CHAPTER37
using BuIk soil RadiocarbonMeasurements to Estimatesoil organic Matter Turnover Times Kevin G. Harrison
I. Introduction This chapteroutlines a.strategyfor using bulk soil radiocarbonmeasurementsto estimate soil organic matter tumover in native, cultivated, and recovering soil. The turnover of soil organic nitrogen can also be inferred from thesemeasurements. Knowing soit carbon turnover times allows a f;st step toward understandinghow soil carbon,which containJthreetimes the amomt of carbon present in the preindushial atmosphere,will respondto anthr.opogenicperturbations,including changing land use, anthropogenicnihogen deposition,changing climati, -ico, i.niii#ion. The radiocarbon results suggest that soil "utlol.(exclusive oftitter; exchangesriinin"uniu.ounts of carbon with the atmosphere(-20-25 c/ve1!, thus having tire potentiat to iespond to perturbations. The increase !t "*'"t storagedue to co, fertilizatioi may potentially explain most of the so-called ,,missing $;]l Many workers have-estimated the global inventoryof soil carbonusing a variety of techniques: Schlesinger(1977) usedvegetationtyp'rr to estimatean inventoryoi i+so Gt. c; post et ar.(r9g2) estimatedsoil humus to hold 1395 G; c using climatic rir*on"r; e*aran et al. (1gg3) used soil orders to estimatean inventoryof 1576Gt c;-and Batjes(lgg6) rouna tqaz-t54g Gt c using data from 4'353 soil profiles' Aithough there are differencesbetweenthese techniques, the results generallyagree' How this large pooiof carbon influencesu*otptr..i. co, is uncertain, becausethe individual turnover times of the multitude of soil compounds are noi wett known. These turnover times may range from days to millennia. The purpose of this chapter is to introduce a strategy for estimating the turnover.time of soil organic matter using burk soil iadiocarbon measurements. The techniquebuilds on the approachesof otherr, th" c""t"ry ""a Rothamstedmodels, mass balance studies, and fractionation of soil humus. The century and Rothamsted models use measurementsof soil carbondecomposition as the.foundationfoi soptristlcateaecosystem models. The model structuresare similar, having soil organicmaterial consist'ingof fast, active, and passive fractions. The century model (parton Jt ur., 1gg7, lggg, rss:; schim;l et ar., lg94) has an active carbon turnover time ranging from 20 to 50 years, and, it assignsa turnover of g00 to 1200 years to passive carbon' The Rothamstedmodel (Jenkinson, 1990)usesa 2L-yearturnovertime for active carbon and a near infinite turnover time for passive carbon. o'Brien and Stout (1978)-use a sophisticated model to interpret their New Zealand soil radiocarbon measurements. Their model includes carbon input, a'".o.position rates, and soil diffusivity, which is constrained by the depth distribution or *Jil.*uon and total carbon, and it ISBN0-8493-7441_3 @1997 byCRCpress, Inc. 54g
550
Kevin G. Harrison
Table l. Prebombsoil radiocarbonvaluesfor uncultivatedsoil % modern
96 96 82 96 90 94 6 .
llI
Depth
year
Reference
1 6
1959 1959 1927 1962 1962 1959
O'Brien,1986 Trumbore,1993 Trumboreet al., 1990 Campbelletal.,1967 Campbell et al.,l96j Vogel,1970
l-8 cm 0-23cm 0-12cm A-horizon A-horizon 0-2 cm
Mollisol New Zealand T. forest California Spodosol USSR ChernozemicCanada Mollisol Canada Forest
includescarbon input' decomposition,and soil diffusivity. They assign a 50-yeartumover time for active carbon and a near infinite time for passivecarbon. Researchershave tried to separateactive and passivecomponentsusing physical and chemical fractionationtechniques(Paul et al.,1964;Cambeliet al., l96i;Martel and paul, 1974b;Gohet al., 1976, 1977,1984;Scharpenseel et al., l96ga,b;Trumboreet al.,19g9,1990). Trumbore(iss:) summarizes the results of various fractionation techniques. one way to test the effectiveness of fractionation is to see if the amountof bomb radiocarbonis distributed in the soil as predictedby estimatesof the residencetime for soil organicmatter. To date,none of the availablefractionation schemesreproducethe expandeddistributionof bomb carbon(Trumbore. 1993).
