Integrating microbial physiology and physio-chemical principles in

Biogeosciences 11 3899ndash3917 2014wwwbiogeosciencesnet1138992014doi105194bg-11-3899-2014copy Author(s) 2014 CC Attribution 30 License

Integrating microbial physiology and physio-chemical principles insoils with the MIcrobial-MIneral Carbon Stabilization (MIMICS)modelW R Wieder1 A S Grandy2 C M Kallenbach2 and G B Bonan1

1National Center for Atmospheric Research Boulder Colorado 80307 USA2Department of Natural Resources and the Environment University of New Hampshire Durham NH USA

Correspondence toW R Wieder (wwiederucaredu)

Received 19 December 2013 ndash Published in Biogeosciences Discuss 17 January 2014Revised 12 June 2014 ndash Accepted 12 June 2014 ndash Published 24 July 2014

Abstract A growing body of literature documents the press-ing need to develop soil biogeochemistry models that moreaccurately reflect contemporary understanding of soil pro-cesses and better capture soil carbon (C) responses to en-vironmental perturbations Models that explicitly representmicrobial activity offer inroads to improve representationsof soil biogeochemical processes but have yet to considerrelationships between litter quality functional differences inmicrobial physiology and the physical protection of micro-bial byproducts in forming stable soil organic matter (SOM)To address these limitations we introduce the MIcrobial-MIneral Carbon Stabilization (MIMICS) model and evalu-ate it by comparing site-level soil C projections with obser-vations from a long-term litter decomposition study and soilwarming experiment In MIMICS the turnover of litter andSOM pools is governed by temperature-sensitive MichaelisndashMenten kinetics and the activity of two physiologically dis-tinct microbial functional types The production of micro-bial residues through microbial turnover provides inputs toSOM pools that are considered physically or chemically pro-tected Soil clay content determines the physical protectionof SOM in different soil environments MIMICS adequatelysimulates the mean rate of leaf litter decomposition observedat temperate and boreal forest sites and captures observedeffects of litter quality on decomposition rates MoreoverMIMICS better captures the response of SOM pools to exper-imental warming with rapid SOM losses but declining tem-perature sensitivity to long-term warming compared witha more conventional model structure MIMICS incorporatescurrent microbial theory to explore the mechanisms by which

litter C is converted to stable SOM and to improve predic-tions of soil C responses to environmental change

1 Introduction

The response of the terrestrial carbon (C) cycle to projectedenvironmental change remains highly uncertain in Earth sys-tem models (Arora et al 2013 Friedlingstein et al 2006)Some of this uncertainty results from challenges in repre-senting biological processes that drive exchanges of wa-ter energy and C between the land surface and the atmo-sphere Above ground Earth system models rely on empir-ical differences in plant physiology and life history strate-gies to represent the biogeochemical and biogeophysical ef-fects of vegetation dynamics in global simulations (Bonan2008 Roy et al 1993) Although imperfect these and otherdata (eg Kattge et al 2011) are improving and refiningthe autotrophic or ldquogreenrdquo representations of the terrestrialC cycle (Bonan et al 2012) Comparatively less attentionhas been given to revising biologically driven representa-tions of the soil heterotrophic or ldquobrownrdquo C cycle Accord-ingly Earth system models display wide variation in theirsoil C projections (Todd-Brown et al 2013) Given the sizeof global soil C pools (Hugelius et al 2013 FAO et al 2012Jobbaacutegy and Jackson 2000) and the potential magnitude ofsoil Cndashclimate feedbacks (Jones et al 2003 2005 Jenkinsonet al 1991) greater attention should be directed towards crit-ically evaluating and improving the theoretical and numerical

Published by Copernicus Publications on behalf of the European Geosciences Union

3900 W R Wieder et al Soil carbon dynamics with MIMICS

representation of soil biogeochemistry models that are usedon multiple scales

A growing body of literature calls for significant revi-sions to the theoretical basis for modeling soil C dynamics(Cotrufo et al 2013 Dungait et al 2012 Conant et al2011 Schmidt et al 2011) Traditional soil C stabilizationconcepts do not explicitly simulate microbial activity or soilmicrobial communities but instead strongly emphasize therelationship between litter chemical recalcitrance and soil Cstorage By contrast new theoretical and experimental re-search shows that soil microbes strongly mediate the for-mation of soil organic matter (SOM) through the productionof microbial products that appear to form mineral-stabilizedSOM (Wallenstein et al 2013 Miltner et al 2012 Wick-ings et al 2012 Kleber et al 2011 Grandy and Neff 2008Six et al 2006) These insights suggest that basic physi-ological traits such as microbial growth efficiency (MGE)and growth kinetics have direct influences on litter decom-position rates and net microbial biomass production whilethe subsequent turnover of microbial biomass strongly influ-ences input rates to SOM Furthermore the ultimate fate ofSOM also depends upon the mineral stabilization of thesemicrobially derived inputs However despite wide recogni-tion that microbial physiology and soil mineral interactionsfacilitate the formation of stable SOM this theoretical in-sight has not been adequately represented in process-basedmodels

These emerging concepts highlight the need to simulateexplicitly the microbial processes responsible for decompo-sition and stabilization of organic matter (Todd-Brown et al2012 Treseder et al 2012 Allison et al 2010 Lawrenceet al 2009 Bardgett et al 2008 Schimel and Weintraub2003) even if the magnitude of microbial control over soil Cdynamics in mineral soils remains poorly defined (Schimeland Schaeffer 2012) Microbially explicit approaches in re-cent models range in complexity from simple fungal to bac-terial ratios (Waring et al 2013) microbial guilds specializ-ing in different litter C substrates (Miki et al 2010 Moor-head and Sinsabaugh 2006) and complex community dy-namics (Allison 2012 Wallenstein and Hall 2012 Loreau2001) These models incorporate the complexity of micro-bial physiology and competitive interactions in litter decom-position dynamics and provide valuable insight into our un-derstanding of ldquoupstreamrdquo soil C inputs but focus less at-tention on the stabilization of microbial residues in mineralsoils Other recent work demonstrates that simple non-linearmicrobial models are feasible on the global scale and resultsin divergent responses compared with traditional soil biogeo-chemistry models (Wieder et al 2013) Again however thismicrobial modeling framework does not adequately capturehow microbial physiology and activity may facilitate the sta-bilization of SOM

Microbial attributes that regulate microbial residue inputsto SOM and their interactions with the soil matrix are ef-fectively absent in traditional soil biogeochemical models

and poorly accounted for in current microbially based mod-els Thus our chief motivation is to develop a process-basedmodeling framework to explore the potential role of micro-bial physiology and the stabilization of microbial biomassat the soilndashmineral interface as key drivers in the formationof SOM (Miltner et al 2012 Liang et al 2011 Bol et al2009 Grandy et al 2009) In this model litter and SOMturnover are governed by microbial biomass pools whichcorrespond to different microbial functional types Inherentphysiological differences between microbial functional typesprovide a basis to begin simulating below-ground biologicaland metabolic diversity and explore how the relative abun-dance of different microbial functional groups regulates bio-geochemical processes (Miki et al 2010) Microbial growthrates growth efficiency and turnover are subject to intrin-sic physiological constraints (Beardmore et al 2011 Mole-naar et al 2009 Dethlefsen and Schmidt 2007) but arealso sensitive to external forces such as resource chemistry(Manzoni et al 2012 Keiblinger et al 2010 Steinweg etal 2008 Rousk and Baringaringth 2007 Thiet et al 2006) suchthat both community composition and the soil environmentshould determine the optimization of physiological traits andtheir downstream influence on SOM dynamics

We use observational and theoretical insights to guideour incorporation of microbial physiological processes intopredictions of SOM stabilization Physiological differencesacross species have been linked to life-history strategies opti-mized for different resource environments (Resat et al 2012Beardmore et al 2011 Russell and Cook 1995) For in-stance in resource-rich environments fast-growing r strate-gists (copiotrophs) are typically characterized by a lowerMGE but higher growth rates relative to slower-growing Kstrategists (oligiotrophs Fierer et al 2012 Ramirez et al2012 Fierer et al 2007 Klappenbach et al 2000 Pianka1970) This physiology gives copiotrophs a competitive ad-vantage under resource-rich conditions such that they tendto dominate in these environments On an individual specieslevel MGE growth rates and turnover are expected to in-crease as resource quality increases However selection for acopiotroph-dominated community may drive up community-level growth rates and turnover at the expense of a lowerMGE The effect of this tradeoff on total biomass produc-tion and microbially derived inputs to physically protectedSOM is uncertain and remains a challenging often missingaspect of microbially based soil C models

In order to more rigorously evaluate dynamics betweenmicrobial physiology soil environmental conditions andSOM formation we introduce a new soil biogeochemistrymodel MIMICS (MIcrobial-MIneral Carbon Stabilization)MIMICS incorporates the relationships between microbialphysiology substrate chemical quality and the physical sta-bilization of SOM (Wang et al 2013 Goldfarb et al 2011Fontaine and Barot 2005) with the long-term aim of rep-resenting these interactions in global simulations using theCommunity Land Model (Lawrence et al 2011) Thus here

Biogeosciences 11 3899ndash3917 2014 wwwbiogeosciencesnet1138992014

W R Wieder et al Soil carbon dynamics with MIMICS 3901

SOMpLITm MICr

Vmax Km

LITs MICK SOMc

τ

fi met

fi struc

fmet

fc

1-MGE

MGE

1-fc

I1-fmet

(1)

(2)

(6)

(7)

(5)

(8)

(4)

(10)

(3)

(9)

Figure 1 Litter microbial biomass and soil organic matter (SOM)pools and carbon flows represented in MIMICS Litter inputs (Iblack lines) are partitioned into two litter pools based on litterquality (fmet) Litter pools in the model correspond to metabolicand structural litter (LITm and LITs respectively) Rates of de-composition (red lines) are controlled by temperature-sensitiveMichaelisndashMenten kinetics derived from observational data (Ger-man et al 2012) that are modified by microbial functional typeand C-substrate pool quality Microbial functional types corre-spond to copiotrophic and oligiotrophic growth strategies (MICrand MICK respectively Fierer et al 2007) Microbial growth ef-ficiency (MGE) determines the partitioning of C fluxes enteringmicrobial biomass pools vs heterotrophic respiration Turnoverof the microbial biomass pools (τ ) (blue lines) depends on mi-crobial functional type which are partitioned into physically andchemically protected SOM pools (SOMp and SOMc respectivelybased onfc) Decomposition of C from SOM pools also followsMichaelisndashMenten kinetics with the clay fraction increasing thehalf-saturation constant for both pools but especially SOMp Afraction of litter inputs (fi) dashed black lines) bypasses litter andmicrobial biomass pools and is transferred directly to SOM pools

we explore how microbial physiological traits can be appliedto a process-based soil biogeochemistry model furthermorewe evaluate how incorporating these physiological traits intomodels may improve our ability to predict soil C responses toglobal change scenarios compared with the more traditionalDAYCENT model

2 Methods

21 Model configuration

To develop MIMCS we modified the CLM microbial modela soil biogeochemistry model that explicitly represents mi-crobial activity and microbial physiology (Wieder et al2013) to simulate two plant litter microbial biomass andSOM pools (LIT MIC and SOM in Fig 1 respectively)This structure blends aspects of traditional and microbiallyexplicit models In particular MIMICS simulates physicallyand biochemically protected SOM pools that are also repre-sented in the MEND model (Wang et al 2013) and multi-ple substrate pools that are represented in the EEZY model(Moorhead et al 2012) while simplifying the overall model

structure by eliminating explicit enzyme pools (Wieder et al2013) A vertical dimension could be added to this basic six-pool model structure (Koven et al 2013) but here we fo-cus on the dynamics simulated within a single soil layer (0ndash30 cm)

The representation of plant litter pools in MIMICS isbased on well-established paradigms of litter chemistry anddecomposition dynamics (Melillo et al 1982) We partitionfresh litter inputs into high- and low-quality pools (LITmand LITs respectively) that correspond to the metabolic andstructural pools used in CENTURY and DAYCENT (Partonet al 1994 Parton et al 1987) As in DAYCENT parti-tioning into these pools is based on a linear function of litternitrogen to lignin ratios (fmet Table 1) A fraction of inputs(fi) bypasses litter and microbial biomass pools and is trans-ferred directly to corresponding SOM pools For metaboliclitter inputs this fraction is analogous to dissolved organicmatter fluxes that leach out of leaf litter or root exudates thatquickly become sorbed onto mineral surfaces For structurallitter inputs this is analogous to a relatively small proportionof the structurally complex compounds that could be incor-porated into SOM before microbial oxidation Thusfi forstructural litter inputs is inversely related to litter quality (Ta-ble 1)

