ArticlePDF AvailableLiterature Review

Abstract

There is growing international interest in better managing soils to increase soil organic carbon content to contribute to climate change mitigation, to enhance resilience to climate change and to underpin food security, through initiatives such as international “4p1000″ initiative and the FAO's Global assessment of soil organic carbon sequestration potential (GSOCseq) programme. Since soil organic carbon content of soils cannot be easily measured, a key barrier to implementing programmes to increase soil organic carbon at large scale, is the need for credible and reliable measurement/monitoring, reporting and verification (MRV) platforms, both for national reporting and for emissions trading. Without such platforms, investments could be considered risky. In this paper we review methods and challenges of measuring SOC change directly in soils, before examining some recent novel developments that show promise for quantifying SOC. We describe how repeat soil surveys are used to estimate changes in SOC over time, and how long‐term experiments and space‐for‐time‐substitution sites can serve as sources of knowledge and can be used to test models, and as potential benchmark sites in global frameworks to estimate SOC change. We briefly consider models that can be used to simulate and project change in SOC and examine the MRV platforms for soil organic carbon change already in use in various countries / regions. In the final section, we bring together the various components described in this review, to describe a new vision for a global framework for MRV of soil organic carbon change, to support national and international initiatives seeking to effect change in the way we manage our soils.
Glob Change Biol. 2019;00:1–23.    
|
 1
wileyonlinelibrary.com/journal/gcb
Received:13July2019 
|
  Accepted:22Augus t2019
DOI : 10.1111/gcb .1481 5
RESEARCH REVIEW
How to measure, report and verify soil carbon change to realize
the potential of soil carbon sequestration for atmospheric
greenhouse gas removal
Pete Smith1| Jean‐Francois Soussana2| Denis Angers3| Louis Schipper4|
Claire Chenu5| Daniel P. Rasse6| Niels H. Batjes7| Fenny van Egmond7|
Stephen McNeill8| Matthias Kuhnert1| Cristina Arias‐Navarro2| Jorgen E. Olesen9|
Ngonidzashe Chirinda10| Dario Fornara11 | Eva Wollenberg12| Jorge Álvaro‐Fuentes13|
Alberto Sanz‐Cobena14| Katja Klumpp15
1InstituteofBiological&EnvironmentalSciences,UniversityofAberdeen,Aberdeen,UK
2INRA ,ParisCedex07,France
3AgricultureandAgri‐FoodCanada,Quebec,QC,Canada
4EnvironmentalResearchInstitute,UniversityofWaikato,Hamilton,NewZealand
5INRA ,AgroParisTech.,Thiverval‐Grignon,France
6NorwegianInstituteofBioeconomyResearch(NIBIO),Ås,Norway
7ISRIC–WorldSoilInformation,Wageningen,TheNetherlands
8ManaakiWhenua–LandcareResearch,Lincoln,NewZealand
9DepartmentofAgroecology,AarhusUniversity,Tjele,Denmark
10InternationalCenterforTropicalAgriculture(CIAT),Cali,Colombia
11Agri‐FoodandBiosciencesInstitute,Belfast,UK
12CGIARCCAFSProgramme,UniversityofVermont(UVM),Burlington,VT,USA
13SoilandWaterDepartment,SpanishNationalResearchCouncil(CSIC),Zaragoza,Spain
14ResearchCenterfortheManagementofEnvironment alandAgriculturalRisks(CEIGRAM),UniversidadPolitécnicadeMadrid,Madrid,Spain
15INRA ,VetA gro‐Sup,UCA,ClermontFerrand,France
ThisisanopenaccessarticleunderthetermsoftheCreativeCommonsAttributionLicense,whichpermitsuse,distributionandreproductioninanymedium,
providedtheoriginalworkisproperlycited.
©2019TheAuthors.Global Change BiologypublishedbyJohnWiley&SonsLtd
Correspondence
PeteSmith,InstituteofBiological&
EnvironmentalSciences,Universityof
Aberdeen,23StMacharDrive,Aberdeen
AB243UU,UK.
Email:pete.smith@abdn.ac.uk
Funding information
AGRISOST‐CM,Grant/AwardNumber:
S2018/BAA‐4330;MinisteriodeEconomia
yCompetitividad,Grant/AwardNumber:
AGL2017‐84529‐C3‐1‐R;NUEVA;
GlobalResearchAllianceonAgricultural
GreenhouseGases;REMEDIA;Danish
MinistryofClimate,EnergyandUtilities,
Grant/AwardNumber:SINKS2;New
ZealandA griculturalGreenhouseGas
ResearchCentre;EuropeanUnion,Grant/
Abstract
There is growing international interest in better managing soils to increase soil
organiccarbon(SOC)contenttocontributetoclimatechangemitigation,toenhance
resiliencetoclimatechangeandtounderpinfoodsecurity,throughinitiativessuchas
international‘4p1000’initiativeandtheFAO'sGlobalassessmentofSOCsequestra
tion potential (GSOCseq) programme. SinceSOC contentofsoils cannot be easily
measured,akeybarriertoimplementingprogrammestoincreaseSOCatlargescale,
istheneedforcredibleandreliablemeasurement/monitoring,reportingandverifica‐
tion(MRV)platforms,bothfornationalreportingandforemissionstrading.Without
suchplatforms,investmentscouldbeconsideredrisky.Inthispaper,wereviewmeth‐
odsandchallengesofmeasuringSOCchangedirectlyinsoils,beforeexaminingsome
recent novel developments t hat show promise for quantif ying SOC. We describe
2 
|
   SMITH eT al.
AwardNumber:774378;DEVIL ,Grant/
AwardNumber:NE/M021327/1;Soils‐R‐
GRREAT,Grant/AwardNumber:NE/
P019455/1;CGIARTrustFund;UKERC
1 | INTRODUCTION
Soilorganiccarbon(SOC)representsastockofaround1,500–2,40 0
GtC(~5500–8800GtCO2)inthetopmetreofsoilsglobally(Batjes,
1996;Sanderman,Hengl,&Fiske, 2017). The lowerestimate in the
range isapproximately threetimesthe stockof carbon (C)in veg
etation andtwice thestock of Cin the atmosphere (Smith, 2012).
SmallchangesinCstockscanthereforehavesignificantimpactson
theatmosphereandclimatechange.Since the onset of agriculture
around 8 ,000 years a go (Ruddiman, 20 05), soils have lost aro und
140–150GtC(~510–550Gt CO2; Sandermanetal.,2017)through
cultivation.Itisknownthatbestmanagementpracticescanrestore
someatleastsomeofthislostcarbon(Laletal.,2018),soithasbeen
suggestedthatsoi lCsequestrati oncouldbeasignif ica ntgreenhouse
gas(GHG) removal strategy(also called negative emissiontechnol
ogy,orcarbondioxideremovaloption;Smith,2016).Globalestimates
ofsoilCsequestrationpotentialvaryconsiderably,butarecentsys
tematicreviewbyFussetal.(2018)suggestsanannualtechnicalpo
tentialof2–5GtCO2/year.Estimates ofeconomicpotentialsareat
thelowerendofthisrange(Smith,2016;Smithetal.,2008).
Anin c o m p let e u n d ers t and i n g o n h owSO C c h a n gesa r e i n f lue n c e d
byclimate,land use, management and edaphic factors (Stockmann
etal., 2013),adds complexity to designing appropriate monitoring,
repor ting and verifi cation (MRV) p latforms. Fo r instance, pr ocess‐
le velk nowl e dgeo nho wthes evar i able sinf l uencechan gesi nCs t ocks
andfluxesremainsincomplete(Bispoetal.,2017).Furthermore,the
reversibilityofCsequestration,whenpracticesthatretainCarenot
maintained,orduetoclimatevariabilityorclimatechange,increases
uncertaintyinthetimeframesneededtomonitorSOCenhancement
activities (Rumpel et al., 2019). In addition, the large background
stocks,inherentspatialandtemporalvariabilityandslowsoilCgains
make the dete ction of shor t‐term changes (e.g. 3–5 years) in S OC
stocks a nd the design of re liable, cost‐ef fective and e asy to apply
MRVplatformschallenging(Post,Izaurralde,Mann,&Bliss,1999).
Smithetal.(2012)describedaframework,buildingonavailable
models,data sets andknowledge, to quantifythe impactsof land
useandmanagementchangeonsoilcarbon.Thatpaperconcluded
bypresentingafuture visionfor aglobalframeworktoassess soil
carbo n change, based on a co mbination of math ematical mod els,
spatialdatatodrivethemodels,short‐andlong‐termdatatoevalu
atethemodels,andanetworkofbenchmarkingsitestoverifyesti
matedchanges.Here,wereviewthenewknowle dgesincethen,and
fu r t herd eve l opt h isv isio nint hel ight oft h ene e dto prov idec redi ble
and robust MRVcapabilitiesto support the growing International
andNationalinitiatives to increase SOC, suchasthe International
‘4p1000’initiative(Chabbietal.,2017;Rumpeletal.,2018,2019).
Wefocus onmethodstomeasure and/or estimateSOCchange,
butthesemeasurement/estimationmethodsalsoformthebasisof
howchangesinSOCcanbemonitored and reportedatplo t tona t i onal
(andevenglobal)scales,andhowreportedchangescouldbeverified.
Webegin byreviewing the methods and challenges ofmeasuring
SOCchangedirectlyinsoils(Section2),beforeexaminingsomere
centdevelopments thatshow promisefor quantifyingSOC stocks
(and thereforechange) using flux measurements,non‐destruc tive
field‐based spectroscopic methods and thepossibilityinfutureof
estimatingSOCchange throughearthobservation/remote sensing
(Sec tion3).Wet henreviewhowr epeatso ilsur veysareusedtoesti
mateterritorialchangesinSOCovertime(Section4),andhowlong‐
termexperiments and space‐for‐timesubstitution sites can serve
as sources of k nowledge and c an be used to tes ting models , and
as potentia l benchmark s ites in global pl atforms to es timate SOC
change(Section5).Section6summarizesrecentreviewsonmodels
availableforsimulatingand predicting changeinSOC,afterwhich
Section7 describes MRV platforms for SOC changealready inuse
invariouscountries/regions.Wefinishthereview(Section8)byde
scribingan ewvisionforaglobalfram eworkforMRVofSOCchange
tosupportnationalandinternationalinitiatives.
2 | DIRECT MEASUREMENT OF SOC
STOCK CHANGES
AccurateestimatesofSOCstocksrelystronglyonbaselineSOCval
ues,whicharedeterminedbyphysicalsamplingandsoilCcontent
how repeat soilsurveys areusedto estimate changesin SOC overtime,and how
long‐termexperimentsandspace‐for‐timesubstitutionsitescanserveassourcesof
knowledgeandcanbeusedtotestmodels,andaspotentialbenchmarksitesinglobal
frameworkstoestimateSOCchange.Webrieflyconsidermodelsthatcanbeusedto
simulateandprojectchangeinSOCandexaminetheMRVplatformsforSOCchange
alreadyinuseinvariouscountries/regions.Inthefinalsection,webringtogetherthe
various componentsdescribed inthis review,todescribe a newvisionfor a global
frameworkforMRVofSOCchange,tosupportnationalandinternationalinitiatives
seekingtoeffectchangeinthewaywemanageoursoils.
KEYWORDS
measurement,monitoring,MRV,reporting,soilorganiccarbon,soilorganicmatter,verification
    
|
 3
SMITH eT a l.
measurements.Thisapproachtraditionallyinvolves thequantifica‐
tionof(a)fineearth(<2mm)andcoarsemineral(>2mm)fractionsof
thesoil;(b)organiccarbon(OC)concentration(%) ofthefineearth
fraction;and(c)soilbulkdensityorfineearthmass(FAO,2019a).In
someinstances,suchasgrasslandsorforestsoils,itmaybeofinter‐
esttoquantifyandaccountforthecoarsefractionofbelowground
OC(FAO,2019a).Thechallengeremainstoaccuratelyestimatethe
rockcontentofsampledsoils,whichcansignificantlyaffectsoilbulk
density(Page‐Dumroese,Jurgensen,Brown,&Mroz,1999;Poeplau,
Vos, & Don, 2017; Throop, Archer, Monger, & Waltman, 2012).
Changes inmanagement thatinfluence carbon content alsoaffect
thebulkdensityofthesoil(Haynes&Naidu,1998),andtherebythe
amount of soil thatissampled within a givensampling depth. It is
thereforerecommendedtousean‘equivalentmassbasis’approach
whencomparingSOCstocksacrosslandusesanddif ferentmanage
ment regimes(Ellert& Bettany,1995;Upson, Burgess, &Morison,
2016;Wendt&Hauser,2013).
Directmeasurementsalsorelyonappropriatestudydesignsand
samplingprotocolstodealwithhighspatialvariabilityofSOCstocks
(Minasny et al., 2017).Toreduce potentialsourcesoferror in SOC
stockestimationandminimizetheminimumdetectable difference
(i.e.thesmallestdifferenceinSOCs to ckthatcanbedetectedassta
tisticallysignificant between twosamplingperiods;FAO,2019a),a
largenumberofsoilsamplesisoftenrequired(Garten&Wullschleger,
1999;Vanguelovaetal.,2016).Sufficientsamplingdepthisacrucial
factor forproperly evaluating changes in soil C content(IPCC rec‐
ommendsaminimumdepthof30cm).Severallong‐termagronomy
experimentssufferfromanincreaseinploughingdepthduringmore
recentdecades, as agriculturalmachinery becamemore powerful.
Insuff icient inform ation on histor ical sampli ng depth can al so add
uncertaintytotheresults.
Several methods for increasing soil C content require deeper
samplingfor confirmingtheexpectedeffect.Thepositiveeffectof
no‐tillonsoilCcontentmeasuredinthesurfacesoilmaynotbeap
parentwhenmeasuringto60 cmdepth(Angers&Eriksen‐Hamel,
2008 ; Blanco‐Canq ui & Lal, 2008 ). Crops with deep ro ot pheno‐
typesareconsideredapromisingmethod to increase Csequestra‐
tioninsoils(Paustianetal.,2016),thoughdemonstratingtheireffect
requiresdeepsoilsampling.Deepersoilsampling(100cm)isrecom
mended (FAO,2019a),butoftenrequiresspecificmachinery and is
cos tly.