II. Estimating Soil carbon Turnover Times Using Bulk Radiocarbon Measurements My researchusesa time-stepone-boxmodel and bulk soil radiocarbon measurements to estimate tumover times and inventoriesof active and passivecarbon. The model has atmosphericC-14 varues and CO, concentrationsfor every year from 1800 until the present. The user selectsthe carbon inventory and the turnover time. The turnover time equalsihe carbon inventory divided by the exchange flux. The exchange flux equals the amount of carbon that is added io the box (from photosynthesis)or lost fromthe box (respiration). Lossesthrough erosionand dissolution are thought to be small (Schlesinger,1986)and are not considered.The model can be run in either a steadystate mode (where the flux in equalsthe flux out) or in a nonsteadystatemode (in which carbon is either accumulatingor decreasing).My researchusesthis model and soil radiocarbon data to show that soil carbon has more than one component and to estimatethe turnover time of the passivellaction, the proportionsofactive and passivecarbonin surfacesoil, and the active soil carbonresidencetime. Many researchershave concludedthat soil consistsof a complex mix of organic moleculeswhose turnover times rangefrom a few yearsto thousandsofyears. Tiris pool cannot be characterizedby a singleturnovertime (O'Brien and Stout,1984;Balesdent,1988;parton, lgg7, l9gg,l993; Jenkinson and Raynor, 1977)' For example,six publishedprebomb values for the averageradiocarbon content of surfacesoil are 92o/omodem (Table l). "Modern" meansthe amount of radiocarbonrelativeto 1850wood. This 92%omodern value indicatesa 650-yeartumover time, which would show very little increasein bomb radiocarbon with time. Further,diilerent types of vegetation and soil rypeswithin the sameclimate will often have differentvalues. In reality, ihe soil rJdiocarbon valuesincreasein the 1960sand then level off(Figure 1). This increasesuggeststhat soil organicmaterialcontainsan active componentwith a turnover time significantlylessthan 650 years. This active componentmust be diluted with a passivecomponenthaving a turnovertime of greater than 650 years. The turnover time of passive soil carbon can be estimatediom soil radiocarbonmeasurements made at depthswhere little or no activesoil carbonis present. Sometropical soils have activesoil carbon several meters below the surface(Nepsted et-a|., 1994 and Fisher et al., |994);howeveil
I
( I
I c s fi
I !
n G. Harrison
using Bulk soil Radiocarbon Measurements to Estimatesoil organicMatterTumover Times
rcation >wZealand ilifornia ]SR urada nada ,rmany
81,4 o o 3
-.-
\J
551
650-yearcarbonpool nativesoil
Et,z
G'
o \f
ver time for rd chemical ; Gohet al., bore(1993) ;tivenessof rredicted by ractionation
I
91.0
g
o.
tr G
^on a
(J"'e $ I
o I
0.6 I 950
to estimate l-14 values the carbon ided by the box (from are thought steadystate ron is either ow that soil action, the :nce time. :uleswhose terizedby a ; Jenkinson lon content r relative to w very little ypes within increasein containsan ronentmust asurements r activesoil ); however,
1970 yeor
1990
Figure l' Soil radiocarbonvaluesvs. time' Measuredsoil radiocarbonvaluesfor noncultivated soils are plotted againsttime from 1950to 1991. The valuestend to increaseduring the 1960s and then leveloff' Moderresults p.1111.:rrl.a c*uonfool timeareshownby laving" 650J,;;;;;sidence the line' Data for thesefiguresare from Tables I and,2,andHarrison(rg94\.