Decomposition of litter and SOM pools is based ontemperature-sensitive MichaelisndashMenten kinetics (Allison etal 2010 Schimel and Weintraub 2003) through the basicequation

dCs

dt= MIC times

Vmaxtimes Cs

Km + Cs (1)

where Cs is an individual C substrate pool (LIT or SOM) andMIC corresponds to the size of the microbial biomass poolboth in mg C cmminus3 Thus rates of C mineralization dependon donor C (either LIT or SOM) and receiver (MIC) poolsizes as well as kinetic parametersVmax andKm The maxi-mum reaction velocity (Vmax mgCs (mg MIC)minus1 hminus1) andhalf-saturation constant (Km mg C cmminus3) are respectivelycalculated as

Vmax = eVslopemiddotT +Vint middot av middot Vmod (2)

Km = eKslopemiddotT +Kint middot ak middot Kmod (3)

whereT represents soil temperature which we assumed tobe 15C unless otherwise noted The temperature sensitiv-ity of kinetics parameters (described in Table 1) are derivedfrom observational data (German et al 2012) with modifi-cations based on assumptions regarding microbial functionaltypes litter chemical quality and soil texture effects (VmodandKmod Table 1) The equations governing soil C turnoverin MIMICS are provided in the Supplement Soil moisture isnot presently considered in MIMICS but work by Davidsonand others (2012) presents a tractable framework that we canapply to this model structure in the future

wwwbiogeosciencesnet1138992014 Biogeosciences 11 3899ndash3917 2014


Table 1Model parameter descriptions values and units used in the MIMICS model

Parameter Description Value Units

fmet Partitioning of litter inputs to LITm 085ndash0013 (lignintimes Nminus1) ndashfi Fraction of litter inputs directly transferred

to SOM002 03times e(minus4timesfmet) a ndash

Vslope Regression coefficient 0063b ln(mg Cs (mg MIC)minus1 hminus1) Cminus1

Vint Regression intercept 547b ln(mg Cs (mg MIC)minus1 hminus1)

aV Tuning coefficient 8times 10minus6 b ndashVmod-r ModifiesVmax for each substrate pool

entering MICr

10 2 6 2c ndash

Vmod-K ModifiesVmax for each substrate poolentering MICK

2 2 2 2d ndash

Kslope Regression coefficient 0007b ln(mg C cmminus3) Cminus1

Kint Regression intercept 319b ln(mg C cmminus3)

aK Tuning coefficient 10b ndashKmod-r ModifiesKm for each substrate pool

entering MICr

0125 05Pscalar Cscalarc ndash

Kmod-K ModifiesKm for each substrate poolentering MICK

05 025Pscalar Cscalard ndash

Pscalar Physical protection scalar used inKmod 1(25times e(minus3timesfmet)) ndashCscalar Chemical protection scalar using inKmod 1(14+ 02(fclay)) ndashMGE Microbial growth efficiency for

substrate pools06 03 06 03e mg mgminus1

τ Microbial biomass turnover rate 6times 10minus4times e(09timesfmet) 3times 10minus4 f hminus1

fc Fraction ofτ partitioned to SOMc 02times e(minus2timesfmet) 04times f(minus3timesfmet) f ndash

a For metabolic litter inputs entering SOMp and structural litter inputs entering SOMc respectivelyb From observations in German et al (2012) as used in Wieder et al (2013)c For LITm LITs SOMp and SOMc fluxes entering MICr respectivelyd For LITm LITs SOMp and SOMc fluxes entering MICK respectivelye For C leaving LITm LITs SOMp and SOMc respectivelyf For MICr and MICK respectively

Two microbial functional types are represented in MIM-ICS that roughly correspond to copiotrophic and oligotrophicgrowth strategies (MICr and MICK respectively Lipson etal 2009 Dethlefsen and Schmidt 2007 Fierer et al 2007)We have intentionally classified our microbial functionaltypes based on these broad ecological life-history traits be-cause they explicitly parameterize the growth physiologieswe are exploring in MIMICS and avoid the exclusivity offungal bacterial ratios (Strickland and Rousk 2010) We as-sume that the copiotrophic community (MICr) has a highergrowth rate when consuming metabolic litter and physicallyprotected soil C because of the relatively low C N ratio andthe chemical complexity of these pools (LITm and SOMprespectively) whereas the kinetics of the oligotrophic com-munity (MICK) are comparatively more favorable when con-suming structural litter and chemically protected soil C (LITsand SOMc respectively Fig 1 Table 1) relative to MICr Werecognize the uncertainties in these classifications but arguethat they provide a tractable starting point to begin represent-ing microbial metabolic diversity in regional- to global-scalemodels We consider the SOMp pool to be largely derived oflow C N labile materials that are either microbial products or

highly processed litter (Grandy and Neff 2008) whereas themore low-quality SOMc pool consists of litter that is higherin structural C compounds such as lignin

We implement the physical protection of SOM throughenvironmental scalars that increase theKm of SOM poolswith increasing soil clay content This environmental scalaris more dramatic for the physically protected SOM pool(SOMp) than it is for the chemically protected pool (SOMc)and strongly reduces microbial access to substrates in min-eral soils (Schimel and Schaeffer 2012) Other aspects ofsoil mineralogy certainly regulate SOM stabilization (Heck-man et al 2013 Kramer et al 2012 Kaiser et al 2011 vonLuumltzow et al 2008 Jastrow et al 2007) but in order to rep-resent microbially driven soil biogeochemical processes onglobal scales we constrain the complexity of our parameter-ization to clay content shown to be a primary factor in phys-ical stabilization mechanisms (Kleber et al 2011 Mikutta etal 2006 Sollins et al 1996)

Microbial growth efficiency determines the fraction of as-similated C that builds microbial biomass (del Giorgio andCole 1998) We have incorporated new experimental in-sights into the modelrsquos MGE dynamics by first accounting



for substrate quality which is positively related to MGE(Frey et al 2013 Keiblinger et al 2010 Steinweg et al2008 Table 1) Second we explore the theoretical evi-dence that there is differential MGE for each microbial func-tional type whereby for a given substrate fast-growing co-piotrophic communities should have lower growth efficiencythan slower growing oligotrophic communities (Sinsabaughet al 2013 Lipson et al 2009 Fierer et al 2007 Pfeiffer etal 2001 Russell and Cook 1995) We recognize the impor-tance of considering MGE sensitivity to changes in temper-ature and nutrient availability in refining our understandingof microbial physiological responses to perturbations (Leeand Schmidt 2013 Tucker et al 2013 Wieder et al 2013Manzoni et al 2012 Bradford et al 2008 Frey et al 2008Steinweg et al 2008) although these theoretical considera-tions are not addressed in this manuscript

A fixed fraction of the microbial biomass pools turns over(τ) at every time step with partitioning into physically andchemically protected SOM pools dependent on the chem-ical quality of litter inputs (fc Table 1) We assume thatthe turnover rates of copiotrophic microbial communitieswill be greater than their oligotrophic counterparts (Fierer etal 2007) and that the turnover of MICr will increase withhigher quality litter inputs We also assume that the majorityof τ will be partitioned into the physically protected SOMpool especially from MICr and that partitioning into chemi-cally protected pools will be inversely related to litter quality

22 Model evaluation litter decomposition study

Validating assumptions and parameterization in MIMICSpresents unique challenges Given the difficulty in obtain-ing empirical data on MGE and microbial turnover and themethodological limitations in resolving the flow of micro-bial C into SOM (Six et al 2006) our model evaluationis restricted to the litter fluxes and decomposition dynamicsthat are represented in the left portion of our model (Fig 1)Leaf litter decomposition studies provide process-level eval-uation of soil C dynamics across biomes and with multiplelitter types (Bonan et al 2013 Yang et al 2009) We takea similar approach in evaluating MIMICS using data fromthe LIDET study LIDET was a decade-long multisite studydesigned to investigate climatendashlitter quality dynamics in de-caying litter (Currie et al 2010 Harmon et al 2009 Adairet al 2008 Parton et al 2007 Gholz et al 2000) Herewe used a subset of LIDET data (from Parton et al 2007)that has been used previously to evaluate litter decomposi-tion dynamics in Earth system models (Bonan et al 2013)Although similar exhaustive evaluations of leaf litter decom-position dynamics are outside the scope of this paper weused data from six leaf litters of varying chemical qualitydecomposed at two LIDET sites (Harvard Forest and Bo-nanza Creek) to begin evaluating process-level simulationsprovided by our non-linear microbial model

Litterbag studies are relatively simple to replicate usingtraditional soil biogeochemistry models based on first-orderkinetics (Bonan et al 2013) In these donor control modelspool size has no bearing on rates of litter decay so decompo-sition dynamics can be simulated by adding a fixed amountof litter to appropriate pools subject to environmental scalars(eg soil temperature and soil moisture) that modulate baserates of decomposition over time However in MIMICS de-composition does not follow simple first-order kinetics be-cause the size of both the donor and receiver pools modu-lates decay rates via environmentally sensitive microbial ki-netics parameters (Eq 1) Since the size of the microbialbiomass pools exerts a strong influence over rates of litterdecay MIMICS had to be equilibrated at steady state beforeadding a cohort of litter to track over the experimental periodSecond augmenting litter pools initially increased rates ofdecomposition and enlarged microbial biomass pools whichfurther accelerated decomposition rates (see Eq 1) To over-come these complications we took several steps to facilitatethe evaluation of leaf litter decomposition studies using non-linear microbial models

We applied a NewtonndashRaphson approach to analyti-cally calculate steady-state C pools using the stode func-tion in the rootSolve package in R (R Development CoreTeam 2011 Soetaert 2009) We calculated steady-statepools and site productivity estimates at the Harvard For-est and Bonanza Creek Long-Term Ecological Researchsites (Knapp and Smith 2001) Mean annual soil tempera-tures were estimated as 107 and 39C at Harvard Forest(May 2001ndashOctober 2010) and Bonanza Creek (June 1984ndashDecember 2004) respectively (data from the Climate andHydrology Database Projects ndash a partnership between theLong-Term Ecological Research program and the US ForestService Pacific Northwest Research Station Corvallis Ore-gon httpclimhylternetedu accessed August 2013) Pro-ductivity estimates provided litter inputs that were distributedat hourly intervals evenly throughout the year We createda daily climatology from a decade or more of soil temper-ature records at each site and calculated steady-state poolswith hourly litter inputs and mean annual soil temperatureWe assumed soils at both sites had 10 clay content andmetabolic litter inputs were 30 of total litter inputs Fromtheir steady states models for each site were run for an ad-ditional thirty years with hourly litter inputs and daily soiltemperature climatologies allowing all C pools to equili-brate to seasonally fluctuating temperatures Data for controlsimulations continued beyond this equilibration period foran additional decade In experimental simulations we added100 g C mminus2 to litter pools on October first with partition-ing between LITm and LITs dependent on litter quality (fmetTable 1) To avoid changing results by augmenting microbialbiomass pools through this litter addition (Eq 1) we forcedthe experimental simulation to maintain the same microbialbiomass pool as the control simulation We calculated thepercent mass remaining as the difference in litter pools from



experimental and control simulations Model parameteriza-tions were modified to provide the best fit for the HarvardForest data and independently evaluated using the resultsfrom Bonanza Creek

221 Model evaluation Soil warming experiment

Litter decomposition studies provide validation for process-based representation of models under steady-state condi-tions but the Community Earth System Model into which weaim to integrate MIMICS is used to project C cyclendashclimatefeedbacks in a changing world It is in these non-steady-state simulations that differences in model structures as-sumptions and parameterizations become important (Wiederet al 2013) Thus we compare the soil C response of MIM-ICS and DAYCENT to experimental warming in simulationsdesigned to replicate the experimental warming manipula-tions at Harvard Forest (Melillo et al 2011 2002)