Costs associated with collecting, processing and storing soil
samplesandCcontentmeasurementsusing,forexample,common
drycombustionmethods(Nelson&Sommers,1996)canmakelarge
scaledirect measurementsofsoilSOCstocksprohibitivelyexpen
sive. It was estimated that to detect meaningful changes in soil C
stocks across forest ecosystems in Finland (i.e. 3,000 plots at the
nationalscale)mightcost4millionEuroforonesamplingcampaign
(e.g.baseline measurement from 1year)andthenagainforthefol‐
lowing sampling interval (e.g. 10 years later; Mäkipää, Häkkinen,
Muukkonen, & Peltoniemi,2008). Thus, there is the needto eval‐
uate these costs against the value of soil C sequestered (Mäkipää
etal.,2008;Smith,2004b)andsearchfortrade‐offsbetweencosts
involvedand alternative SOC estimationmethodsincludingdiffer
entmodellingapproaches.
A combinat ion of direct measu rements (at the pl ot scale) and
modelling(atlarger spatial scales) can greatly helpdefiningthe ef‐
ficacy of different landmanagement practices in enhancing soil C
sequestrationandhasbeenusedforestimatingsoilCchangeinna‐
tionalGHGinventoryplatforms(e.g.VandenBygaartetal.,2008).It
is,therefore,crucialtoevaluatethecost‐effectivenessofmeasuring
andsequesteringCacrossdifferent landuses andsocio‐economic
conditions(Alexander,Paustian,Smith,&Moran,2015).
3 | NOVEL METHODS OF MEASURING
SOC CHANGE
3.1 | Inferring SOC stock changes from flux
measurements
Analternativetorepeatedmeasurementsistodrawupafullcarbon
budget.Thisindirectapproachaccountsfortheinitialuptakeofcar
bon through photosynthesis (gross primary production),its subse‐
quentpartiallossesthroughrespiration(soil,plantandlitter)togive
net ecosystem excha nge (NEE) or net ecosys tem production and
furtherCinputs(organic fertilization)and outputs (harvest)toand
fromthesystem(seeSmith,Lanigan,etal.,2010;Soussana,Tallec,&
Blanfort,2010).ThemeasurementsofthenetbalanceofCfluxesex‐
changed(i.e.estimatingNEE)canbeachievedbychambermeasure‐
mentsorbytheeddycovariance(EC)method(e.g.Baldocchi,2003).
Duringr ecentdec ades ,es timatesofCseques tratio nfromflu xmeas
urementshavebeenreportedtobecomparativelyuncertaindueto
(a)necessar yassumptionsassociatedwithdataprocessing(e.g.foot
print,spectralcorrections,i.e.Aubinet,Vesala,&Papale,2012);the
fact th at (b) this metho d is a point‐in‐space meas urement; and (c)
netchangesinsoilCpoolsarerelativelysmallcomparedtoCstored
in biomass a nd litter whe n measured over sh ort time pe riods (i.e.
<5years).
Despitethis,recentdevelopmentsininstrumentation(analyser
performance andset‐ups, e.g. Rebmann et al.,2018), dataacquisi‐
tionandprocessing(i.e.dataloggers,software,QA/QCchecks)have
greatly improvedthe reliabilityofestimates(e.g.Fratini& Mauder,
2014).Furthermore,harmonizednetworksoflong‐termobservation
sites, created to provide access to standardized dataand to quan‐
tifytheeffectivenessofcarbonsequestrationand/orGHGemission
at Europea n (Integrated C arbon Obser vation Syste m, ICOS; Franz
etal.,2018)andglobalscale(FLUXNETglobalnetwork,e.g.Baldocchi,
Housen,&Reichstein,2018;Figure1),havegreatlyreduceduncer‐
taintiesinfluxandsupplementarymeasurements.Moreover,ongo
inganalysesonpeculiaritiesofflux measurement likelytoincrease
uncertaintiesinfluxmeasurements,suchasintegration of(moving)
point sources, that is, grazing animals (Felber, Münger, Neftel, &
Ammann,2015;Gourlez de la Motteetal.,2019),ditches(Nugent,
Strachan, Strack, Roulet, & Rochefort, 2018) and fallow periods,
havebeenstudiedthoroughlyandhaveallowedroutinedataanaly‐
sestobeupdated(e.g.Sabbatinietal.,2018).
4 
|
   SMITH eT al.
Concerning the comparison between C sequestration deter
mined via the EC technique (i.e. full C balance) and soil C stock
changes , some studies have s hown poor agreem ent (Jones et al.,
2017),butanumberof studies have showncomparable estimates,
whenappliedfortimeframes>10yearandwithsoildataincluding
atleastbothtopandmediumsoildepths(i.e.0–60cm;e.g.grassland:
Leifeld,Ammann,Neftel,&Fuhrer,2011;Skinner&Dell,2014;Stahl
etal.,2017;cropland:Emmeletal.,2018;Hoffmannetal.,2017;for
est:Ferster,Trofymow,Coops,Chen,&Black,2015).CouplingofEC
with soil Cstock change studies hasbecomeafavouredapproach
tounderstandbothshort‐andlong‐termeffectsofprincipaldrivers
(e.g.management,climate)onecosystemfunctioning(i.e.Eugster&
Merbold,2015),in natura measurementandmodellingapproaches
(e.g.Beeretal.,2010;Besnardetal.,2018;Williamsetal.,2009).
3.2 | Spectral methods for measuring SOC stocks
NewspectralmethodsformeasuringSOCconcentrationandstocks
arerapidlybecomingavailablefordirectpointmeasurementsinfield
andinthelab,butalsoformeasurementofpatternsatlargerscales
across landscapes and regions. Each comeswith aspecific associ‐
atedaccuracyandcost(Bellon‐Maurel&McBratney,2011;England
&Viscarra Rossel, 2018;Nayak et al., 2019). A smart combination
oftheseandmoretraditionalmethodscaneitherbringdowncosts
(Nocita e t al., 2015), provide more e xhaustive spati al patterns of
SOC stocks (Aitkenhead, 2017; Rosero‐Vlasova, Vlassova, Pérez‐
Cabello,Montorio,&Nadal‐Romero,2019)orprovideindicationsfor
changeinstocks(Lietal.,2018;Zhao,Ye,Li,Yu,&Mcclellan,2016).
ThemethodsformeasuringSOCconcentrationmainlyrelyonthe
reflectance oflight on soilintheinfraredregion. Theorganic bonds
and minerals in the soilabsorblight atspecific wavelengths, result
ingisasoilcontent‐specificabsorbanceorreflectancespectrum.This
spectrumismeasuredwithhighlevelofspectraldetail(hyperspectral,
ofteninthelab)orlimitedlevelofdetailinwiderbands(multispectral,
often fro m satellites or c heaper fiel d instrume nts). Using a stat isti
calmodelbasedonaspectrallibrary,thesoilcarbonpercentagecan
bepredictedfrom spectralmeasurements of theunknown samples.
Thespectrallibraryisderivedfromsamplesonwhichsoilproperties
havebeendeterminedbytraditionallaboratorymethods,suchasdry
combustion, alongside reflectance measurements. Relevant wave
lengthsforsoilandSOCaremainlyinthemid‐(4,000–600cm−1)and
thenear‐orshort‐waveinfraredregion(2,000–2,500nm).Otherkey
soilproperties can also be simultaneouslydeterminedifpresentin
the spectral libraries, including fractions of OC and vulnerability of
soilcarbontoloss(Baldock,Beare,Curtin,&Hawke,2018;Baldock,
Hawke,Sanderman,&Macdonald,2013),soiltexture,pHandothers
(Stenberg,ViscarraRossel,Mouazen,&Wetterlind,2010),whichcan
beusedtoinformmodellingapproaches.Partialleastsquaresregres‐
sion(PLSR) isastatisticalmethodthatiscurrentlymostwidelyused
topredict soilpropertiesfrom spectra. These machine learningap
proaches (e.g. Cubist, Random Forests, SupportVector [regression]
Machinesandothers)arerapidlydeveloping,andnewtechniquesare
becomingavailable,currentlyreferredtoasdeeplearning(Padarian,
Minasny, & Mcbra tney, 2019) and memo ry based lea rning (Dangal, 
Sanderman, Wills, & Ramirez‐Lopez, 2019; Ramirez‐Lopez et al.,
2013).Thesetechniques,suchaslocallyweightedPLSR,uselocalcal
ibrationsbasedonspectrallysimilarsubsetsofaspectrallibrary.This
willlikelyleadtoconsiderableimprovement,reducingtheprediction
errors.Thisdoesnotresolvetheinherentlaboratory measurement
uncertaintiesassociatedwithbothreferenceandspectraldata.
Standardizationofreferencelaboratorymethods,spectralmea
surementsandsoildataexchangetosomeextentnegatestheseis‐
sues,andtheyareaddressedinseveralinternationalco‐operations,
oneofwhichisPillar 5 of theGlobalSoil Partnership (GSP,2017).
Ifstandardizationandcalibrationtransferchallengescanbesolved,
combiningspectrallibrariescanprovideavastdataresourcefornot
onlylocalbutalsomoreregionalandglobalSOCanalyses(England&
ViscarraRossel,2018;ViscarraRossel,Behrens,etal.,2016;Viscarra
Rossel,Brus,Lobsey,Shi,&Mclachlan,2016).
LaboratorycostscouldbereducedbyusingFouriertransformmid‐
infrared(MIR)diffusereflectancespectroscopyforestimationoftotal
carbon,OC,claycontentandsandfraction(ViscarraRossel,Walvoort,
Mcbratney, Janik, & Skjemstad, 2006; Wijewardane, Ge, Wills, &
Libohova, 2018). Several commercial laboratories use near‐infrared
FIGURE 1 Mapoffluxtowersand
availabletimeseriesworldwide
Source:Fluxnet,2019
    
|
 5
SMITH eT a l.
(N I R )fort h ispur p o s ebuto n c eas u f f i c ients p e c t r a llib r a r yo rcalibr a t i on
setiscompiled, MIR outperformsNIR(Reeves,2010; ViscarraRossel
etal.,2006;Vohland,Ludwig,Thiele‐Bruhn,&Ludwig, 2014).In such
studie s or applicatio ns, bigger libr aries are spike d or subselect ed to
buildlocal(spectralorgeographical)predictionmodelsusingmachine
learningtechniques (Janik, Skjemstad, Shepherd,&Spouncer,2007).
Samplepreparationisverysimple(dry,sieveto<2mm,finegrind(Soil
Survey Staff,2014)andafter alibrary is built,the measurementsare
fastandinexpensive,andcanassessallofthelistedpropertiesatthe
sametime(Nocitaetal.,2015).
These spectrallibrariescanalsobeused tocalibratefieldspec
trom eters,al tho ughaccur acywilloftenb elower,m ost lyduetomois
ture and surfaceroughness of thesoil. Higher cost in situsystems
areavailableforbothNIRand MIR (Dhawaleetal., 2015;Hutengs,
Lu d wig,J u ng,Ei s ele,& Vo h land,2 018 ).A l t ernat i vesa r eche a pin‐f ield
NIRspectrometersforpointmeasurements(Tang,Jones,&Minasny,
2019)which tend tohave low(er)accuracies dueto hardware con
straintsandwhichmayhavebias.On‐the‐gosystemswith2–5wave
lengthsareonthemarketaswellaspenetrometerswithvisibleand
near‐infrared reflectance spectroscopy (VNIR), which also provide
ameasurefor penetration resistanceor compacted soil (Ackerson,
Morgan , & Ge, 2017; Al‐Asadi & Mouazen , 2018; Poggio, Brown,
& Bricklemyer, 2017; Wetterlind, Piikki, Stenberg, & Söderström,
2015).A final possibilityis a coresampler which measures theex
tractedsoil core in field withVNIRand activegammaradiationfor
(total)bulkdensity(Lobsey&ViscarraRossel,2016).
AnimportantpropertyforcalculatingSOCstocksissoilbulkden
sitywhichisdifficulttomeasureaccuratelyinfield(Bellon‐Maurel&
McBratney,2011).A method used in a number ofset‐ups is gamma
attenuation.Thiscanbemeasuredontheextractedsoilcore(England
&ViscarraRossel,2018;Lobsey & ViscarraRossel, 2016) or directly
inthesoil(Jacobs,Eelkema,Limburg,&Winterwerp,2009).Withthis
technique,the attenuationby matter of gammaradiationoriginating
fromasmallradioactivesourceismeasuredoveraknownvolumebe
tween source and detector.Thematter inthis caseconsistsofboth
soilandmoisture.ThevolumeissimulatedusingMonteCarlosimula
tions.Thisprovidesameasureofdrybulkdensityaftercorrectionfor
moisturecontentasmeasuredforinstancewithatimedomainreflec
tometr y(Jacobse ta l.,20 09)orVNIR(Lobsey&Visc arr aRos sel,2016).
Thebenefitof thesetechniquesisthepossibilitytoacquiremore
samp lesan d/ormorein‐fie ldmea surem ents,allowingausertoaddress
thepotentialofcarbonsequestrationof the soiladequately.Some of
thesetechniquesaremostsuitablefordescribingthespatialdistribu
tionofsoilcarbon,whileothersaresuitableforquantitativeestimates
ormonitoring(intime,allowingtheimpactsofmanagementonsoilcar
bontobedetected).Choicescanbemadebasedoncostandrequired
accuracyofthepurpose(valueofinformationordecisionanalysis).
Atlarger scales, remote sensingoffersaddedpossibilities.This
caneitherbebyrelatingUAV,airplaneorsatellitedatadirectlytosoil
properties,orbyinferringchangesinSOCbyvegetationchanges,or
byusingremoteimageryasacovariateindigitalsoilmappingofSOC.
Directinterpretationcanbeperformedonhyperspectralimageryin
combinationwithspectrallibrariesfordirectquantificationofbare
soil patterns (top1 cm; Gomez, Lagacherie,&Bacha, 2012; Jaber,
Lant, & Al‐Qinna,2011),or by usingmultivariateimageryfor map
pingbaresoilpatternsasindicationofSOCorsoilclassdifferences
either using raw or enhanced imagery su ch as by multi‐temporal
composites(Galloetal.,2018;Roggeetal.,2018).
Changes in ve getation pat terns visible in r emote imagery c an be
usedtodetect(changesin)landuseand thusinfersoilpropertiesand
SOCchange.Analysisofland‐usechange,netprimaryproductivityand
SOCstocksareinstrumentalforidentif yinghot spotsofSOCseque str a
tionpotential(Caspari,Lynden,&Bai,2015;vanderEschetal.,2017).