Table2. Deepsoilradir %omod,em % modern Depth n"ntn (cm) r"R"f.."n.. 43 60 60 62
74-94 60-140 60-140 s s - ll 0
O'Brien and Stout.7g ScharpenseelandBecker-Heidmann, g9 Becker-Hiedmann,89
Tsutsuki et al.,88
Mollisol Vertisol
Udic
New Zealand Israel India
radiocarbonvaluesdecreasewith depth, which showsa decreasein the proportion of active to passive carbon (Hanison et al., 1993a). At ,orn. depth, the soil radiocarbon
minimum, where varues tendrodecrease vert'srowry. These var'es #"#JrirTi:#ffiff:"l locations(seeTable2)' The
averageualu" 6, the siteslistedin iuit" z is 55vomodern,which corresponds to a 4700-yearturnovertime for passivesoil carbon. For theprebombcondition,onecanestimate theproportionsof activeandpassivecomponents in surfacesoilusingthe55%o modernpassivesoilradiocarbon uutu.unJih. gzN ^od"mvalue measured for thebulk soil (Tablel)' If theu"tiu. .oo,fo"*ll-r overquickly(<100years)enoughsothat canassumethat its radiocarbon we valueis almost100percentmodern(radiocarbon hasa half life of
552
Kevin G. Harrison
5,700 years), a mixture of 11% passive and 83%oactive leads to the observed averageradiocarbon value of 92ohmodern (Figure 2). The post-bombincreasein soil radiocarbonvalues can be used to estimatethe active soil carbon turnover time (Hanison, 1993a). A25-year turnover time producesthe best fit to the available data. Most of the points are for temperateecosystems,and warmer tropical ecosystemsmay have faster turnover times, while cooler boreal turnover times may be slower. For example, Bird et al. (1996) found faster soil carbontumover times in the tropics and Trumbore et al. (1996) found that soil carbon turnover times decreasedwith increasingtemperature.The proportionsof active and passivesoil may differ for tropical and boreal climates. This approach can be validated by looking at a specific site where field data can be used to comparemodel predictionswith soil radiocarbonmeasurements.O'Brien and Stout (1978) reported measurementsfor a New Zealand grasslandsoil that included a deep soil value and a time series of surfacesoil valuesthat extendedfrom prebombtimes into the mid-1970s. The model that best fit, the data consistedof a l2%opassive,88% activeportion that tumed over every 25 years. Thesevalues are very similar to those derived from the available data for soil radiocarbon globally, with the slightly shorter turnover time being the most significant difference. Further, Figure 3 shows the excellent agreement befween the model and the data. This model reproducesthe prebomb soil radiocarbon values and the post-bomb increasein radiocarbon values in native soil. The model can be further validated by seeing if derivatives could explain radiocarbon measurements in cultivated and recoveringsoil.
III. Soil Carbon Dynamics in Cultivated Ecosystems Soil loses about 25Yo of its carbon when cultivated (Schlesinger,1986; Post and Mann, 1990; Davidson and Ackermen, 1993). This loss stemsfrom reducedinputs of organic matter and increased rates of organic matter mineralization. Cultivated soil generally has lower radiocarbon values than native soil that hasbeen sampledat the sametime (Figure 4; Martel and Paul, 1974). Hsieh (1992 nd 1993)has developeda two-componentmodel that reproducestemporal changesin radiocarbonvalues in cultivated soil. Using a similar approach,Harrison et al. (1993b) assumedthat carbon lost from soil due to cultivation would be lost from the activecarbonpool. However,the oxidation and loss of a fiaction ofthe active-soilcarboncould only explainabouthalfthe observedradiocarbondepletion. Mixing subsurfacesoil with the shallowsurfacesoil throughcultivation (i.e., the plow mixes up the soil) can accountfor the remainingdepletionin radiocarbonvalues. The model, which included mixing and oxidation, produced good agreementwith the available data for changes in soil radiocarbonupon cultivation (Figure4).
IV. Soil Carbon Turnover in RecoveringEcosystems To further test the model, the baselinemodel was modified to explore the carbon dynamics of a recoveringsoil that was increasingits carbonstores. One exampleincludesa recoveringtemperate forest locatedin the CalhounNational Forestof South Carolina that is describedby Richter et al. (1994, 1995),Hanison (1994), and Harrisonet al. (1995). This site containsLoblolly pine (Pinus taeda)thatwas plantedin 1959on land that had beencultivatedfor 150 years. From 1962to 1968, the surfacecarbon concentrationincreasedfrom 5.9 Mglha to 8.0 Mg,&a. The native soil carbon model was modified to take into accountthis carbonaccumulationby increasingthe flux of carbon into the active-soilcarbonpool. The tumover time that best reproducedthe observedradiocarbon measurementswas 12 years, which is about twice as fast as carbon turnover in native ecosystems. Figure 5 shows the agreementbetweenthe model and the data for this accumulating ecosystem.
n G. Harrison
UsingBulk Soil Radiocarbon Measurements to EstimateSoil OrganicMatterTumover
rimes
553
82,0 o
radiocarbon
=
: soil carbon 'ailable data.
o
l
B I . 6J
/ have faster et al. (1996) I soil carbon ;ive soil may
:
l
O
J
+ - l T ( ) 1 . 2I1
n be used to 78) reported ime seriesof t best fit, the rsevalues are r the slightly the excellent radiocarbon an be fuither Lltivated and
o
l
A
L
G
I
E
t--
!0.8 J
o
l
:
.