We calculated steady-state soil C pools (0ndash30 cm) forDAYCENT and MIMICS models using site-level productiv-ity estimates (Knapp and Smith 2001) mean annual soiltemperature (here from Harmon 2013) soil texture data(S Frey personal communication 2014) and litter qual-ity estimates for deciduous forests from the TRY database(Brovkin et al 2012 as in Wieder et al 2014) For DAY-CENT simulations we used a soil pH= 38 (Melillo et al2002) and solved the steady-state C pools as described inWieder et al (2014) with modifications to simulate 0ndash30 cmsoils (Metherell et al 1993) with a climate decompositionindex (CDI) of 027 This CDI value was calculated using thetemperature scalar formula described by Metherell and oth-ers (1993) and allows us to revise CDI values in the warmingexperiment Despite omitting effects of soil moisture in ourCDI calculations our initial CDI value is close to CDI esti-mates for Harvard Forest used previously (025 Parton et al2007 and Bonan et al 2013) From our steady-state poolswe simulated control and 5C warming experiments for 500years focusing on results from the first 20 years Experi-mental evidence also shows that MGE decreases with warm-ing approximately 1 Cminus1 although temperature-inducedchanges in MGE likely depend on the chemical quality ofsubstrates and potential acclimation of the microbial com-munity to extended warming (see Frey et al 2013) Here weconsider the potential effects of changing MGE in warmingexperiments by repeating our simulation with a 5 reductionin MGE for all fluxes simulated by MIMICS and DAYCENTrespectively

We compare changes in SOM pools and total soil C poolssimulated by MIMICS and DAYCENT with observed resultsfrom two sets of warming experiments from Harvard Forest(Melillo et al 2011 2002) We present results from the SOMpools represented by each model and the total change insoil C storage with warming Analysis of SOM pools ignoreschanges in litter and microbial biomass pools (Fig 1 see alsoBonan et al 2013 for the DAYCENT pool structure) Ob-

servational data were extracted from published figures usingthe DataThief software package (httpdatathieforg) Weassumed that 80 of the soil respiration values reported byMelillo and others (2002) were attributable to heterotrophicrespiration The cumulative soil C losses we calculated withthis approach are identical for the Melillo et al (2011) re-sults (1300 g C mminus2 over seven years) and slightly greaterthan those reported in the 2002 study (we calculate cu-mulative C losses over the ten-year study of 1071 g C mminus2

vs 944 g C mminus2 reported by Melillo et al 2002)

23 Steady-state soil C pools and sensitivity analysis

Initial parameter values were evaluated with data fromLIDET sites to explore how steady-state litter microbialbiomass and soil C pools vary with soil texture and litterchemical quality We calculated steady-state conditions forall MIMICS pools using the stode function Soil texture ef-fects on turnover of SOM pools (Pscalar and Cscalar Table1) provide a strong influence on steady-state SOM poolsWe chose values for these parameters assuming that low-clay soils would provide low physical protection of soil C(ie lowKm) which increases exponentially with increasingsoil clay content (Table 1) We furthermore constrained ini-tial parameterizations to keep the ratio of total soil microbialbiomass to SOM roughly within observational constraints(Serna-Chavez et al 2013 Xia et al 2012) This providesuseful bounds because much larger SOM pools can be simu-lated with this model by adjusting soil texture effects on thehalf-saturation constant for C fluxes from SOM to microbialbiomass pools (using thePscalarandCscalar Table 1) Fromthese initial conditions (Table 1) we modified individual pa-rameters by 10 to illustrate important model assumptionscharacteristics and uncertainties

Recent analyses have criticized the unrealistic dynamicsproduced in the numerical application of microbially explicitmodels notably their oscillatory behavior in response to per-turbations (Li et al 2014 Wang et al 2014) To begin ex-ploring these dynamics in MIMICS we replicated the partof the numerical simulations employed by Wang and oth-ers (2014) We calculated steady-state C pools (0ndash30 cm) fora hypothetical site with a mean temperature of 15C thatreceives litter inputs of 3723 g C mminus2 yminus1 Using parame-ter values described in Table 1 we prescribe the fraction ofmetabolic litter inputs and the soil clay fraction (fmet = 03andfclay = 02) From their steady-state values we reducedmicrobial biomass and SOM pools by 10 at time zero andran MIMICS for a fifty-year simulation to track changes inlitter microbial biomass and soil C pools We stress that ouraim with this simulation is not to provide a rigorous mathe-matical analysis of MIMICS but to illustrate the magnitudeof the oscillatory response produced by the parameters ap-plied in the model



0 2 4 6 8 10

020

4060

80100

time (y)

Mas

s re

mai

ning

()

(a)

0 2 4 6 8 10

time (y)

(b)

Figure 2 Observed and modeled leaf litter decomposition dynam-ics at the(a) Harvard Forest and(b) Bonanza Creek Long-TermEcological Research sites Closed circles show the mean percentmass remaining of six leaf litter types decomposed over ten yearsas part of the LIDET study (Parton et al 2007 Bonan et al 2013meanplusmn1 SD) Similarly solid lines indicate the mean percent massremaining (plusmn1 SD shaded region) of the six leaf litter types pre-dicted by the microbial model forced with a climatology of ob-served mean daily soil temperature at each study site Model param-eters were calibrated to fit observations from Harvard Forest (Table1) The same parameters were used to evaluate model output at theBonanza Creek site In observations and simulations the range ofvariation shows the effects of litter quality on rates of litter massloss

3 Results

31 Model evaluation litter decomposition

In Fig 2 we show the mean percentage mass remaining(plusmn 1 SD) of six different leaf litter types decomposed atHarvard Forest and Bonanza Creek for LIDET observa-tions (points and error bars) and MIMICS simulations (linesand shaded area) After calibrating model parameters to fitLIDET observations from Harvard Forest (Table 1) MIM-ICS can replicate litter decomposition dynamics well at bothsites Beyond capturing the mean state that reflects broad cli-matic influences on leaf litter decomposition the observedlitter quality effects on litter decomposition rates (error bars)are simulated well in the model (shaded area) These resultsindicate that the parameterization of litter decay dynamics inMIMICS can replicate real-world observations and demon-strate that microbially explicit models of moderate complex-ity can be parameterized to preform just as well as stan-dard soil biogeochemistry models based on first-order kinet-ics (eg Bonan et al 2013)

The final litter mass remaining was lower than that ob-served by LIDET at the warmer Harvard Forest site (Fig 2a)suggesting that a third litter C pool corresponding to leaf lit-ter lignin may be necessary to capture the long tail of leaflitter decomposition dynamics (Adair et al 2008) Some of

SO

M c

hang

e (g

C m

minus2)

-3000-2500-1000

-500

0

(a)

year

Tota

l C c

hang

e (g

C m

minus2)

0 5 10 15 20 500

-4500

-2500

-1500

-500

(b)

Figure 3 Changes from steady-state(a) SOM pools and(b) to-tal soil C pools at Harvard Forest simulated by DAYCENT (black)and MIMICS (blue) Lines show model results from a 5C warm-ing (solid lines) and warming with 5 decreases in MGE (dashedlines sensu Frey et al 2013) Points are for observations reportedby Melillo et al 2002 (open circles) and Melillo et al 2011 (stars)Note the breaks on both thex andy axes showing long-term pro-jections from each simulation

our model assumptions we made to facilitate good agree-ment with LIDET observations exerted negligible effects onsteady-state SOM pools while other assumptions had a sig-nificant influence For example our assumption that soils atboth sites contained ten percent clay had no bearing on lit-ter decomposition dynamics because when parameterizedthe clay fraction only modifies steady-state SOM pools Bycontrast assumptions about the metabolic fraction of litterinputs exert a significant influence on litter decompositiondynamics by modifying the steady-state size of all simulatedpools Specifically decreasing thefmet generates compara-ble total microbial biomass pools but with a larger propor-tion of biomass in the oligotrophic (MICK) community Theoligotrophic community decomposes leaf litter more slowlyresulting in lower rates of mass loss (Wieder unpublisheddata) More broadly any parameter that modifies the steady-state size of microbial biomass pools will strongly influencethe temporal dynamics of the model although such modi-fications may have little or no effect on steady-state soil Cstorage

32 Model evaluation soil warming experiment

Initial total soil C pools simulated for Harvard Forest us-ing MIMICS are significantly smaller than those projected



by DAYCENT (42 and 117 kg C mminus2 respectively 0ndash30 cm) Observations from these same sites report soil Cpools of 84 kg C mminus2 (0ndash60 cm Melillo et al 2002) and67plusmn 03 kg C mminus2 (0ndash20 cm Frey et al 2013) Neithermodel project changes in SOM pools with warming thatare large enough to match observed results (Fig 3a) Sim-ulated total soil C losses from each model approach obser-vations over the first decade but differ greatly on longertimescales (Fig 3b) Initial soil C losses (0ndash5 years) simu-lated by MIMICS may be too rapid but decreasing effectsof soil warming on longer timescales show better agree-ment with observations from Harvard Forest (Melillo et al2002) compared with DAYCENT By contrast DAYCENTdemonstrates good agreement with data from the observa-tional period but continues to lose soil C over the followingdecades to centuries and has still not reached equilibrium af-ter 500 years (data not shown)

Model responses to reductions in MGE elicited opposingresponses in the soil models we examined In MIMICS re-ducing MGE reduces the size of the microbial biomass poolconcurrently decreasing rates of litter and SOM turnover(Eq 1) These effects are negligible for changes in SOMpools especially over longer timescales (Fig 3a) becauseturnover of microbial residues that contribute to SOM forma-tion are also reduced Reducing MGE and microbial biomasspools however has larger implications for litter C dynam-ics in MIMICS (evident in Fig 3b) Reductions in micro-bial biomass serve as a counterbalance to the accelerated mi-crobial kinetics associated with warming (see also Wiederet al 2013 and Allison et al 2010) and reduce total soilC losses Such feedbacks are absent from models like DAY-CENT thus warming-induced decreases of MGE acceler-ates C losses from traditional soil biogeochemistry models

33 Steady-state soil C pools influence of litter inputs

Soil C pools in MIMICS vary as a function of litter qualityand soil texture (Fig 4) with a high metabolic fraction oflitter providing the widest range of steady-state values (from48 to 23 mg C cmminus3 in low-clay and high-clay soils respec-tively) In soils with low clay content (lt 03 clay fraction) re-ceiving low-quality litter inputs (lt 02fmet) the chemicallyprotected SOM pool was larger than the physically protectedSOM pool however in all other cases the majority of SOMwas found in the physically protected SOM pool Reducinglitter inputs by 10 directly reduced the size of microbialbiomass pools by 10 with no changes to the size of steady-state litter or SOM pools

At steady state the total litter pool size (the sum of LITmand LITs) was inversely related to the fraction of metabolicinputs ranging from 17 to 32 mg C cmminus3 (with high andlow fmet respectively) The proportion of total litter foundin the metabolic litter pool increased exponentially with in-creasingfmet Total microbial biomass pool size was rel-atively invariant with litter quality and soil clay content

Figure 4 Steady-state soil organic matter pools (mg C cmminus3 0ndash30 cm) that are simulated by MIMICS across hypothetical sites witha range of clay content and litter quality at 15C with litter inputsof 160 g C mminus2 yminus1 Low-clay soils store more C when receivinglow-quality litter inputs whereas high-clay soils store more C withhigh-quality litter inputs

(010ndash011 mg C cmminus3) totaling about 1ndash2 of SOM poolsThe relative abundance of MICr increased from approxi-mately 13 of the total microbial biomass pool with low-quality litter inputs to nearly 54 with high-quality litterinputs (Fig 5a)

34 Steady-state soil C pools influence of microbialphysiology

We assumed that microbial functional types control theMichaelisndashMenten kinetics of litter mineralization By con-trast the physical soil environment exerts a strong controlover the half-saturation constant of SOM mineralizationwith modest differences in theVmax driven by microbialfunctional types (Table 1) Thus MIMICS illustrates howfunctional differences between microbial functional typescan regulate the fate of C substrates and affect steady-stateSOM pools either directly or indirectly We illustrate thesedynamics with a series of sensitivity analyses where we per-turb individual parameters by 10 and document their effecton the steady-state soil C pool simulated by MIMICS