The thirdoptionis to usesatelliteimagery productsas covari‐
atesindigitalsoilmapping,wheretherelationbetweensoilproper‐
tiesandsatelliteinformationisusedtopredictSOCmapsatvarious
depths using point observations and satellite imagery products
(Hengletal.,2017;McBratney,MendonçaSantos,&Minasny,2003;
Minasny&McBratney,2016).
Remote sens ing offers a range of p ossibilities, de tail and spa
tialscalesthatarenot feasiblewithpointmeasurementsalone(Ge,
Thomasson,& Sui,2011;Mulder,Bruin,Schaepman,&Mayr,2011).
Thatsaid, acombinationofremoteandinsituorpointdatawillre
main nece ssary to der ive high resolu tion and accur ate SOC maps.
Ap artfromtheli mit edp enetrat iondepth(to p1cmw hil easoi lpr ofi le
wouldbedesirable),thisisalsoduetothefactthatinmanyregions,
baresoilisnevervisible,orareasaretoooftencoveredinclouds.At
thesametime,thehightemporalfrequencyandhighspatialresolu
tionofremoteimageryofferanunprecedentedpossibility to study
andmonitor space–timedynamicsofSOCchangeifusedincombi
nationwith(long‐term)monitoringstations(Chabrillatetal.,2019).
4 | REPEATED SOIL SURVEYS—NATIONAL/
SUB NATIONAL
Repeatsoilsamplingprogrammeshavebeenconductedinanumberof
countries,suchasEnglandandWales(Bellamy,Loveland,Bradley,Lark,
&Kirk,2005;Kirkbyetal.,2005),Denmark(Heidmann,Christensen,&
Olesen,2002;Taghizadeh‐Toosi,Olesen,etal.,2014),Belgium(Sleutel,
Neve,&Hofman,2003) and NewZealand (Schipper et al.,2014—see
below).Theserelyonresamplingofpreviouslysampledlocationsafter
va r ying peri o ds.Adva ntage sar etha tre p eat samp l ing sche mesm easu re
actualsoilcarboncontentsoverlargespatial scalesandoverlongpe
riods(Bellamyetal.,2005),butthemaindisadvantageisthatland‐use
change and l and manageme nt between sa mpling perio ds are mostly 
unknown,makingattributionofanyobservedchangesinsoilcarbonto
specificdrivers (such as managementorclimate change)verydifficult
(Sm it hetal.,2007).Insomec as es,recordsoflanduseandmanagement
havebeenavailableallowingtheeffectofmanagementchangestobe
assessedforbetterverificationofmodellingapproachestoquantifying
SOCstockchanges(Taghizadeh‐Toosi,Olesen,etal.,2014).
Resampling of soil survey sites originally sampled in the
1970s–1990sinNewZealandhasplayedanimportantroleinidenti
fyingchangesinsoilcarbonstocksingrazedpastures(Schipperetal.,
2014).Thedif ficultywiththesehistoricalresamplingeffor tswasthat
6 
|
   SMITH eT al.
siteswerenotchosenwithnationalsurveypurposesinmind,sotheir
representativenesswasquestionable.Additionally,samplingefforts
were not carriedout uniformlyover spaceand time, soresampling
waspotentially confounded by theeffects of soiltype, climateand
otherfactors.However,thesedatahavebeencentraltodevelopment
andsubsequentimplementationofmorerobustsamplingdesignsof
grazed lands. Alongside,resampling of siteimpacts of management
practices on carbonstock hasbeen exploredthrough the sampling
ofadjacentlong‐termmanagementpractices(e.g.Barnett,Schipper,
Taylor,Balks,&Mudge,2014;Mudgeetal.,2017).
Inthecase of Europe,differencesexist in the availabilityofsoil
surveys a mong countries . As highlighted i n the final repor t of the
ENVASSO project, soil monitoring networks are much denser in
northernandeasternEuropeancountries compared withcountries
located in the southern part of the continent (Kibblewhite et al.,
2008). For example, countries such as France, Sweden or Poland
maintai n systematic soil m onitoring syste ms at national level wit h
different density of monitoring sites and sampling frequencies.
In the cas e of France, diffe rent soil monito ring system le vels exist
whichoper atestoeithe rforestandnon‐forestareas.TheSoilQuality
Monitoring Network was created 20 years ago for non‐forested
area s,coveringthem ainlanduse sinFranceina16×16k mgrid(King ,
Stengel, Ja magne, Le Bas, & A rrouays, 20 05). Similar ly,in Swe den,
soilmonitoringisperformedattwogeographicallevels(nationaland
regional ) and with diffe rent levels of applic ation: forest l and, inte
grated monitoring (areaswith minorimpactofforest management),
intensivemonitoringplots(223forestplots)andarablelandmonitor
ing(Olsson,2005).Polandhasalsodifferentsoilmonitoringsystems
forforestand cropland soils. For thecase of croplands, monitoring
soils startedin1994andsince then soilshavebeen sampledevery
8yearswithdifferentsoils'propertiesmeasured(Białousz,Marcinek,
Stuczyński, & Turski, 2005). In Denmark, soils are sampled every
8–10yearsto1mdepthonaregular7kmgridcoveringbothagricul
turalandforestsoils(Taghizadeh‐Toosi,Olesen,etal.,2014).
In contrast, EU Mediterranean countries such as Italy, Spain or
Greece a re examples of Euro pean regions wh ere systemati c national
soil monitoringsys temsare underdeveloped or non‐existent, despite
therisksofSOC losses, andsoil erosion eventsresulting fromacom
bination of cropmanagement and regional impacts of climate change
(Trnkaetal.,2011).Forexample,inthecase ofItaly,thereisnomoni
toring system, but thereis willingness to developit. In Spain, over the
last 20 years, two independent soil national inventories have been
performed;onetoassesssoilerosionandtheothertoassessoilheavy
metalpollution(Ibáñez,SánchezDíaz,deAlba,LópezArias,&Bioxadera,
2005). However,theinventories have notbeen linkedand thereis no
firmscheduleforfutureresamplingyetinplace.
5 | LONG‐TERM EXPERIMENTS OF SOC
CHANGE
Since changes in bulk soil carbon occur slowly (Smith, 2004a),
long‐termmeasurementsarerequiredtoshowtherelativelysmall
change aga inst the large back ground carbo n stock. To this end,
long‐ter mfieldexper iment sex istinvariouspartsoftheworld,w ith
somedatin gfromthe19thcentury.A lth oughm anyofth eseex peri
ment swer eo riginallysetu ptoexaminetheef fec tsofmanagem ent
(oftenfertilization)oncrop or grassyield,manyhaveahistoryof
measurements of soil carbon and nitrogen change. Over recent
decades,results from these fieldexperiments have been central
totestingtheaccuracyofmodelsofturnoverofSOC.Asnotedby
Smith et al. (2012),thelong‐termexperiments in various parts of
th eworld exi stedlar gel yin iso lat ionofe achothe r,bu tinthe 1990s,
therewereattemptstobringthevariousexperimentstogetherinto
shared networks (Barnett,Payne, &Steiner,1995),withtwosuch
networ ks focussing on soil C ; the Soil Organic Matter Network
(SOMNET)and EuroSOMNET (themoredetailed European com
ponent of the larger global network) weretwo attempts to cou
ple SOC models with observations from long‐term experiments
(Smith et al., 1997), withtheaimsorboth testingmodelsandthe
sharing, comparing and useof data from across theexperiments
toestimatecarbonsequestrationpotential(Smith,Powlson,Smith,
Falloon,&Coleman,2000).SOMNETlaterevolvedintoanonline,
real‐time inve ntory project w ith a website known as Long‐Term
Soil‐Ecosystems Experiments,which now has collectedmetadata
onwell over200 long‐termsoil experiments Richter,Hofmockel,
Callaham,Powlson,andSmith(2007),withthemetadatacurrently
hosted by the International Soil Carbon Network (iscn.fluxdata.
org/partner‐networks/long‐term‐soil‐experiments/). Smith et al.
(2012)showe dt helocat ionsandpurposeofthes el ong‐termex per
iments.Most(>80%)oftheworld'slong‐termfieldstudiesaddress
agricultural research questions,andmost of the fieldstudiestest
agriculturalquestionsinthetemperatezone.Nonagriculturalsites
andexperimentsinthebioclimaticzonesotherthanthetemperate
regionareunder‐represented(Smithetal.,2012).
Long‐termfieldstudieshaveprovedextremelyvaluableforun
derst anding the lo ng‐term dynamic s of SOC and wider i ssues of
soilsustainability(Richteretal.,2007).IntermsofMRV,thelong‐
term experimentsserveas(a)along‐termrecord of change; (b) a
testbedforSOCmodels;(c)locations wherenewpracticescould
betestedandmeasured;and(d)siteswhereshorterterm(e.g.flux
measurements)couldbetakentobetterunderstandshorterterm
processes. Such experiments could therefore form vital compo
nentsofnationalandinternationalMRVplatformsforSOCchange.
Existing long‐term monitoring sites are extremely valuable but
donot exist in every global region, making a compellingcase for
star ting new long‐term ex perimental/ m onitoring sites in u nder‐
representedregions.
6 | MODELS OF SOC CHANGE
The soilorganic matter (SOM) dynamics canbe described by dif
ferentmathematical formulations(Parton,Grosso,Plante,Adair,&
Luz,2015),aspresentedinTable1,anddifferentmodelapproaches
(Campb ell & Paustian, 2015; M anzoni & Porporato, 20 09). Most
    
|
 7
SMITH eT a l.
common SOM models are compartment models, which use be
tween two and five carbon pools (Falloon & Smith, 2000). While
thestability andcomplexityoftheorganiccompoundsisnotrep
resentedexplicitlyinmodels,itisrepresentedbyvaryingturnover
and resid ence times of OC in d ifferent c arbon pools (Sto ckmann
etal.,2013).Theresidencetimesarecontrolledbythedecayrate
ofthecarbon in the differentpools,whichis usuallydescribedby
the first‐order kinetics (e.g. Falloon & Smith, 2000; Parton et al.,
2015;Paustian,1994).Awiderangeofdifferentmodelsshowthis
structure, eitherasindependentSOMmodel oraspart ofaneco
system model,dynamic vegetation model or a general circulation
model(Campbell&Paustian,2015;Ostleetal.,2009;Partonetal.,
2015).Manzoni andPorporato(2009)identified about250differ
entmodels,buttherearestillnewdevelopments,astherearestill
unresolvedchallenges.
Despite the developmentof different approaches that allow
the measurement of different carbon pools in the models (e.g.
Janik et al., 2007; Skjemstad, Spouncer, Cowie, & Swift, 2004;
Zimmermann,Leifeld,Schmidt,Smith,&Fuhrer,2007),SOCpools
areoften stillinitializedinaspin‐up run(Nemo et al.,2017).This
isapracticalapproachifinformationaboutthefractionationisnot
available,but it relies on ideal assumptions of equilibrium (Smith,
Smith, Monaghan, & MacDonald, 2002) which impacts the re
sults (Bruun & Jensen, 2002).Furtherm ore, the residence times
of most pools exceed the duration of available measurements,
whichmakesthecalibrationandvalidationofthemodelsdifficult
(Campbell&Paustian,2015;Falloon&Smith,2000).Additionally,
notallrelevantproce sses(e.g.priming )arerepresente dinthemod
els (Guen et, Moyano, Peyl in, Ciais, & Jans sens, 2016; Wutzle r &
Reichstein,2013).Recently,therehasbeenadiscussionaboutthe
abilit yofexistingmod elstoreflectch angesintemp erature(Cona nt
etal.,2011;Moyano,Vasilyeva,&Menichetti,2018),whichismost
relevant to simulateclimate changeimpacts (Conantetal.,2011).
Inshort,it is notclear,iftheslower,morestablepoolsgetdiffer
entlyaffectedbytemperaturechanges(e.g.Campbell& Paustian,
2015;Conantetal.,2011).Fortheseandotherpurposes,thereare
an increa sing number of new m odel approa ches and hypot heses
(e.g. Cotrufo, Wallenstein, Boot, Denef, & Paul, 2013; Lehmann
&Kleber,2015;Wieder,Bonan,&Allison,2013;Wutzler,Zaehle,
Schrumpf,Ahrens,&Reichstein,2017).Therefore,long‐termdata
sets(Section5)areneeded totest theperformanceoftheestab
lishedandthenewmodels.
ManyoperationalSOC modelsonlysimulateturnover and de
composition of theSOC pools and the added OC (Toudert etal.,
2018).Thesemodelsthusrelyheavy onproperestimationof car
boninputsinresiduesandorganicamendments(manure,compost,
etc.) as well as o n information on t he biologica l quality of th ese
input s. Most modellin g approaches us ed for inventory pu rposes
rely oninput datafrom harvest residuesor decaying plantparts
andexternal organic amendments. Theplant C inputs aremostly
derivedfrommeasuredagriculturalyieldsusingsimpleallometric
equations,wheretheC inputsisrelatedlinearlyornonlinearlyto
cropyield (Keel,Leifeld,& Mayer,2017).Comparisonofdifferent
publishedapproachesofestimatingCinput ,butusingthesamede
compositionmodel,hasdemonstratedlargeuncertaintiesinsimu
latedchangesinSOC(Keeletal.,2017).Theselectionofallometric
functionsfor estimatingCinput isthereforeacriticalstep inthe
choiceofmodelap proach.Recentresearchhasalsoquestionedthe
appropriatenessofusingsimpleallometricfunctionssuchasfixed
shoot:rootratiosforestimatingCi nput(e.g.Hueta l.,2018).Rathe r
than assuming a fixe d shoot:root ratio, u sing a fixed amount of
belowgroundCinputdependingonsiteandcropmayprovidethe
TABLE 1 ListofdifferentfunctionstosimulatethedecompositioninmodelsfollowingthediscussionofPartonetal.(2015).The
publicationslistedrefertotheexamplemodels.Theabbreviationsdescribethecarbon(C)atthestart(C0)andatacertaintime(t)step
(Ct),thedecompositionrate(k),theMichaelis–Mentenconstant(Km)andthemaximumreactionvelocityfortheprocess(Vm),thecarbon
demandbythemicrobes(X0),theMonodconstant(Kt)andthemaximumgrowthrate(µmax).ThegraphsshowCtinatimeseriesforoneset
ofarbitraryparameters
Approach Equation Graphical relation (C(t)) Example model Publications
Zero‐orderkinetics
Ct
=
C0
kt
First‐orderkinetics
Ct
=C
0
e
kt
RothC,ICBM JenkinsonandRayner(1977),
AndrénandKätterer(1997)
Enzymekinetics
dC
dt=Vm
C
K
m
+
C
CLM,SEAM Wiederetal.(2013),Wutzleret
al.(2017)
Microbialgrowth
dC
dt=𝜇max
C
K
t
+C
C0+X0C
NICA BlagodatskyandRichter(1998)
8 
|
   SMITH eT al.
mostrobustestimate(Hirte,Leifeld,Abiven,Oberholzer,&Mayer,
2018;Taghizadeh‐Toosi,Christensen,Glendining,&Olesen,2016).
