonlandfromtheatmosphere.After the initialdepositionFRNsprovideabaseline toassess thesoil redistribution–soilerosion, transportanddeposition– innaturalaswellasmanaged (orcultivated)landscapes.

Thefurtherdevelopmentofthe137Csmethodrequiresmorewidespreaddisseminationofknowledgetoscientistsandstudentsspecialisedinsoilerosionresearchaswellastowider audience of specialists e.g. agricultural decision makers and farmers. For thispurpose,thereisademandforashortdocumentthatcanexplaintheprinciplesofthemethod clearly and briefly and to provide ideas about its applications for variousresearch objectives. This publication attempts to provide the readers with thisinformation.Additionally,itallowsFAOMemberCountriesandIAEAMemberStatestobenefit from the use of 137Cs method for sustainable land management and soilconservation.

3

2 Purposeandmethodsmeasuringsoilerosion

2.1SoilerosionaslandscapeprocessLandscapeprocessesofair,water,rocks,soil,nutrients,pollutantsandlivingorganismredistributionandtransformationaretheresultsofcomplexinteractionofavarietyofindividual processes from very dynamic processes such as air andwater circulation,throughmedium-termprocesses,forexampletheseasonaldynamicsofvegetation,uptolong-lastingprocessessuchastectonicmovementandrockweathering.Thestudyofthemagnitudes,spatialandtemporaldistributionoftheseprocessesisoneofthebasicobjectivesofEarthSystemSciences.

Soilerosionisoneofmostcommonandmostimportantgroupoflandscapeprocesseswithanon-siteandoff-siteimpact.Itiscloselyconnectedwiththewatercycleandthecirculationofsolublebasicnutrients,traceelementsandpollutants.Severalparticularerosion processes participate in soil redistribution.Most important among them arewatererosion(sheet,rillandgully)andwinderosion(Figure2.1).

Figure2.1.Examplesofthemainerosionprocesses:sheet,rill,gullyandwinderosion.

Erosionprocesses showawidegeographicaldistribution (Figure2.2)andaffect largeportionof landespeciallycultivatedlandandpasturesoccurringinaridandsemi-aridareas.Erosionaffectsmorelandthanallotherdegradationprocessestogether.Whilemost of the latter (e.g. contamination, compaction, salinization, etc.) result in

4

deterioration of some soil properties, soil erosion result in complete removal of thewholeupper,oftenmostfertilesoillayerandredistributesitdown-slopeinlandscapesassediments.Thereforetheprogrammesfordevelopmentofsustainablelanduseforagricultureandforestryhavegreatimportanceforsecuringtheoveralldevelopmentofsociety,sustainablelandexploitationandfoodsecurity.

Figure2.2.Geographicaldistributionoflanddegradation(UNEP,1997).

The most basic task for erosion research is to assess soil redistribution rates, theirspatial distribution, and temporal dynamics. These data are serving as primaryinformation for all further investigations of erosion, its mechanisms, modelling ofscenarios,predictions,andanyotherappliedactivitiesaimedatsoilconservation.

Thereare fivemajor thematic groups requiringquantitativedataon soil erosionandsedimentationrates:

• Basicresearch(understandingofthemechanismoferosionprocesses);

• Identificationofcriticalareasforlandconservation;

• Testingoftheefficiencyofsoilconservationmeasures;

• Derivingtheinputparametersformodels;and

• Modelcalibration,validationandimplementationfordecision-making.

5

2.2Soilerosionmeasurementmethods

The measurements of soil redistribution is not an easy task because soil erosion iscausedbyseveralagents(mainlywater,wind,animalsandhuman),whichcaninteractandare responsible forawide rangeofprocesses runningatdifferent temporal andspatialscales.Watererosionsuchassplash,sheet(interrill),rill,gullyandlateralfluvialerosion (e.g. river bank erosion) are stages of erosion causedby surface runoff fromsmallertolargerrunoffvolumes(RenschlerandHarbor,2002).Transportofsedimentsinwater(andalsowind)occursas:creep,saltationandsuspension(Figure2.3).

Figure2.3.Timeandspaceextentofatmospheric,topographic,soilandvegetationphenomenonimportantfordominantsoilerosionprocesses.Themanagementunitsindicateextentofhumaninterestandimpact

(RenschlerandHarbor,2002).

Most important processes affecting cultivated land are sheet and rill erosion. Ingrassland, important impactsmayhaverillandgullyerosionandinaridareasrillandgullyerosionoftencombineswithwinderosion.

Differenterosionprocessesnotonlyresult informationofdifferenterosionfeatures,butrequirealsovariousapproachesofmeasurement(Laletal.,1994).Forthisreasonvariousapproachesoferosionmeasurementsweredeveloped.

Themostcommonarethefollowingmethods:

• Volumetricmethods:thevolumeofrillsandgulliesismeasuredeitherbysimpleinstruments(measuringtape,framewithprofilepins)orbyscanners;

6

• Erosionplots:asmallpartofslopeisfencedtoformseparatedrectangularplot,protectedfromrunoffgeneratedinsurroundings(Figure2.4).Atthelowestendoftheplot,therunoffgeneratedwithintheplotandthetransportedsedimentis trappedby furrowandderived to someof following collectingor recordingdevices:o Thetankcollectingandstoring thewholevolumeofwaterandsediment

(totalcollectionmethod);o Thedevicesplittingrunoffandsedimenttoseveralequalparts(multi-slot

divisor) among which only one is further collected (thus only a smallfractionoftherunoffandsedimentiscollectedandstoredinordertosavespaceandlabourforsedimentprocessing);

o The dynamic devices (filled and emptied) measuring the discharge andsedimentconcentration(tippingbuckets,Coshoctonwheels);

Figure2.4.Exampleoferosionplotswithtotalcollectionofsediment(Osikov,Slovakia).

• Hydrologicalmethods:thesedimenttransportedbywatercoursescanbeeitherfloatinginthewater(insuspension)orrollingontheriverbed(asbedload).Foreachofthesetwogroupsofsediments,differentmethodsareneededastheyarecontrolledbydifferentmovementprinciples:

o Thedeterminationofsuspendedsedimentloadisbasedonmeasurementsof water discharges and concentration of suspended sediments. Thedischarge (l s-1 in small water courses andm-3 s-1 in rivers) is measuredusing the standard hydrological approach measuring the mean flowvelocity in known hydrological profile, which are either the selectedprofilesinnaturalbeds(inrivers),orartificialbuiltuphydrologicalprofilessuch as sheet-iron Parshall flumes (Figure 2.5) used for very small linear

7

flowsandstonyorconcretehydrologicalprofilesbuiltinriverbedsofsmallwatercourses.Thesuspendedsedimentload(expressedasconcentrationof insoluble suspended soil material in g l-1) is determined in watersamples collected periodically using either hand sampling or automaticsamplers.

o The bed load is measured with the aid of various mechanical bed loadtrapsorbedloadsamplers.

Figure2.5.ExampleofParshallflumeusedformeasurementsofsuspendedsediment(Lukacovce,Slovakia).

As soil erosion is a complex phenomenon to investigate, each assessment ormeasurementmethodhasitsconceptualandtechnicaladvantagesandlimitationsandprovides a different picture of the erosion rates. Different methods are mutuallycomplementaryandtheircombinationimprovestheoverallunderstandingoferosion.Theselectionofaparticularmethodortheireventualcombinationshouldreflectthepurposeofthestudy.Thespecificadvantagesanddisadvantagesofthe137CsmethodincomparisontoconventionalmethodswillbediscussedinChapter9.

8

9

3Principlesof137Csmethod

3.1Originandbasiccharacteristicsof137CsCaesium(Cs)istheheaviestalkalimetal(atomicnumberof133),thatoccursinnature.Alkalimetalsareknownasthemostelectropositiveelementsofperiodictableandaretherefore very reactive. Caesium is rare component of rocks and does not have animportantroleinsoilandlifeprocesses.Whenoccurringinsoil,itispresentedmostlyaspositivelychargedcationsboundedtomostlynegativelychargedclaymineralsandorganicmatter(soilcolloids).

Thenatural stable isotopeofCs is 133Cs.Throughhumanprocessesawhole rangeofartificial radioactive isotopes (with atomic mass varying from 125 to 145) can becreated.Themostimportantare134Cs(half-lifeoft1/2=2.06years)and

137Cs(t1/2=30.17years). The 134Cs resulted from past fallout from nuclear weapon tests and nuclearpowerplant(NPP)accidentssuchasChernobylandFukushima-Daiichi.Becauseof itsshorthalf-lifethe134Csisotopeisnotrelevantforsoilerosionassessment.

The137CswasformedduringnuclearbombtestsaswellasNPPaccidentsasaproductofUraniumdecay. It furtherdecays resulting in the formationof 137Ba that is a finalproduct of this decay chain. The decaying 137Cs emits gamma-rays of high energy(662.66Kev).Becauseofitsrelativelylonghalf-life(30.17years),the137Csisanoptimalerosion tracer. This is an important advantage of 137Cs as compared to many otherenvironmentalradionuclides.Itsmeasurementinenvironmentalsamplesusinggammaspectroscopy is relatively easy and accurate without the need of special chemicalseparation(RitchieandMcHenry,1990).

In summary, the main sources of the 137Cs present in the environment were theatmospheric nuclear weapon tests (bomb-derived 137Cs, or bomb 137Cs) and NPPaccidents among which the most important was Chernobyl accident; therefore thetermChernobyl 137Cs is used (Carter andMoghissi, 1977;Wise,1980) (Figure3.1). In2011,anotherreleaseof137CswascausedbytheFukushimaDaiichiNPPaccident,butthis137CsfalloutaffectedmostlyJapanandPacificOceananddoesnothavesignificantimportanceforerosionstudiesaroundtheworld(Miller,2014).

3.2Spatialandtemporaldistributionof137CsfalloutIncontrasttotheNPP-derived137Csinjectedintothefast,dynamicweatherprocessesofthetroposphere,alargerportionofthebomb-derived137CsgotintothemorestablestratosphereandthereforecirculatedaroundtheEarthbeforegraduallyfallingtotheEarth’s surface. The latter fallout was associated primarily with precipitation (wetfallout) and the dry falloutwas important only locally, around the nuclear test sites(RitchieandMcHenry,1990).

The spatial distribution of the bomb-derived 137Cs fallout was determined by acombinationofa)thelocationofthenuclearweaponstestsites,b)theaircirculationandc)therainfalldistributionpattern.Becausemostoftheweapontestswerecarried

10

out in the Northern Hemisphere and the air exchange between the Northern andSouthern hemisphere is limited, the soil contamination by bomb-derived 137Cs isconsiderablyhigherintheNorthernHemisphere.Ineachhemisphere,thedistributionfollows a latitudinal zoning corresponding to the rainfall zones (Sutherland and deJong,1990;Davis,1963;Bernardetal.,1998).

Figure3.1.Originof137Csintheenvironment.

The global pattern of bomb-derived 137Cs fallout indicates that inputs ranged fromabout160 toabout3200Bqm-2dependingon latitude (GarciaAgudo,1998). In theSouthernHemisphere the values rangeup to a fewhundreds,while in theNorthernHemisphere they reach from 1 000 to more than 3000 Bq m-2. These values aresufficientlyhighforgammaspectroscopymeasurements;therefore,the137Csmethodcanbeusedallovertheglobe.The137Cs inventories intheSouthernHemisphereareusuallysufficienttobemeasurableifusingappropriatedetectorsandcountingtimes.However,withtimetheyaregraduallyreducedbytheradioactivedecay.Infuture,theuseofthe137CsmethodinSouthernHemispherewillbecomemoreandmoredifficult.In the last years, the alternative use of 239+240Pu as erosion tracer was successfullytested (Alewell et al., 2014). This new soil tracer will become promising soil lossevaluationtechniqueintheever-decreasing137Csactivityareas.

11

An important parameter used for the calculation of soil redistribution rates is thetemporal dynamic of the 137Cs fallout. Although the first nuclearweapon testsweredone already in 1945, theworldwide detectable 137Cs fallout began in 1954 and thehighestfalloutrecordoccurredin1963.Duringthesecondhalfofthe1960s,thefalloutintensitydecreasedabruptly.Thenduringthe1970s,aslowdecreasingcontinuedandduringthefirsthalfofthe1980s,thefalloutdecreasedtobelowdetectionlimit.

TheChernobyl 137Cs releasedduring theNPPaccident in1986was injectedonly intotroposphereand it furtherspreadoverRussiaandEurope inmuchlowerheightthanthebomb-derived137Cs. Itdidnotpersist inatmosphereas longasthebomb-derived137Cs but it fell out within a fewmonths. The distribution of Chernobyl 137Cs is veryheterogeneousas its falloutwasdeterminedbytherainfalldistributionfromtheendofApril-toMay1986.ThisirregularfalloutmakestheerosionassessmentinChernobylaffectedareasmoredifficult.TheChernobyl137Csfalloutincreasedtheexistingbomb-derived inventories by several orders of magnitude in some European areas (WHO,1986; Anspaugh et al., 1988; Cambray et al., 1987, 1988; Mabit et al., 1999). Forexample inCentralRussia the 137Cs inventoriesafter theChernobyl accident reached500000Bqm-2(Golosovetal.,1999).Anotherexampleoflocal137CsfalloutfromNPPaccidenthappenedduringFukushima-DaiichiNPPaccidentin2011(Fukushima-Daiichi137Cs).