Litter quality determines the relative abundance of micro-bial functional types in MIMICS These results however de-pend on the competitive dynamics between microbial func-tional types that are directly related to assumptions madeabout the catabolic potential MGE and the turnover ratesof the MICr and MICK communities (Fig 5a) For examplereducing the catabolic potential of litter mineralization for



Figure 5Proximal controls over the production and turnover of mi-crobial biomass (MGE andτ ) in MIMICS have larger influence overthe relative abundance of microbial functional types and steady-state SOM dynamics(a)The relative abundance of the copiotrophiccommunity (MICrtotal microbial biomasstimes 100) as a function oflitter quality (fmet) in base simulations (black line parameters asin Table 1) and in response to 10 reductions in catabolic poten-tial of the MICr community consuming litter C substrates (reducingVmax red line increasingKm green line) simulating substrate andcommunity effects on MGE (by reducing MGE of MICr 10 blueline) and increasing turnover (τ) of the MICr community by 10 (cyan line)(b) Differences between the base simulation and modi-fications described in panel A on the percent change in steady-stateSOM pools vs the change in the MICr relative abundance

the copiotrophic community MICr either by reducingVmaxor increasingKm reduces the relative abundance of MICr by2ndash5 across soil textures This decline in MICr abundanceindirectly feeds back to steady-state SOM pools (Fig 5b)because we also assumed that the turnover of MICr is greaterthan that of MICK Thus reducing the relative abundance ofMICr indirectly reduced inputs of microbial residues to SOMpools reducing total soil C storage Reductions in soil C stor-age associated with the kinetics of litter C mineralizationhowever are distal to the production of microbial residuesthat build SOM in MIMICS Instead parameters like MGEand microbial turnover are proximal to the production of mi-crobial biomass and exert a greater influence over soil C dy-namics (Fig 5)

Carbon substrates and microbial physiology likely deter-mine the efficiency by which different microbial functionaltypes convert assimilated C into microbial biomass (Sins-abaugh et al 2013 Lipson et al 2009 Pfeiffer et al 2001Russell and Cook 1995) We explore this theory by de-creasing the MGE of MICr communities by 10 (see Leeand Schmidt 2013) This modification concurrently reducesthe relative abundance of the MICr community by 7ndash14with greater reductions in high-substrate quality environ-ments (Fig 5a) These results indicate that assumptions madeabout tradeoffs between microbial growth rates and MGEmay be important in structuring competitive interactions be-tween microbial functional types The relative abundance of

Figure 6 Temporal response of(a) litter (b) microbial biomassand (c) soil C pools to a 10 reduction of steady-state MIC andSOM pools at time zero of the experiment (solid lines) Steady-statepools were calculated for a hypothetical site at 15C as describedin the methods and are indicated by dashed lines Litter and micro-bial biomass pools oscillate following this perturbation while SOMpools increase monotonically until they reach their steady state Theamplitude of the oscillation is comparable with results from Wang etal (2014) however the period of oscillation and the time requiredto return to steady-state values is shorter in this parameterization ofMIMICS

microbial functional types relates to the community physio-logical function which in turn influences soil C dynamicsIn this example reducing the relative abundance of the MICrcommunity by reducing its MGE can either increase or de-crease soil C storage (Fig 5b) In low-quality resource en-vironments (fmet lt 025) reducing MICr abundance reducesrates of SOM turnover and can lead to modest increases inSOM pools by as much as 2 In high-quality resource en-vironments (fmet gt 025) reducing the relative abundance ofMICr communities reduces inputs of microbial residues toSOM pools with more than 3 declines in steady-state soilC storage

Modifications that reduce the kinetic capacity and growthefficiency of the MICr community reduce the relative abun-dance of this community Shifts in community compositionlargely influence SOM dynamics via interactions with theproduction of microbial residues that govern the fate of C inMIMICS Not surprisingly increasing the microbial turnoverof the MICr community by 10 also reduces the relativeabundance of this functional type by 6ndash18 (Fig 5a) Thisdirectly increases inputs of microbial residues to SOM poolsand generally increases soil C storage with greater SOC



accumulation in clay-rich environments that stabilize micro-bial residues (Fig 5b) Thus in MIMICS we assume thatmicrobial functional types can govern the fate of C sub-strates assimilated Microbial growth efficiency and turnoverare proximal to the production of microbial biomass and mi-crobially derived inputs to SOM pools Accordingly theseparameters have a larger influence on the relative abundanceof microbial functional types and soil C stabilization

As with other microbially explicit models MIMICS pro-duces an oscillatory response to perturbations of initial poolsizes (Figs 3 and 6 Li et al 2014 Wang et al 2014) InMIMICS only litter and microbial biomass pools oscillatewhile SOM pools show a monotonic response Total LIT andMIC pools increase from their steady-state conditions by asmuch as 11 and 31 respectively and the magnitude oftheir oscillations decreases exponentially with time (Fig 6)These results align with findings from Wang et al (2014)that analyzed a three-pool model based on the same modelstructure applied in MIMICS (modified from Wieder et al2013) The frequency of the oscillation and time required toreturn to steady state is shorter in MIMICS than the three-pool model analyzed by Wang et al (2014) which could bedue to differences in model structures or could be an artifactof model parameterizations These results indicate that theadditional complexity and number of non-linear terms thatgovern soil C turnover in MIMICS do not affect model sta-bility and may even reduce the frequency of oscillations inresponse to disturbance

4 Discussion

We outline a framework for integrating litter quality func-tional differences in microbial physiology and the physicalprotection of microbial byproducts in forming stable SOM ina process-based numerical model (Fig 1 Table 1) Our ap-proach simulates observed climate and litter quality effectson average rates of leaf litter decomposition (Fig 2) pro-viding a robust validation for the MIMICS parameterizationsgoverning the kinetics of litter decay Moreover the struc-ture and parameterization of MIMICS better captures the ac-climatization of soil C losses seen at Harvard Forest and else-where (Fig 3 Melillo et al 2002 Luo et al 2001 Oechel etal 2000) While initial results suggest MIMICS is a promis-ing predictive tool they also provide a framework for eval-uating the microbial processes underlying SOM dynamicsInitial results from MIMICS suggest that soil C stabiliza-tion may be greatest in environments with high metabolicinputs and clay-rich soils (Fig 4) results that align with re-cent experimental evidence and conceptual models of SOMformation that highlight the production and stabilization ofmicrobial residues on mineral surfaces (Cortufo et al 2013Miltner et al 2012 Kleber et al 2011 Grandy and Neff2008 Marschner et al 2008) Furthermore our results sug-gest that proximal controls over the production of microbial

biomass and residues MGE and microbial turnover providean important mechanism by which microbial communitiesmay influence SOM dynamics in mineral soils (Fig 5)

Given the sensitivity of terrestrial C storage to changesin C turnover times (Friend et al 2014 Todd-Brown etal 2014 Arora et al 2013 Jones et al 2013) improv-ing the process-level representation of soil biogeochemistrymodels in environmental change remains a critical researchpriority The Earth system models used to project future Cndashcycle feedbacks currently use soil biogeochemical modelswith structures and parameterizations similar to DAYCENT(Todd-Brown et al 2013) Total soil C losses simulated byDAYCENT in response to experimental warming at Har-vard Forest are three times greater than those projected byMIMCS after 20 years (Fig 3b) Moreover MIMICS seemsto better capture observed attenuation of soil C losses withprolonged warming while allowing us to begin exploringtheoretical interactions between substrate quality microbialcommunity abundance and the formation of stable SOMThe extent to which microbial physiological variation andresponse to environmental changes can be constrained by ob-servations remains uncertain especially for MGE and micro-bial turnover but overcoming this technical challenge may becritical to resolving the potential effects of microbial func-tional diversity in soil biogeochemical models

While MIMICS provides insights into the interaction be-tween SOM dynamics and microbial physiology it alsopresents a platform for evaluating the key differences be-tween traditional and microbial modeling approaches (sum-marized in Table 2) Traditional soil biogeochemistry modelssimulate the turnover of SOM based on the implicit represen-tation of microbial activity (Schnitzer and Montreal 2011Berg and McClaugherty 2008) Soil C turnover in thesemodels is typically parameterized via first-order kinetics andmodified by environmental scalars (Wieder et al 2014 ToddBrown et al 2013) an approach that overlooks potentialwarming-induced substrate limitation andor thermal accli-mation of soil microbial communities (Fig 3 Bradford etal 2008 Kirshbaum 2004 Luo et al 2001) In microbiallyexplicit models like MIMICS substrate concentration par-tially determines rates of C turnover (Eq 1 see also Wiederet al 2013 and Allison et al 2010) Initially warming ac-celerates SOM decomposition rates through accelerated ki-netics but as the concentration of substrate pools declinesso do rates of soil C loss Future investigations into the po-tential acclimation of microbial physiological traits andorshifts in community abundance present a unique opportu-nity to evaluate and refine our mechanistic understandingof soil microbial and biogeochemical responses to environ-mental perturbations For example observations show de-creases in microbial biomass and evidence of microbial com-munity shifts with experimental warming (Bradford et al2008 Frey et al 2008) Rigorous evaluation of our process-level representation can improve our confidence projectionsof soil C feedbacks and should include cross-site analyses



and syntheses that look at soil biogeochemical and microbialresponses to experimental manipulations across environmen-tal and edaphic gradients

41 Litter inputs and SOM formation

The number of litter inputs is positively related to steady-state SOM pool sizes in traditional biogeochemistry models(Todd-Brown et al 2013) By contrast the quantity of lit-ter inputs is completely unrelated to steady-state SOM poolsize in microbially explicit models This feature appears tobe characteristic of microbially explicit models (Li et al2014 Wang et al 2014) Instead litter quantity determinesthe size of microbial biomass pools in agreement with obser-vational data (Fierer et al 2009) In MIMICS larger micro-bial biomass pools increase input rates of microbial residuesto SOM pools but they also accelerate rates of C turnoverin a ldquopriming effectrdquo This phenomenon occurs when newC inputs result in the accelerated turnover of native SOM(Phillips et al 2011 Kuzyakov 2010) Recognizing the po-tential for priming to have a disproportionately strong effecton SOM dynamics in microbially explicit models we cre-ated a physically protected SOM pool (Fig 1) This providesa mechanism whereby increasing litter inputs could increaseinputs of microbial residues to SOM pools to a greater extentthan larger microbial biomass pools could mineralize extantSOM at least in clay-rich soils However microbial biomassstill directly affects inputs and losses from SOM suggest-ing that MIMICS likely overemphasizes the role of biolog-ical processes in what should be physically dominant SOMstabilization mechanisms How frequently priming decreasesSOM with increasing litter inputs is unknown but there is anincreasing number of studies showing that increases in litterinputs do not increase or may even decrease soil C (Sulz-man et al 2005 Nadelhoffer et al 2004 Bowden et al2014 Lajtha et al 2014 but see also Leff et al 2012) Whilethe relationships between C inputs microbial biomass poolsand SOM are a controversial element of microbially explicitmodels and require clarifications they do have their basis inboth experimental evidence and theory

In traditional models increasing litter quality typicallycauses declines in soil C storage with greater partitioninginto pools with faster turnover times (Wieder et al 2014Schimel et al 1994) As parameterized in MIMICS increas-ing the chemical quality of litter inputs increases the relativeabundance of the copiotrophic microbial community (MICr)

with faster kinetics (Table 1 Fig 5a) a result that quali-tatively aligns with observations (Waring et al 2013 Ne-mergut et al 2010 Rousk and Baringaringth 2007) Our results in-dicate that the combined effects of accelerated litter turnoverand increasing microbial inputs on SOM depend on soil tex-ture with maximum soil C storage occurring in high-claysoils receiving high-quality litter inputs (Fig 4) These di-vergent projections between traditional and microbial mod-els highlight the importance of considering potential inter-

actions between microbial physiology and the physical soilenvironment

Explicit representation of the physical protection of SOMin MIMICS provides a mechanism to form stable SOM viamineral stabilization of microbial byproducts (Six et al2002 Sollins et al 1996) A high turnover of MICr com-munities can thus actually build stable SOM in resource-richenvironments when those microbial byproducts are physi-cally stabilized in finely textured soils (Fig 4) Currentlywe combine the physical protection mechanisms of aggrega-tion and mineral associations (Grandy and Robertson 2007Mikutta et al 2006 Six et al 2002 Hassink 1997) in thesame functional pool (SOMp Fig 1) Given the potentialdifferences in the long-term stabilization of various protec-tion mechanisms further analyses are needed to evaluate themodel structures and parameterizations to simulate diversestabilization mechanisms better across larger spatial and tem-poral scales