This has implications for modelling application where changes in
cropproductivityareamaindriverofCinputs.
7 | WHAT MRV PLATFORMS ARE
CURRENTLY IN USE
AnumberofGHGemissionandsoil carbonchangequantification
schemeshave been developed in various parts of the world. For
example, the Australian Carbon Farming Initiative/Emission re
ductionfundhas guidancerelatingtosamplingandmeasurement
ofSOC andestimating andreportingSOC stock changefor SOC
management projects (Australian Government,2018). In Alberta
inCanada,thereisaConservationCropping Protocol,atoolused
to quantify GHG emission reductions from conservation crop
ping(Alberta Government,2012).Forcertain productionsystems
(e.g.livestock production),FAOhaspublishedguidanceon meas
uring and m odelling soil c arbon stocks a nd stock changes ( FAO,
2019a).Int hissection,weexaminemethodsalreadyinuseincoun
triesparticipating in theGlobal Research AllianceofAgricultural
GreenhouseGases(GRA).
7.1 | Operational soil MRV systems in use in
GRA countries
Wefirst searched the GRA publications library (https://globalrese
archalliance.org/publication‐library/) for operational soil MRV
systems/procedures, giving limited results (e.g. Minamikawa,
Yamaguchi,Tokida,Sudo,&Yagi,2018).Subsequently,wesearched
theWeb‐of‐Scienceusing“((soilANDcarbon)ORsoc)AND((moni‐
toring ORreporting OR verification) OR mrv),” giving 91 potential
sources.Adding the GRAcountry names (56as of October 2018)
totheinitialsearchreducedthis to14papers. Thesestudiescover
partsofacountry(McHenry,2009;Nerger,Funk,Cordsen,&Fohrer,
2017;Steinmannetal.,2016;Wilson,Barnes,Koen,Ghosh,&King,
2010), consider selected agro‐ecosystems or agricultural prac
tices (A llen, Pringle , Page, & Dalal, 2010; d e Gruijter et al. , 2016;
McHenry,2009;Scottetal.,2002;Wu,Clarke,&Mulder,2010),out‐
linethebasisforapossiblenationalsoilmonitoringsystem(Spencer,
Ogle,Breidt,Goebel,&Paustian,2011;Visschers,Finke,&Gruijter,
2007),werediscontinuedduetolackoffunding(Goidts,Wesemael,
& Oost, 2 009; Taghizadeh‐Toosi, Olesen, e t al., 2014; Yagasak i &
Shirato,2014)or,alternatively,concernmeasurementsystemsthat
are in thei r first (Mäkip ää, Liski, Gue ndehou, Malim bwi, & Kaaya,
2002; Nijbroek et al., 2018) or second round (Orgiazzi, Ballabio,
Panagos,Jones,&Fernández‐Ugalde,2018;Spenceretal.,2011).
Much early work has been done in Australia (McKenzie,
Henderson, & Mcdonald, 2002), and in 2014, the Australian
Government approved the first methodology for soil carbon se‐
questration for use at farm level (de Gruijter etal., 2016); recom‐
mended procedures of stratification and sampling, however, may
varybetweencountries(e.g.AustraliaandNewZealand,seeMalone
et al.,2018). Overall,a lack of common procedures between (and
within)countriesaf fectsthesuitabilityofusingtheSOCstockasab
soluteindicatorformonitoringchangesinlandqualityandsoildeg‐
radation,forexample,inrelationtotheSDGmonitoringframework
(Sims et al., 2019).Earlier reviews (Batjes & van Wesemael, 2015;
deBrogniez,Mayaux,&Montanarella,2011;Lorenz,Lal, & Ehlers,
2019)alsoindicatedthatbasicsoildataandSOCstockchangemon
itoringsystemsarenotavailable,orinconsistent(Jandletal.,2014),
formanyregionsandnations.WithintheGRAandtheCGIARCCAFS
programm e,theinitialfocushasbeenonMRVresour ce sforthelive
stocksector(Wilkes,Reisinger,Wollenberg,Van,&Dijk,2017).
Therearethreemainapproaches(experimentalfieldtrials,chro
nosequencestudiesorpairedland‐usecomparisons,andmonitor
ing networks) to determine relationships between environmental
and manage ment factor s, and SOC dyna mics and GHG emi ssions
(Batjes & van Wesemael, 2015; McKenzie et al., 2002; Morvan
etal., 2008; Spencer etal., 2011) or changes insoil quality/health
(Bai et al., 2018;Leeuwen et al., 2017). An overview of long‐term
terrestrialsoilexperiments(LTEs)ismaintainedbytheInternational
SoilCarbonNetwork,includingthosefromaEuropeanNetworkof
long‐termstudiesforsoilorganicmatter (SOMNET,Powlsonetal.,
1998). Exampl es of chronoseq uence studie s include thos e carried
outinBrazil(Cerrietal.,2007;deMoraesSáetal.,2009),Ethiopia
(Lemenih,Karltun,&Olsson,2005)andChina(He,Wu,Wang,&Han,
2009), whi le paired land‐u se comparison s have been reviewe d by
variousresearchers(Baietal.,2018;Murphy,Rawson,Ravenscroft,
Rankin,&Millard,2003;Oliveretal.,2004).
Following u p from the review of Eu ropean soil mon itoring net
works (Morvan et al., 2008), the Joint Research Centre of the
European Commission launched aninitiative to sample the topsoil
at22,000 points ofthe Land Use/Cover AreaSurvey (LUCAS proj
ect, see Montanarella, Tóth, & Jones, 2011). The first soil sampling
round(2009),basedonstandardsamplingandanalyticalprocedures,
followed a stratified samplingdesign toproducerepresentativesoil
samplesfor majorlandformsand types ofland cover ofthe partici
pating countries. A new LUCAS sampling round is presently under
way,providingthebasisforalongertermmonitoringsystem(Orgiazzi
etal.,2018).Similarly,fortheUnitedStates,Spenceretal.(2011)dis
cussthede signofanatio nalso ilmonitoringn etworkforcarbononag
riculturallands,includingdeterminationofsamplesize,allocationand
site‐scaleplotdesign .Tengetal.(2014)indicatedthatforaccuratesoil
monitoringinChina,itwillbenecessarytosetuproutinemonitoring
systemsatvariousscales(national,provincialandlocalscales),taking
intoconsiderationmonitoringindicatorsandqualityassurance.
Table2 s e r v e s toill u s t r a t e t h ediver si t yi n s o ilmonit o r i n gnet works
and samp le designs in sele cted GRA co untries. The m ost common
samplingdesignfor networks aimed atmonitoringregional/national
SOCstocks iseitherstratified(accordingto soil/landuse/climate)or
grid based.Largecountrieswith a lowsamplingdensity(<1site per
100 km2) generally adoptastratifieddesignto include all important
units (va n Wesemael et al., 2011). Th e (expected) va riability w ithin
these un its should be d etermined to as sess the opti mal number of
    
|
 9
SMITH eT a l.
TABLE 2 ExamplesofsoilmonitoringnetworksandsampledesigninselectedGRAcountriesa
Belgium Brazil China Mexico New Zealand Sweden
Objective NationalSOCmonitoring SOCresponsetolanduse/
managementchange
RegionalSOCmonitoring NationalSOCmonitoring NationalSOCmonitoring NationalSOCmonitoring
Region covered Croplandandgrasslandin
southernBelgium
Rodônia,MatoGrosso,
CentralAmazonia
Northeast(120sites),
North(241),East(356),
South(119),Northwest
(148),Southwest(97)
Forestandnon‐forestland
inparticularpastureand
shrubs
Allregionsandlanduses Cropland~3Mha
Startingdate NationalSoilSurvey
1950–1970;resampled
2004–2007
~20 07 78%startedbefore1985
and87.5%continued
untilatleast1996
Startedin2003;eachyear
one‐fifthofthesiteswill
beresampled
Nationalsoilsdatabase
from1938;Landuseand
carbonanalysissystem
startedin1996c
Fullscalein1995,some
datafrom1988
Sitedensity(km2
persite)
18km2N/A N/A 78km2202km210km2
Siteselection Stratified Stratified Stratified Grid Stratified Grid
Soilsampling
Subsamples Composite Composite Composite Composite Single Composite
Depth 0–30 and 0–100 cm 0–10,10–20,20–30,and
30–40 cm
0–2 0 cm 0–30and30−60cm Variable,sampledbysoil
horizon;in2009,1,235
samplesto30cm
0–20 and 40–60 cm
Frequency Futuresamplingrounds
largelydependonfunding
(Goidtsetal.,2009)
Once(chronosequences
andpairedsites)
Annualsamplingfrom
2010,seeTengetal.
(2014)b
Every5years Afit‐for‐purposemethodis
beingdesignedtomonitor
SOCstocksat~5year
intervalsoverupcoming
decades
1995 and 2005 round
completed;inprinciple
repeatedevery10years
Abbreviation:SOC,soilorganiccarbon.
aAdaptedfromVanWesemaeletal.(2011).
bForaccuratesoilmonitoringinChina,itwillbenecessarytosetuproutinemonitoringsystemsatvariousscales(national,provincialandlocalscales),takingintoconsiderationmonitoringindicatorsand
qualityassurance(Tengetal.,2014).
cForrecentdevelopments,seehttps://soils.landcareresearch.co.nz/index.php/soils‐at‐manaaki‐whenua/our‐projects/soil‐organic‐carbon.
10 
|
   SMITH eT al.
samplesforeachstratum(Brus&deGruijter,1997;DeGruijter,Brus,
Bierkens , & Knottter s, 2006; Lo uis et al., 2014). Such an app roach
willallowa(geo)statisticalanalysisofSOCstockchangesforthesoil/
landuse/climateunitsunde rconsiderationasanalternativeortestfor
process‐b ased models . Continuous s oil monitorin g for SOC at time
intervalsof10yearisoftenproposedasacompromisebet weenmini
mumdetectabilityofchanges(Garten&Wullschleger,1999)andtem
poralshiftsintrends(Bellamyetal.,2005;Schrumpf,Schulze,Kaiser,
&Schumacher,2011;Steinmannetal.,2016).Thismaybelongerthan
thedurationofmanyland‐usemanagementprojectsthatinvolvethe
measurementofSOCstockchanges(Milneetal.,2012).
NewZealandhasdevelopedamodel‐based approach (McNeill,
Golubiewski,&Barringer,2014;Tateetal.,2005)totrackSOCstock
changeswithtimeassumingthatSOCstockvaluesvarybysoiltype,
climateandlanduse,andthatthekeydriverforlong‐term(decadal)
changesinSOCstocksareduetochangesinlanduse,withallother
changes due to soil, climate or erosion assumed constant. This
country‐specific(Tier2)empiricalmethodwasinitiallydescribedin
Tateetal.(2005)reflectingland‐usechangeissuesrelevanttoNew
Zealand.As furthersoilprofiledatawerecollected(currently2050
profiles)themodelwasincreasinglyimproved(McNeilletal.,2014)
adding datafromspecificland‐useclasses(notably indigenousand
exotic forest, cropland, horticulture and wetlands). The approach
was alsorefined to account for spatial autocorrelation to improve
the assessmentof the overall significance of land‐use change and
reportsthreevalidationstudiesforthemodel(McNeilletal.,2014).
Usinglow‐producing grassland on ahigh‐activityclayIPCC default
soilandmoist‐temperateIPCC defaultclimate classasa reference,
the0–30cmSOCstockis133.1tonnes/ha,thechangeasaresultof
landusecanbedetermined,alongwiththemarginalsignificance.For
example,atransitiontohigh‐producinggrasslandresultsinachange
of−0.22 tonnes/ha(notsignificant),whileatransitiontoperennial
croplandresultsinachangeof−19.5tonnes/ha(significant).
While cha nges in national or la rge regional scal e carbon stock
measurements c an be addressed usi ng geostatisti cal sampling ap
proache s, aligned targeted approaches (such as sampling of chro
nosequencesandpairedlanduses)can directly determine land‐use
changefactors, whilecontrollingforother spatially dependent vari
ables,thatis,theycandeterminethecarbongain/lossthatwilloccur
withachangeinlanduseormanagement.Whencoupledwithmon
itored changesinland areaundergoingthesechanges,estimatesof
nationalscalecarbonstockchangescanbecalculated.Thechangein
carbonstocksdeterminedfrompairedsitesamplingcanalsobeused
tovalidateinterpretationsderivedfromnationalscalemeasurements.
7.2 | Methods used by GRA countries for estimating
SOC changes for the ‘cropland remaining cropland’
category in national inventories
All countries that are party to the United Nations Framework
ConventiononClimateChange(UNFCCC)arerequiredtoprovidena
tional inventoriesof emissions and removals of GHG due to human
activities.TheIPCCmethodol ogiesareintendedtoyieldnati onalGHG
inventoriesthataretransparent,complete, accurate,consistentover
time and co mparable acros s countries. Bec ause differen t countries
have diffe rent capacitie s to produce inventorie s, the guideline s lay
outtiersofmethodsforeachemissionssource,withhighertiersbeing
morecomplexand/orresourceintensivethanlowertiers.Inthecon
textofagriculturalGHGemissions,inventoriesremainthemaintool
connectingpolicywithmitigation.
Figure 2 sh ows the categorie s of methods used by G RA coun
tries forestimating the changes in mineral soilcarbon stockforthe
‘Cropland remaining Cropland’ category. Countries listed as non‐
annexIfacemajorchallengeswitheithernon‐existentdata(15coun
triesdonothavecountry‐specificinformationtheycanusetodevelop
theirinventoryandeightcountriesdonotconsiderforSOCchanges
incroplandsbecause do nothavethe technicalcapacity to monitor
thesesources)oralackofrelevantdata(withtheexceptionofGhana
andMalaysia)GRAnon‐annexIcountriesuseaTier1approachtore
portSOCchangesasso ciatedwit ha reasdef inedasC roplandlanduse .