l
a
l l
-
o.4F 1950
1970 y e ar
1980
1990
Figure 2. Soil radiocarbonmodel predictionsand soil radiocarbonobservationsvs. time. A mixture of 83% active and 17% passivesoil carbon producesthe best visual fit to the native soil radiocarbon measurement(open circles). The concentrationof atmosphericradiocarbon almost doubled because of nirclear bomb testing. Mann, 1990; nd increased n values than eh (1992 and arbon values lost from soil and lossofa on depletion. mixes up the rich included mges in soil
!,
1.4
o o =
o ro
1.2
@
o lf I
o g
1.0
CL
E
(g
o
o ynamics of a ing temPerate Richter et al. y pine (Pinus 1962to 1968, 'e soil carbon lux ofcarbon I radiocarbon I ecosystems. cosystem.
0.8
.l I
o 0.6 1950
1960
1970
1980
1990
y s ar Figure 3. New Zealandtest case.O'Brien and Stout (1978) publishedradiocarbondata for a New Zealandgrasslandsite comprisedof a time seriesof surfacesoil, including one prebomb and one deep soil radiocarbon value. This information was used to to attempt to validate the model for a specific site.
Kevin G. Hanison
554
1,4
g
o o -
3t.z
@
o
natlvg
ll
\ u.ry...-r-l-\-
a g l t l
Iq. E (lt
\---\ \
o
mlxlng
6o.a rl I
o 0.6 1950
1970 yeor
1960
1980
990
Figure 4. Native vs. cultivated soil radiocarbonvalues. Cultivated soil has lower radiocarbon values tha*n native soil. This difference is caused by mixing and oxidation. The plow mixes deeper radiocarbondepletedsoil with radiocarbonrich surfacesoil, diluting the amount of active soil carbon in the surface. Increased soil organic matter oxidation further reduced the inventory of active soil carbon. T'
o o =
3 t.z @ o
o CL
E (lt
at,
60.8 tr
a
o
modelwithaccumulation soilradiocarloon measured modelwithoutaccumulation
0.6
r 970
1990
Year results that Figure 5. Radiocarbon measurementsand model results for South Carolina. The model that took model accumulation an from those were Uestnt the surfacesoil radiocarbonmeasurements Mg ha-'. to 8.0 ha' Mg from 5.9 increased carbon The into accountthe increasein carbon inventory. soils. for native global average as the as fast twice is about time The active reservoir turnover
n G. Harrison
Using Bulk Soil Radiocarbon Measurements to EstimateSoil OrganicMatterTurnoverTimes Table 3. Soil carbon
555
fertilization
due to
r99l
Postet al.,1982 Eswaranet al.. 1993 1400 1600
Total soil carbon(Gt C)
1500
Nonwetlandsoil carbon (Gt C)
1000
1200
r250
Activesoil carbon(Gt C)
s00
600
625
Exchangeflux (Gt C yr')
20
.,^
PotentialC sequestration dueto CO, fertilization (ct C yr-')
0.5
0.6
25 0.'7
7o "miss
sink" Thistableusesestimatesoftheinventoryofactive'oil"arb estimatethepotentialcarbonsequeshation in soil dueto CO, fertilization. Only nonwetlandsoil was
rrbonvalues rixesdeeper : soil carbon rf activesoil
included in the calculation. The inventory of active soil was esimtatedassumingthat a soil profile containsabout 50% active carbon; aCOrfertilization factor of 0.35 was used in the CO, fertilization modelto calculatethe amountof carbonsequestered during an averageyear in the 1980s(Harrison, 1993a).Dixon et al.'s (1994)missingsink estimateforthe 1980sof l.l Gt C/yearwas used.
v. Determining the Global Inventory of Active Soil organic Matter
I results that lel that took 8.0 Mg ha-'. r soils.