3.3137CsbehaviourinsoilThe early investigations of radionuclide occurrence and transformation in landscape,carriedout in the1960sand1970s,were focusedon theirenvironmental impacts.Anumberofhuman-made radionuclideswere investigated, 137Csbeingamong them. Itwas found out that 137Cs is strongly bound to soil colloids and is in principle non-exchangeable (Davis, 1963; Lomenick and Tamura, 1965; Eyman and Kevern, 1975;Ritchie and McHenry, 1990) and its uptake by plants is negligible (Dahlman et al.,1975).The137Cs,ifabsorbedbythevegetation,isreleasedtosoilafterplantsdieanddecay (Davis, 1963; Rogowski and Tamura, 1970 a,b; Dahlmanet al., 1975). A lot ofattention was paid to the investigation of the 137Cs vertical distribution (see typicalexamplesinFigure3.2).

Theinvestigationoflandundisturbedbycultivationshowedthatthemost137Csoccursin the uppermost layer of soil and its content exponentially decreases with the soildepth.Many researchers reported such 137Cs distribution profiles that are similar tothoseshownintheFigure3.2(WallingandHe,1993;Wallingetal.,1999).

As 137Cs is non-exchangeable, it is not released into soil solution and cannotsignificantlymigrateor takepart in thechemicalprocesses running in soil.However,some limited depth redistribution reaching usually 10 to 20 cm results frombioturbation carriedbyburrowing animals such as earthwormsdrilling up anddownalongthesoilprofile.Thiscausestheslightdiffusionof137Cstosoillayerimmediatelybelowthesoilsurface.Apartfromthat,someotherphysical-chemicalprocessessuchas freezing-thawing or wetting-drying of soils can contribute to dynamics in verticaldiffusionof137Cs.

12

On cultivated land, the redistribution of 137Cs is primarily the result of mechanicalmixing associated with cultivation (e.g. tillage erosion). The 137Cs is homogenouslydistributedoverthewholedepthofploughhorizonandbelowthatanabruptdecreaseof137Cscontentappears(comparewithFigure3.2).

Figure3.2.Typicalverticaldistributionof137Csinsoilprofileofundisturbedandcultivatedsites

(Rabesirananaet.al.,2016).

3.4Generalprinciplesof 137Csmethodused for soilerosionandsedimentationassessment

Exceptinsomeveryacidsandysoils,therelativeimmobilityof137Csinagriculturalsoilsunderusualenvironmental conditions is thebasicconditionenabling itsuseasa soilerosion tracer. The 137Cs method is based on following key assumption about 137Csbaselinedepositionandredistribution:

13

• The 137Cs after (mainly wet) fallout from atmosphere is homogenouslydistributed over the landscape and is strongly bound to soil colloids in theuppermostsoillayer(Figure3.3);

• Itisnotsignificantlyleachedbywater;

• Itdoesnotmigrateasaresultofchemicalprocesses;

• Itsuptakebyplantsisnegligible;and

• It moves only as a result of mechanical processes mobilizing soil particles towhichitisbound.

Figure3.3.Homogenousdistributionofthefallen137Csinareanotaffectedbysoilerosion.

Several mechanical processes of soil redistribution occur in landscape such asbioturbation, erosion, geomorphological processes, or human activities (tillage,digging,mining,construction,etc.).Inmostlandscapesapartfromurbanareas,erosionisbyfarthemostimportantamongthesemechanicalprocesses.The137Csmethodcanprovideanassessmentofallmajorerosionprocessescausedbymostcommonerosionagents such as water, wind and tillage. Most of these processes (splash erosion byraindrops, interrill/sheet erosion through surface water flow, rill erosion by linearrunoff, all processes of wind erosion and tillage erosion) contribute to 137Csredistribution and are represented by erosion rates estimated by the 137Csmethod.Themostadvancedprocessesoflinearwatererosion,whichtransportlargestvolumesof soil and rockmaterial (gully erosion, and fluvial/lateral riverbankerosion) canbe

14

assessedindirectlybymorecomplexmethods,e.g.sedimentfingerprintingusing137Cstogetherwithseveralotherparameters(Walling,2003).

Theprincipleof the soil erosionassessment is basedon the comparisonof soil 137Cscontent at landscape positions affected by the soil redistribution dynamics (erosion,transport and sedimentation) to relatively stable landscape position which is notaffectedbysuchsoil redistributionprocesses.Providedthat 137Cscanbemovedonlytogether with soil particles, then soil enrichment or depletion of 137Cs at somelandscape positions corresponds to removal or input of soil material from thesurroundings(Figure3.4).Toassesshowmuchof137Cswaslostorgained,informationontheoriginal137Csinputisneeded.

Figure3.4.Schemeofthe137Csandsoilredistributionbyerosion:undisturbed,erodedanddepositionsite.

Toobtainsuchinformationaconceptofreferencesitewasintroduced.Areferencesiteisastableundisturbedareawhereneithererosion,noraccumulationtookplaceso itrepresents the initial 137Cs fallout input (only reduced by radioactive decay). Thesestable locations are selected at flat surfaceswherewater runoff and associated soilerosion cannot originate. Most suitable are somewhat elevated landscape positions

15

such as terraces or denudation plateaus. The lowest landscape positions such asalluvialplainsarenot suitablebecause theymaybeaffectedbydeposition.At somealluvialplainstherecentevidencesofinundationmaybelackingbutitisnotsureifandwhenitmightoccurinthepast.Thereforeifthereisnootherchoicethantoselectthereference site at alluvial plain the reliable information on the history of flood is ofhighestimportance.

The137Csinventoriesatstudysiteswheresoilredistributionisexpectedarecomparedtothatofthestablereferencesite.Thelandscapepositionswherethe137Csinventoriesare smaller than at the reference site are interpreted as erodedand thosepositionswhere they are greater are interpreted as deposition sinks. Those with 137Csinventories similar to reference site are either stable have a long-term balancedequilibriumoferosionanddeposition.

An example of 137Cs redistribution (at cultivated land) along the slope transect(comprisingof stable site, erosion site anddeposition site) is provided at Figure 3.5.Thetransectprofilesindicatethereductionof137Cscontentattheslope(ascomparedtostableprofileattheplateau)andincreaseof137Cscontentinvalleybottom.Thesoilerosion and deposition rates can be calculated by conversionmodelswhich expresstherelationbetweenthe137Csredistributionandsoilredistribution(seeChapter6).

Figure3.5.Exampleofthe137Csredistribution(atcultivatedland)alongtheslopetransect,Bohunicesite,Slovakia(Fulajtar,2000).

Thedetaileddescriptionofthe137Csmethodfortheassessmentofsoilredistributionisprovidedby several guidebooks (Walling andQuine, 1993; Zapataet al., 2002; IAEA,2014).Moreinformationanddiscussionontheadvantagesandlimitationsofthe137CsmethodisprovidedbyMabitetal.(2013).Acomprehensivebibliographyontheuseofthe 137Csmethod forvariouspurposeswaspreparedbyRitchieandRitchie (2008).Asimplevideodemonstratingtheprinciplesofthe137Csmethod(FalloutRadionuclidesinSoilErosion)wasrecentlydevelopedbytheJointFAO/IAEADivisionandisavailableathttp://www-naweb.iaea.org/nafa/resources-nafa/Soil-Erosion-web.mp4.

16

17

4Siteselectionandsamplingstrategy

4.1Purposesof137CssamplingThe sampling design is a critical step for the successful implementation of the 137Csmethod in soil erosion and sedimentation studies. The sampling strategy mainlydependson the studyobjectives, thegeographical focus of the studiedareaand theheterogeneityoftherelief,soilscapeandlanduse.

Accordingtostudyobjectives,threetypesofstudiescanberecognised(Zapataetal.,2002):

• Descriptivestudies identifyonlybasiccharacteristicsof thesitesuchasvaluesof 137Cs activity and basic statistical parameters of datasets, while the spatialaspectisnotreflected;

• Analyticalstudiescomparingtwodatasetsortwostudiedsites;and

• Spatialdistributionstudiesaimedonmappingofspatialdistributionoferosionandsedimentationandcalculatingspatiallyrelatedstatisticalparameters(grossandneterosionanddepositionrates).

Accordingtothegeographicalfocus,thesefourtypesofsoilredistributionstudiescanberecognised:

• Plot,fieldorhillslopestudies;

• Floodplainstudies;

• Catchment(orwatershed)studies;and

• Reservoirsedimentationstudies.

Themoststudiesareaimedatmappingoferosionredistributionatfield(orhillslope)scale. Inorder toevaluatea sedimenthillslopebudget, they includea reference sitewithnearlynoerosion,asedimentsourcearea(erosion>deposition),atransportarea(erosion=deposition)andasinkarea(erosion<deposition).

The investigations of reference sites can be classified as descriptive studies. In thischapteronlysamplingstrategiesforreferencesiteandsoilredistributionatfieldscalewillbepresentedanddiscussed.

4.2Studysiteselection,scaleanddatacollectionFor each intended study, amain objectivemust be formulated. Before selecting thestudy site, it should be investigated whether the erosion is active there. Erosionphenomena (for example rills and gullies, sink depressions, soil particles sorted bysheet runoff, etc.) can be observed and the farmers, municipal representatives orother local stakeholders familiarwith the area should be interviewed. It is useful toselectanareawherethebasicclimatedataareaccessible,somesoilsurveyhasalreadybeenperformed,thetopographicaldataforDTMareavailableandlandusehistoryisknown.Theselectedstudysiteshouldnotbetoolargetolimittheheterogeneityand

18

complexity of processes. Its size should correspond to research budget and staffnumberavailabilitytoensureappropriatesampledensity.Laterthestudyareacanbeextended.

The collection of environmental andmanagement data should be done like for anyother fieldpedologicalstudy(suchassoilsurvey, landmanagementorcropnutritionfield experiments, etc.). Available pedological literature (including unpublishedmanuscriptsandgreyliterature)relatedtotheareaofinterestshouldbeinvestigated.Interviewing of land managers or stakeholders (e.g. landowners or agricultural andmunicipalemployees) intheareaaboutcurrentandhistoricextremeweathereventsand land management practices is highly recommended. They can provide keyinformationon landuse/coverchanges, landmanagement, climate,unusualweatherevents, erosion features, landslides, etc. The data collection should be dedicated atthose pedological, climatic, land use and socio-economic data that are related toerosionandsedimentationprocesses.

Reconnaissance survey of the selected study area should 1) verify and interpret thecollected background information; 2) complementmissing geographical, climate, soiland land use information; 3) assess to which extend the area was/is disturbed byhumanactivities,and4) identifypotential referencesites.Especially the latter task isveryimportant.Ifanappropriatereferencesiteisnotavailablenearbythestudysite,itshouldbeconsideredtoselectotherstudysite locations inareawherethereferencesitewouldbeavailable. Especially in areas affectedbyChernobyl 137Cs the referenceandstudysitesshouldbeveryclosetoeachotherbecausethefalloutheterogeneityofChernobyl137Csismuchhigherthanthatofbomb137Cs.

4.3SelectionofundisturbedreferencesiteThe reference site should be selected at flat surfaces where neither erosion norsedimentationtookplacesincethebeginningof137Csfalloutinearly1950s.Thesesitesshouldbepreferentiallyat leastslightlyelevates (plateaus)asthe lowelevatedareas(suchasalluvialplains)couldbeaffectedbyfloodandsedimentation.

The reference site should be as close to the study site as possible. According to theoriginal concept of the 137Cs method the reference sites have to be selected ingrassland(Zapataetal.,2002;Mabitetal.,2014).Thegrasslandwasunderstoodasthemostappropriate landuse typeas it is considered tobeundisturbedandsufficientlyhomogenous.

In some areas no undisturbed land suitable for reference sampling is available. Thishappensusuallyinthefollowingenvironments:

• Mountainousareaswithslopingtopography;

• Areasstronglyaffectedbywinderosionincludingsanddunemovements;

• Heavily managed or engineered landscapes such as urbanised, residential,industrial,miningordenselypopulatedmixeduseareaswheremostof land isdisturbedbyvarioushumanactivities.

19

In such cases when reference site is not available the following approaches can beused:

• The reference value can be estimated using information on 137Cs fallout. Insome countries, themonitoring of radionuclide fallout has been recorded. Inthose countries where the data on fallout distribution are not available, the137Cs fallout can be estimated using the yearly precipitation amount and thelatitudeof thestudysite.Severalequationsdirectly relating137CsatmosphericfallouttotheaverageannualprecipitationofstudyareahavebeendevelopedinCanada,EuropeandAustralia(Bernardetal.,1998;Basher,2000);

• TheinitialfalloutcanbeevaluatedusingthesoftwarepackageofWallingetal.(2002). The parameters needed are: longitude, latitude and the annualprecipitation of the area under investigation. However, such estimation offallout is not possible for countries affected by the Chernobyl or other NPPaccidentfalloutassuchfalloutwasveryheterogeneousanddidnotdependonthelongtermglobalaircirculation;

• Adopting the reference values from studies carried in surrounding or nearbyareaswithsimilarconditions.