SoilC stocksare influencedbymultiplefactors that affectpri‐
mary productionanddecomposition,includingchangesinland use
and management and feedbacksbet weenmanagement activities,
climate and s oils. However, only a few count ries have taken into
FIGURE 2 Tiermethodsusedby
GlobalResearchAllianceofAgricultural
GreenhouseGasescountriesfor
estimatingthechangesinmineralsoil
carbonstockforthe‘Croplandremaining
Cropland’category.NAindicatesthat
thecountryhasnotdevelopedaGHG
inventory.NEindicatesthatthecountry
hasnotincludedsoilorganiccarbon
changesincroplandsintheinventory.
Countriesreportingcarbonstockchange
associatedwithagriculturallanduseand
managementactivitiesareindicatedby(*)
    
|
 11
SMITH eT a l.
TABLE 3 MethodologyusedtoestimatechangesinsoilCstocksforcroplandremainingcropland,includingagriculturallanduseand
managementactivitiesonmineralsoils
GRA country Tier Land management activities Reference
Australia
TheFullCarbonAccountingModel(FullCAM) Tier3 Croptypeandrotation(includingpasture
leys)
Richards(2001)
Stubblemanagement,includingburning
practices
Tillagetechniques
Fertilizerapplicationandirrigation
Applicationofgreenmanures(particularly
legumecrops)
Soilameliorants(applicationofmanure,com‐
postorbiochar)
Changesinlandusefromgrassland
Crop‐specificcoefficientssourcedfromtheliterature
combinedwithABSagriculturalcommoditiesstatistics
Tier2 Changesintheareaofperennialwoodycrops
Canada
Processmodel(CENTURY)basedontheNationalSoil
DatabaseoftheCanadianSoilInformationSystem
Tier3 Changeinmixtureofcroptype(increasein
perennialcropsandincreaseinannualcrops)
McConkeyetal.
(2014)
Changeintillagepractices
Changeinareaofsummerfallow
Landuse,tillage,typeandamountofinput
Cropresidue,farmyardmanureandpresence
orabsenceofvegetativecover
Perennialandorganicmanagementsystems
Denmark
AverageSOCcalculatedannuallypersoiltypeand
regionbasedonprocess‐basedmodel(C‐TOOL)using
dataontemperatureandestimatedCinputfromcrop
residuesandmanureusingnationaldatabases
Tier3 Croptypeandcropyield Taghizadeh‐Toosiand
Olesen(2016)
Covercrops
Residuemanagement
Manureapplication
Grasslandmanagement
France
TheIPCCGuidelinesandOMINEAdatabase Tier1 Landuse CITEPA(2019)
Tillage
Typeandamountofinput
Japan
Averagecarbonstockchangesineachyearbyland‐use
subcategory(ricefields,uplandfields,orchardsand
pasturalland)calculatedbytheRothCmodelbythe
mineralsoilareaofeachprefectureobtainedfrom
statisticalmaterial,mapdataandquestionnairesur vey
Tier2 Carboninputfromcropresidue ShiratoandTaniyama
(2003)
Farmyardmanure
Presenceorabsenceofvegetativecover
Lithuania
Nationalstatisticsforwoodycropsandavailabledata
ofarablelandcertifiedasorganicinFAOSTATand
ecologicalagriculturallandstatistics.
Tier2 Croptype(perennialcrops,certifiedorganic
crops,othercrops)
StatisticsLithuania
(2018)
Amountofinput
Norway
Referencestockandstockchangefactorsestimated
bytheIntroductoryCarbonBalanceModel(ICBM)
inastudywhereCO2emissionswereestimatedfor
Norwegiancropland
Tier2 Croprotations Borgenetal.(2012)
Carboninputs
Tillage
(Continues)
12 
|
   SMITH eT al.
accountcroplandmanagementactivities.Table3providesanover
viewof the methodsused in GRA countriesforestimating carbon
stockchangeandemissionsassociatedwithagriculturallanduseand
managementactivitiesonmineralsoil.
Therearestillhighlevelsofuncer taintyintheestimates;however,
uncertaintiesarerelat ivelylowfo rA nnexIcountriesduetotheirwell‐
developedstatistical systems and capacityto use higher tier meth
ods. In cont rast, natio nal inventories of m any developing cou ntries
generally have greater uncertaintyand are not sufficientlyrigorous
toenablemonitoringofemissions.ForTier2inventorydevelopment,
countriescouldusetheexpertiseofotherGRAmembers,forinstance
fromthosecountriesthathaveadoptedaTier3method(seeTable4)
toestimatesoilorganicCstockchangesinagriculturalland.
With increased obligations for reporting on GHG emissions
and Nationally Determined Contributions (NDCs) under the
Paris agreement,it is important thatall countries areable to es
timatetheirGHG emissions to maximize transparency,accuracy,
completeness and consistency. Improving inventories requires
enhancednationalcapabilit ytogatherrelevantactivit ydatatode
velo pc ountry‐sp ecificem issio nfa cto rs.The reisane edtoi mprove
theevidencebase and to betterconnectgovernmentsand rele
vantexpertisetosubsequentlyimprovethequalityofagricultural
NDCs and t he way their ach ievements ar e reflected by n ational
GHGinventories.
8 | PROPOSED GLOBAL SOIL MRV
PL ATFO R M
The sec tions above descri be the methods ava ilable to measure
and monitor carbon; models that can be used to simulate and
projectchangesinSOC,differenttypesofexperimentalplatform
andthedataneededtotestmodelsandallowthemtoberunfrom
plot to glob al scale; and met hods/platforms t hat could be used
toverifyanysimulatedchange inSOC (summarized in Figure 3).
TheseformthecomponentsofasystemsuitableforMRVofSOC
change(Figure3).
Centraltothesystemarebenchmarksites,whichcouldbelo
catedatexistingornewlong‐termexperiments(Figure3,item2;
Richteretal.,2007),orcouldconsistofwell‐characterizedchro
nosequencesorpairedsamplingsites(e.g.Heetal.,2009;Oliver
et al., 2004). Thebenchmark sites would preferably be located
onrepresentativelandcover/land‐usetypes,soiltypesandwith
representativemanagement. At these sites,proposedpractices
to increas e SOC could be tes ted in fully ran domized block de
signs,andSOCchangemeasuredovertime(measurementsevery
fewyears),whilemeasuringshortertermprocesses(suchasGHG
emissions)morefrequently (continuouslywith ECfluxtowersor
fre quentlywithautomatedchambers;Figure3,item2;Baldocchi,
2003).Thesamesitescouldbeusedtotestnovelspectralmeth
ods for measuring SOC change against traditional direct SOC
measure ment (England & Vis carra Rossel, 2 018). Careful align
mentofsiteselectionandexperimentaldesignwithothergoals
of land owne rs, manager s and regulator s (e.g. quantif ication of
soilqualitychangeornutrientuseefficiency)willpromotestron
geruptakeofaninternational suite of benchmark siteswithad
ditionalbenefits.
Since it would be prohibitively expensiveto set up benchmark
sites covering all possible combinations of land use, climate, soil
type and management practice, models of SOC change are re‐
quired tointerpolateand inferchangeacross allcombinations,and
to project changes into the future, across landscapes and under
GRA country Tier Land management activities Reference
Spain
SOCvaluescalculatedbyuseandprovince,together
withthereferencevaluesofthemanagementfactors
providedbytheIPCCGuidelines
Tier1 Landuse Roviraetal.(2007)
Croprotations
Amountofinput
Tillage
UnitedKingdom
ReviewUKrelevantliteratureontheeffectsofcrop
landmanagementpracticesonsoilcarbonstocksto
modelUK‐specificemissionfactors(methodology
developedinDefraprojectSP1113)
Tier1 Manure Moxleyetal.(2014)
Residueinputs
Croptype(perennial,cropland,set‐aside)
Tier2 Tillage
UnitedStates
Publishedliteraturetodeterminetheimpactof
managementpracticesonSOCstorage.Activitydata
basedonthehistoricallanduse/managementpatterns
recordedinthe2012NRI(USDA,2018)
Tier2 Tillage Ogle,Breidt,Eve,and
Paustian(2003);
Ogle,Breidt,and
Paustian(2006)
Croppingrotations
Intensification
Land‐usechangebetweencultivatedand
uncultivatedconditions
Abbreviation:ABS,AustralianBureauofStatistics;GRA,GlobalResearchAllianceofAgriculturalGreenhouseGases.
TABLE 3 (Continued)
    
|
 13
SMITH eT a l.
FIGURE 3 Componentsofasoil
measurement/monitoring,reporting
andverificationframework,indicating
whichcomponentscontributeto
measurement/monitoring(M),reporting
(R)orverification(V).SeetextinSection
8forexplanationoflinkagesbetweenthe
components
TABLE 4 Modelsusedtoestimatecarbondioxideemissionsandremovalsfromthecroplandremainingcroplandsoilscomponent(Tier3
method)inGRAcountries
GRA country Model Reference
Australia TheFullCarbon
AccountingModel
(FullCAM)
Estimatesemissionsfromsoilthroughaprocess
involvingallon‐sitecarbonpools(livingbiomass,
deadorganicmatterandsoil)onapixelbypixel
(25m×25m)level
Richards(2001)
Canada CENTURY ProcessmodelusedforestimatingCO2emissions
andremovalsasinfluencedbymanagement
activities,basedontheNationalSoilDatabaseof
theCanadianSoilInformationSystem
Parton,Schimel,Cole,andOjima
(1987),Parton,Stewart,andCole
(1988)
Denmark C‐TOOL 3‐Pooldynamicsoilmodelparameterizedand
validatedagainstlong‐termfieldexperiments
(100–150years)conductedinDenmark,United
Kingdom(Rothamsted)andSwedenandis
‘State‐of‐the‐art’
Taghizadeh‐Toosi,Christensen,etal.
(2014)
Finland Yasso07soilcarbonmodel TheparameterizationofYasso07usedincropland
wastheonereportedinTuomi,Rasinmäki,Repo,
Vanhala,andLiski(2011)
Palosuo,Heikkinen,andRegina(2015)
Japan SoilCarbonRothCmodel InordertoapplythemodeltoJapaneseagricul‐
turalconditions,themodelwastestedagainst
long‐termexperimentaldatasetsinJapanese
agriculturallands(Shirato&Taniyama,2003)
Colemanetal.(1997),Coleman,and
Jenkinson(1987)
Sweden SoilCarbonmodel
ICBM‐region
CalculateannualCbalanceofthesoilbasedon
nationalagriculturalcropyieldandmanuresta‐
tistics,andusesallometricfunctionstoestimate
theannualCinputstosoilfromcropresidues
AndrénandKätterer(20 01)
Switzerland SoilCarbonRothCmodel TheimplementationofRothCintheSwissGHG
inventoryisdescribedindetailinWüst‐Galley,
Keel,andLeifeld(2019)
Colemanetal.(1997),Colemanand
Jenkinson(1987)
UnitedKingdom CARBINESoilCarbon
Accountingmodel
(CARBINE‐SCA)
SimplifiedversionoftheECOSSEmodel(Smith,
Gottschalketal.,2010),coupledwithalitterde‐
compositionmodelderivedfromtheForClim‐D
model(Liski,Perruchoud,&Karjalainen,2002;
Perruchoud,Joos,Fischlin,Hajdas,&Bonani,
1999)
Matthewsetal.(2014)
UnitedStates DAYCENTbiogeochemical
model
UtilizesthesoilCmodellingframeworkdeveloped
intheCenturymodel(Partonetal.,1987,1988,
1994;Metherell,1993),buthasbeenrefinedto
simulatedynamicsatadailytimestep
Parton,Hartman,Ojima,andSchimel
(1998),DelGrossoetal.(2001),Del
GrossoandParton(2011)
Abbreviation:GRA,GlobalResearchAllianceofAgriculturalGreenhouseGases.
14 
|
   SMITH eT al.
novelcombinations(Figure3,item3;e.g.Richardsetal.,2017).To
establishconfidence that thechosenmodel or models are capable
ofaccuratelyand reliably simulatingSOC change, they need to be
tested ac ross the full r ange of paramete r space (i.e. mult iple soils
types,climate zones,land‐use types andsoilmanagementoptions;
Ehrhardtet al.,2018; Smith et al., 1997). If necessary,the models
canbefurtherdevelopedorparameterizedusingdatafromthesame
long‐termexperiments, orfrom shorter term experiments, before
beingevaluatedagainagainstadatasetnotusedindevelopmentor
parameterization(Smith&Smith,2007).
Whenthemodel(s)aredeemedtobereliable,theycouldbeap
plied(a)toderiveIPCCTier2emissionorSOCstockchangefactors,
whicharespecifictotheregionandconditionsrepresentedwithin
theregion (e.g.Begum etal.,2018); or (b) spatially over thewhole
landscape (or theentire land area ofacountry) usingspatial data‐
bases of soi l character istics, an d land cover, management a nd cli‐
mate data (Figure 3, item 4), to directly simulateSOC change and
GHGemissions,therebydeliveringaTier3methodologytoreport
emissions (Smithet al.,2012).Data onchangesinsoilmanagement
are necessary for estimating changes inSOC/GHG emissions, and
thiscouldalsobeprovidedbyself‐reportedorfarmsurvey‐derived
activitydata(Figure3,item5).
Ifself‐reportedactivity dataareusedastheprimarymechanism
fo r reporting,s u chactivi t yd a t acoul d b ever i f i e dthrou g h spotc h e c k s/
farmvisitsorcouldbedoneusingremote sensing(Figure 3, item 7),
whichcanshow,forexample,thepresenceofbarefallow,covercrop
orresidueretention(Galloetal.,2018;Roggeetal.,2018).Inaddition
toproviding a mechanismfor verification of activity data, remotely
sensedearthobservationproductscouldalsoprovidespatialdatato
run the SOC/ GHG models. For example, earth observation canbe
usedtoestimatechangesincarboninputtosoils,throughchangesin
NPP/GPP(Chenetal.,2019;Neumann&Smith,2018),landdegrada
tion(Simsetal.,2019)andcanalsobeusedtodeterminelandcover/
landcoverchange(e.g.Chenetal.,2019).