Table 3 lists the estimatesfor the global inventory of soil organic matter in nonwetland ecosystems. Thesevaluescan be usedto estimatethe global inventory of active soil carbon. The active-to-passive proportions found in surfacesoil cannot be applied to all carbon present in nonwetland soil because the proportion of active to passivecarbondecreaseswith increasingdepth. The integratedinventories suggest that the global pool is about 50% passive and 50Yo active (Harrison, tee3a;. While extrapolatingthesedistributionsinvolvesuncertainty(i.e., the proportions are likely to diffei for other climatesand types of vegetation),the global inventory of active carbon ranges from 500 to 625 Gt C. A way to confirm the model is to seeif measuredfluxes from the soil agreewith values predicted by the model. A 500 Gt. C pool turning over every 25 years emits 20 Gt. C from the soil annually. This translatesinto a 150 gC/m2/yr from the world's uplandssurface. The 600 and 625 Gt. C. values result in 24 GtClyeat (180 g Clm2lyr) and25 Gt C/year(190 g Clm2lyr). All are still significantly lower than the observedflux from a temperateforest soil ranging from 400 to 500 g C/m2lyear(Raich and Schlesinger,1992). Yet, the measuredvaluesinclude sourcesof carbon dioxide besidesmicrobial respiration of soil organic matter, such as root respiration and oxidation of liuer and fme roots (perhapsas much as 50%;o of NPP). It is impossibleto separatethese CO, sources,but it is unlikely that they are greaterthan 50% ofthe total. Also, land having low organiccarbon contents,such as desertsoil, make it diffrcult to compareglobal and regionalvalues. It would be impossibleto get better agreementbetweenthe fluxes predictedby the model and measuredfluxes for a temperateforest ecosystembecauseofthese dif[erences. However, ifone considersthe measuredfluxes tobe an upper limit, the predictedvaluesfall well below it.
Kevin G. Harrison 556
VI. Carbon DioxideFertilization of fastcyclingsoil carbon'it is possibleto estimate Having estimatedthetumovertime andinventory CO, fertilizationoccurs ;iored in soil bicauseof CO, fertilization' theamounlof carbonp";;t to elevatedcarbondioxidelevels(StrainandCure' tn#s"*rtt *rten expo.sed whenplantsincrease co' way or expressing.co,fertilizationis with a tgg5; BazzazandFajei T,ial a .onnrni.nt concentration)' co' in growttlror-adouutingof rncrease fertilizationfactor(i.e.,thepercentage haveshownincreasedgrowthat elevatedCO' experimlnts Althougtrmanyindoor iOrt"rtit2ution is highlycontroversial tir.r. resultsto naturalvegetation levels(Strainarra C,rre,lSgii "*t upotuti.rg going to co, fertilizationfactors the carbonflux (BazzazandFajer,r ss2). F;tlhrt, apryitr! tn"sa grown plants under carbon soil the .-6nti, i^i'"t ui' irqssl havesirownthat into soil is speculation pairedcounterparts' nonelevated their in than was-greater chambers in doubledco2 in open-top foundevidenceof (1992) etal' Norby significant'Also' althoughtheir results*"r. iot statistically lendingcredibilityto the CO' concentrations' fmeroot densityfor treesg.o*o intttuut"a increased will soil carbonstorage' belief that if plant growthis stimulated'so to estimatethe additionalamountof u",n developed For this study,a Co,ferti|izationmodelt'u, mattercanbe fertilizatlon' itre flux of carboninto soil organic carbonstoredin ,oif U"iuur" of CO, (0'35 after p * deltanc9'- ; en' p is the co'- fertilizationfactor increasedby addingfl* ;; flux. exchange ct,angein carbondioxide,andEF is the Harrison,|993a), deltapCo, is thefractionat amount the times flux (i.e-,thedecayconstant thedec"ay As theflux of carbon^ri rr,. uo* increases, cgbol dioxidestopsincreasing'the uittosptt"tic r"u"ioi of carbonin thebox) ulro i."."ar"r. If the soilwillattainahighersteadystatecarboncontentwithane-foldingtimeof25years. in soil becauseof cot of carbonsequestered Tabte 3 lists the modelpredictions.The amount estimatethat the (1994) al' et ior the 1980s' Dixon fertilizationrangesfrom 0.j to 0.7 Gt Clyear canbe storing fertilization ..missingsink, is 1.1Gt C/yearfor thistimeperi,oJ.Thus,carbondioxide "missingsink" in soil' muchoithe VII. Conclusion and Future Research ThischapterpresentsaSffategyforestimatingtheglobalinlgnto.vandturnovertimefornonwetland time and trruiirtiu" soil carbonhasa 25-yearturnover datasuggests soil. Theavailableradiocarbon perturbations to activesoil carbonmayrespondsignificantly a 500to 625Gt.c inventory.Therefore, deposition'CO' fertilization changingclimate,unJ*,t'.opogenic nitrogen suchas COrfertilization, of the"missingsink'" Grc/yearin t"il,;;; "xpiui"i"g u ,,'uiotportion canpotentiallystore0i;0,
Acknowledgments that andmanydiscussions improvingthemanuscript for helpfulcomments.for I thankBill schlesinger calhoun providing and data io,.u""J* to calhoun providedthe"ont"r.tiorlt is studyandDanni"*". Zambellaforher supportandencouragement' Ann Beth thank I *.urur"n'.nis. soil for radiocarbon this research' Th" NutionalScienceFoundationsupported
References Balesdent,J.,G.H.Wagner,andA.Mariotti.lg88.Soilorganicmatterturnoverinlong-termfield soi/ sci. soc-Am.J, 52l.118-124' c-13 naturalabundance. ", ,;;;ilty experiments
in G. Harrison
Leto estimate zation occurs rin and Cure, ; with a CO, ntration). elevatedCO, ;ontroversial rn flux going plants grown ;ounte{parts, I evidenceof libility to the al amount of natter can be r (0.35 after lchangeflux. s the amount creasing,the :auseof CO, nate that the rn be storing
nonwetland ver time and rcrturbations fertilization rissingsink."