Ifnoinformationoninitial137Csinputcanbegained,the137Csmethodcannotprovidequantitativeestimatesofsoilerosionratesbutitcanbeusedforrelativeorqualitativecharacterizationofoveralltrendsinsoildistributionwithinastudyarea.Anexampleofsuch study in mountainous area affected by Chernobyl fallout was provided byFroehlichetal.(1993).

4.4ReferencesitedatacollectionWhen starting any soil sampling for 137Cs determination, the thickness of 137Cscontaminated layer should be determined to ensure that the whole 137Cs inventorywould be involved in the samples. This is achieved through depth incrementalsampling.Thedepthincrementalsamplingenablesalsotoassessthe137Csdistributionprofile.

According to this profile, it can be distinguished whether the soil was disturbed byhumanactivitiesor affectedbyerosionanddeposition.At theundisturbed land, the137Csshouldbeconcentratedwithintheupperfewcentimetresofthesoilprofileandthe decrease with soil depth should be exponential. At the cultivated site the 137Csactivity is homogeneous throughout the whole thickness of ploughed horizon andbelowitthe137Csconcentrationdropsabruptly.

Whenthedepthdistributionof137Csisknown,thebulkcoresamplescanbetaken.Thedepthofcoresamplesshouldexceedtheactualdepthof137Cscontaminatedlayerby10 cm (or at least 5 cm), to ensure that all 137Cs is included in the sample. Thereference site should be sampled along a regular grid. It is recommended that eachsampling point should be represented by three replicate cores,which are bulked toconstituteonesampleforanalysis.Thenumberofsamplesdependsonthevariabilityof 137Cs inventories at the reference site. It depends on random variability of soil

20

properties such as soil bulk density, infiltration capacity, cracking and stoniness, theeffects of vegetation cover and roots, etc. It is recommended that the variationcoefficientofreferencesamplesetsdoesnotexceed30percent(Shutterland,1996).

4.5SamplingstrategyatfieldscaleThe sampling density depends on the study area, relief variability and the budgetavailable.Threesamplingdesignstrategiescanbeadopted(Figure4.1):

• Individualtransect(A);

• Multipletransects(B);and

• Regulargrid(C).

Figure4.1.Basicsamplingdesigns(Mabitetal.,2014).

Theindividualtransectapproachisseldomused.Itisusefulforreconnaissancesurveysor in other cases of preliminary investigations and other situations when morecomprehensive sampling is not possible (for time or budget constraints, etc.). Themultiple transectapproach isbasedontheassumptionthat thevariabilityoferosionanddepositionprocesses is loweracrosstheslopesthanalongtheslopes.Therefore,thedensityofsamplingpointsacrosstheslopedoesnotneedtobeasdenseasalong

21

the slopes. This approach allows a more rational use of resources than the gridapproachand isusefulespeciallyonsteep,shortandhomogeneousslopeswherenosignificant across-slope curvature exists. The number of samples along transectdependsontheslopelengthandshape.Thenumberoftransectsdependsontheslopewidth,butgenerallyat least threetransectsshouldbesampled.Thegridapproach isused when the topography is more complex (e.g. slopes with significant planarcurvature).

Morecomplexsamplingdesigns (suchas fan-likemultiple transectdesign, composeddesigns involving both regular grids and multiple transects or regular grids withdifferentdensities)canbeusediftherelieforstudyobjectivesrequireamorecomplexpictureon137Csdistribution.

4.6SamplecollectionandsamplingtoolsTwomethodsarecommonlyusedtocollectsoilsamples:

• Bulksampling;

• Depthincrementalsampling.Bulksamplescanbecollectedbysteelcylinderthat is insertedtosoilthroughoutthewholedepthof 137Cscontaminated layer.Thisdepthshouldbe investigatedbydepthincrementalsamplingthatshouldbedonepriortobulksampling.Thediameterofthetube is usually 7 to 10 cm (Wallbrink et al., 2002) with a wall thickness of 5 mmadaptedto investigatestonyandcompactedsoil.Tofacilitatethecoreextrusion,thecuttingedgeof thetubeshouldhaveasmaller internaldiameter thanthetube itself(Walling and Quine, 1993). The tube can be inserted manually by hammering ormechanicallyifusingamotorizedsoilcolumncylinderaugerset(Figure4.2).

The depth incremental sampling helps to provide information on the 137Cs depthdistribution (vertical distribution over the soil profile). This is a key knowledgerequestedatthereferencesiteandalsoatthestudysiteswithuncultivatedlandastheconversionmodelsforuncultivatedlandarebasedon137Csdepthdistribution.

The depth incremental sampling requires special devices designed for this purpose,which are able to collect thin layers of soil (1, 2 ormaximum5 cm thick). Themostcommon tool for depth incremental sampling is the scraper plate (Campbell et al.,1988; Loughran et al., 1992; Walling and Quine, 1993; Loughran et al., 2002). Itcomprisesametalframethatisfixedonthesoilsurfaceandthemetalplatewhichcanmovewithintheframe.Thesoillayerscanbescrapedsuccessivelybythismetalplate.Theplatecanbeshiftedsuccessivelyfollowingtheselecteddepthintervals.

More precisely the soil layers can be cut by the Fine Increment Soil Collector (FISC)developed by the SWMCN Laboratory (Mabit et al., 2014). This device can cut thinlayerswithinafewmmthick.Ifspecialsamplingtoolsforincrementalsamplingarenotavailable,theincrementalsamplescanbeobtainedthroughslicingthebulkcores.

22

Figure4.2.Soilsamplescollectionusingamotorizedsoilcolumncylinderauger.

4.7CollectionofadditionalbackgroundinformationThe important part of the erosion investigation is to study and collect data onenvironmental or management processes that impact the erosion factors andtherefore,erosionratesandspatialsoilredistribution.Erosioniscontrolledespeciallyby rainfall, soil properties, topography, and vegetation cover. At cultivated land theland management and human activities are other factors to be considered. Theinterpretation of the 137Cs data and erosion rates calculated from 137Cs inventoriesrequiresinformationonselectedcharacteristicsoferosionfactors.Variousapproachescanbeusedtocoverthisitem.Thefollowingminimaldatasetisrecommended:

23

Topographicaldata:

• Geographical coordinates: latitude, longitude and altitude should be recordedby usingGlobal Positioning Systems (GPS) or ideally byGPS-enabled samplingdevices.

• GPS-indexed photographs of study and each sampling locations: the detailedphotographicaldocumentationisextremelyvaluable.Itcanhelptoillustratethelocationsinadditiontoanyquantitativedata(Figure4.3).Itisrecommendedtomakephotographsofeachlocationfromvariousanglesandifpossiblefromanelevated viewpoint before and after sampling (use of flags). In addition thedetailedphotographsofeachparticulartransectorsamplingpointisuseful.

• Slopeinclination:Thefieldmeasurementsoftheinclinationofeachormultipleslope segments between particular sampling points within transect arerequired.Exceptofuniformslopesegments, it isnotsufficienttocalculatetheinclination from GPS data of variable slope segments as this approach mayresultinsignificanterror(dependingontheaccuracyoftheGPSused).

• Slope length: The field measurements of the distances between particularsampling points within transect are required. One may use distancemeasurement between sample locations and/or slope break points usingmeasurement tapes or distance lasers. As for slope inclination, its calculationusing slope distances and altitudes determined by GPS only is notrecommended.

Figure4.3.GPS-referencedphotodocumentationofsamplelocation.

24

Climate/Weatherdata:

• Precipitation: For erosive events the most detailed sequence of rainfallintensitiespertimeperiod(year,month,day,hour,minute),eventpluviographs(e.g. rainfall intensity graph recorded with a tipping bucket) or the totalprecipitation amounts per event duration are the most important data sincerainfall intensities determine the amounts of soil being detached. Infiltrationrates will determine surface runoff rates for erosion and transport. Standingwateronsoilwillactasprotectionreducingthedetachmentofsoil.

• Otherclimateparameters:Variousparameterscanbeconsidereddependingonthe purpose of the study, use of models, etc. Maximum and minimum airtemperature during an event will especially help to consider the effects offreezing and thawing of the precipitation and water in/on the soil. Togetherwithdewpointtemperatureandwindvelocitytheinfluenceofevaporationandevapotranspirationwouldhelptoassesstheantecedentsoilmoisturebeforeaprecipitationevent.Thesolarradiationmightalsobehelpfultodeterminethesoilsurfacetemperatureaswellasrelatedparametersforvegetationgrowth.

• Station location(s) and measurement method(s): the spatial and temporalvariabilityofweatherdatawillassisttodeterminethevariabilityofprecipitationand weather events along a hillslope or with a catchment. In the latter casethere might be multiple weather stations (Figure 4.4) available andinterpolation methods for each climate parameter might be required(arithmeticmean,Thiessenpolygons,isohyetalmethodology).

Figure4.4.Weatherstation,Prague,CzechRepublic.

25

Soildata:

• Soil classification: soil sample location and soil name with profile depth anddescription of each horizon will allow evaluating soil surface and subsurfacepropertiesandprocesses.

• Soil parameters for each horizon: colour, soil texture, organic matter, pH,carbonatecontentandrockcontent.Cationexchangecapacityandsoilalbedomight also be recorded. Soil survey instructions can be found in soil surveymanualssuchastheUSSoilsSurveyManual(SoilScienceDivisionStaff,2017).

Landuseandmanagementdata:

• Landuseorcover:historyorcroprotationsor landusessincethe137Csfalloutwill enable toevaluate temporal variabilityof soil erosionprocesses impactedby the vegetation. Besides qualitative measures gathered through simplesurveys,quantitativemeasurementsthroughsatellite,airborneorground-basedremotesensingsuchasleafareaindices(LAI)orbiomassindicatorssuchasthenetprimaryproduction(NPP)canbeveryuseful.QuantitativemeasuressuchascropyieldsforaspecificfieldorprecisionfarmingdataofhigherresolutioncropyieldswhenharvestingwithGPS-cropyieldmonitorsareuseful tocapturethevegetation response to indicate themain components of the local hydrologicwaterbalanceandsoilredistributionprocesses.

• Landmanagement:anydataoncroprotationand landmanagementpracticesthatmayhave impact on infiltration and runoff aswell as small scale surfaceroughnessormicrodepressions(e.g.furrowdepthandwidth)areuseful.Veryimportantistheevaluationoftillageerosionthatmighthaveoccurredovertheyears. Crop rotation sequences and changes in the crop rotation should berecorded if possible. Interviews with land managers and farmers especiallyabout extreme events such as erosion or sedimentation events and theirlocationareveryuseful.Remotesensingmapscancontributetoreconstructionoflandusehistory.

26

27

5Analysing137Csdata:gammaspectroscopy

5.1SamplespreparationThe gamma radiation emitted by environmental radionuclides occurring in soil isusuallymeasuredbylaboratorygammadetectors(Figure5.1)insoilsampleswhicharecollectedinfieldandtransportedtolaboratory.Thesamplepreparationshouldinvolvethefollowingsteps(WallingandQuine,1993;PennockandAppleby,2002):

• Airdryingorovendryingat60°C;

• Weighing;

• Grindingoflargeaggregatestopassthrough2mmmesh;

• Samplehomogenization;

• Sievingthefineearth(0-2mm);

• Weighingbothfineandcoarsefractions;

If thesampleexceedsthequantityrequiredforanalysis,arepresentativesub-sample(e.g. 50 to 1 500 g depending on the counting geometry) of the fine fraction issubmittedforanalysis.

Figure5.1.Laboratorygammadetectorinleadshieldandwithnitrogencooling(CNESTEN,Morocco).

28

However,itispossibletousealsoportablegammadetector(Figure5.2),whichcanbecarried to field and measure the in situ gamma radiation. The advantage of thisapproach is the acquiring of mean activity values representing large area. Forassessmentof soil redistributionprocesses it is better andmoreprecise to interpretthe soil samplesmeasured in laboratory, but the portable detector can be used forreconnaissance surveys to check theoverall rangeof activity levels. It is veryhelpfulespeciallyinChernobylaffectedareaswherethe137Csinventoriesareveryvariable.

Figure5.2.Portablegammadetectorinstalledatselectedstudysite(CNESTEN,Morocco).

5.2BasicprinciplesofgammaspectrometryThe 137Cs is one of the radionuclides easiest analysed by gamma spectroscopy. Itsmeasurement is facilitated by its clear energy peak at relatively high energy (662.66Kev)thatdoesnotinterferewithotherexistingradionuclides.Therefore,137Cscanbemeasuredbystandardgammadetectors,whicharenotsensitivetolowenergies.

As such measurement is not destructive, the soil samples analysed by gammaspectrometry can then be used further to perform other analyses such as basic soilanalyses(e.g.soiltexture,pH,organicmattercontent).