42 Microbial physiology

Microbial physiological traits related to growth and biomassproduction are key elements in determining input rates of mi-crobial residues to SOM The deficit of empirical data re-lating soil C stabilization to MGE and microbial turnoverpresents challenges to moving beyond these theoretical con-cepts however MIMICS introduces a framework to ex-plore how microbial physiological tradeoffs may influencerelative community abundance and soil C dynamics Al-though highly reductionist simplifying the metabolic andlife-history strategies of below-ground communities intobroad categories relating to microbial physiology providesa tractable path forward to begin exploring how microbialcommunity structure and abundance may affect soil biogeo-chemical processes (Miki et al 2010) While the temper-ature responses of MichaelisndashMenten kinetics are based onobservational data (German et al 2012) the model pre-sented here provides numerous avenues to explore how lesswell-defined microbial characteristics respond to the physi-cal and chemical soil environment and how changes in thoseresponses may mediate biogeochemical processes For ex-ample we have made certain assumptions about the effec-tiveness of microbial functional types in mineralizing dif-ferent C substrate pools (Table 1) While broadly based onmicrobial physiological theory (Fierer et al 2007) these as-sumptions establish competitive interactions between copi-otrophic and oligotrophic microbial communities that struc-ture the relative abundance of microbial functional types atsteady state (Fig 5)

Key physiological parameters in MIMICS include theMichaelisndashMenten kinetics of substrate mineralization (Vmaxand Km) the efficiency by which microbial communitiesturn C substrates into biomass (MGE) and the rate of mi-crobial turnover (τ) The importance of microbial physi-ology for determining soil C dynamics remains uncertain



Table 2Main effects of major model components in traditional soil biogeochemical models based on theories of chemical recalcitrance andthe MIMICS microbial model

Component Traditional model MIMICS model

Litter quality Determines partitioning to pools with differ-ent turnover times SOM pools decline with in-creasingfmet

Determines partitioning to LIT pools and the relative abun-dance of MIC communities Variable SOM pool response tofmet

Litter quantity Determines SOM pool size Determines MIC pool size

Soil texture Modulates turnover constants and partitioningof SOM between pools No explicit representa-tion of physical protection

Explicitly represents physical protection of SOM Providesa mechanism for microbial byproducts to build stable SOM

Reaction kinetics Environmentally sensitive Determines turnoverof C pools

Temperature sensitive Along with MIC pool size deter-mines substrate turnover Structures competitive dynamicsbetween MICr and MICK

MGE Determines fraction of C lost between pooltransfers no effect on rates of C mineralization

Determines fraction of C lost in transfers to MIC pools andMIC pool size Thus MGE affects rates of C mineralizationand competitive dynamics between MICr and MICK

τ Implicitly simulated as part of reaction kinetics Explicitly simulated Determines microbial control overSOM formation in mineral soils

especially in mineral soils where physical access to C sub-strates and not microbial catabolic potential limits rates ofSOM mineralization (Schimel and Schaeffer 2012) Thusthe extent to which physiological differences between micro-bial functional groups determine the fate of C remains specu-lative (Cotrufo et al 2013 Schimel and Schaeffer 2012) butMIMICS suggests that proximal factors controlling the pro-duction and turnover of microbial biomass (MGE andτ) willmediate soil C dynamics (Fig 5) These characteristics of mi-crobially explicit models show similarities to key drivers de-termining steady-state SOM pools in traditional models (en-vironmentally sensitive turnover rates the number of litterinputs and MGE Todd-Brown et al 2013 Xia et al 2013)although these parameters can elicit contrasting responses indifferent modeling frameworks

In traditional and microbial models MGE influences soilbiogeochemical responses to perturbations by determiningthe fraction of assimilated C that enters receiver pools anobservation that initiated a surge of interest in quantifyingand understanding factors that influence MGE (Frey et al2013 Sinsabaugh et al 2013 Tucker et al 2013 Manzoniet al 2012 Allison et al 2010 Bradford et al 2008) Al-though MGE is typically fixed in traditional models (Man-zoni et al 2012) reducing MGE increases the fraction ofassimilated C lost to heterotrophic respiration rates withoutmodifying rates of C mineralization from donor pools (Freyet al 2013 Tucker et al 2013) causing an overall reduc-tion in SOM pools (Fig 3) By contrast MGE and micro-bial turnover regulate both pool size and rates of litter andSOM turnover in microbially explicit models Thus reduc-ing MGE also increases the fraction of mineralized C lostto heterotrophic respiration and reduces the size of micro-

bial biomass pools in MIMICS Reducing microbial biomasspools concurrently slows rates of substrate mineralization(Eq 1) and may result in no net change in steady-state SOMpool size (Fig 3a) Changes in MGE may affect steady-stateSOM pools by influencing the relative abundance of micro-bial functional types (Fig 5) which determines both micro-bial turnover and SOM mineralization kinetics This featureis absent in microbial models lacking explicit microbial func-tional types (Wieder et al 2013) and shows that understand-ing the response of MGE to perturbations may be importantfor resolving questions of microbial competition physiolog-ical tradeoffs community composition and soil biogeochem-ical function on multiple scales

Physiological differences in catabolic potential betweenmicrobial functional types become less important in mineralsoils where soil texture determines the half-saturation con-stant for SOM mineralization (Table 1) Instead the alloca-tion of microbial biomass to the chemically and physicallyprotected pools becomes more important in determining themicrobial influence on SOM dynamics (Schimel and Schaef-fer 2012 Fig 5) In sandy soils with low physical protectionof SOM microbial communities have easier physical accessto SOM pools and biochemical protection is more importantin stabilizing SOM In MIMICS low litter quality environ-ments favor oligotrophic microbial communities which haveslower kinetics and turnover While MICK still allocate C tothe physically protected pool the combination of litter chem-ical recalcitrance and lower microbial kinetics results in morelitter byproducts entering the chemically protected pool Thisleads to greater C storage in low-clay soils receiving low-quality litter inputs (Fig 4) With increasing clay content mi-crobial access to C become restricted as physical protection



increases Accordingly soil C storage is maximized in high-clay soils that have greater microbial turnover that is allo-cated to physically protected pools (ie high-quality litterinputs)

These examples highlight the importance ofτ in determin-ing the amount of microbial control over soil C cycling al-though our inability to quantify the flow of C from microbesinto SOM (Simpson et al 2007) and rates of microbialgrowth and turnover (Blagodatskaya and Kuzyakov 2013Rousk and Baringaringth 2011 Blazewicz and Schwartz 2011)limits numerical approaches to simulating microbial physi-ology (sensu Elliott et al 1996) Thus estimating microbialturnover and its potential response to the soil environment re-mains a huge source of uncertainty in microbial models thathas not been readily assessed with current experimental tech-niques

Although our aim with this paper is to document thetheoretical underpinnings that generated the model assump-tions and structures that are applied in MIMICS we brieflydiscuss the broader implications of applying this type ofmodel in global change scenarios Microbially explicit mod-els produce notably divergent responses to perturbationswhen compared with traditional soil biogeochemistry mod-els (eg Wieder et al 2013 Wang et al 2014) For exampletemperature-sensitive reaction kinetics govern rates of soil Cturnover in both classes of models so that warming temper-atures drive losses in soil C Different assumptions about thetemperature sensitivity of SOM model structures and modelparameterizations (Todd-Brown et al 2013a Davidson andJanssens 2006 Jones et al 2005) drive large discrepanciesin model projections of soil C responses to warming tem-peratures (Todd-Brown et al 2013b) These effects are com-pounded in microbially explicit models like MIMICS be-cause the accelerated SOM turnover associated with morerapid kinetics also builds more microbial biomass which fur-ther accelerates decomposition rates (Eq 1) These dynamicsmay be dampened if the effects of temperature-driven kinet-ics are offset by concurrent declines in microbial biomasseither through reduced microbial efficiency or higher micro-bial turnover (Wang et al 2014 Wieder et al 2013) De-spite recent advances (sensu Frey et al 2013) we stress thatour theoretical and empirical understanding of these dynam-ics remains poorly quantified especially across environmen-tal gradients We hypothesize that spatiotemporal variationin microbial community dynamics and their interaction withthe physio-chemical soil environment will have strong site-level effects on soil C response warming MIMICS presents aframework to begin generating and testing these hypotheses

While microbially explicit models like MIMICS presentadvancements in our ability to represent contemporary eco-logical theory their numerical application can also gener-ate oscillatory behaviors that are not desirable in modelsused on global scales (Li et al 2014 Wang et al 2014)The oscillations evident in Fig 2 are generated by temper-ature variability and are not evident when forcing the model

with constant soil temperature (W Wieder unpublished data2014) however preliminary results indicate that MIMICShas an oscillatory behavior similar to the three-pool modelanalyzed by Wang and others (2014) (Figs 3 and 6) Ouraim with MIMICS however is to marry the strengths of mi-crobially implicit and microbially explicit approaches andfurther evaluation is needed to determine which model struc-ture and parameterizations are needed to improve our rep-resentation of contemporary soil biogeochemical theory andthe predictability of microbially driven soil C dynamics oncontinental and global scales

Evaluating the patterns and processes that emerge frommicrobially explicit models poses significant challenges andopportunities In particular MIMICS would benefit froma stronger theoretical and empirical understanding of mi-crobial physiological responses to perturbations A grow-ing body of literature documents site-level microbial com-munity shifts and physiological responses to environmentalchange drivers (Lee and Schmidt 2013 Stone et al 2012Dijkstra et al 2011 Manzoni et al 2011 Nemergut et al2010 Carney et al 2007 Waldrop et al 2004) Metage-nomic analyses may provide insight into the relative abun-dances of microbial functional types and their associatedphysiological traits (Portillo et al 2013 Verberk et al 2013Fierer et al 2012) Pairing microbial data that are sam-pled across wide geographic gradients (Ramirez et al 2012)with meta-analyses of ecosystem responses to perturbations(Janssens et al 2010 Liu and Greaver 2010) could pro-vide useful observational constraints and present broad tar-gets that models should be expected to replicate Moreoverdata-rich observations from global change experiments (egSierra et al 2012) provide additional resources to evaluatemodel assumptions structures and parameterizations Con-fidence in model projections can be improved if parameter-izations for microbial physiology traits across gradients orin response to perturbations draw on robust empirical rela-tionships This goal seems feasible for parameterizations ofMichaelisndashMenten kinetics and MGE (Frey et al 2013 Leeand Schmidt 2013 German et al 2012) Developing simi-lar empirical relationships to resolve community effects onthe fate of microbial turnover products remains a challengealthough new approaches may offer key insights (eg Aan-derud and Lennon 2011 Blazewicz and Schwartz 2011)The degree to which microbial communities affect soil C dy-namics in mineral soils (Schimel and Schaeffer 2012) likelydepend on such data

MIMICS provides a tractable test bed for exploring theimplementation of microbially based soil biogeochemicalconcepts across scales Our new microbially based modelMIMICS demonstrates how to incorporate the effects ofbelow-ground metabolic and biological diversity into biogeo-chemical cycles through the explicit representation of micro-bial functional types parameterized by functional tradeoffsin physiological strategies We also introduce a frameworkfor simulating the effects of the litter chemical quality and



physical stabilization of SOM in microbially explicit mod-els Further model developments should include soil environ-mental drivers that modify rates of biogeochemical processesand microbial community composition (eg soil moistureoxygen availability nitrogen availability soil pH land man-agement practices etc) These developments demand col-laboration between observational and modeling communi-ties and will benefit from the synthesis of data sets that canbe used to parameterize and evaluate processes simulatedacross gradients and in response to perturbations Despitethese challenges we see the potential for significant stepsforward in advancing and refining our theoretical understand-ing of soil biogeochemical cycles and the implementation ofthat theory in processed-based models

The Supplement related to this article is available onlineat doi105194bg-11-3899-2014-supplement