;ussionsthat ing Calhoun curagement.
g-term field 4.
Using Bulk Soil Radiocarbon Measurements to EstimateSoil OrganicMatterTurnoverTimes
557
Batjes, N. H. 1996. Total carbonand nitrogen in the soils of the world. European J. Soit Sci. 47:l5l163. Bazzaz,F.A. and E.D. Fajer. 1992.Plantlife in a COr-richworld. Scr.Amer.266:68-74. Bird, M. I., A. R. Chivas,and J. Head. 1996.A latitudinal gradient in carbon hrnover times in forest soils.Nature 381:143-146. Campbell, C.A., E.A. Paul,D.A. Rennie,and K.J. McCallum. 1967.Applicability of the carbon-dating method of analysisto soil humus studies.Soil Sci. 104:217-223. Davidson, E. A. and Ackerman, l.L. 1993. Changesin soil carbon inventories following cultivation of previously untilled soils. Biogeochemistry20:16l-193.. Dixon, R.K., S. Brown, R.A. Houghton,A.M. Solomon,M.C. Trexler, and J. Wisniewski. 1994. Carbonpools and flux ofglobal forest ecosystems. Science263:I 85-190. Eswaran,H., E.V. Den Berg, and P. Reich. 1993.Organiccarbonin soils of the world. Soil Sci. Soc. Am. J. 57:192-194. Fisher, M. J., I.M. Rao, M.A. Ayarza,C.E. Lascano,J.I. Sanz,R.J. Thomas,and R.R. Yera. 1994. Carbon storage by introduced deep-rooted grassesin the South American savannas.Nature 371:236-238. Goh, K.M, T.A. Rafter, J.D. Stout,and T.W. Walker. 1976. Accumulation of soil organic matter and its carbon isotopecontentin a chronosequence ofsoils developedon aeoliansandin New Zealand. J. Soil Sci.27:89-100. Goh, K.M., J.D. Stout,and T.A. Rafter. 1977.Radiocarbonenrichmentof soil organic fractions in New Zealandsoils.Sori Sci. 123:385-390. Goh, K.M. J.D. Stout, and J.O'Brien. 1984. The significanceof fractionationdatingthe age and turnover of soil organicmatter.New ZealandJ. Soil Sci.35:.69-72. Harrison, K. G. 1994.The impact of CO, fertilization,changingland use, and N-depositionon soil 'carbon storage.ColumbiaUniversity Ph.D. Thesis. Harrison, K. G., W. M. Post, and D. D. Richter. 1995.Soil carbonhrnover in a recovering temperate forest. Global Biogeochemical Cycles 9 :449-454. Harrison, K. G., W. S. Broecker, and G. Bonani. 1993aA strategyfor estimating the impact of CO, fertilization on soil carbon storage.Global Biogeochem.Cycles 7 :69-80. Harrison, K.G., W.S. Broecker, and G. Bonani. 1993b. The effect of changing land use on soil radiocarbon.Science262:725-726. Becker-Heidmann,P. 1989.Die Teifenfunktionender naturlichenKohlenstoff-Isotopengehalte von vollstandig dunnschichtweisebeprobten Parabraunerdeund ihre Relation zur Dlmamic der organischenSubstanzin diesenBoden." Ph.D. Thesis,HamburgUniversity. Becker-Heidmann,P., Liu Liang-wu, and H.W. Scharpenseel. 1988.Radiocarbondating of organic matter fractionsof a Chinesemollisol. Z. Pflanzenernahr.Bodenk. 151:37-39. Hsieh,Y-P., l992.Pool sizeandmeanageof stablesoil organiccarbonin cropland,Sol. Sci. Soc.Am. J.,56:460-464. Hsieh, Y.P. 1993.Radiocarbonsignaturesof turnoverratesin active soil organiccarbonpools. Soll Scr.Soc.Am. J. 57:1020-1022. Jenkinson, D.S. and J.H. Raynor. 1977. The turnover of organic matter in some of the Rothamsted classicalexperiments.Soil Sci. 123:298-305. Jenkinson,D.S. 1990.The turnover of organic carbon and nitrogen in soil. Phil. Trans. R. Soc.Lon. 8,329,361-368. Martel, Y.A. and E.A. Paul. l974a.Effects of cultivationon the organicmatterof grasslandsoils as determinedby fractionation and radiocarbon dating. Can. J. Soil Sci. 54:419-426. Martel, Y.A. and E.A. Paul. 1974b.Use of radiocarbondatingof organicmatter in the study of soil genesis.Soil Sci.Soc.Amer. Proc.38:501-506.