Theprincipleofgammaraydetectionisbasedontheinteractionofgammarayswithgermanium crystal,which emits electric signalswhenbeing exposed to gamma rays.Thesesignalsarecharacteristicsforparticularradionuclidesandarerecognizedbythe

29

analyser and presented as peaks at the screen of computer connected to analyser.Each radionuclide can have one or several characteristic peaks at certain energeticlevel.The137Cs ischaracterizedbywell identifiablepeakatenergyof662keV(Figure5.3). The collected spectral data are converted into activity of gamma radiationexpressed in Bq kg-1. The radionuclide activity per weight unit (mass activity) isconvertedintoinventory(activityperunitarea)expressedinBqm-2.

Themethodologyofgamma radiationdetectionand theequipmentusedunderwentsignificant development since the mid-20th century. Traditionally, gammaspectroscopic analytical set comprises of high purity germanium semiconductordetector (HPGe detector) equipped with amplifier and connected to amultichannelanalyser (MCA) and computer with software for data assessment. The detectorsoperateatverylowtemperaturesandneedefficientcooling.Usuallyliquidnitrogenisusedintheseprocessesbutelectricalcoolingsystemisalsoaviableoption.Theyneedalsoleadshieldingbecausetheradiationofenvironmentalsamplesisverylowandthebackgroundradiationinlaboratoryneedstobeeliminated.

Figure5.3.Exampleofthemain137Cspeakat662keVingammaspectrumofsoilsample.

Operating a gamma spectroscopic analytical set requires an experienced and skilledstaff.Theresultofthemeasurementanditsaccuracydependsonseveralkeyfactorssuch as the sample quantity and its activity, the counting time and the samplegeometry.Agammadetectorshouldbecalibrated(usingspecialmultigammasourcesproducedforthispurpose)andtheassessmentsofthemeasurementerrorsshouldberegularlychecked(usingspecificreferencematerials).

30

31

6Conversionof137Csdatatosoillossvalues

The last key stepof soil erosionanddepositionassessmentwith the 137Csmethod istheconversionof 137Csdatasets intosoilerosionanddeposition rates.Although theamount of eroded soil is directly proportional to 137Cs activity redistribution, thequantitative mathematical expression of this relation is complex. The methodologyunderwentlongdevelopment;severaldifferentapproachesweretestedandvalidated.Theycanbeseparatedintotwogroups:

• Empiricalmodels;

• Theoreticalmodels.

The empirical models are based on calibration of 137Cs activities with erosion ratesmeasuredatexperimentalplots.Thesearestatisticalmodels; therefore, theyrequiretheuseofequationswithparametersthatfittheobservations.Thederivedequationsaresiteandtimespecific;theyexpressthecorrelationbetweenreductionoftotal137CsinventoryX (%)andsoil lossY (tha-1y-1)andusuallytheyhaveaformofexponentialequations such as Y=aXb or Y= aX (with a, b as constants). These equations weredevelopedintheearlystageofthe137Csmethoddevelopment(RitchieandMcHenry,1975;Elliotetal.,1990;LoughranandCampbell,1995).

Thedisadvantageofempirical relationswasovercomebytheoreticalmodels thatarebasedon the logical assumptionsandalgorithms for calculationof soil redistributionrates. Different theoretical assumptions are needed to express the conditions atundisturbedandatcultivatedland.Atundisturbedsitesthe137Csisconcentratednearthesoil surfaceand itdecreasesexponentiallywith thedepth.Atcultivated land the137Cs is distributed homogeneously over the plough horizon and decreases abruptlyimmediately below the lower boundary of ploughed horizon (Figure 6.1). Thisdifferencerequiresdifferentalgorithmsforsoillosscalculation.Therefore,themodelsareusuallygroupedintwodifferentgroups:

• Modelsforundisturbedland;

• Modelsforcultivatedland.

Several models were developed at Exeter University, UK (Walling and Quine, 1990;Walling andHe, 1999) and they represent an integrated set thatwasworkedout asuser-friendly PC software (Figure 6.2). The software package (Excel add-in) can bedownloaded from the website of the SWMCN Subprogramme http://www-naweb.iaea.org/nafa/swmn/models-tool-kits.html. Thehandlingof this IT tool is veryuser-friendly.Thedialoguewindowofferstoselectaparticularmodel.Thenextstepsare to fill the input data to predefined cells and finally it provides estimate of soilredistributionrates.

Thecalculationofsoilerosionanddepositionratesatundisturbedsiteisbasedon137Csinventories and the 137Cs depth distribution. Two approaches are used. The moresimple approach considers the fixed 137Cs fallout input and stable 137Cs profiledistribution.Thedepthdistributionoverthesoilprofileismathematicallydescribed.Ifthisdistributionischaracterizedbysimplemathematicalfunctionthenthesoillosscan

32

beestimatedbytheproportionofthe137Csreferencevalueremovedattheexaminedsite. This approach is used for the Profile Distribution Model (PDM). A morecomprehensive approach takes into consideration time variant processes of: 1) the137Csfalloutand2)thegradualpost-depositionalredistributionof137Cswithinthesoilprofile, which is caused predominantly by bioturbation and several other processes.ThisapproachisusedbytheDiffusionandMigrationModel(DMM).

Figure6.1.Exponentialdepthdistributionof137Csatundisturbed(left)andhomogenousatcultivated(right)

siterequiringdifferentconceptsofconversionmodels.

Thecalculationofsoillossatcultivatedlandcanbebasedontwotheoreticalconcepts.The first one called proportional concept is very simple and it presumes that theremoval of soil and 137Cs are directly proportional. Model based on this concept iscalledtheProportionalModel(PM).

More complex approaches involve a mass balance concept, which considers thetemporal dynamics of 137Cs inputs and outputs resulting in the time-variantconcentrationof 137Cs in soil.Thesechanges inconcentrationaffect considerably therelationbetweenthe137Cslossandsoillosscausedbyerosion.The137Csconcentrationin soil is controlled by several processes. Most important are 1) time variant 137Csfallout,2)radioactivedecayof137Cs(i.e.30.17years),3)removaloffreshlydeposited137Cs by erosion prior to its incorporation into plough horizon by tillage, and 4)incorporationofsubsoilmaterialfreeof137Csorhavinglow137Cscontentintoerodedploughedhorizonbytillage.

TheMassBalanceModels(MBM)weredevelopedinmid-1980s.Differentmodelsusedifferent approaches to handle the 137Cs time-variant concentration in soil andconsider some but not all processes and factors controlling it. More recently threemassbalancemodelsdevelopedbyWallingandHe(1999)areused,e.g.MassBalanceModel1(MBM1),MassBalanceModel2(MBM2)andMassBalanceModel3(MBM3).

33

More information on conversion models is provided by Walling and He (1999) andMabitetal.(2014).

Figure6.2.Exampleoftheuseof137Csconversionmodelsoftware:assessmentoftheoriginal137Csfalloutat

areferencesite.

34

35

7Dataanalysis,interpretationandpresentation

7.1ReferencesiteThevalidationofestimatedinitial137Csfalloutforanundisturbedreferencesiteisakeyrequirement toget reliablevaluesof soil redistribution rates fornearby sample sitesaffectedbyerosionordeposition.

Thefirstcriteriaforreferencesiteselectionareflattopographyandknowledgeonlanduse/land cover history. The data interpretation at the reference site should besupported by descriptive statistics (average value, standard deviation, coefficient ofvariation) of the area 137Cs inventories and by the 137Cs depth distribution. Twofeatures typical for distributionof 137Cs at landundisturbedbypost-depositional soilredistribution can help to validate the selection of the reference site and therepresentativeness of the obtained reference value: (1) the 137Cs vertical depthdistribution and (2) the horizontal spatial variability (the homogeneity of the initial137Csfallout).

Avalidselectionofareferencesiteshouldfulfilatleastthesetwoconditions:

• Thespatialvariabilityofthe137Csfalloutatthelocalscale(highandlowvaluesin a small area) should be as low as possible. It should be characterised bydescriptive statistic as suggested by Sutherland (1991, 1996), Owens et al.(1996)andMabitetal.(2012).Avalueofthecoefficientofvariation(CV)below30percent isagood indicatorofa stableandundisturbed referencesite.Thenumberofsoilsamplesneededtoassessaccuratelythereferencesite(usuallycca. 10 samples)dependson the variability and canbe calculated consideringthevalueoftheCV.

• At reference site a clear exponential decreaseof the 137Cs activitywithdepth(around 80 to 90 percent of the 137Cs included in the top 20 cm soil layer) isexpected.

7.2PresentationoftheresultsofstudiesatfieldscaleAtastudiedfield,thefollowing137Csinformationshouldbeestablished:

• Inventoriesof137Cs(Bqm-2)representingthetotal137Csactivityperareaunitatthesamplingpoints;and

• Depthdistributionof137Csalongtheverticalsoilprofiles(activityinBqkg-1perselecteddepthintervalsincm).

The inventories are calculated from the 137Cs mass activity (see Chapter 5.2).Comparisonbetweentheinventoriesatthestudiedfieldandthereferencesiteallowsidentifyingtheerosionanddepositionzones.This137Csspatialdistributionpermitstorepresent a soil erosion and sedimentation pattern. Its variability can be studied inrelation to land use, soil proprieties and especially relief, e.g. the 137Cs distributionalongtheslopetransects(Figure7.1).

36

Figure7.1.Exampleofthe137Csdistributionalongtheslopetransects(A-H)atJaslovskeBohunice,Slovakia.The137Csinventoriesaredecreasingfromplateaustoslopesandincreasinginvalleybottomwherethey

reachtheirmaximumvalues(Fulajtar,2003).

The 137Cs depth distribution profiles, associated with cultivated and/or uncultivatedland,allowtoseetheprofileshapesandtoverifythattheyconformtotheexpectedbehaviourof137Csinerosionanddepositionareas.

Assuming an adequately representative sampling design, the 137Cs inventory errordependsmostlyon thedetermination foundbygammaspectrometry.Soilerosionordepositionrateforparticularsamplingpoint isquantifiedusingconversionmodelstoconvertthe137Csinventories(Bqm-2)toerosionordepositionrates(tha-1yr-1),whichwere described in Chapter 6. The obtained values of soil redistribution rates can berepresented with the aid of the following parameters (Quine and Walling, 1991;WallingandQuine,1993;Mabitetal.,2014):

• Erosionzone:thepartofthestudiedfieldaffectedbyerosion(areaofstudiedfield multiplied by the ratio of eroded sampled profiles and all sampledprofiles);

0

1000

2000

3000

4000

5000

600013

7 Cs i

nven

tory

Slope position

A

B

C

D

E

F

G

H

I

Mean

37

• Deposition zone: the part of the studied field affected by deposition (area ofstudiedfieldmultipliedbytheratioofsampledprofileswithdepositionandallsampledprofiles).

• Mean erosion in the erosion zone: the total mass of eroded soil materialdividedbytheareaaffectedbyerosion;

• Mean deposition in the deposition zone: the total mass of deposited soilmaterialdividedbytheareaaffectedbydeposition;

• Grosserosion inentire field: thetotalmassoferodedsoilmaterialdividedbytotalareaofstudiedfield;

• Gross deposition in entire field: the total mass of deposited soil materialdividedbytotalareaofstudiedfield;

• Totalerosion:thesumofallerosionvalues;

• Totaldeposition:thesumofalldepositionvalues;

• Neterosion:theamountofsoilleavingthestudiedfield;

• Sedimentdeliveryratio:theratioofnetandgrosserosion.

7.3DatainterpolationandsoilredistributionmappingThemeasured 137Cs inventoriesand soil redistribution rates calculatedby conversionmodels represent point datawhich are arranged in a grid over the area of interest.Suchadatasetprovidesvaluableinformationnotonlyonthesoilredistributionratesbut on the spatial variability of soil redistribution as well. The quality of thisinformationdependsontheprocessingofpointdataandcreationofsoilredistributionmaps. If a digital elevation model (DEM) is available the 3D-visualisations of soilredistributioncanbemade(Figure7.2).

Various simpleormore sophisticated interpolation approaches canbe employed. Toobtain more realistic results, it is recommended to use advanced approaches forspatial interpretation of data. Geostatistical methods such as kriging can take intoaccount thespatial variabilitybasedon thedistancebetweenpointsandevenbasedondirectionsofprocesses(e.g.soilmovementalongaflowdirection).

7.4Datainterpretation,utilizationandcommunicationFor interpretation and utilization of erosion assessments for practical purposes it iscrucialtoagreeuponandimplementaterminologythatiseasytounderstandandthatenablesstakeholderstotakeactions.Theconceptofatolerable(ortarget)soilloss(T-value)(Figure7.3)describestheboundarybetweenacceptable(legendofquartersofTin green) and unacceptable soil loss (multiple of T in red). Even though there is nodefined tolerable negative soil loss, the amount of deposition is illustrated in twoclasses:belowandabovethenegativeTvalue(yellowcolours).

38

Figure7.2.Exampleof3-Dvisualisationofsoilredistribution,Marchouch,Morocco.(Benmansouretal.,

2013).

The identification of likely source areas such as upslope or upstream areas (on-site)andlikelydepositionareassuchasdownslopeordownsteamareas(off-site)enablesaspatial identification of stakeholder responsibility (Figure 7.3). This enablesstakeholders to put 137Cs erosion and deposition values into perspective andstakeholders canbetter communicate to find an agreementonhow to interpret theresults and turn them into actionable items such as policies and regulations torecognizetheproblemandfindbestmanagementpractices(BMP)solutions.