AcknowledgementsThe National Center for Atmospheric Re-search is sponsored by the National Science Foundation Thiswork was supported by National Science Foundation grant AGS-1020767 Financial support for A S Grandy was provided by theUnited States Department of Agriculture Soil Processes programgrant 2009-65107-05961 Financial support was also providedby the US DOE Office of Science (DE-FCO2-07ER64494)and the Office of Energy Efficiency and Renewable Energy(DE-ACO5-76RL01830) funding for the Great Lakes BioenergyResearch Center Finally C M Kallenbach was funded by theNSF (BIO-1311501) and the USDA-NIFA (2014-67011-21569)Significant funding for data accessed through Climate and Hy-drology Database Projects was provided by the National ScienceFoundation Long-Term Ecological Research program and theUSDA Forest Service

Edited by S Zaehle

References

Aanderud Z T and Lennon J T Validation of heavy-water sta-ble isotope probing for the characterization of rapidly respond-ing soil bacteria Appl Environ Microbiol 77 4589ndash4596doi101128aem02735-10 2011

Adair E C Parton W J Del Grosso S J Silver W L Har-mon M E Hall S A Burkes I C and Hart S C Simplethree-pool model accurately describes patterns of long-term lit-ter decomposition in diverse climates Global Change Biol 142636ndash2660 2008

Allison S D Wallenstein M D and Bradford M A Soil-carbonresponse to warming dependent on microbial physiology NatGeosci 3 336ndash340 doi101038ngeo846 2010

Allison S D A trait-based approach for modelling microbial litterdecomposition Ecol Lett 15 1058ndash1070 doi101111j1461-0248201201807x 2012

Arora V K Boer G J Friedlingstein P Eby M Jones CD Christian J R Bonan G Bopp L Brovkin V Cad-ule P Hajima T Ilyina T Lindsay K Tjiputra J Fand Wu T Carbon-concentration and carbon-climate feedbacksin cmip5 earth system models J Climate 26 5289ndash5314doi101175jcli-d-12-004941 2013

Bardgett R D Freeman C and Ostle N J Microbial contri-butions to climate change through carbon cycle feedbacks TheISME Journal 2 805ndash814 2008

Beardmore R E Gudelj I Lipson D A and Hurst L DMetabolic trade-offs and the maintenance of the fittest and theflattest Nature 472 342ndash346 2011

Berg B and McClaugherty C Plant litter Decomposition hu-mus formation carbon sequestration 2nd ed Springer 338 pp2008

Blagodatskaya E and Kuzyakov Y Active microorganisms insoil Critical review of estimation criteria and approaches SoilBiol Biochem 67 192ndash211 2013

Blazewicz S and Schwartz E Dynamics of18O incorporationfrom H18

2 O into soil microbial DNA Microb Ecol 61 911ndash916 doi101007s00248-011-9826-7 2011

Bol R Poirier N Balesdent J and Gleixner G Molecularturnover time of soil organic matter in particle-size fractionsof an arable soil Rapid Commun Mass Sp 23 2551ndash2558doi101002rcm4124 2009

Bonan G Ecological climatology Concepts and applications 2ed Cambridge University Press Cambridge UK New York550 pp 2008

Bonan G B Oleson K W Fisher R A Lasslop G and Re-ichstein M Reconciling leaf physiological traits and canopyflux data Use of the try and fluxnet databases in the communityland model version 4 J Geophys Res-Biogeo 117 G02026doi1010292011jg001913 2012

Bonan G B Hartman M D Parton W J and Wieder WR Evaluating litter decomposition in earth system models withlong-term litterbag experiments An example using the commu-nity land model version 4 (clm4) Global Change Biol 19 957ndash974 doi101111gcb12031 2013

Bowden R D Deem L Plante A F Peltre C m Nadel-hoffer K and Lajtha K Litter input controls on soil car-bon in a temperate deciduous forest Soil Sci Soc Am Jdoi102136sssaj2013090413nafsc 2014

Bradford M A Davies C A Frey S D Maddox T RMelillo J M Mohan J E Reynolds J F Treseder K Kand Wallenstein M D Thermal adaptation of soil microbialrespiration to elevated temperature Ecol Lett 11 1316ndash1327doi101111j1461-0248200801251x 2008

Brovkin V van Bodegom P M Kleinen T Wirth C CornwellW Cornelissen J H C and Kattge J Plant-driven variationin decomposition rates improves projections of global litter stockdistribution Biogeosciences 9 565ndash576 doi105194bg-9-565-2012 2012

Carney K M Hungate B A Drake B G and MegonigalJ P Altered soil microbial community at elevated CO2 leadsto loss of soil carbon P Natl Acad Sci 104 4990ndash4995doi101073pnas0610045104 2007

Conant R T Ryan M G Aringgren G I Birge H E DavidsonE A Eliasson P E Evans S E Frey S D Giardina C PHopkins F M Hyvoumlnen R Kirschbaum M U F Lavallee J



M Leifeld J Parton W J Megan Steinweg J WallensteinM D Martin Wetterstedt J Aring and Bradford M A Temper-ature and soil organic matter decomposition rates ndash synthesis ofcurrent knowledge and a way forward Global Change Biol 173392-3404 doi101111j1365-2486201102496x 2011

Cotrufo M F Wallenstein M D Boot C M Denef K andPaul E The microbial efficiency-matrix stabilization (mems)framework integrates plant litter decomposition with soil or-ganic matter stabilization Do labile plant inputs form sta-ble soil organic matter Global Change Biol 19 988ndash995doi101111gcb12113 2013

Currie W S Harmon M E Burke I C Hart S C Parton WJ and Silver W Cross-biome transplants of plant litter showdecomposition models extend to a broader climatic range but losepredictability at the decadal time scale Global Change Biol 161744ndash1761 doi101111j1365-2486200902086x 2010

Davidson E A and Janssens I A Temperature sensitivity of soilcarbon decomposition and feedbacks to climate change Nature440 165ndash173 doi101038nature04514 2006

del Giorgio P A and Cole J J Bacterial growth efficiency innatural aquatic systems Annu Rev Ecol Syst 29 503ndash541doi101146annurevecolsys291503 1998

Dethlefsen L and Schmidt T M Performance of the translationalapparatus varies with the ecological strategies of bacteria J Bac-teriol 189 3237ndash3245 doi101128jb01686-06 2007

Dijkstra P Thomas S C Heinrich P L Koch G W SchwartzE and Hungate B A Effect of temperature on metabolicactivity of intact microbial communities Evidence for al-tered metabolic pathway activity but not for increased mainte-nance respiration and reduced carbon use efficiency Soil BiolBiochem 43 2023ndash2031 2011

Dungait J A J Hopkins D W Gregory A S and Whit-more A P Soil organic matter turnover is governed by acces-sibility not recalcitrance Global Change Biol 18 1781ndash17961doi01111j1365-2486201202665x 2012

Ehleringer J R and Field C B Scaling physiological processesLeaf to globe edited by Roy J Academic Press Inc San DiegoCA 388 pp 1993

Elliott E Paustian K and Frey S Modeling the measurable ormeasuring the modelable A hierarchical approach to isolatingmeaningful soil organic matter fractionations in Evaluation ofsoil organic matter models edited by Powlson D Smith Pand Smith J Nato asi series Springer Berlin Heidelberg 161ndash179 1996

FAO IIASA ISRIC ISSCAS and JRC Harmonized world soildatabase (version 12) in 12 ed edited by FAO Rome Italyand IIASA Laxenburg Austria 2012

Fierer N Bradford M A and Jackson R B Toward an eco-logical classification of soil bacteria Ecology 88 1354ndash1364doi10189005-1839 2007

Fierer N Strickland M S Liptzin D Bradford M Aand Cleveland C C Global patterns in belowground com-munities Ecol Lett 12 1238ndash1249 doi101111j1461-0248200901360x2009

Fierer N Lauber C L Ramirez K S Zaneveld J BradfordM A and Knight R Comparative metagenomic phylogeneticand physiological analyses of soil microbial communities acrossnitrogen gradients ISME J 6 1007ndash1017 2012a

Fierer N Leff J W Adams B J Nielsen U N Bates S TLauber C L Owens S Gilbert J A Wall D H and Capo-raso J G Cross-biome metagenomic analyses of soil microbialcommunities and their functional attributes P Natl Acad Sci109 21390ndash21395 doi101073pnas1215210110 2012b

Fontaine S and Barot S Size and functional diversity of microbepopulations control plant persistence and long-term soil carbonaccumulation Ecol Lett 8 1075ndash1087 doi101111j1461-0248200500813x 2005

Frey S D Drijber R Smith H and Melillo J Microbialbiomass functional capacity and community structure after 12years of soil warming Soil Biol Biochem 40 2904ndash2907 2008

Frey S D Lee J Melillo J M and Six J The temperature re-sponse of soil microbial efficiency and its feedback to climateNat Clim Change 3 395ndash398 doi101038nclimate17962013

Friedlingstein P Cox P Betts R Bopp L von Bloh WBrovkin V Cadule P Doney S Eby M Fung I Bala GJohn J Jones C Joos F Kato T Kawamiya M Knorr WLindsay K Matthews H D Raddatz T Rayner P ReickC Roeckner E Schnitzler K G Schnur R Strassmann KWeaver A J Yoshikawa C and Zeng N Climate-carbon cy-cle feedback analysis Results from the c4mip model intercom-parison J Climate 19 3337ndash3353 doi101175jcli38001 2006

Friend A D Lucht W Rademacher T T Keribin R Betts RCadule P Ciais P Clark D B Dankers R Falloon P D ItoA Kahana R Kleidon A Lomas M R Nishina K OstbergS Pavlick R Peylin P Schaphoff S Vuichard N Warsza-wski L Wiltshire A and Woodward F I Carbon residencetime dominates uncertainty in terrestrial vegetation responses tofuture climate and atmospheric CO2 P Natl Acad Sci 1113280ndash3285 doi101073pnas1222477110 2014

German D P Marcelo K R B Stone M M and Allison S DThe michaelisndashmenten kinetics of soil extracellular enzymes inresponse to temperature A cross-latitudinal study Glob ChangeBiol 18 1468ndash1479 doi101111j1365-2486201102615x2012

Gholz H L Wedin D A Smitherman S M Harmon M E andParton W J Long-term dynamics of pine and hardwood litterin contrasting environments Toward a global model of decom-position Glob Change Biology 6 751ndash765 2000

Goldfarb K C Karaoz U Hanson C A Santee C A Brad-ford M A Treseder K K Wallenstein M D and Brodie EL Differential growth responses of soil bacterial taxa to carbonsubstrates of varying chemical recalcitrance Front Microbiol2 doi103389fmicb201100094 2011

Grandy A S and Robertson G P Land-use intensity effects onsoil organic carbon accumulation rates and mechanisms Ecosys-tems 10 59ndash74 doi101007s10021-006-9010-y 2007

Grandy A S and Neff J C Molecular c dynamics downstreamThe biochemical decomposition sequence and its impact on soilorganic matter structure and function Sci Total Environ 404297ndash307 doi101016jscitotenv200711013 2008

Grandy A S Strickland M S Lauber C L Bradford M Aand Fierer N The influence of microbial communities man-agement and soil texture on soil organic matter chemistry Geo-derma 150 278ndash286 2009



Harmon M E Silver W L Fasth B Chen H U A BurkeI C Parton W J Hart S C Currie W S and Lidet Long-term patterns of mass loss during the decomposition of leaf andfine root litter An intersite comparison Global Change Biol 151320ndash1338 doi101111j1365-2486200801837x 2009

Harmon M E Lter intersite fine litter decomposition experiment(lidet) 1990 to 2002 Long-term ecological research edited byBank F S D Corvallis OR 2013

Hassink J The capacity of soils to preserve organic c and n by theirassociation with clay and silt particles Plant Soil 191 77ndash87doi101023a1004213929699 1997

Heckman K Grandy A S Gao X Keiluweit M Wickings KCarpenter K Chorover J and Rasmussen C Sorptive frac-tionation of organic matter and formation of organo-hydroxy-aluminum complexes during litter biodegradation in the presenceof gibbsite Geochim Cosmochim Ac 121 667ndash683 2013

Hugelius G Tarnocai C Broll G Canadell J G Kuhry P andSwanson D K The northern circumpolar soil carbon databaseSpatially distributed datasets of soil coverage and soil carbonstorage in the northern permafrost regions Earth Syst Sci Data5 3ndash13 doi105194essd-5-3-2013 2013