Ii l
558
Kevin G. Harrison
Nepsted,D.C., D.R. de Carvalho,E,A. Davidson,P.H. Jipp, P.A. Lefebwe, G.H. Negreiros,E.D. da Silva, T.A. Stone,S.E. Trumbore,and S. Vieira. 1994.The role of deep roots in the hydrological and carbon cycles of Amazon forests and pastures.Nature372:666-669. Norby, R.J., C.A. Gunderson,S.D. Wullschleger,E.G. ONeill, and M.K. McCracken. 1992. Productivity and compensatoryresponsesof yellow-poplar treesin elevatedCOt. Nature 357:322' 324. O'Brien, B.J. and J.D. Stout. 1978.Movementand turnover of soil organicmatter as indicatedby Soil Biol. Biochem.l0:309-3 17. carbonisotopemeasurements. O'Brien, B.J. 1984. Soil organic carbon fluxes and turnover rates estimatedfrom radiocarbon enrichments.SoilBiol. Biochem. l6: I l5-120. O'Brien, B.J. 1986.The use of natural and anthropogenicC-14 to investigatethe dynamicsof soil organic carbon. Radiocarbon 28:358-362. parton,W.J., C.V. Cole, J.W.B. Stewart,D.S. Ojima, and D.S. Schimel. 1989.Stimulatingregional patternsof soil, C,N and P dynamicsin the U.S. central grasslandsregion. p. 99-108. In: M. Clarholm and L. Bergstr0m (eds.), Ecologt of Arable Land. Kluwer Academic Publishers. Norwell, MA. parton, W.J., D.S. Schimel,C.V. Cole, and D.S. Ojima. 1987.Analysis of factors controlling soil Sol/ Sci.Soc.Am. J. 5l:.1173-1179. organicmatterlevels in GreatPlainsGrasslands. Parton,W.J., M.O. Scurlock,D.S. Ojima, T.G. Gilmanov,R.J. Scholes,D.S. Schimel,T. Kirchner, J-C. Menaut,T. Seastedt,E. GarciaMoya, A. Kamnalrut,and J. I. Kinyamario. 1993. Observations and modeling of biomass and soil organic matter dynamics for the grasslandbiome wordwide. Global BiogeochemicalCycles7:785-809. paul, 8.A., C.A. Campbell,D.A. Rennie,and K.J. McCallum. 1964.lnvestigationsof the dynamics of soil humusutilizing carbondatingtechniques.In: Transactions8th Int. Soil Sci. Soc., Bucharest, Romania, 201-208. post, W.M. and Mann, L.K. 1990. Changesin soil organic carbon and nitrogen as a result of cultivation.p.401-401.In: Soils qnd the GreenhouseEffect. Joht Wiley & Sons,NY. post, W.M., W.R. Emanuel,P.J.Zinke, and A.G. Stangenverger. 1982. Soil carbonpools and world life zones.N ature 298:156-159. Raich, J.W. and W.H. Schlesinger.1992.The global carbondioxide flux in soil respirationand its relationshipto vegetationand climate.TellusB, 44,81-89. Richter, D.D., D. Markewitz, C.G. Wells, H.L. Allen, J. Dunscombe,K. Harrison, P.R. Heine, A. B. Unego, and G. Bonani. 1995.Carboncycling in a Loblolly pine forest: implications Stuanes, for the missing carbon sink and for the concept of soil. p. 233-251. In: W. MacFee (ed.), Proceedings,Eighth North American Forest SoilsConference,Univ.of Florida, Gainesville. fuchter,D.D., Markewitz,D., Wells, C.G., Allen, H.L', April, R., Heine, P.R., and B' Unego' 1994' Soil chemical changeduring three decadesin an old-field loblolly pine ecosystem.Ecologt 7 5:1463-14'73. H.W., C. Ronzani,and F. Pietig. 