Thedataobtainedthroughthe137Csmethodhaveahistorical informationvaluewhathappened accumulatively since mid-1950s until the day of the measurement. Thislong-termdatacanthenbeusedforvariouspurposesamongwhichthefollowingaremostimportant:

• Long-termerosionratemonitoring(thiscouldbeacoupleofpoints);

• Investigation of the relative long-term temporal dynamic and spatialdistribution of erosion rates (this requires more points; ideally in a rasterpattern);

• Study of impact of erosion factors on long-term erosion rates atdifferent/neighbouringsites;

• Verification,calibrationandvalidationofsoilerosionmodels;

• Testing the long-term efficiency of soil and water conservation measures (orBMPs).

39

Figure7.3.Theconceptoftolerablesoillossforstakeholderstoputmeasurements(e.g.137Cs)andmodelling

(e.g.WEPPsimulations)intoactionablemanagement.

The last purpose is very important to evaluate BMPs in landmanagement (e.g. zerotillage,bufferstrips,stripcropping,etc.)andtodevelopcomprehensive,long-termsoilconservation strategies and policies. If stakeholders want to assess short-termprocesses,the137Csdatashouldbedonerepetitively(adjacenttotheprevioussamplelocation)combinedwithshort-termobservationsofsoilredistributionsuchasplotorcatchment studies. In contrast to total annual (or average annual) on-site soil loss,event-basedrunoffandsedimentyielddataiscollectedatthelowerboundaryofaplotor even at the catchment outlet. A nestedwatershed approachwithmultiple outletmeasurement points would be exceptional for researchers, but would be hard toimplement at multiple locations due to costs. This event-based observation datatogetherwith long-termpointdataof 137Cs samples,however,enable to successfullydevelop, validate and apply spatially distributed, process-based, continuous soil lossmodels.

Oncesuchaprocess-basedmodel iscalibratedandvalidatedforaparticularhillslopesite it can be used to assess soil redistribution based on a variety of scenariosrepresentingchanges insoilorvegetationmanagement. If therearesufficientevent-basedclimatedataavailablethemodelscanbeusedtoderivestatisticalparameterstogenerate realistic long-termweather patterns (Meyer et al., 2008). There is also theopportunitytointroduceclimatechangepatternandgeneratescenariostoassesstheriskofrunoffandsedimentdischargesofcertainreturnperiodsofprecipitationevents(e.g. 100-year return periods of events without and with climate changes intemperatureorprecipitation).

40

41

8 137Cs method – validation and use of erosionmodels

In contrast to just mathematically interpolating 137Cs sample points, scientists havecollectedinformationonerosionprocessesanddevelopedprocess-basedmodelsthatenable them to fill the information gaps between sample points on the basis ofunderstandingtheprocessesinlandscapes.

Since thedevelopmentof the first,widelyappliedsoilerosionmodel– theUniversalSoil Loss Equation (USLE) byWischmeyer and Smith (1958) - numerous regional andnational versionsof theempirical-USLE-typemodelsweredeveloped (e.g. BoardmanandPoesen,2006).Theseempirical soil lossequationswerederivedbasedon runoffand sediment yielddata collected at the endof unit plot areasof a certainhillslopelength.Alternatively,observationsof theexposuredepthofaphysicalmarkeroveraparticulartimeperiod,e.g.erosionpins,wouldquantifytheerodedordepositedsoilataparticularpointalongorattheendofalonger,sometimescomplexhillslopewithnoinflowattheupperboundary.Suchlong-termempiricalmodelsweredevelopedbasedonalargeamountofquantitativedataforaspecificregion,butwhentheyshouldbeappliedinotherregiontheyneedtobecalibratedandvalidatedbysimilaramountsofdatacollectedinthatregion.

The 137Cs method does the same as the erosion pins, but without the intensiveinstrumental setup for a fixed starting point and it covers a long period (sincemid-1950s or after a NPP fallout). Once at studied slope the 137Cs measurement isestablished,thefollowupmeasurementcampaigns(e.g.every5to10years)areableto providemulti-temporal data indicating the dynamics of erosion rates. Combiningthe137CsmethodandempiricalUSLE-type,long-termsoilerosionmodelsenableustocalibrateandvalidatethemodelsforareferenceslopeandfurthermorethemodelcanbe used to extrapolate the information to surrounding fields. However, the originaldevelopmentofthemodelreliesonaspecific,uniformplotscale(USLEstandardslopelength is 22.13 m) that is usually much shorter than the variable, complex slopesprevailinginreallandscapes.Thereisnoaccountingforconfluenceordiversionofflowpatternswithin thoseplots. Therefore, 137Csdata ina rastergrid samplepatternhasthe potential to validatemodels for complex hillslopes and validatedmodels can beappliedatneighbouringhillslopesandinanentirecatchment.

The137Csmethodisauniquewaytocreatelandscapeerosionpatternsincemid-1950sthat can be compared not only with single hillslope models, but also spatially-distributed soil erosion models. Models that are capable of simulating soilredistribution along complex hillslopes and/or flow patterns have been developedsince half a century. These soil loss and redistribution patterns can then be usedtogetherwithrunoffandsedimentdischargedataandgagingstationstocreateentirelandscapebudgetsofwaterandsediment.Besidesplotandhillslopedata, floodplainandreservoirdataof137Cssamplesandtotalmeasuresofsedimentsorsedimentfluxeshelptodescribetheselandscapebudgets.

42

Thepossibilityofusing soil redistribution ratesestimated for 137Csdataandprocess-based models in coordinated research effort, enables the international scientificcommunity to create new higher resolution spatial and temporal modellingtechnologiestosimulatenotonlyhistoriceventsortimeperiodssincethepast.Oncevalidated for different land use histories, it is possible to predict the spatial andtemporaldistributionoferosion for timeperiodswithalternative landuses.Realisticscenarios for process-basedmodels would help to assess the impacts of changes insoil,waterandcropmanagementand/orinclimate.

Therefore, continuous process-based and catchment-based soil erosionmodelsweredeveloped over the past half-century that are now able to simulate scenarios andenableanevent-basedanalysis.Theseprocessmodelscanbeverifiedandvalidatedforameasurement sitewith long-term (137Csdata) and short-termdata (event-basedorhigher resolution data) and then transferred to neighbouring areas with minimumcalibration of parameters. Process-basedmodels require users to carefully study theconsiderationofprocessesthatarerepresented(Figure8.1);e.g.somelocationswouldneedtoconsidergullyerosionprocesseswhiletillageerosionprocessescould impactothers. The latter two examples would not be necessarily included in a model andwouldneedtobeconsideredwheninterpretingtheresultsofthe137Csmeasurements.

Figure8.1.Scalesofinterest,spatial,andtemporalvariablepropertiesimportantfordominantprocessesat

anindicatedscale(RenschlerandHarbor,2002).

43

8.1137Csdata:findingrightmodelandscale

Giventhecriticalimportanceofassessingtheimpactoftemporalandspatialvariabilityin soil erosion parameter data at different scales (Figure 8.1), surprisingly littleemphasis has been placed on this in soil erosion simulation tool development andevaluation for practical decisionmaking (Renschler and Harbor, 2002). Inmodelling,shorter time scales are typically associated with smaller spatial scales because finertime resolution requires more detailed modelling of hydrological and sedimenttransportprocesses,whichusuallymeansconsiderationofvariabilityatmoredetailedspatial scales (Kirkby, 1998). At different scales, different groups of processes aredominant, so theeffective focusof themodelalsochangeswithscale.Mostmodernwatershedmodelsfocusonthepredictionofwaterfluxesforparticularspaceandtimescale(Figure8.1;RenschlerandHarbor,2002).

Hydrologic processes are oftendescribed inways that dependon themodel’s scale.Butthescaleinwhichthemodelisdesignedtooperateinfluencesdecisionsaboutthelevelofspatialaggregationininputdata,analysis,andmodeloutput.However,whentheproblemof interest isnotatthescaleofthemodel,scalingbothdataandmodelresultsisverycomplex(Renschleretal.,1999).Forexample,empirical,process-basedapproaches to predict sheet and rill erosion rates can provide reasonable results toevaluate environmental and economic impacts of agricultural policy on a large scale(Carriquiryetal.,1998),butdonotyieldaccuratepredictionsoferosionvariabilityonthefieldscale.

Nearingetal. (2005)providesanoverviewof a seriesof state-of-the-art soil erosionmodelsandillustratesresultsofamodelperformancecomparisonforaparticularsitebasedonscenarioswithchangesinvegetationcoverandclimatechange.ModelsthatarewidelyusedincludetheRevisedUniversalSoilLossEquation(RUSLE)–anupgradedversionof theempirically-basedUSLE (Renardetal., 1997), the Limburg Soil ErosionModel (LISEM)(Jettenetal.,1998),thekinematicrunoffanderosionmodelKINEROS(Smithetal.,1995),thewidely-usedSoilWaterAssessmentTool(SWAT)(Arnoldetal.,1999) for largerwatersheds,and theprocess-basedWaterErosionPredictionProject(WEPP) model for single or multiple hillslopes in small watersheds (Flanagan andNearing,1995).

Renschler (2003) developed the Geospatial Interface of WEPP (GeoWEPP) enablinggeospatial application of the WEPP model using GIS data to assess the impacts ofspatial and temporal land management and climate change through scenarios.Geospatial tools like the GeoWEPP use the idea of a T-value (Figure 7.3) to assiststakeholders in the quest for understanding the history of soil erosion, floods andsedimentation and assess the predicted land use, land management and climatechangescenarios.

The soil losspattern can thenbecomparedwithpoint-based 137Csmeasurementsaswellas runoffandsedimentyielddataatacatchmentoutlet (Figure8.2).Theevent-basedrunoffandsedimentyieldmeasurementsenabletocomparemodelresultswithobservationsatthewatershedoutlet.

44

Figure8.2.137Cs-basedsoillossandsimulatedsoillossbasedongeneratedweather.

8.2137Csdataandprocessmodelling:powerfulnewtools

Validatedprocess-basedmodelsenablescientistsandpractitionersthentocreatenew,powerful scientific soil loss prediction tools to support better soil and waterconservation decision-making and policy-making through “what if”-scenarios (Figure8.3). They also help to better plan and deal with contaminated soils after a NPPdisasteroraccidentwithradioactivitybeingreleased.

Figure8.3.Modellingofsoilerosionforselectedcroprotationscenarios,GuadaltebaWatershed,Andalusia,Spain(Renschler,1996).

45

It can be concluded that the experience with combined use of the 137Csmethod oferosion assessment and erosion modelling brought up to now the following newdevelopments:

• The137Csmethodcanbeusedforsoilconservationincombinationwitherosionmodels.Thiscombinedapproachmayprovidemorecomprehensiveassessmentofsoilerosion involving informationon long-termmeansoilerosionratesanddetailed spatial patterns of soil redistribution (provided by 137Cs) as well asvariousscenariosofshort-termsoilerosiondynamicsunderdifferentlandusesandmanagements(providedbyerosionmodels).

• The137Csmethodcanbeusedforcalibrationandvalidationoferosionmodels.Most easily it can be used for calibration and validation of empirical modelssuchastheUSLEbecausetheUSLEandUSLE-typeempiricalmodelsareworkingwith empirical factors, such as rainfall erosivity or R-factor, that can beexpresses in annual cumulative values. This approach is close to long-termmean soil erosion rates derived from 137Cs inventories. Despite of theirsimplicity the USLE is still very useful for soil conservation because it is notdemandingforalotofinputdataandthuscanbeappliedforlargeterritories.

• Very helpful for calibration and validation of erosionmodels is the combineduse of 137Cs and conventional methods (experimental plots, hydrologicalprofiles)tocreatecatchment(orwatershed)models.Thesemodelscanprovidewater and sediment budget calculations that are especially useful for waterresourcemanagementandplanningsuchaswaterreservoirs.

• Apart from the information on erosion rate the 137Cs data provides veryvaluable information especially on spatial distribution of soil redistributionrates.Calibrationandvalidationoferosionmodels isusuallybasedonerosionmeasurements at experimental plots and suspended sediment loadmeasurements at water courses in small catchments. These data representusually very few spatial locations. This limited spatial information can besignificantlyextendedby137Csdatawhichareusuallyrelatedtolargenumberoflocationscreatinggridovertheareaofinterest.

• Useof 137Csdata for calibrationandvalidationofprocess-based (orphysicallybased) models, which are built on event-based data are more challenging.However 137Cs data points in the landscape provide the ultimate measuredpatternforprocess-basedmodelsforaparticulartimeperiod(sincemid-1950sor after a NPP disaster). The complex calibration and validation of detailedprocess-basedmodelsmay require amuch larger, coordinateddata collectionof input parameters. However, those validated model could then be usedwithout further calibration when transferred into neighbouring areas withsimilarclimateandlanduse.