Well‐calibratedmodels,supportedbymeasurements,canalso
beusedtoestablishrelationshipsbetweenamanagementchange
inaparticularsituation(combinationorsoil,climate,landuseand
management)an dachangeinSOC/GHGemissions,includingest i
matesofuncertainty(Fittonetal.,2017).Thiswouldallowactivit y
data(Figure 3,item5),self‐reportedbythefarmer/landmanager,
tobeusedastheprimarysourceofdataforreporting,inplaceof
theneedtodirectlymeasureSOCofGHGemissionchange(Smith,
20 04b).Mo r ebro a d ly,unce r t a i ntie s andp o tent i albi a sesin allc om
ponentsofthe MRVframework, including allmeasurementsand
modellingschemes,needtobeaddressed.Fortransparency,there
isaneedforunifiedprotocolsforsuchuncertaintyassessments.
Intermsofverification,changeinSOCstocks,spatialsoilmoni‐
toringnetworks(Figure3,item6)couldbeusedtoground‐truthSOC
changesestimatedbytheTier2methodorTier3modelprojections
over time. If r esampled ever y few years, the so il monitoring net‐
work(onagridasshowninFigure3item7,e.g.Bellamyetal.,2005,
or using a st ratified sam pling protocol; Mo ntanarella et al ., 2011)
could provide independent estimates of large‐scale SOC change.
Some basic methodological requirements and recommendations
can be formulated for ‘good SOC‐monitoring and MRV practice’
tosupportscientific and policy decisions (Batjes & van Wesemael,
2015; Desau les, Ammann, & S chwab, 2010; Morv an et al., 2008 ;
Spencer etal., 2011).These include: (a) theprovision of long‐term
continuity and consistency under changing boundary conditions,
suchasbiophysicalsiteconditions,climate change,methodologies,
socio‐economicset tingandpolicycontext;(b)adoptionofascientif
icallyandpolitically(e.g.forGRA,UNFCCC,UNNCCD)appropriate
spatial andtemporalresolutionfor the measurements; (c)ensuring
continuousqualityassuranceat all stagesofthemeasurement and
monitoring process; (d) measurement/observation and documen
tation of allpotential drivers of SOCand GHG change; and (e) soil
monitoringnetwork‐collated,georeferencedsamplesarchivedand
the associated (harmonized) data madeaccessible through distrib
uteddatabasestoenhancethevalueofthecollateddataformultiple
uses.Inadditiontothis,soilmonitoringnetworksshouldbeincluded
ina broadercross‐method validationprogramme toultimatelyper‐
mitspatiallyandtemporallyvalidatedcomparisonsbothwithinand
betweencountries.Anopen‐accessdatabase,whereshort‐orlong‐
termsoilCmeasurementscouldbeuploadedandshared(e.g.https://
dataverse.o rg/ or an online col laborative plat form as used in the
CIRCASA project: https://www.circasa‐project.eu/),would also be
ofgreatbenefitforprogressingaglobalMRVsystem.
As indic ated, the implem entation of soil mon itoring network s
posesseveralscientific,technicalandoperationalchallenges.From
anoperationalpointofview,toimplementanintegratedmonitoring
system,itwillbecrucialtoovercomeinitializationcostsandunequal
access to monitoring technolo gies. For developing countrie s, this
will require international cooperation, capacity building and tech
nologytransfer(deBrogniezetal.,2011),whichcouldbefacilitated
withinGRA,CCAFSandsimilarorganizations,insynergywithrele‐
vantfundingmechanisms,orviatherecentlyestablished‘GSOCseq’
programmeoftheUNFAO(FAO,2019b).
While other components of a soil MRV framework could be
added,thecomponentsoutlinedinFigure3couldcertainlyfulfilall
ofthefunctionsnecessaryforanMRVsystem.AsseeninSections
4–7,the existingcapacity in termsof existing benchmark sites,soil
monitoringprogrammesandaccesstomodelsindifferentcountries
varies greatly.Whilesomecountriesarealready usingTier2and3
monitoringofsoil C change,othershavebarelybegun to buildca‐
pacity.Recently, theUN FAO hasestablished a programme called
‘GSOCseq’(FAO,2019b)whichaims to buildthis capacityinterna‐
tionally.Programmessuchasthiscouldpavethewayformakingthis
proposedMRVframeworkareality.
ACKNOWLEDGEMENTS
P.S.,J.F.S., C.C., N.B., M.K.,C.A. and J.O. acknowledgesupport from
the CIRCASA project which received funding from the European
Union's Horizon 2020 Research and Innovation Programme under
grantagreementno774378.The inputofP.S.also contributesto the
projects:DEVIL(NE/M021327/1),Assess‐BECCS(funded by UKERC)
    
|
 15
SMITH eT a l.
and Soils‐R‐GRREAT (NE/P019455/1). AS‐C acknowledges support
fromtheAGRISOST‐CMproject(S2018/BAA‐4330)andMACSUR‐JPI
initiative,as wellasthe inspiration and support fromthe Spanish re
searchnetworksREMEDIAandNUEVA.J.A.‐F.acknowledgessupport
fromMinisteriodeEconomiayCompetitividadofSpain(projectnum
berAGL2017‐84529‐C3‐1‐R).TheparticipationofN.C. andE.W.was
funded as part of the CGIAR Research Programon Climate Change,
AgricultureandFoodSecurity(CCAFS),whichiscarriedoutwithsup
port fromtheCGIARTrustFundandthrough bilateral fundingagree
ments(https://ccafs.cgiar.org/donors).J.E.O.wasfundedbytheDanish
MinistryofClimate,EnergyandUtilitiesaspartoftheSINKS2project.
L.S.andS.M.acknowledgesupportfromtheNewZealandAgricultural
Greenhouse Gas Research Centre and Global Research Alliance.
This paper contributes tothe work of theSoil Carbon Sequestration
network of the Integrative Research Group of the Global Research
Alliance on Agricultural Greenhouse Gases (https://globalresearcha
lliance.org/).Theviewsexpressedinthisdocumentcannotbetakento
reflecttheofficialopinionsofthefundingorganizations.
ORCID
Pete Smith https://orcid.org/0000‐0002‐3784‐1124
Louis Schipper https://orcid.org/0000‐0001‐9899‐1276
Daniel P. Rasse https://orcid.org/0000‐0002‐5977‐3863
Cristina Arias‐Navarro https://orcid.org/0000‐0002‐5125‐4962
Dario Fornara https://orcid.org/0000‐0002‐5381‐0803
REFERENCES
Ackerson,J.P.,Morgan,C.L.S.,&Ge,Y.(2017).Penetrometer‐mounted
VisNIR spectroscopy: Application of EPO‐PLS to in situ VisNIR
spectra. Geoderma, 286, 131138. https://doi.org/10.1016/j.geode
rma.2016.10.018
Aitkenhead,M.J.(2017).MappingpeatinScotlandwithremotesensing
andsite characteristics.European Journal of Soil Science, 68,28–38.
https://doi.org/10.1111/ejss.12393
Al‐Asadi, R.A., & Mouazen, A. M. (2018). A prototypemeasuring sys‐
tem of soil bu lk density wit h combined freq uency domain re flec‐
tometryand visibleand nearinfraredspectroscopy.Computers and
Electronics in Agriculture, 151, 485–491. https://doi.org/10.1016/
j.compag.2018.06.045
Alberta Government. (2012).Quantificationprotocolfor conservation
cropping(version1.0). Retrievedfrom https://open.alberta.ca/publi
c a t i o n s / 9 7 8 0 7 7 8 5 9 6 2 8 8
Alexander,P.,Paustian,K.,Smith,P.,&Moran,D.(2015).Theeconomics
ofsoilCsequestrationandagriculturalemissionsabatement.Soil,1,
331–339.https://doi.org/10.5194/soil‐1‐331‐2015
Allen, D. E ., Pringle , M. J., Page, K . L., & Dalal , R. C. (2010). A r eview
ofsampling designs for the measurementof soil organiccarbon in
Australiangrazinglands.The Rangeland Journal,32,227–246 .https://
doi.org/10.1071/RJ09043
Andr én,O.,&K ätter er,T.(1997).ICB M:Theintrod uctor ycarbo nb alance
modelforexplorationofsoilcarbonbalance.Ecological Applications,
7(4),1226–1236.https://doi.org/10.2307/2641210
Andrén,O.,&Kätterer,T.(2001).Basicprinciplesforsoilcarbonseques
trationandcalculatingdynamiccountry‐levelbalancesincludingfu
turescenarios.A ssessment Methods for Soil Carbon,495–511.
Angers,D.A.,&Eriksen‐Hamel, N. S. (2008). Full‐inversiontillage and
organi c C distribu tion in soil pr ofiles: A met a‐analysis. Soil Science
Society of America Journal,72,1370–1374.https://doi.org/10.2136/
sssaj2007.0342
Aubinet,M.,Vesala, T.,& Papale,D.(2012).Eddy covariance: A practical
guide to measurement and data analysis.Dordrecht,theNetherlands/
Heidelberg,Germany/London,UK/NewYork,NY:Springer.https://
doi.org/10.10 07/978‐94‐0 07‐2351‐1
Australian Government. (2018). The supplement to the carbon credits
(carbon farming initiative—measurement of soil carbon sequestration in
agricultural systems) methodology determination 2018.Retrievedfrom
h t t p : / / w w w . e n v i r o n m e n t . g o v . a u / s y s t e m / f i l e s / c o n s u l t a t i o n s / 0 7 2 b 4 
825‐ec0f‐49d9‐991e‐42dfa1fbeae3/files/supplement‐soil‐carbon‐
agricultural‐systems.pdf
Bai,Z.,Caspari,T.,Gonzalez,M.R.,Batjes, N.H.,Mäder,P.,Bünemann,
E.K.,…Tóth,Z.(2018).Effectsofagriculturalmanagementpractices
on soil quality: A review of long‐term experiments for Europe and
China. Agriculture, Ecosystems & Environment, 265, 1–7. ht tps://d oi.
org/10.1016/j.agee.2018.05.028
Baldocchi, D. (2003). Assessing the eddy covariance technique for
evaluation carbon dioxide exchange rates of ecosystems: Past,
presen t and future . Global Change Biology, 9,479–492. ht tps://doi.
org/10.1046/j.1365‐2486.2003.00629.x
Baldocchi,D.,Housen,C.,&Reichstein,M.(2018).Inter‐annualvariabil‐
ityofnetandgrossecosystemcarbonfluxes: Areview. Agricultural
and Forest Meteorology, 249, 520–533. https://doi.org/10.1016/
j.agrformet.2017.05.015
Baldock,J.A.,Beare,M.H.,Curtin,D.,&Hawke,B.(2018).Stocks,com‐
positio n and vulnera bility to loss of s oil organic c arbon predic ted
usingmid‐infraredspectroscopy.Soil Research, 56,468. https://doi.
org /10.1071/SR17221
Baldoc k, J. A., Hawke, B ., Sanderman , J., & Macdonald , L. M. (2013).
Predicting contents of carbon and its component fractions in
Austr alian soils fro m diffuse ref lectance m id‐infrared sp ectra. Soil
Research,51,577–595.https://doi.org/10.1071/SR13077
Barnet t, A. L ., Schipper, L . A., Taylor, A., Balk s, M. R., & Mu dge, P.L .
(2014).SoilCandNcontentsinapairedsurveyofdairyanddrystock
pasturesinNewZealand.Agriculture Ecosystems & Environment,185,
34–40.https://doi.org/10.1016/j.agee.2013.12.008
Barnett,V.,Payne,R.,&Steiner,R.(1995).Agricultural sustainability. New
York,NY:JohnWiley&Sons.
Batjes, N. H. (1996). Total carbon and nitrogen in the soils of the
world. European Journal of Soil Science, 47, 151163. https://doi.
org/10.1111/j.1365‐2389.1996.tb01386.x
Batjes,N.H.,&VanWesemael,B.(2015).Measuringandmonitoringsoil
carbon.InS.A.Banwart,E.Noelmeyer,&E.Milne(Eds.),Soil carbon:
Science, management and policy for multiple benefits (pp. 188–201).
Wallingford,UK:CABI.
Beer,C.,Reichstein,M.,Tomelleri,E.,Ciais,P.,Jung,M.,Carvalhais,N.,…
Papale,D.(2010).Terrestrialgrossc arbond ioxideuptake:Gl obaldis
tributionandcovariationwithclimate.Science,329,834–838.https
://doi.org/10.1126/science.1184984
Begum, K ., Kuhner t, M., Yeluripat i, J., Ogle, S. , Parton, W., Pan , G.,
…Smith, P.(2018).Modelling greenhousegas emissions and mit
igation potentials in fertilized paddy rice fields in Bangladesh.
Geoderma, 341, 206–215. https://doi.org/10.1016/j.geode
rma.2019.01.047
Bellamy,P.H.,Loveland,P.J.,Bradley,R. I.,Lark, R.M.,& Kirk,G. J.D.
(2005).CarbonlossesfromallsoilsacrossEnglandandWales1978–
2003. Nature,437,245–248.https://doi.org/10.1038/nature04038
Bellon‐Maurel,V.,&Mcbratney,A.(2011).Near‐infrared(NIR)andmid‐
infrared(MIR)spectroscopictechniquesforassessingtheamountof
carbonstockinsoils–Criticalreviewandresearchperspectives.Soil
Biology and Biochemistry, 43, 1398–1410. https://doi.org/10.1016/
j.soilbio.2011.02.019
16 
|
   SMITH eT al.
Besn a r d , S . ,C a r v a l h a is,N.,A rain , M . A . , B l a c k ,A . , d eBruin,S . , B u c h m a n n ,
N.,…Reichstein,M.(2018).Quantifyingtheeffectof forestagein
annualnetforestcarbonbalance.Environmental Research Letters,13,
124018.https://doi.org/10 .1088/1748‐9326/aaeae b
Białousz,S.,Marcinek,J.,Stuczyński,T.,&Turski,R.(2005).Soilsurvey,
soil moni toring and soil d atabases in Pol and. In R. J. A . Jones, B.
Houšková,P.Bullock&L.Montanarella(Eds.),Soil resou rces of Europe.
EuropeanSoil Bureau Research ReportNo. 9,EUR 20559 EN (2nd
ed., pp. 26 3–273). Luxemb ourg: Off ice for Off icial Publi cations of
theEuropeanCommunities.