Janssens I A Dieleman W Luyssaert S Subke J A Re-ichstein M Ceulemans R Ciais P Dolman A J GraceJ Matteucci G Papale D Piao S L Schulze E DTang J and Law B E Reduction of forest soil respirationin response to nitrogen deposition Nat Geosci 3 315ndash322doi101038ngeo844 2010

Jastrow J Amonette J and Bailey V Mechanisms con-trolling soil carbon turnover and their potential applicationfor enhancing carbon sequestration Clim Change 80 5ndash23doi101007s10584-006-9178-3 2007

Jenkinson D S Adams D E and Wild A Model estimates ofco2 emissions from soil in response to global warming Nature351 304ndash306 1991

Jobbaacutegy E G and Jackson R B The vertical distribution ofsoil organic carbon and its relation to climate and vegeta-tion Ecological Applications 10 423ndash436 doi1018901051-0761(2000)010[0423TVDOSO]20CO2 2000

Jones C McConnell C Coleman K Cox P Falloon P Jenkin-son D and Powlson D Global climate change and soil carbonstocks predictions from two contrasting models for the turnoverof organic carbon in soil Global Change Biol 11 154ndash166doi101111j1365-2486200400885x 2005

Jones C Robertson E Arora V Friedlingstein P ShevliakovaE Bopp L Brovkin V Hajima T Kato E Kawamiya MLiddicoat S Lindsay K Reick C H Roelandt C Segschnei-der J and Tjiputra J Twenty-first-century compatible CO2emissions and airborne fraction simulated by cmip5 earth sys-tem models under four representative concentration pathways JClimate 26 4398ndash4413 doi101175jcli-d-12-005541 2013

Jones C D Cox P and Huntingford C Uncertainty inclimatendashcarbon-cycle projections associated with the sensitiv-ity of soil respiration to temperature Tellus B 55 642ndash648doi101034j1600-0889200301440x 2003

Kaiser M Walter K Ellerbrock R H and Sommer M Effectsof land use and mineral characteristics on the organic carboncontent and the amount and composition of na-pyrophosphate-soluble organic matter in subsurface soils Europ J Soil Sci62 226ndash236 doi101111j1365-2389201001340x 2011

Kattge J Diacuteaz S Lavorel S Prentice I C Leadley P BoumlnischG Garnier E Westoby M Reich P B Wright I J Cornelis-sen J H C Violle C Harrison S P Van Bodegom P M Re-ichstein M Enquist B J Soudzilovskaia N A Ackerly DD Anand M Atkin O Bahn M Baker T R Baldocchi DBekker R Blanco C C Blonder B Bond W J BradstockR Bunker D E Casanoves F Cavender-Bares J ChambersJ Q Chapin Iii F S Chave J Coomes D Cornwell WK Craine J M Dobrin B H Duarte L Durka W ElserJ Esser G Estiarte M Fagan W F Fang J Fernaacutendez-Meacutendez F Fidelis A Finegan B Flores O Ford H FrankD Freschet G T Fyllas N M Gallagher R V Green WA Gutierrez A G Hickler T Higgins S I Hodgson J GJalili A Jansen S Joly C A Kerkhoff A J Kirkup D Ki-tajima K Kleyer M Klotz S Knops J M H Kramer KKuumlhn I Kurokawa H Laughlin D Lee T D Leishman MLens F Lenz T Lewis S L Lloyd J Llusiagrave J LouaultF Ma S Mahecha M D Manning P Massad T MedlynB E Messier J Moles A T Muumlller S C Nadrowski KNaeem S Niinemets Uuml Noumlllert S Nuumlske A Ogaya ROleksyn J Onipchenko V G Onoda Y Ordontildeez J Over-beck G Ozinga W A Patintildeo S Paula S Pausas J GPentildeuelas J Phillips O L Pillar V Poorter H Poorter LPoschlod P Prinzing A Proulx R Rammig A Reinsch SReu B Sack L Salgado-Negret B Sardans J Shiodera SShipley B Siefert A Sosinski E Soussana J F Swaine ESwenson N Thompson K Thornton P Waldram M Wei-her E White M White S Wright S J Yguel B ZaehleS Zanne A E and Wirth C Try ndash a global database of planttraits Global Change Biol 17 2905ndash2935 doi101111j1365-2486201102451x 2011

Keiblinger K M Hall E K Wanek W Szukics U Haumlm-merle I Ellersdorfer G Boumlck S Strauss J SterflingerK Richter A and Zechmeister-Boltenstern S The effectof resource quantity and resource stoichiometry on microbialcarbon-use-efficiency FEMS Microbiol Ecol 73 430ndash440doi101111j1574-6941201000912x 2010

Kirschbaum M U F Soil respiration under prolonged soil warm-ing Are rate reductions caused by acclimation or substrateloss Global Change Biol 10 1870ndash1877 doi101111j1365-2486200400852x 2004

Klappenbach J A Dunbar J M and Schmidt T M Rrna operoncopy number reflects ecological strategies of bacteria Appl En-viron Microbiol 66 1328ndash1333 doi101128aem6641328-13332000 2000

Kleber M Nico P S Plante A Filley T Kramer M SwanstonC and Sollins P Old and stable soil organic matter is notnecessarily chemically recalcitrant Implications for modelingconcepts and temperature sensitivity Global Change Biol 171097ndash1107 doi101111j1365-2486201002278x 2011

Knapp A K and Smith M D Variation among biomes in tempo-ral dynamics of aboveground primary production Science 291481ndash484 2001

Koven C D Riley W J Subin Z M Tang J Y Torn M SCollins W D Bonan G B Lawrence D M and SwensonS C The effect of vertically resolved soil biogeochemistry andalternate soil C and N models on C dynamics of CLM4 Biogeo-sciences 10 7109ndash7131 doi105194bg-10-7109-2013 2013



Kramer M G Sanderman J Chadwick O A Chorover Jand Vitousek P M Long-term carbon storage through re-tention of dissolved aromatic acids by reactive particles insoil Global Change Biol 18 2594ndash2605 doi101111j1365-2486201202681x 2012

Kuzyakov Y Priming effects Interactions between living anddead organic matter Soil Biol Biochem 42 1363ndash1371doi101016jsoilbio201004003 2010

Lajtha K Bowden R D and Nadelhoffer K Litter androot manipulations provide insights into soil organicmatter dynamics and stability Soil Sci Soc Am Jdoi102136sssaj2013080370nafsc 2014

Lawrence C R Neff J C and Schimel J P Does adding mi-crobial mechanisms of decomposition improve soil organic mat-ter models A comparison of four models using data from apulsed rewetting experiment Soil Biol Biochem 41 1923ndash1934 doi101016jsoilbio200906016 2009

Lawrence D Oleson K W Flanner M G Thorton P E Swen-son S C Lawrence P J Zeng X Yang Z-L Levis S Sk-aguchi K Bonan G B and Slater A G Parameterization im-provements and functional and structural advances in version 4of the community land model J Adv Model Earth Syst 3 27pp doi1010292011ms000045 2011

Lee Z M and Schmidt T M Bacterial growth efficiency variesin soils under different land management practices Soil BiolBiochem 69 282ndash290 2014

Leff J W Wieder W R Taylor P G Townsend A R Ne-mergut D R Grandy A S and Cleveland C C Experimentallitterfall manipulation drives large and rapid changes in soil car-bon cycling in a wet tropical forest Global Change Biol 182969ndash2979 doi101111j1365-2486201202749x 2012

Li J Wang G Allison S Mayes M and Luo Y Soil carbonsensitivity to temperature and carbon use efficiency comparedacross microbial-ecosystem models of varying complexity Bio-geochemistry 1ndash18 doi101007s10533-013-9948-8 2014

Liang C Cheng G Wixon D and Balser T An absorb-ing markov chain approach to understanding the microbial rolein soil carbon stabilization Biogeochemistry 106 303ndash309doi101007s10533-010-9525-3 2011

Lipson D Monson R Schmidt S and Weintraub M Thetrade-off between growth rate and yield in microbial commu-nities and the consequences for under-snow soil respiration ina high elevation coniferous forest Biogeochemistry 95 23ndash35doi101007s10533-008-9252-1 2009

Liu L L and Greaver T L A global perspective on below-ground carbon dynamics under nitrogen enrichment Ecol Lett13 819ndash828 doi101111j1461-0248201001482x 2010

Loreau M Microbial diversity producer-decomposer interactionsand ecosystem processes A theoretical model P Roy SocLond B 268 303ndash309 doi101098rspb20001366 2001

Luo Y Wan S Hui D and Wallace L L Acclimatization ofsoil respiration to warming in a tall grass prairie Nature 413622ndash625 2001

Manzoni S Schimel J P and Porporato A Responses of soilmicrobial communities to water stress Results from a meta-analysis Ecology 93 930ndash938 doi10189011-00261 2011

Manzoni S Taylor P Richter A Porporato A and AringgrenG I Environmental and stoichiometric controls on micro-

bial carbon-use efficiency in soils New Phytol 196 79ndash91doi101111j1469-8137201204225x 2012

Melillo J M Aber J D and Muratore J F Nitrogen and lignincontrol of hardwood leaf litter decomposition dynamics Ecol-ogy 63 621ndash626 1982

Melillo J M Steudler P A Aber J D Newkirk K Lux HBowles F P Catricala C Magill A Ahrens T and Morris-seau S Soil warming and carbon-cycle feedbacks to the climatesystem Science 298 2173ndash2176 2002

Melillo J M Butler S Johnson J Mohan J Steudler P LuxH Burrows E Bowles F Smith R Scott L Vario C HillT Burton A Zhou Y-M and Tang J Soil warming carbon-nitrogen interactions and forest carbon budgets P Natl AcadSci 108 9508ndash9512 doi101073pnas1018189108 2011

Metherell A K Harding L A Cole C V and Parton W JCentury soil organic matter model environment Technical doc-umentation agroecosystem version 40 Great plains system re-search unit technical report no 4Usda-ars Fort Collins COUSA 1993

Miki T Ushio M Fukui S and Kondoh M Functional diversityof microbial decomposers facilitates plant coexistence in a plant-microbe-soil feedback model P Natl Acad Sci 107 14251ndash14256 doi101073pnas0914281107 2010

Mikutta R Kleber M Torn M and Jahn R Stabilization ofsoil organic matter Association with minerals or chemical re-calcitrance Biogeochemistry 77 25ndash56 doi101007s10533-005-0712-6 2006

Miltner A Bombach P Schmidt-Bruumlcken B and Kaumlstner MSom genesis Microbial biomass as a significant source Biogeo-chemistry 111 41ndash55 doi101007s10533-011-9658-z 2012

Molenaar D van Berlo R de Ridder D and Teusink B Shiftsin growth strategies reflect tradeoffs in cellular economics MolSyst Biol 5 doi101038msb200982 2009

Moorhead D L and Sinsabaugh R L A theoretical model of lit-ter decay and microbial interaction Ecol Monogr 76 151ndash174doi1018900012-9615(2006)076[0151atmold]20co2 2006

Moorhead D L Lashermes G l and Sinsabaugh R L A theo-retical model of c- and n-acquiring exoenzyme activities whichbalances microbial demands during decomposition Soil BiolBiochem 53 133ndash141 2012

Nadelhoffer K J Boone R D Bowden R D Canary J DKaye J P Micks P Ricca A Aitkenhead J A Lajtha Kand McDowell W H The dirt experiment Litter and root in-fluences on forest soil organic matter stocks and function inForests in time The environmental consequences of 1000 yearsof change in new england edited by D Foster and Aber J YaleUniversity Press New Haven 300ndash315 2004

Nemergut D R Cleveland C C Wieder W R WashenbergerC L and Townsend A R Plot-scale manipulations of organicmatter inputs to soils correlate with shifts in microbial commu-nity composition in a lowland tropical rain forest Soil BiolBiochem 42 2153ndash2160 doi101016jsoilbio2010080112010

Oechel W C Vourlitis G L Hastings S J Zulueta R C Hinz-man L and Kane D Acclimation of ecosystem CO2 exchangein the alaskan arctic in response to decadal climate warming Na-ture 406 978ndash981 2000

Parton W Silver W L Burke I C Grassens L Harmon M ECurrie W S King J Y Adair E C Brandt L A Hart S C



and Fasth B Global-scale similarities in nitrogen release pat-terns during long-term decomposition Science 315 361ndash3642007