• The use of the 137Cs method for erosion models calibration and validationrequiresspecialemphasisonthesystematiccollectionofinputdatainadditionto137Csdata.Thedatasetshouldbeaccommodatedaccordingtotheneedsof

46

eachparticularmodeltobeused.Forthatreasonsitisimportanttoknowpriorto developing study designwhether the intended 137Cs studywill be used forerosionmodelsand forwhichmodel.Theuseofolder 137Csdata sets isoftenproblematic because some complementary geographical data aremissing andmustbeestimatedorgeneratedusingsecondarymethods,whichreducetheiraccuracyandrepresentativeness.

• Theuseof the 137Csmethodeither incombinationwitherosionmodelsor fortheir calibration and validation can be done by two approaches: 1) theretrospective approach (using 137Cs data fromone timehorizon and assessingtheperiodbetweenthe137Cs falloutandthetimeof 137Cssampling),or2) themonitoring approach collecting 137Cs data from several (at least two) timehorizonsandassessing theperiodsbetweenthese timehorizons). Ifusing thisapproach, it is recommended to repeat the 137Cs samplingonceupon5or 10years. This approach is very demanding for precise recording of the samplingpositions, because if the later sampling would not fit to the position of theformersamplingthevalueswouldnotbecomparable.

• When using 137Cs data for calibration and validation of erosion models, it isimportanttousemultipletransectdesignandensurethattransectsfollowtheslope direction (and hence the runoff direction). The regular grid is lessappropriate as the erosion models consider the relations along the slope orrunoffcontributingareas. It ispossibletousealsodatasetsfromregulargridsbutinsuchcasetheoperatorshouldtrytoidentifythemutualrelationbetweenthedatapointsconsideringtheslopedirectionsandtherepresentativenessofthe results is lower as the points from the regular grid will mostly not bearrangedinlinesalongtheslopedirections.

• Whenusing137Csdataincombinationwithothermethods(suchasconventionalmeasurements of soil erosion rates, erosion modelling, remote sensing orexperimental rain simulation) the interpretation should always carefullyconsiderthedifferenterosionprocesses,spatialandtemporalscalescoveredbyparticularmethods.Mosterosionmodelsestimatewatererosionprocessesatslope,plotorsmallcatchmentscales.Somemodelsareaimedsolelyeitheronwind or on tillage erosion. Each conventional erosion measurement methodcovers different processes. The experimental plots represent the sheet andinitial stages of rill erosion, the volumetric method represents rill and gullyerosion and the suspended sediment load measurements at hydrologicalprofiles represents erosionby those rills and gullieswhich are interconnectedwith water courses and bank erosion by water course. Most measurementsrepresentthetemporalscalesofevent,season,yearorseveralyears.The137Csmethod represents overall rates of soil redistribution by all processes and itcoverslongperiodseithersincethe137Csfallout(recently64years;1954-2018)orvariousmediumtermperiods(e.g.theperiodsbetweentwosamplingsifthe137Cs is sampledrepeatedly).Thesedifferences inprocessesandtemporalandspatial scales represented by different methods should be always carefullyconsideredwhencomparingtheresultsfromdifferentmethods.

47

8.3Exampleofusing137CsmethodforWEPPmodelverificationinMorocco

To illustrate thebenefits for a combined 137Csmeasurement transect and ahillslopemodellingstudy,arecentlystudiedtransectsalongasingle90m-longhillslope(withina1-hectarefieldsectionofasmallagriculturalwatershednearMarchouch,Moroccowasused(Benmansouretal.,2013).Likemostofthe137Csmeasurementstakensofar,themeasurement design was not planned with a modelling exercise in mind. Acoordination of modelling exercise and data sampling in the field would enable toconsiderflowpatternandaccumulationaswellasconversionanddiversionofflowinthe landscape. The flow path can be determined by using contour lines of atopographic map (Figure 8.4) or a flow accumulation algorithm on a DEM in a GIS.Thereforethefollowingstudyshouldbeseenasanexampletoillustratewhatcanbedonewithalreadyexistingdata.

The90m-transectwitha10m-distancebetweeneachof thenine 137Csmeasurementlocationisanidealdistributionforalandscapesectionwithparallelsheetflowtoverifyand validate a process-based erosion model at the hillslope scale. The previouslymentionedWaterErosionPrediction(WEPP)model(FlanaganandNearing,1995)wasusedinthisstudyforillustrationpurposes.

Figure8.4.DirectionofcontourlinesandcroprowsatMarchouchstudysite,Morocco.

48

Besidesthetopographicdataoftheslope(slopesegmentlengthandslopeinclination),the WEPP model requires long-term climate data (in this case generated 100-yearworth of data based on the precipitation dynamic of a couple of years of standardmeteorologicaldatafromanearbystationwereutilized),soil-samplingdata(standardsoilsurveydatasuchassoilprofileswithsoiltexture,organicmatter,rockcontentaswell as some estimatedmodel parameters such as hydraulic conductivity and shearstress), and land use data (a continuous winter wheat rotation was the bestassumptionsincedetailedcropdataforthepast60yearsaremissing).Moredetailsondatagathering,modelparametersestimationandmodelequationscanbetakenfromFlanaganandNearing(1995).

Figure8.5illustratesthedifferencebetweenhillslopeprofile(redline)andtherelativeerosion (green line). The grey-shade curve at the bottom of the graph illustrateserosionalong theslope ranging fromnearly25 tha-1yr-1 in theupper sectionof theslopeanddepositionvaluesofmorethan25tha-1yr-1atthebottomoftheslope.

Figure8.5.SoilRedistribution–WEPPvs.IsotopeDerivedTransectatMarchouch,Morocco.

The137Csmeasurementsforeachoftheninesamplelocations(Table8.1)wereusedtomake the erosion rate estimation. Gained values were categorized to four classes:slighterosionupto10tha-1yr-1(orange),mediumerosionbetween10and20tha-1yr-1 (red),strongerosionmorethan20tha-1yr-1 (boldred)anddeposition(green).Thesamplingandmodelsimulationdonotmatchexactly,buttheymatchtheoverallsoilredistributionpatternandtheorderofmagnitude.

ForamoredetailedanalysisoneshouldusetheGISandmodeltodeterminetheflowpath firstand then sample the 137Csmeasurementpointsaccordingly.This illustratesthat GIS, sampling andmodelling can and should be ideally used in combination tocontributetomorecomprehensiveunderstandingofsoil redistribution.Aspreviously

49

indicateda rangeofvaluesof inputparameters suchasclimateand landusecanbeapplied to evaluate various scenarios thatmayhelp to identify the optimal landusepolicy.

Table8.1.LanduseandClimateChangeanalysisusing100-yearsofgeneratedweather,Marchouch,Morocco.

a) Winterwheat(conventionaltillage)

Period Precipitation(mmy-1)

Runoff(mmy-1)

SoilLoss(kgm-2y-1)

SedimentYield(tha-2y-1)

Current 415.8 42.4 2683 26.1

2030 395.9 41.7 2869 27.8

2060 374.6 40.7 2955 28.7

2090 348.2 42.2 3521 34.1

b) Fallout(withoutvegetation)

Period Precipitation(mmy-1)

Runoff(mmy-1)

SoilLoss(kgm-2y-1)

SedimentYield(tha-2y-1)

Current 415.8 107.0 5874 58.7

2030 395.9 102.5 5631 56.3

2060 374.6 96.3 5341 53.4

2090 348.2 90.3 4990 49.9

50

51

9Advantagesof137Csmethod

The methodology of soil erosion rate assessment and measurements is a result ofalmost 100 years of research. Several methods were developed and refined (e.g.volumetric methods, geodetic methods, erosion plots, hydrological methods,radionuclide methods, etc.) by the contributions of the international scientificcommunity.Eachmethodhasdifferentassumption,principle,coversdifferentspatialand temporal segments of erosion-sedimentation cycle, and reflects particular soilredistributionprocessesparticipatinginthiscycle.Thereforethedataobtainedcannotbe schematically compared. Each presented dataset has to be specified in terms ofmethod used, temporal and spatial scale covered and particular processesrepresented.Moreover,eachmethodhasitsadvantagesandlimitation,whichshouldbeconsideredalreadybeforestartinganystudyandthemethodmostappropriateforthe study objective should be selected. Somemethods are complementary and canprovideusefulresultsifusedjointly.

The 137Cs is an excellent tracer for studying soil erosion processes. It is the moremature andmostwidely used FRN for soil erosion assessment. Itswide applicabilitywas demonstrated by the bibliography provided by Ritchie and Ritchie (2008). The137Csmethodhasthefollowingadvantages,modifiedaccordingtoMabitetal.(2008)andZupancandMabit(2010):

§ Theobtaineddatarepresentlong-term(recentlyoversixdecadessincethe137Csfalloutuntil the samplingperiod) soil redistribution rates. This isprobably themain asset providedby the 137Csmethod. Long- ormedium-termdataon soilredistributionmagnitudesaredifficulttoacquirebyconventionalmethodsandtheyareveryvaluableforstudyingthedevelopmentoflandscapeandsoilscape.Datacanbeusedforsoilconservationprogrammesandforsoilerosionmodelvalidation.

§ Theobtainedvaluesrepresentanoverall soil redistributionrate integratingallprocesses responsible for mechanical movement of soil particles (e.g. water,wind and tillage erosion), eventually other processes, such as gravitationprocesses (creep, landslides,etc.). This isanadvantage for studies focusedonoverall assessment of erosion impact on landscape development. Data onoverallerosionratesisveryscarce.Amongtheconventionalmethods,onlythegeodetic method (erosion pins) reflects the impact of all erosion processes.Mostotherconventionalmethodsreflectparticularprocessesofwatererosion.Measurementsatexperimentalplotsprovideinformationonlyoninitialstagesofwatererosion(rillandsheeterosion).Volumetricmethodsreflectadvancedstages of rill erosion and gully erosion. Hydrological methods representadvanced stages of rill and gully erosion and linear erosion by permanentwatercourses such as riverbed erosion and riverbank erosion. Other specificmethods have been designed for specific investigation of erosion processesotherthanwatererosion(e.g.tillageandwinderosion).

§ The obtained values provide retrospective information. It is a unique featurethat past erosion rates can be estimated from samples collected at present

52

time.Retrospectivedatacannotbegainedbyconventionalmethods.The137Csmethod can provide also perspective data if the 137Cs sampling is periodicallyrepeatedintime.

§ The obtained data provide information on the spatial distribution of soilredistributionratesifthesamplingdesignisbasedongridormultipletransects(Figure 9.1). This is a clear advantage as compared to conventional methods(suchaserosionplotsorhydrologicalmethod),whichprovidedataattributedtooneorfewprofileswithinstalledmeasurementequipment.

§ Thereisnoneedforcostlyandlabourintensivelong-termmonitoringprograms.The fieldwork required iseasy tomanage.The resultscanbeobtained fromasinglesitevisitorshortfieldcampaigns.

Figure9.1.Samplinggridfor137Csdeterminationandtheresultingmapofsoilredistributioncreatedbyordinarykriging,BoyerSite,Canada(Mabitetal.,2008)

53

10 Activities performed by the Joint FAO/IAEADivisionandfurtherperspectives

The 137Cs method is today widely used by the international erosion researchcommunityandhasgainedmuchpopularitysincethe1990swhenuser-friendlymodelsforconverting137Cs inventories intosoil redistributionratesbecamewidelyavailable.Since 1995 – when the Joint FAO/IAEA Division launched its first CRP on theassessment of erosion by FRNs, it has become one of the leading institutionsdevelopingtheseFRNmethods.ProbablythemostimportantresultsofupgradingtheFRNmethodology to a higher level are themethodological handbooks published byZapata (2002) and IAEA (2014). The Joint FAO/IAEA Division is refining the 137CsmethodalsothroughappliedresearchconductedatitslaboratoriesinSeibersdorf.

Apart from research activities, the Joint FAO/IAEA Division, through IAEA technicalcooperationprojects(TCPs),disseminatesthe137CsmethodtoFAOandIAEAMemberStates. Until now, this specific technical support was provided to more than 70countriesworldwide.

All necessary information about the activities of the Joint FAO/IAEA Division inresearchandtechnicalcooperationonthe137CsmethodsandotherrelatedFRNsandisotopic techniques is available through the SWMCN Subprogramme websitehttp://www-naweb.iaea.org/nafa/swmn/index.html, the Soils Newsletter and theAnnual Reports of the SWMCN Laboratory, Seibersdorf http://www-naweb.iaea.org/nafa/swmn/public/newsletters-swmcn.html.

In future, particular attention will be paid towards extending the use of the 137Csmethodtotheassessmentofsoilconservationefficiencyandtoerosionmodelling.Theassessmentof soil landuseandmanagement impactandsoil conservationefficiencyhasbeenextensivelydeveloped.CRPsprovideexamplesofsuchstudies, includingtheassessment of terracing in Morocco and Madagascar (Benmansour et al., 2013;Rabesirananaetal.,2016).