Bispo, A., Andersen, L., Angers, D. A ., Bernoux, M., Brossard, M.,
Cécillo n, L., … Egli n, T.K . (2017). Account ing for carb on stocks in
soilsandmeasuringGHGsemissionfluxesfromsoils:Dowehavethe
necessarystandards?Frontiers in Environmental Science,5,41.https
://doi.org/10.3389/fenvs.2017.00041
Blagodatsky,S.,&Richter,O.(1998).Microbialgrowth insoilandnitro‐
genturnover:A theoreticalmodelconsideringtheactivitystateof
microorganisms. Soil Biology and Biochemistry, 30(13), 1743–1755.
https://doi.org/10.1016/S0038‐0717(98)00028‐5
Blanco‐ Canqui, H. , & Rattan La l, R. (200 8). No‐tillage and s oil‐profile
carbon sequestration: An on‐farmassessment. Soil Science Society
of America Journal, 72, 693–701. https://doi.org/10.2136/sssaj
2007.0233
Borgen, S. K., Grønlund, A., Andrén, O., Kätterer, T., Tveito, O. E.,
Bakken, L. R., & Paustian, K. (2012). CO2 emissions from crop
land in Norway estimated by IPCC default and Tier 2 methods.
Greenhouse Gas Measurement and Management,2(1), 5–21. https://
doi.org/10.1080/20430779.2012.672306
Brus,D.J.,&DeGruijter,J.J.(1997).Randomsamplingforgeostatistical
modelling?Choosingbetweendesign‐basedandmodel‐basedsam
plingstrategiesforsoil(with discussion).Geoderma,80,1–44.https
://doi.org/10.1016/s0016‐7061(97)00072‐4
Bruun,S., &Jensen,L.S.(2002).Initialisationofthesoilorganicmatter
poolsoftheDaisymodel.Ecological Modelling,153(3),291–295.https
://doi.org/10.1016/S0304‐3800(02)00017‐0
Campbell,E.E.,&Paustian,K.(2015). Currentdevelopmentsin soilor‐
ganic matter modelingand the expansion of model applications: A
revi ew. Environmental Research Letters,10 (12),123004. https://doi.
org /10.1088/1748‐9326/10/12/1230 04
Cas pari,T.,vanLynden,G.,&Bai,Z.G.(2015).Landdegradationneutral
ity:Anevaluationofmethods.EnvironmentalresearchoftheFederal
Ministry for the Environment, Nature Conservation, Building and
NuclearSafety,ProjectNo.46658ReportNo.(UBA‐FB)0 02163/E.
Dessau‐Roßlau,Germany:Umweltbundesamt.53pp.
Cerri, C.E.P.,Easter,M.,Paustian,K.,Killian, K., Coleman,K.,Bernoux,
M., … Cer ri, C. C. (20 07). Simulati ng SOC changes i n 11la nd use
change chronosequences from the Brazilian Amazon with RothC
andCenturymodels.Agriculture, Ecosystems and Environment, 122,
46–57.https://doi.org/10.1016/j.agee.2007.01.007
Chabbi,A.,Lehmann,J.,Ciais,P.,Loescher,H.W.,Cotrufo,M.F.,Don,A.,
…Rumpel,C.(2017).The4‰initiative:Anexusbetweenagriculture
and climatepolicyto offset CO2 emissions. Nature Climate Change,
7,307–309.
Chabrillat, S., Ben‐Dor, E., Cierniewski, J., Gomez, C., Schmid, T., &
Van Wesemael , B. (2019). Imaging spectroscopy for soil ma pping
and monitoring. Surveys in Geophysics, 40, 361–399. https://doi.
org/10.1007/s10712‐019‐09524‐ 0
Chen,C.,Park,T.,Wang,X.H.,Piao,S.L.,Xu,B.D.,Chaturvedi,R.K.,…
Myneni,R.B.(2019).Chinaand Indialead ingreeningofthe world
through land‐use management. Nature Sustainability, 2, 122–129.
https://doi.org/10.1038/s41893‐019‐0220‐7
CITEPA.(2019).Rapport OMINEA – 16 ème édition.Organisationetméth
odes des inventairesnationaux des émissions atmosphériques en
France. Paris, France: CITEPA. Retrieved from https://www.citepa.
org/en/activities/emission‐inventories/ominea
Coleman,K.,&Jenkinson,D.S.(1987).RothC – A model for the turnover
of carbon in soil. Model description and users guide, ROTHC manual.
Harpenden,Herts,UK:LawesAgriculturalTrust.
Coleman, K., Jenkinson, D. S., Crocker, G. J., Grace, P. R., Klír, J.,
Körschens,M., … Richter,D.D. (1997). Simulating trends in soilor‐
ganiccarboninlong‐termexperimentsusingRothC‐26.3.Geoderma,
81(1),29–44.https://doi.org/10.1016/S0016‐7061(97)00079‐7
Conant , R. T., Ryan, M. G. , Ågren, G. I ., Birge, H. E. , Davidson, E. A .,
Eliasson,P.E.,…Bradford,M.A.(2011).Temperatureandsoilorganic
matterdecompositionrates–Synthesisofcurrentknowledgeanda
wayforward.Global Change Biology,17(11),3392–3404.https://doi.
org /10.1111/j.1365‐2486.2011.02496.x
Cotruf o, M. F.,Wal lenstein, M . D., Boot, C . M., Denef, K. , & Paul, E.
(2013).T heMic rob ialEf ficien c y‐ Mat rix St a biliz ati on(ME MS)fr ame
workintegratesplantlitter decompositionwithsoilorganicmatter
stabilization:Dolabileplantinputsformstablesoilorganicmatter?
Global Change Biology, 19(4), 988–995. https://doi.or g/10.1111/
gcb.12113
Dangal,S.,Sanderman,J.,Wills,S.,&Ramirez‐Lopez,L.(2019).Accurate
and precise prediction of soilproperties from a large mid‐infrared
spectral library. Soil Systems, 3, 11. https://doi.org/10.3390/soils
ystems3010011
De Brogniez, D.,Mayaux, P.,& Montanarella, L. (2011). Monitoring, re
porting and verif ication systems for carbon in soils and vegetation in
African, Caribbean and Pacific countries. Luxembourg: Publications
OfficeoftheEuropeanUnion.
De Gruijter, J. J.,Brus, D. J., Bierkens, M. F.P., & Knottters, M. (Eds.).
(2006). Sampling for natural resource monitoring. Heidelberg,
Germany:Springer.
DeGruijter,J.J.,Mcbratney,A.B.,Minasny,B.,Wheeler,I.,Malone,B.P.,
&Stockmann, U.(2016).Farm‐scale soilcarbonauditing.Geoderma,
265,120–130.https://doi.org/10.1016/j.geoderma.2015.11.010
deMoraesSa,J.C., Cerri, C.C., Lal,R., Dick,W.A ., de CassiaPiccolo,
M., & Feigl , B. E. (2009). S oil organic ca rbon and fer tility inte rac‐
tions affected by a tillage chronosequence in a Brazilian Oxisol.
Soil and Tillage Research, 10 4, 56–64. ht tps://doi.org/10.1016/
j.still.2008.11.007
DelGrosso,S.,&Parton,W.J.(2011).Quantifyingnitrousoxideemis
sionsfromagriculturalsoilsandmanagementimpacts.InL.Guo,A.
Gunasekara, & L. McConnell (Eds.), Understanding greenhouse gas
emissions from agricultural management,ACSSymposiumSeries,1072
(pp.3–13).Washington,DC:AmericanChemicalSociety.
DelGrosso, S., Parton,W.J.,Mosier,A.R., Hartman,M. D.,Brenner,J.,
Ojima,D.S.,&Schimel,D.S.(2001).Simulatedinteractionofcarbon
dynamicsandnitrogentracegasfluxes usingtheDAYCENTModel.
InM. Schaffer,L .Ma,& S.Hansen(Eds.),Modeling carbon and nitro
gen dynamics for soil management(pp.3 01–332).BocaRaton,FL:CRC
Press.
Desaules, A.,Ammann, S., & Schwab, P.(2010). Advances in long‐term
soil‐pollutionmonitoringofSwitzerland.Jour nal of Plant Nutritio n and
Soil Science,173,525–535.https://doi.org/10.1002/jpln.200900269
Dhawale,N. M., Adamchuk, V.I., Prasher,S. O., Viscarra Rossel, R. A.,
Ismail, A . A.,&Kaur,J.(2015). Proximal soilsensingofsoil texture
and organic matter with a prototype portable mid‐infrared spec‐
trometer.European Journal of Soil Science, 66,661–669. https://doi.
org /10.1111/ejss.12265
Ehrhardt,F.,Soussana,J.F.,Bellocchi,G.,Grace,P.,McAuliffe,R.,Recous,
S.,…Zhang,Q. (2018).Assessinguncertainties in crop andpasture
ensemble model simulations of productivity and N2O emissions.
Global Change Biology, 24, e603–e616. https://doi .org/10.1111/
gcb.13965
Ellert, B. H., & Bettany, J. R. (1995). Calculation of organic matter
and nutrients stored in soils under contrasting management re‐
gimes. Canadian Journal of Soil Science, 75, 529–538. https://doi.
org/10.4141/cj ss 95‐075
    
|
 17
SMITH eT a l.
Emmel, C.,Winkler,A.,Hörtnagl, L.,Revill, A.,Ammann,C.,D'Odorico,
P.,… Eu gster, W. (2018). Integrat ed managemen t of a Swiss crop‐
land is not s ufficient to pr eserve its so il carbon pool in t he long
term. Biogeosciences, 15, 5377–5393. https://doi.org/10.5194/
bg‐15‐5377‐2018
England,J.R.,&Viscarra Rossel,R. A. (2018).Proximal sensing for soil
carbon accounting. SOIL, 4, 101–122. https://doi.org/10.5194/
soil‐4‐101‐2018
Eugster,W.,&Merbold,L.(2015).Eddycovarianceforquantifyingtrace
gas fluxes fromsoils.SOIL, 1(1),187–205.https://doi.org/10.5194/
so i l‐1‐187‐20 15
Falloon,P.D.,&Smith,P.(2000).Modellingrefractorysoilorganicmatter.
Biology and Fertility of Soils,30(5),388–398.https://doi.org/10.1007/
s003740050019
FAO. (2019a). Measuring and mode lling soil carbon stock s and stock
changesinlivestockproductionsystems:Guidelinesforassessment
(Versio n 1). In Livestock environmental assessment and performance
(LEAP) partnership.Rome, Italy:FAO.170pp.Licence:CCBY‐NC‐SA
3.0IGO.
FAO.(2019b).SubmissionbytheFoodandAgricultureOrganizationofthe
UnitedNations(FAO)totheUnitedNationsFrameworkConvention
onClimateChange(UNFCCC)inrelationtotheKoroniviajointwork
onagriculture(4/CP.23)onTopics2(b)and2(c).Retrievedfromhttps
: / / w w w 4 . u n f c c c . i n t / s i t e s / S u b m i s s i o n s S t a g i n g / D o c u m e n t s / 2 0 1 9 0 
5031649–‐FAO%20Submission%20on%20KJWA_2(b)_(c).pdf
Felber,R.,Münger,A.,Neftel,A.,&Ammann,C.(2015).Eddycovariance
methaneflux measurements overagrazed pasture:Effectof cows
asmovingpointsources.Biogeosciences,12,3925–3940.https://doi.
org /10.5194/ bg‐12‐3925‐2015
Ferste r, C. J., Trofym ow,J . A., Coops, N . C., Chen, B. , & Black, T. A.
(2015). Comparison of carbon‐stock changes, eddy‐covariance
carbonfluxesand model estimates incoastalDouglas‐fir stands in
BritishColumbia.Forest Ecosystems,2,13.https://doi.org/10.1186/
s40663‐015‐0038‐3
Fitton,N., Datta, A., Cloy, J. M.,Rees, R.M., Topp, C.F.E., Bell, M. J.,
… Smith, P. (2017). Modelling spatial and inter‐annual variations
ofnitrousoxideemissionsfromUKcropland and grasslands using
DailyDayCent. Agriculture, Ecosystems and Environment, 250, 1–11.
https://doi.org/10.1016/j.agee.2017.08.032
FLUXNET.(2019).Flux ne tdataset slist.Retrievedfromht tps://daac.ornl.
gov/cgi‐bin/dataset_lister.pl?p=9
Franz,D.,Acosta,M.,Altimir,N.,Arriga,N.,Arrouays,D.,Aubinet,M.,…
Vesala,T.(2018).Towardslong‐termstandardisedcarbonandgreen
housegas observationsfor monitoringEurope'sterrestrial ecosys
tems: A review. International Agrophysics, 32, 439–455.https://doi.
org /10.1515/intag‐2017‐00 39
Fratini, G.,& Mauder,M. (2014).Towardsaconsistent eddy‐covariance
process ing: An intercom parison of Eddy Pro and TK3. Atmospheric
Measurement Technology, 7, 2273–2281. https://doi.org/10.5194/
amt‐7‐2273‐2014
Fuss,S.,Lamb,W.F.,Callaghan,M.W.,Hilaire,J.,Creutzig,F.,Amann,T.,
…Minx,J.C.(2018).Negativeemissions– Part2:Costs,potentials
andsideeffects.Environmental Research Letters,13,0630 02.ht tps://
doi.org/10 .1088/1748‐9326/aabf9 f
Gallo, B., Demattê, J., Rizzo,R., Safanelli, J., Mendes, W., Lepsch, I., …
Lacerda,M.(2018).Multi‐temporal satellite imagesontopsoil attri‐
butequantificationandtherelationshipwithsoilclassesandgeology.
Remote Sensing,10(10),1571.https://doi.org/10.3390/rs10101571
Garten,C.T.,&Wullschleger,S.D.(1999).Soilcarboninventoriesunder
a bioenergy crop (Switchgrass): Measurement limitations. Journal
of Environmental Quality, 28, 1359–1365. https://doi.org/10.2134/
jeq1999.00472425002800040041x
Ge,Y.,Thomasson,J.A.,&Sui,R.(2011).Remotesensingofsoilproper
tiesinprecisionagriculture:Areview.Frontier s of Earth Science,5(3),
229238.https://doi.org/10.1007/s11707‐011‐0175‐0
GlobalSoilPartnership.(2017).Implementation plan for Pillar Five of the
Global Soil Partnership, harmonization of methods, measurements and
indicators for the sustainable management and protection of soil re‐
sources, providing mechanisms for the collation, analysis and exchange
of consiste nt and comparable g lobal soil data an d information.Retrieved
from http://www.fao.org/global‐soil‐partnership/pillars‐action/5‐
harmonization/en/
Goidts,E.,Wesemael,B.V.,&VanOost,K.(2009).Drivingforcesofsoil
organiccarbonevolutionat the landscape and regional scale using
data from a stratified soil monitoring. Global Change Biology, 15,
2981–3000.https://doi.org/10.1111/j.1365‐2486.2009.01940.x
Gomez,G.,Lagacherie,P.,&Bacha,S.(2012).UsingVis‐NIRhyperspec‐
tral data to map topsoil properties over bare soils in the Cap Bon
region,Tunisia.InB.Minasny,B.P.Malone,&A.B.McBratney(Eds.),
Digital sail assessments and beyond (pp. 387–392). New York, NY:
Springer.