1968a.Comparativeage determinationson different Scharpenseel, humic-matterfractions.p.67-74.In: Proc. Symposiumon the use of isoptopesand radiation in soil organic matter studies,Vienna. Intemational Atomic Energy Commission. H.W., M.A. Tamers,and F. Pietig. 1968b.Altersbestimmunbon Boden durch die Scharpenseel, Zeitschr. Pflaraenemernahr Bodenkunde l19:34-52. Radiokohlenstoffdatierungsmethode. H.W. and P. Becker-Heidmann.1989. Shifts in C-14 patternsof soil profiles due to Scharpenseel, bomb carbon. Radiocarbon3l :627'636. Schimel,D.S.,Braswell,8.H., Holland,E.A., McKeown, R., Ojima, D.S., Painter,T.H., Parton,W.J., and Townsend,A.R. 1994.Climatic, edaphic,and biotic controls over storageand turnover of carbonin soils. Global BiogeochemicalCycles8:279'273. Schlesinger,W. H. 1991.Biogeochemistry:An Analysis of Global Change.New York: Academic Press,433 pp.
l
in G. Harrison iros, E.D. da hydrological rcken. 1992. ure357:322indicated by radiocarbon amics of soil ting regional - 1 0 8 .I n : M . ; Publishers. ntrolling soil 9. T. Kirchner, Observations re wordwide. he dynamics :., Bucharest, s a result of
ry. rls and world :ation and its R. Heine, A. implications tacFee (ed.), ainesville. Inego.1994. em. Ecologt ; on different Jiation in soil en durch die ,:34-52. 'ofiles due to Parton,W.J., I turnover of
MatterTumoverTimes to Estimate SoilOrganic Measurements UsingBulkSoilRadiocarbon
559
W.H. 1977.CarbonBalancein TenestrialDehitus.Ann.Rev.Ecol. Syst.8:51-81. Schlesinger, propertieswith disturbance W.H. 1986.Changesin soil carbonstorageandassociated Schlesinger, andrecovery,p.194-220.In:J.R.TrabalkaandD.E.Reichle(eds.),TheChangingCarbonCycleNY. A GlobalAnalysrs,Springer-Verlag, carbondioxideon vegetation.DOE/ERStrain,B.R. andJ.D.Cure.1985.Directeffectsof increasing 0238. Trumbore, S. E. 1993.Comparisonof carbondynamicsin tropical and temperatesoils using . GIobal Bi ogeochemi cal CycIes 7:275-290. radiocarbonmeasurement Trumbore,S.E.,G. Bonani,andW. Wolh. 1990.The ratesof carboncyclingin severalsoils from soilorganicmatter.p.407-4l4.In:A.F.Bouwman(ed.), of fractionated AMS C-14measurements Effect,JobnWiley & Sons,NY. Soilsand the Greenhouse of fractionated soil Trumbore,S.E.,J.S.Vogel,andJ.R.Southon.1989.AMS C-14measurments organicmater.Radiocarbon, 3 | :644-654. betweensoil carbonand Trumbore,S.E., O. A. Chadwick,andR. Amundson.1996.Rapidexchange change.Science272:393-396. carbondioxidedrivenby temperature atmospheric 1988. and H.W. Scharpenseel. Tsutsuki,K., C. Suzuki,S. Kuwatsuka,P. Becker-Heidmann, Bodenk 151 :87-90. Pflaruenernahr. in mollisols. S. of thehumus of thestabilization Investigation IAEA-SM-129/15:225-237. Vogel,J.G.1970.Carbon-14DatingofGroundwater.IsotopeHydrologt and D.L. Randlett.1993.Elevated R. Fogel, J.A. Teeri, P.S. Curtis, Zak, D.R., K.S. Pregitzer, Plant andSoil 151:105-l17. nitrogen cycles. carbon and between COrfeedback atmospheric