Thepotential of using the 137Csmethod for validating themodelshasnotbeen fullyrecognized yet. Erosion rates estimated by the 137Cs method can be used to cross-validate the erosion models and test the scenarios to find optimal land and soilmanagementstrategiesandpoliciesforsustainablefoodproduction.The137Csmethodcanbeusedalsotomonitorsoilerosionandsedimentationdynamicsifthesamplingisrepeatedatthesamesiteeitherperiodicallyoroccasionally(forexampleafterextremeerosionevents).Combiningthe137CsmethodwithotherFRNs(suchas210Pband7Be)can provide more comprehensive information on erosion dynamics and spatialdistribution.

54

Glossary

Activity - concentration of radionuclide in measured sample detected through itsradiationandexpressedinBecquerelsperunitmass.

Alkalimetals–groupofelements(columnintheMendeleev’speriodictable)involvinglithium (Li), sodium (Na),potassium (K), rubidium (Rb) caesium (Cs)and francium(Fr) characterized by high reactivity at normal atmospheric temperatures andpressures.Alkalimetalsreadilylosetheiroutermostelectronandformcationswith+1charge.

Bomb-derived 137Cs - 137Cs unintentionally released to the atmosphere by the pastsurfacenuclearweaponstests.

Chernobylaccident–accidentofnuclearpowerplantinChernobyl,Ukraine(26thApril1986).

Chernobyl137Cs-137CsreleasedtoatmospherebytheChernobylNuclearPowerPlant(NPP)accident.

Cosmogenicradionuclides–naturalradionuclides(suchas3H,7Be,14C,32P)originatedas a result of cosmic ray spallation of atoms (expelling of neutrons and protonsfrom the atom nucleus by high energy cosmic rays). The only cosmogenicradionuclide important for erosion research is 7Be that is formedby spallationofnitrogenandoxygen.

Deposition – complex of geomorphological process depositing soil or rock materialmobilised and transported from other areas. Deposition takes place usually (butnot necessarily) in depressions or low elevated, flat areaswhen the transportingagent loses the kinetic energy and the transported material exceeds thetransportingcapacityofthetransportingagent.Thedepositionbywindcanoccuralsoonelevatedandconvexlandforms.

Erosion - complex of geomorphological processes driven by several erosion agents(water,wind,tillage,snow,ice,animals,etc.)accusingdetachmentandremovalofsoilorrockparticlesfromlandsurfaceinaffectedarea.

Erosionrate–amountofsoilorrockmaterialerodedfromunitareaduringunittime.Usuallyit isexpressedeitherasmassintonsperhectareperyearorasvolumeinmmperyear.

Erosionplot–mostcommonlyusedconventionalapproachoferosionmeasurements.Theerosionisquantifiedforcertainlandareafencedbyartificialboundaries(suchasmetalsheets).Thesoilerodedbysurfacerunoffiscapturedtocollectiontanks.

Fallout radionuclides – radionuclides that are getting to terrestrial and aquaticenvironment by fallout from atmosphere. This term was introduced in erosionresearchandreferstothoseradionuclideswhichareusableaserosiontracers(i.e.137Cs,210Pb,7Be,239+240Pu).Thefalloutisdryandwet(washingbyrainfall).

Gamma detector – analytical equipment aimed on measurement of the gammaradiation.

55

Gamma radiation –one of three types of electromagnetic radiation (alfa, beta andgammaradiation)emittedbyunstableradioactivenucleiofatoms(radionuclides).Gammaradiationcorresponds toemissionofphotonsand is the strongestof thethreemain radiation types. It does not havemass and charge and travelsmuchfurtherthatalfaandbetaradiation.

Gamma spectrometry – analytical method for determination of gamma radiationemittedbyradionuclidesbasedonassessmentoftheirenergyspectra.

Geogenic radionuclides –natural radionuclidesoriginating frombedrock substratum.For erosion research the most important geogenic radionuclide used as erosiontraceris210Pb,whichoriginatesasaproductof238Uradioactivedecayseries.Partofthe222Rn,whichisoneofthedecayproductofthisseriesevaporatesfromsoiltoatmosphereandisdecomposingto210Pb,whichfallstolandsurfaceandocean.

Half-life –period duringwhich half of the amount of radioactive isotope decays (itsactivityishalfoftheoriginalactivity).

Human-inducedradionuclides–radionuclidesintroducedintoenvironmentbyhumanactivity. For erosion research the most important human-induced radionuclides,which may be used as erosion tracers are 137Cs, 239Pu and 240Pu released fromsurfacenuclearweapontests.

Inventory - activity of radionuclide per surface area representing the amount ofradionuclideoccurringinsoilwithintheunitsurfacearea.

Isotope – atom that differs from the other atom of the same chemical element bynumberofneutrons.Isotopescanbeeitherstableorradioactive.

Kriging- widely used method in spatial analysis of interpolation modelling using aGaussianprocessutilisingdistance-weightedaverages.

Multislotdivisor–deviceusedforsplittingrunoffaterosionplotsinordertosavethestoragespaceofcollector.Usingthisdeviceenablestocollectonlysmallfractionoftherunoff.

Parshal fume – hydrological device serving as artificial hydrological profile on smallwatercourses for measurements of discharge and picking samples formeasurementsofsuspendedload.

Radioactivedecay–processofdecompositionofunstableisotopes(radionuclides)byemitting radiation and gradual losing part of the energy and mass until theradionuclideistransformedtostableisotope.

Radionuclide–isotopewhichisunstableduetohavinganexcessenergyinitsnucleus.It emits radiation and gradually loses energy andmass until it is transformed tostableisotope.

Referencesite–well representativesitewhere the 137Cs inventory isnotaffectedbyerosion, accumulation or other geomorphological processesmoving soil materialsuch as gravitation processes (landslides, creep, solifluction) neither by humanprocesses of land mass translocation (construction, mining activities). Thereference site represents the original fallout input of 137Cs reduced only byradioactivedecay.

56

Referenceslope–termusedinerosionmodellingforaseriesofmeasurementpointsalongalinearslopetransectwhichisusedasa“reference”forvalidatingamodelfor a hillslope segment that can be used to extrapolate results to other similarhillslopes.ItshouldnotbeconfusedwithreferencesiteusedforFRNmethods.

Sedimentation – originally referring to processes of deposition of soil and rockparticlesfromwater(sedimentationinnarrowsense).Inwidersense,itrefersalsoto all processes of soil and rock deposition by all transporting agentswhen theylosethetransportingcapacity.

Tippingbucket–deviceusedformeasurementofdischargeandsuspendedsedimentaterosionplots.Itcomprisesofoneortwocontainershavingobliqueshapes.Suchcontainers when filled by collected runoff lose balance and flip over. As aconsequenceof flipping,water ispouredand thecontainer returns to itsoriginalposition.Theflippingmotionisrecordedbyacounter.Whileflippingsmallportionofwaterwithsuspensionistrappedbypipeanddrainedtosamplecontainertobeusedformeasuringsuspendedsedimentload.

Undisturbedsite–landthatisnotdisturbedbyhumanactivities,especiallyagriculturalpractices(e.g.tillage).

57

References

Alewell, C.,Meusburger,K.,Juretzko,G.,Mabit, L.,Ketterer,M.E., 2014. Suitabilityof239+240Pu and 137Cs as tracers for soil erosion assessment inmountain grasslands.Chemosphere103.274-280.

Anspaugh, L.R., Catlin, R.J., Goldman,M., 1988. The global impact of the Chernobylreactoraccident.Science,242.1513–1518.

Arnold, J.G., Srinivasan, R., Muttiah, R.S., Allen, P.M., 1999. Continental spacesimulation of the hydrologic balance. Journal of the American Water ResourcesAssociation,35(5).1037–1051.

Basher, L.R., 2000. Surface erosion assessment using 137Cs: examples from NewZealand.ActaGeologicaHispanica,35.219–228.

Benmansour,M.,Mabit, L.,Nouira,A.,Moussadek,R.,Bouksirate,H.,Duchemin,M.,Benkdad,A.,2013.AssessmentofsoilerosionanddepositionratesinaMoroccanagricultural field using fallout 137Cs and 210Pbex. Journal of EnvironmentalRadioactivity,115.97–106.

Bernard,C.,Mabit,L.,Laverdière,M.R.,Wicherek,S.,1998.Césium-137etérosiondessols.CahiersAgricultures,7.179–186.

Bernard, C., Mabit, L., Wicherek, S., Laverdière, M.R., 1998. Long-term soilredistributioninasmallFrenchwatershedasestimatedfrom137Csdata.JournalofEnvironmentalQuality,27.1178–1183.

Boardman,J.andPoesen,J.(eds).2006.SoilErosioninEurope.Wiley.878p.Cambray, R.S., Cawse, P.A., Garland, J.A., Gibson, J.A.B., Johnson, P., Lewis, G.N.J.,

Newton,D.,Salmon,L.,Wade,B.O.,1987.Observationsonradioactivity fromtheChernobylaccident.NuclearEnergy,26.77–101.

Cambray, R.S., Playford, K., Carpenter, R.C., 1989. Radioactive fallout in air and rain:results to the end of 1988. AERE-R-10155. U.K. Atomic Energy Authority Report,Harwell,UK.

Campbell,B.L., Loughran,R.J.,Elliott,G.L.,1988.Amethod fordeterminingsedimentbudgets using caesium-137. International Association of Hydrological SciencesPublication,174.171–179.

Carriquiry, A.L., Breidt, F.J., Lakshminarayan, P., 1998. Sampling schemes for policyanalyses using computer simulation experiments.EnvironmentalManagement 22(4).505–515.

Carter, M.W. and Moghissi, A.A., 1977. Three decades of nuclear testing. HealthPhysics,33.55–71.

Dahlman,R.C.,Francis,C.W.,Tamura,T.,1975.Radiocesiumcyclinginvegetationandsoil. 462-481. In: Howell, F.G. Gentry, J.B., SmithM.H., (eds.), Mineral cycling insoutheastern ecosystems, USAEC Symposium Series, CONF-740513, US AtomicEnergyCommission,Washington,DC.

Davis, J.J.,1963.Cesiumand its relationship topotassium inecology.539–556. In:V.SchultzandA.W.KlementJr.(eds.),Radioecology,Reinhold,NewYork.

58

Elliott,G.L.,Campbell,B.L.,Loughran,R.J.,1990.Correlationoferosionmeasurementsandsoilcaesium-137content.AppliedRadiationandIsotopes,39.1153-1157.

Eyman,L.D.andKevern,N.R.,1975.Cesium-137andstablecesiuminahypereutrophiclake.HealthPhysics,28.549–555.

Flanagan, D.C. and Nearing, M.A., 1995. USDA-water erosion prediction project:hillslope profile and watershed model documentation. NSERL Report, vol. 10.USDA-ARS National Soil Erosion Research Laboratory, West Lafayette, IN. 1196–47097.

Frere, M.H. and Roberts, H.J., 1963. The loss of strontium-90 from small cultivatedwatersheds.SoilScienceSocietyofAmericaProceedings27.82-83.

Froehlich,W.,Higgitt,D.L.,Walling,D.E.,1993.Theuseof cesium-137 to investigatesoilerosionandsedimentdeliveryfromcultivatedslopesinthePolishCarpathians.271-283.In:S.Wicherek(ed.),Farmlanderosion,Elsevier,Amsterdam.

Fulajtar, E., 2000. Assessment of soil erosion through the use of 137Cs at JaslovskeBohunice,WesternSlovakia,Actageológicahispánica35/3.291-300.

García Agudo, E., 1998. Global distribution of Cs-137 inputs for soil erosion andsedimentation studies. In: Use of Caesium-137 in the Study of Soil Erosion andSedimentation.IAEATECDOC1028.117-121.

Golosov,V.N.,Panin,A.V.,Markelov,M.V.,1999.Chernobyl137CsredistributioninthesmallbasinoftheLoknaRiver,CentralRussia.PhysicsandChemistryoftheEarth,PartA:SolidEarthandGeodesy,24(10).881–885.

Graham,E.R.,1963.FactorsaffectingSr-85andI-131removalbyrunoffwater,Watersewageworks110.407-410.

IAEA, 2014. Guidelines for Using Fallout Radionuclides to Assess Erosion andEffectivenessofSoilConservationStrategies,IAEATECDOCNo.1741.215p.

Jetten,V.G.,DeRoo,A.P.J.,Guerif,J.,1998.SensitivityofthemodelLISEMtovariablesrelated to agriculture. In: Boardman, J., Favis-Mortlock, D., (Eds.),Modelling SoilErosionbyWater,NATOASI Series I,GlobalEnvironmentalChange, vol. 55.339–350.

Kirkby, M.J., 1998. Modelling across scales: the MEDALUS family of models. In:Boardman,J.,Favis-Mortlock,D.(Eds.),ModellingSoilErosionbyWater.NATOASISer.,Ser.I,vol.55.Springer-Verlag,Berlin,Germany.161–173.

Lal,R.,1994.SoilErosionResearchMethods,SoilandWaterConservationSociety(U.S.),CRCPress.352p.

Lomenick, T.F. and Tamura, T., 1965. Naturally occurring fixation of cesium-137 onsedimentsoflacustrineorigin.SoilScienceSocietyofAmericaProceedings29.383–386.

Loughran, R.J. and Campbell, B.L., 1995. The identification of catchment sedimentsources. In: Foster, I.D.L.,Gurnell,A.M.,Webb,B.W., (Eds.), Sediment andWaterQualityinRiverCatchments.Wiley,Chichester.189–205.