GourlezdelaMotte,L.,Dumortier,P.,Beckers,Y.,Bodson,B.,Heinesch,
B.,&Aubinet,M.(2019).HerdpositionhabitscanbiasnetCO2 eco
systemexchangeestimatesinfreerangegrazedpastures.Agricultural
and Forest Meteorology, 268, 156–168. https://doi.org/10.1016/
j.agrformet.2019.01.015
Guenet, B., Moyano, F.E., Peylin, P.,Ciais,P.,& Janssens, I. A. (2016).
Towards a representation of priming on soil carbon decompo‐
sition in the global land biosphere model ORCHIDEE (version
1.9.5. 2). Geoscientific Model Development,9(2),841–855.https://doi.
org/10.5194/gmd‐9‐841‐2016
Haynes,R.J.,&Naidu,R.(1998).Influenceoflime,fertilizerandmanure
applications on soilorganicmattercontent andsoilphysicalcondi‐
tions:Areview.Nutrient Cycl ing in Agroecos ystems,51,123–137.https
://doi.org/10.1023/A:10 097383 07837
He,N.,Wu,L.,Wang,Y.,&Han,X.(2009).Changesincarbonandnitro
geninsoilparticle‐sizefractionsalongagrasslandrestorationchro‐
nosequenceinnorthernChina.Geoderma,150,30 2–3 08.ht tps://doi.
org/10.1016/j.geode rma.2009.02.004
Heidmann,T.,Christensen,B. T.,&Olesen,S.E.(2002).Changesinsoil
CandNcontentindifferentcroppingsystemsandsoiltypes.Report
81.InS.O.Petersen&J.E.Olesen(Eds.),Greenhouse gas inventories
for agriculture in the Nordic countries(pp.77–86).Foulum,Denmark:
Agriculture andFisheries,DanishInstituteofAgriculturalSciences,
MinistryofFood.
Hengl, T., Mendes de Jesus, J., Heuvelink, G. B. M., Ruiperez
Gonzalez, M., Kilibarda, M., Blagotić, A., … Kempen, B.
(2017).SoilG rids250m: Global gridded soil informationbased
on machine learning. PLoS ONE, 12, e 0169748 . ht tps://doi.
org/10.1371/journal.pone.0169748
Hirte , J., Leifeld, J., A biven, S., Obe rholzer, H.‐R., & Mayer, J. (2018).
Below gro und carbo n inputs to s oil via root biom ass and rhizo de‐
positionoffield‐grownmaizeandwheatatharvestareindependent
ofnetprimaryproductivity.Agriculture, Ecosystems and Environment,
265,556–566.https://doi.org/10.1016/j.agee.2018.07.010
Hoffmann,M.,Jurisch,N.,GarciaAlba,J.,AlbiacBorraz,E.,Schmidt,M.,
Huth,V.,…Augustin,J.(2017).Detectingsmall‐scalespatialhetero‐
geneityand temporaldynamicsof soilorganiccarbon(SOC) stocks:
Acomparison between automatic chamber‐derivedC budgets and
repeatedsoilinventories.Biogeosciences,14,1003–1019.https://doi.
org/10.5194/bg‐14‐1003‐2017
Hu,T.,Sørensen, P.,Wahlström,E. M., Chirinda,N., Sharif,B., Li,X., &
Olesen,J.E.(2018).Rootbiomassincereals,catchcropsandweeds
does not depend on aboveground biomass. Agriculture, Ecosystems
and Environment, 251, 141–148. ht tps://doi.org/10.1016/j .
agee.2017.09.024
Hutengs, C., Ludwig, B., Jung, A., Eisele, A., & Vohland, M. (2018).
Comparisonofportableandbench‐topspectrometersformid‐infra
reddiffusereflectancemeasurementsofsoils.Sensors (Basel),18 (4),
E993.https://doi.org/10.3390/s18040993
18 
|
   SMITH eT al.
Ibáñez,J. J.,SánchezDíaz,J.,deAlba,S.,LópezArias,M.,&Bioxadera,
J.(2005). Collectionofsoil informationinSpain: Areviewin2003.
InR. J. A.Jones,B. Houšková, P.Bullock,&L. Montanarella (Eds.),
Soil resources of Europe(2nded.,pp.357–363).EuropeanSoilBureau
Research Report No. 9, EUR 20559 EN. Luxembourg: Office for
OfficialPublicationsoftheEuropeanCommunities.
Jaber,S.M.,Lant,C.L.,&Al‐Qinna,M.I.(2011).Estimatingspatialvari‐
ations insoil organic carbon usingsatellite hyperspectraldata and
mapalgebra.International Journal of Remote Sensing,32,5077–5103.
https://doi.org/10.10 80/01431161.2010.494637
Jacobs,W.,Eelkema,M.,Limburg,H.,&Winterwerp,J.C.(2009).Anew
radiom etric instr ument for in situ m easurement s of physical se di‐
mentproperties.Marine and Freshwater Research,60(7),727.https://
doi.org/10.1071/MF08056
Jandl, R ., Rodeghi ero, M., Mar tinez, C ., Cotrufo, M . F.,Ba mpa, F., van
Wesemael,B.,…Miglietta,F.(2014).Currentstatus,uncertaintyand
futureneeds insoilorganiccarbon monitoring.Science of the Total
Environment, 468–469, 376–383. https://doi.org/10.1016/j.scito
tenv.2013.08.026
Janik,L.J.,Skjemstad,J.O.,Shepherd,K.D.,&Spouncer,L.R.(2007).The
predic tion of soil car bon fractio ns using mid‐infra red‐partia l least
squareanalysis.Soil Research,45(2),73–81.https://doi.org/10.1071/
SR06083
Jenkinson,D.S.,&Rayner,J.H.(1977).Theturnoverofsoilorganicmat
terinsomeoftheRothamstedclassicalexperiments.Soil Science,123,
298–305.https://doi.org/10.1097/00010694‐197705000‐00005
Jones,S.K.,Helf te r,C.,Anderson,M.,Coyle,M.,Campbell,C.,Famulari,D.,
…Skiba,U.M.(2017 ).Thenitrogen,carbonan dgreenhousegasbudget
ofagrazed,cutand fertilisedtemperategrassland.Biogeosciences,14,
2069–2088.https://doi.org/10.5194/bg‐14‐2069‐2017
Keel,S.G.,Leifeld,J.,Mayer,J.,Taghizadeh‐ToosiA.,Olesen,J.E.(2017).
Large uncertainty in soil carbon modelling related to carbon input
calculationmethod. European Journal of Soil Science,117,953–963.
https://doi.org/10.1111/ejss.12454
Kibblewhite, M. G., Jones, R. J. A., Montanarella, L.,Baritz, R ., Huber,
S.,Arrouays,D.,…Stephens,M.(2008).Environmental assessment of
soil for monitoring volume VI: Soil monitoring system for Europe.EUR
23490EN/6.Luxembourg:OfficefortheOfficialPublicationsofthe
EuropeanCommunities,72pp.
King,D.,Stengel,P.,Jamagne,M.,LeBas,C.,&Arrouays,D.(20 05).Soil
mapping a nd soil moni toring: St ate of progress a nd use in Franc e.
InR. J. A.Jones,B. Houskova, P.Bullock,&L. Montanarella (Eds.),
Soil resources of Europe.EuropeanSoil BureauResearchReportNo.
9, EUR 20559 EN (2nd ed., pp. 139–146). Luxembour g: Office f or
OfficialPublicationsoftheEuropeanCommunities.
Kirkby,K.J.,Smart,S.M.,Black,H.I.J.,Bunce,R.G.H.,Corney,P.M.,&
Smithers,R.J.(2005).Long term ecological change in British woodland
(19 71–2 0 0 1). English Nature Research Report 653. Peterborough,
UK:EnglishNature.139pp.
Lal, R., Smith, P.,Jungkunst, H., Mitsch, W., Lehmann,J., Nair,P.K., …
Ravindr anath, N. H . (2018). The car bon seque stration p otential of
terrestrial ecosystems. Journal of Soil and Water Conservation, 73,
145A–152A.https://doi.org/10. 2489/jswc.73.6.145A
Leeuwe n, J. P. V., Saby, N. P.A ., Jones, A ., Louwagie, G ., Micheli, E .,
Rutgers, M., … Creamer, R. E. (2017). Gap assessment in current
soil monitoring networks across Europe for measuring soil func‐
tions. Environmental Research Letters, 12, 124007. https://doi.
org /10.1088/1748‐9326/aa9c5c
Lehmann,J.,&Kleber,M.(2015).Thecontentiousnatureofsoilorganic
matter.Nature,528,60–68.https://doi.org/10.1038/nature16069
Leifeld, J., Ammann, C., Neftel, A., & Fuhrer, J. (2011). A compar‐
ison of repeated soil inventory and carbon flux budget to de‐
tect soil carbon stock changes after conversion from cropland
to grasslands. Global Change Biology, 17, 3366–3375. https://doi.
org /10.1111/j.1365‐2486.2011.02471.x
Leme nih,M .,Kar lt un,E.,&Olsso n,M.(20 05).S oi lo rga nicmatterdyn am
icsafterdeforestationalongafarmfieldchronosequenceinsouth
ernhighlandsofEthiopia.Agriculture, Ecosystems & Environment,109,
9–19.https://doi.org/10.1016/j.agee.2005.02.015
Li, W., Ciais , P., Guenet , B., Peng, S ., Chang, J., C haplot, V., … Yue, C.
(2018).Temporal response of soilorganic carbon after grassland‐
relatedland‐usechange.Global Change Biology,24,47314746.https
://doi. org/10 .1111/gc b.14 328
Liski,J.,Perruchoud,D.,&Karjalainen,T.(2002).Increasingcarbonstocks
intheforestsoilsofwesternEurope.Forest Ecology and Management,
169(1),159–175.https://doi.org/10.1016/S0378‐1127(02)00306‐7
Lobsey,C. R.,&ViscarraRossel,R. A.(2016).Sensing of soilbulk den‐
sity for m ore accurate c arbon accou nting. European Journal of Soil
Science,67,504–513.https://doi.org/10.1111/ejss.12355
Lorenz,K., Lal, R., & Ehlers,K. (2019).Soil organic carbonstock as an
indicator for monitoring land and soil degradation in relation to
UnitedNations'SustainableDevelopmentGoals.Land Degradation &
Development,30(7),824–838.https://doi.org/10.1002/ldr.3270
Louis, B. P.,Saby,N.,Orton,T.G., Lacarce,E.,Boulonne, L.,Jolivet,C.,
… Arrou ays, D. (2014). Statistic al sampling des ign impact on pre ‐
dictivequalityofharmonizationfunctions bet ween soilmonitoring
networks.Geoderma,213,133–143.https://doi.org/10.1016/j.geode
rma.2013.07.018
Mäkipää, R., Häkkinen, M., Muukkonen, P., & Peltoniemi, M. (2008).
Thecostsofmonitoringchangesinforestsoilcarbonstocks.Boreal
Environmental Research,13(suppl.B),120–130.
Mäkipää,R., Liski,J.,Guendehou,S.,Malimbwi,R., & Kaaya,A. (2002).
Soil carb on monitoring using surveys and modelling – General descrip‐
tion and application in the United Republic of Tanzania. FAOForestry
Paper168.Rome,Italy:FAO,46pp.Retrievedfromhttp://www.fao.
o r g / f o r e s t r y / 3 2 5 4 5 ‐ 0 7 8 a d e 9 3 9 2 3 8 3 3 5 d e a 8 5 7 0 f 1 3 4 c d 6 8 f c a . p d f
Malone,B., Hedley,C.,Roudier,P.,Minasny,B.,Jones,E.,&Mcbratney,
A.B.(2018).Auditingon‐farmsoilcarbonstocks usingdownscaled
national mapping products: Examples from Australia and New
Zealand. Geoderma Regional, 13, 1–14. https://doi.org/10.1016/
j.geodrs.2018.02.002
Manzoni,S.,&Porporato,A .(2009).Soilcarbonandnitrogenmineraliza
tion:Theoryandmodelsacrossscales.Soil Biology and Biochemistry,
41(7),1355–1379.https://doi.org/10.1016/j.soilbio.2009.02.031
Matthews,R .,Malcolm,H., Buys,G.,Henshall,P.,Moxley,J.,Morris,A.,
&Mackie, E.(2014).Changes to the representation of forest land and
associated land‐use changes in the 1990–2012 UK Greenhouse Gas
Inventory. Report to Department of Energy and Climate Change,
ContractGA0510,Edinburgh.
Mcbrat ney,A . B., Mend onça Santos, M . L., & Minasny, B. (2 003). On
digitalsoil mapping. Geoderma, 117, 3–52. https://doi.org/10.1016/
S0016‐7061(03)00223‐4
McConkey,B.G., Angers, D. A., Bentham, M., Boehm, M., Brierley, T.,
Cerkowniak,D.,…Worth,D.(2014).Canadian agricultural greenhouse
gas monitoring, accounting and reporting system. Methodology and
greenhouse gas estimates for agricultural land in the LULUCF sector.
Ottawa,Canada:AgricultureandAgri‐FoodCanada.
McHenry,M. P.(2009).Farmsoilcarbonmonitoringdevelopments and
landusechange:Unearthingrelationshipsbetweenpaddockcarbon
stocks,monitoring technologyandnew marketoptions in Western
Australia.Mitigation and Adaptation Strategies for Global Change,14,
497–512.https://doi.org/10.1007/s11027‐009‐9170‐0
Mckenzie, N., Henderson, B., & Mcd onald, W. (2002). Monitoring soil
change: Principles and practices for Australian conditions (112 pp).
CSIROLand&Water,CSIROMathematical&InformationSciences,
Nationa l Land and Water Re sources Aud it. Retrieve d from http://
www.clw.csiro.au/publications/technical2002/tr18‐02.pdf
McNeill,S.J.E.,Golubiewski,N.,&Barringer,J.(2014).Developmentand
calibration of a soil carbon inventory model for New Zealand. Soil
Research,52(8),789–804.https://doi.org/10.1071/SR14020