Loughran, R.J., Campbell, B.L., Shelly, D.L., Elliot, G.L., 1992. Developing a sedimentbudgetforasmalldrainagebasininAustralia.HydrologicalProcesses.6.145–158.

59

Loughran,R.J.,Pennock,D.J.,Walling,D.E.,2002.SpatialDistributionofCaesium-137.In:Zapata,F.(Ed).HandbookfortheAssessmentofSoilErosionandSedimentationUsingEnvironmentalRadionuclides.97-109.

Mabit,L.,Benmansour,M.,WallingD.E.,2008.Comparativeadvantagesandlimitationsof Fallout radionuclides (137Cs, 210Pb and 7Be) to assess soil erosion andsedimentation.JournalofEnvironmentalRadioactivity,99(12).1799–1807.

Mabit, L., Bernard, C., Makhlouf, M., Laverdière, M.R., 2008. Spatial variability oferosion and soil organicmatter content estimated from 137Csmeasurements andgeostatistics,Geoderma145.245–251.

Mabit, L., Bernard, C., Wicherek, S., Laverdière, M.R., 1999. Les retombées deTchernobyl,uneréalitéàprendreencomptelorsdel’utilisationdelaméthodeduCésium-137. In : Paysages agraires et environnement. Principes écologiques degestionenEuropeetauCanada.CNRSed.285–292.

Mabit, L., Chhem-Kieth, S., Dornhofer, P., Toloza, A., Benmansour, M., Bernard, C.,Fulajtar,E.,Walling,D.E., (2014). 137Cs:awidelyusedandvalidatedmedium-termsoil tracer. In: Guidelines for using fallout radionuclides to assess erosion andeffectivenessofsoilconservationstrategies.27–77.IAEA-TECDOC-1741.

Mabit, L., Chhem-Kieth, S., Toloza, A., Vanwalleghem, T., Bernard, C., Amate J.I.,González de Molina, M., Gómez, J.A., 2012. Radioisotopics and physicochemicalbackground indicators to assess soil degradation affecting olive orchards insouthernSpain.Agriculture,Ecosystems&Environment,159.70–80.

Mabit, L.,Meusburger, K., Fulajtar, E., Alewell, C., 2013. Theusefulness of 137Cs as atracer for soil erosion assessment: A critical reply to Parsons and Foster (2011),Earth-ScienceReviews127.300-307.

Mabit, L., Meusburger, K., Iurian, A.R., Owens, P.N., Toloza, A., Alewell, C., 2014.Samplingsoilandsedimentdepthprofilesatafine-resolutionwithanewdevicefordetermining physical, chemical and biological properties: the Fine Increment SoilCollector(FISC).JournalofSoilsandSediments,14(3).630–636.

Menzel,R.,1960.Transportofstrontium-90inrunoff.Science131.499-500.Meyer, C.R., Renschler, C.S., Vining, R.C., 2008. Implementing Quality Control on a

RandomNumberStreamtoImproveaStochasticWeatherGenerator.HydrologicalProcesses,22(8).1069-1079.

Miller, C.W., 2014. Fukushima Accident: Radioactivity Impact on the Environment,HealthPhysics107/3.265-266.

Nearing, M. A., Jetten, V., Baffaut, C., Cerdan, O., Couturier, A., Hernandez, M., LeBissonnais,Y.,Nichols,M.H.,Nunes,J.P.,Renschler,C.S.,Souchère,V.,vanOost,K.,2005.Modellingresponseofsoilerosionandrunofftochangesinprecipitationandcover.Catena,61.131–154.

Owens,P.N. andWalling,D.E., 1996. Spatial variabilityof caesium-137 inventoriesatreferencesites.Anexamplefromtwocontrastingsites inEnglandandZimbabwe.AppliedRadiationandIsotopes,47.699–707.

60

Owens,P.N.,Walling,D.E.,He,Q.,1996.Thebehaviourofbomb-derivedCaesium-137falloutincatchmentsoils.JournalofEnvironmentalRadioactivity,32.169–191.

Pennock,D.J.andAppleby,P.G.,2002.Siteselectionandsamplingdesign. In:Zapata,F., (Ed.), Handbook for the Assessment of Soil Erosion and Sedimentation usingEnvironmentalRadionuclides.KluwerAc.Publ.,Dordrecht,TheNetherlands.15–40.

Quine, T.A. andWalling, D.E., 1991. Rates of soil erosion on arable fields in Britain:quantitative data from caesium-137measurements. Soil use andmanagement, 7(4),169–176.

Rabesiranana,N.,Rasolonirina,M.,Solonjara,A.F.,Ravoson,H.N.,Andriambololona,R.,Mabit, L., 2016. Assessment of soil redistribution rates by 137Cs and 210Pbex in atypicalMalagasyagriculturalfield,JournalofEnvironmentalRadioactivity152.112-118.

Renard, K.G., Foster, G.R.,Weesies, G.A.,McCool, D.K., Yoder, D.C., 1997. Predictingsoilerosionbywater—aguidetoconservationplanningwiththereviseduniversalsoil loss equation (RUSLE). Agricultural Handbook, vol. 703. U.S. GovernmentPrintingOffice,Washington,DC.

Renschler, C.S., 1996. Soil erosion hazard mapping by means of geographicalinformationsystems(GIS)andhydrologicalmodelling.MSc,TechnicalUniversityofBraunschweig,Braunschweig,95p.

Renschler, C.S., 2003. Designing geo-spatial interfaces to scale process models: TheGeoWEPPapproach.HydrologicalProcesses,17.1005–1017.

Renschler, C.S. and Harbor, J., 2002. Soil erosion assessment tools from point toregional scales - The roleof geomorphologists in landmanagement research andimplementation.Geomorphology47.189-209.

Renschler, C.S.,Mannaerts, C., Diekkrüger, B., 1999. Evaluating spatial and temporalvariability of soil erosion risk—rainfall erosivity and soil loss ratios in Andalusia,Spain.Catena34(3–4).209–225.

Ritchie, J.C. andMcHenry, J.R., 1975. FalloutCs-137:a tool in conservation research.JournalofSoilandWaterConservation30.283-286.

Ritchie,J.C.andMcHenry,J.R.,1990.Applicationofradioactivefalloutcesium-137formeasuring soil erosion and sediment accumulation rates and patterns: a review.JournalEnvironmentalQuality,19.215–233.

Ritchie, J.C. and Ritchie, C.A., 2008. Bibliography of publications of 137cesium studiesrelatedtoerosionandsedimentdeposition,USDA-ARS,Beltsville,380p.

Rogowski, A.S. and Tamura, T., 1965. Movement of 137Cs by runoff, erosion andinfiltrationonthealluvialcaptinasiltloam.HealthPhysics11.1333-1340.

Rogowski, A.S. and Tamura, T., 1970a. Environmental mobility of cesium-137.RadiationBotany,10.35–45.

Rogowski, A.S. and Tamura, T., 1970b. Erosional behaviour of cesium-137. HealthPhysics,18.467–477.

61

Smith,R.E.,Goodrich,D.C.,Woolhiser,D.A.,Unkrich,C.L.,1995.KINEROS:akinematicrunoff and erosion model. In: Singh, V.P. (Ed.), Computer Models of WatershedHydrology.WaterResourcesPublications,HighlandsRanch,CO.697–732.

Sutherland, R.A., 1996. Caesium-137 soil sampling and inventory variability inreferencesamples;literaturesurvey.HydrologicalProcesses,10.34–54.

Sutherland, R.A., 1991. Examination of caesium-137 areal activities in control(uneroded)locations.SoilTechnology,4.33–50.

Sutherland, R.A. and de Jong, E., 1990. Quantification of soil redistribution usingcaesium-137, Outlook, Saskatchewan, Canada. In: Bryan, R.B. (ed.). Soil erosion-experimentsandmodels.CatenaSupplement,17.177–193.

Soil ScienceDivision Staff, 2017. Soil surveymanual. Ditzler, C., Scheffe, K.,Monger,H.C.(eds.).USDAHandbook18.GovernmentPrintingOffice,Washington,D.C.

UNEP, 1997. Soil degradation,Map,World Atlas of Desertification, International SoilReferenceandInformationCentre(ISRIC),UNEP/GRID-Arendal.

Walling,D.E.,2002.Recentadvancesintheuseofenvironmentalradionuclidesinsoilerosion investigations. In: Proceeding of International Symposium on NuclearTechniques in Integrated Plant Nutrients, Water and Soil Management.InternationalAtomicEnergyAgencyPublicationIAEA-CSP-11/C.279–301.

Walling,D.E.,2003.Usingenvironmental radionuclidesas tracers insedimentbudgetinvestigations. International Association of Hydrological Sciences Publication No.283:57B78.

Walling, D.E. and He, Q., 1997. Use of fallout 137Cs in investigations of overbanksedimentdepositiononriverfloodplains.Catena,29.263–82.

Walling, D.E., He, Q., 1999. Improvedmodels for estimating soil erosion rates fromcesium-137measurements.JournalofEnvironmentalQuality28(2).611–622.

Walling,D.E.,He,Q., Appleby, P.G., 2002. Conversionmodels for use in soil-erosion,soil-redistributionandsedimentationinvestigations.Chapter7.In:Zapata,F.,(Ed.),Handbook for the assessment of soil erosion and sedimentation usingenvironmentalradionuclides.KluwerAc.Publ.111–164.

Walling, D.E. and Quine, T.A., 1990. Calibration of caesium-137 measurements toprovide quantitative erosion rate data. Land Degradation Rehabilitation 2. 161–175.

Walling, D.E. and Quine, T.A., 1991. Use of 137Cs measurements to investigate soilerosiononarablefieldsintheUK:potentialsapplicationsandlimitations.JournalofSoilScience42.147-165.

Walling, D.E. and Quine, T.A., 1992. The use of caesium-137 measurements in soilerosionsurveys,ErosionandSedimentTransportMonitoringProgrammesinRiverBasins,ProceedingsoftheOsloSymposium,August1992,IAHSPubl.no.210.143-152.

Walling, D.E. and Quine, T., 1993. Use of Caesium-137 as a Tracer of Erosion andSedimentation: Handbook for the Application of the Caesium-137 Technique.DepartmentofGeography,UniversityofExeter.196p.

62

Walling, D.E. and Quine, T.A., 1995. The use of fallout radionuclides in soil erosioninvestigations. In: Nuclear Techniques in Soil-Plant Studies for SustainableAgriculture and Environmental Preservation, International Atomic Energy AgencyPublicationST1/PUB/947.

Wallbrink,P.J,Walling,D.E,He,Q.,2002.HPGeGammaSpectrometry. In: Zapata, F.,(Ed.). Handbook for the Assessment of Soil Erosion and Sedimentation usingEnvironmentalRadionuclides.KluwerAc.Publ.67–95.

Wischmeier,W.H. and Smith, D.D., 1958. Rainfall energy and its relationship to soilloss.TransactionsofAmericanGeophysicalUnion39(2).285–291.

Wise, S.W., 1980. Caesium-137 and lead-210: a review of the technique andapplication to geomorphology, pp. 109–127. In: Cullingford, R.A., Davidson, D.A.,Lewin,J.(eds.),Timescalesingeomorphology,Wiley.

Word Health Organization, 1986. Updated background information on the nuclearreactor accident in Chernobyl, URSS (Copenhagen, 12 June 1986); Word HealthOrganization Summary report of measurement results relevant for doseassessment(UpdatedRevisionN°7,Copenhagen12June1986).

Zapata,F.,(Ed.),2002.Handbookfortheassessmentofsoilerosionandsedimentationusingenvironmentalradionuclides.KluwerAc.Publ.219p.

Zupanc, V. and Mabit, L., 2010. Nuclear techniques support to assess erosion andsedimentation processes: preliminary results of the use of 137Cs as soil tracer inSlovenia.Dela,33.21–36.

63

Fullpagephotos

Pageiv:RillerosiononcultivatedlandatKigwastudysite,UgandaPagevi:Erosionplotwithmultislotdivisor,Harare,ZimbabwePage8:Schemeofreferencesite,erodedsiteanddepositionsite insmallwatershed,

Nanning,ChinaPage 16: The trainees from Soil Laboratory in Seibersdorf performing core sampling

withtheaidofmotorizedsoilcorerPage26:ThetraineesfromSoilLaboratoryinSeibersdorfcuttingcoresampletodepth

incrementsPage30:Gammadetector,SoilLaboratoryinSeibersdorf,IAEAPage34:ThetraineesfromSoilLaboratoryinSeibersdorfpreparingthesoilsamplesfor

gammaspectroscopyPage40: Exampleof study siteof theTechnical CooperationProjectwhere the 137Cs

methodisused;Theterraceswithsubsistencefamilyfarming,UgandaPage50:TechnicalOfficerduringthefieldworkwithprojectpartnersinUgandaPage64:SmallsizedlandusestructureoffamilyfarmsinKigwa,Uganda

I8211EN/1/11.17

ISBN 978-92-5-130050-3

9 7 8 9 2 5 1 3 0 0 5 0 3