Parton W J Schimel D S Cole C V and Ojima D SAnalysis of factors controlling soil organic-matter levels ingreat-plains grasslands Soil Sci Soc Am J 51 1173ndash1179doi102136sssaj198703615995005100050015x 1987

Parton W J Schimel D S Cole C V and Ojima D S A gen-eral model for soil organic matter dynamics Sensitivity to litterchemistry texture and management in Quantitative modelingof soil forming processes edited by Bryant R B and Arnold RW Soil Science Society of America Minneapolis MinnesotaUSA 147ndash167 1994

Pfeiffer T Schuster S and Bonhoeffer S Cooperation and com-petition in the evolution of atp-producing pathways Science292 504ndash507 doi101126science1058079 2001

Phillips R P Finzi A C and Bernhardt E S Enhanced rootexudation induces microbial feedbacks to n cycling in a pine for-est under long-term CO2 fumigation Ecol Lett 14 187ndash194doi101111j1461-0248201001570x 2011

Pianka E R On r- and k-selection The American Naturalist 104592ndash597 doi1023072459020 1970

Portillo M C Leff J W Lauber C L and Fierer N Cell sizedistributions of soil bacterial and archaeal taxa Appl EnvironMicrobiol 79 7610ndash7617 doi101128aem02710-13 2013

R Development Core Team R A language and environment forstatistical computing R Foundation for Statistical ComputingVienna Austria 2011

Ramirez K S Craine J M and Fierer N Consistent effects ofnitrogen amendments on soil microbial communities and pro-cesses across biomes Global Change Biol 18 1918ndash1927doi101111j1365-2486201202639x 2012

Resat H Bailey V McCue L and Konopka A Mod-eling microbial dynamics in heterogeneous environmentsGrowth on soil carbon sources Microb Ecol 63 883ndash897doi101007s00248-011-9965-x 2012

Rousk J and Baringaringth E Fungal and bacterial growth in soil withplant materials of different cn ratios FEMS Microbiol Ecol 62258ndash267 doi101111j1574-6941200700398x 2007

Rousk J and Baringaringth E Growth of saprotrophic fungi and bacteriain soil FEMS Microbiol Ecol 78 17ndash30 doi101111j1574-6941201101106x 2011

Russell J B and Cook G M Energetics of bacterial growth Bal-ance of anabolic and catabolic reactions Microbiol Rev 59 48ndash62 1995

Schimel D S Braswell B H Holland E A McKeown ROjima D S Painter T H Parton W J and TownsendA R Climatic edaphic and biotic controls over storage andturnover of carbon in soils Global Biogeochem Cy 8 279ndash293doi10102994GB00993 1994

Schimel J and Schaeffer S M Microbial controlover carbon cycling in soil Front Microbiol 3doi103389fmicb201200348 2012

Schimel J P and Weintraub M N The implications of exoenzymeactivity on microbial carbon and nitrogen limitation in soil Atheoretical model Soil Biol Biochem 35 549ndash563 2003

Schmidt M W I Torn M S Abiven S Dittmar T Guggen-berger G Janssens I A Kleber M Kogel-Knabner ILehmann J Manning D A C Nannipieri P Rasse D

P Weiner S and Trumbore S E Persistence of soil or-ganic matter as an ecosystem property Nature 478 49ndash56doi101038nature10386 2011

Schnitzer M and Montreal C M Quo vadis soil organic mat-ter research A biological link to the chemistry of humificationin Advances in Agronomy edited by Sparks D L AcademicPress 143ndash217 2011

Serna-Chavez H M Fierer N Van Bodegom P M Globaldrivers and patterns of microbial abundance in soil Global EcolBiogeogr 22 1162ndash1172 2013

Sierra C A Trumbore S E Davidson E A Frey S D SavageK E and Hopkins F M Predicting decadal trends and transientresponses of radiocarbon storage and fluxes in a temperate for-est soil Biogeosciences 9 3013ndash3028 doi105194bg-9-3013-2012 2012

Simpson A J Simpson M J Smith E and Kelleher B PMicrobially derived inputs to soil organic matter Are currentestimates too low Environ Sci Technol 41 8070ndash8076doi101021es071217x 2007

Sinsabaugh R L Manzoni S Moorhead D L and RichterA Carbon use efficiency of microbial communities Stoichiom-etry methodology and modelling Ecol Lett 16 930ndash939doi101111ele12113 2013

Six J Conant R T Paul E A and Paustian KStabilization mechanisms of soil organic matter Implica-tions for c-saturation of soils Plant Soil 241 155ndash176doi101023a1016125726789 2002

Six J Frey S D Thiet R K and Batten K M Bacterial andfungal contributions to carbon sequestration in agroecosystemsSoil Sci Soc Am J 70 555ndash569 doi102136sssaj200403472006

Soetaert K rootSolve Nonlinear Root Finding Equilibrium andSteady-State Analysis of Ordinary Differential Equations Rpackage version 14 2009

Sollins P Homann P and Caldwell B A Stabilization anddestabilization of soil organic matter mechanisms and controlsGeoderma 74 65ndash105 1996

Steinweg J M Plante A F Conant R T Paul E A andTanaka D L Patterns of substrate utilization during long-termincubations at different temperatures Soil Biol Biochem 402722ndash2728 2008

Stone M M Weiss M S Goodale C L Adams M B Fer-nandez I J German D P and Allison S D Tempera-ture sensitivity of soil enzyme kinetics under n-fertilization intwo temperate forests Global Change Biol 18 1173ndash1184doi101111j1365-2486201102545x 2012

Strickland M S and Rousk J Considering fungalBacterial dom-inance in soils ndash methods controls and ecosystem implicationsSoil Biol Biochem 42 1385ndash1395 2010

Sulzman E W Brant J B Bowden R D and Lajtha K Contri-bution of aboveground litter belowground litter and rhizosphererespiration to total soil CO2 efflux in an old growth coniferousforest Biogeochemistry 73 231ndash256 doi101007s10533-004-7314-6 2005

Thiet R K Frey S D and Six J Do growth yield efficien-cies differ between soil microbial communities differing in fun-galBacterial ratios Reality check and methodological issuesSoil Biol Biochem 38 837ndash844 2006



Todd-Brown K Hopkins F Kivlin S Talbot J and Alli-son S A framework for representing microbial decomposi-tion in coupled climate models Biogeochemistry 109 19ndash33doi101007s10533-011-9635-6 2012

Todd-Brown K E O Randerson J T Post W M Hoff-man F M Tarnocai C Schuur E A G and Allison S DCauses of variation in soil carbon simulations from CMIP5Earth system models and comparison with observations Biogeo-sciences 10 1717ndash1736 doi105194bg-10-1717-2013 2013

Todd-Brown K E O Randerson J T Hopkins F Arora V Ha-jima T Jones C Shevliakova E Tjiputra J Volodin E WuT Zhang Q and Allison S D Changes in soil organic carbonstorage predicted by earth system models during the 21st centuryBiogeosciences 11 2341ndash2356 doi105194bg-11-2341-20142014

Treseder K Balser T Bradford M Brodie E Dubinsky EEviner V Hofmockel K Lennon J Levine U MacGregorB Pett-Ridge J and Waldrop M Integrating Microb Ecolinto ecosystem models Challenges and priorities Biogeochem-istry 109 7ndash18 doi101007s10533-011-9636-5 2012

Tucker C L Bell J Pendall E and Ogle K Does declin-ing carbon-use efficiency explain thermal acclimation of soilrespiration with warming Global Change Biol 19 252ndash263doi101111gcb12036 2013

Verberk W C E P van Noordwijk C G E and Hildrew A GDelivering on a promise Integrating species traits to transformdescriptive community ecology into a predictive science Fresh-water Science 32 531ndash547 doi10189912-0921 2013

von Luumltzow M Koumlgel-Knabner I Ludwig B Matzner EFlessa H Ekschmitt K Guggenberger G Marschner Band Kalbitz K Stabilization mechanisms of organic mat-ter in four temperate soils Development and application ofa conceptual model J Plant Nutr Soil Sci 171 111ndash124doi101002jpln200700047 2008

Waldrop M P and Firestone M K Microbial community utiliza-tion of recalcitrant and simple carbon compounds Impact of oak-woodland plant communities Oecologia 138 275ndash284 2004

Wallenstein M and Hall E A trait-based framework for pre-dicting when and where microbial adaptation to climate changewill affect ecosystem functioning Biogeochemistry 109 35ndash47doi101007s10533-011-9641-8 2012

Wallenstein M D Haddix M L Ayres E Steltzer HMagrini-Bair K A and Paul E A Litter chemistry changesmore rapidly when decomposed at home but converges dur-ing decomposition-transformation Soil Biol Biochem 57 311ndash319 2013

Wang G Post W M and Mayes M A Development ofmicrobial-enzyme-mediated decomposition model parametersthrough steady-state and dynamic analyses Ecol Appl 23 255ndash272 doi10189012-06811 2013

Wang Y P Chen B C Wieder W R Leite M Medlyn BE Rasmussen M Smith M J Agusto F B Hoffman Fand Luo Y Q Oscillatory behavior of two nonlinear microbialmodels of soil carbon decomposition Biogeosciences 111817ndash1831 10httpwwwbiogeosciencesnet111817105194bg-11-1817-2014 2014

Waring B G Averill C Hawkes C V and Holyoak M Differ-ences in fungal and bacterial physiology alter soil carbon and ni-trogen cycling Insights from meta-analysis and theoretical mod-els Ecol Lett 16 887ndash894 doi101111ele12125 2013

Wickings K Grandy A S Reed S C and Cleveland C C Theorigin of litter chemical complexity during decomposition EcolLett 15 1180ndash1188 doi101111j1461-0248201201837x2012

Wieder W R Bonan G B and Allison S D Global soil carbonprojections are improved by modelling microbial processes NatClim Change 3 909ndash912 doi101038nclimate1951 2013

Wieder W R Boehnert J and Bonan G B Evaluatingsoil biogeochemistry parameterizations in earth system mod-els with observations Global Biogeochem Cy 28 211ndash222doi1010022013gb004665 2014

Xia J Luo Y Wang Y-P and Hararuk O Traceablecomponents of terrestrial carbon storage capacity in bio-geochemical models Global Change Biol 19 2104ndash2116doi101111gcb12172 2013

Xia J Y Luo Y Q Wang Y P Weng E S and Hararuk OA semi-analytical solution to accelerate spin-up of a coupled car-bon and nitrogen land model to steady state Geosci Model Dev5 1259ndash1271 doi105194gmd-5-1259-2012 2012

Yang X Wittig V Jain A K and Post W Integrationof nitrogen cycle dynamics into the integrated science as-sessment model for the study of terrestrial ecosystem re-sponses to global change Global Biogeochem Cy 23 GB4029doi1010292009gb003474 2009



representation of soil biogeochemistry models that are usedon multiple scales

A growing body of literature calls for significant revi-sions to the theoretical basis for modeling soil C dynamics(Cotrufo et al 2013 Dungait et al 2012 Conant et al2011 Schmidt et al 2011) Traditional soil C stabilizationconcepts do not explicitly simulate microbial activity or soilmicrobial communities but instead strongly emphasize therelationship between litter chemical recalcitrance and soil Cstorage By contrast new theoretical and experimental re-search shows that soil microbes strongly mediate the for-mation of soil organic matter (SOM) through the productionof microbial products that appear to form mineral-stabilizedSOM (Wallenstein et al 2013 Miltner et al 2012 Wick-ings et al 2012 Kleber et al 2011 Grandy and Neff 2008Six et al 2006) These insights suggest that basic physi-ological traits such as microbial growth efficiency (MGE)and growth kinetics have direct influences on litter decom-position rates and net microbial biomass production whilethe subsequent turnover of microbial biomass strongly influ-ences input rates to SOM Furthermore the ultimate fate ofSOM also depends upon the mineral stabilization of thesemicrobially derived inputs However despite wide recogni-tion that microbial physiology and soil mineral interactionsfacilitate the formation of stable SOM this theoretical in-sight has not been adequately represented in process-basedmodels

These emerging concepts highlight the need to simulateexplicitly the microbial processes responsible for decompo-sition and stabilization of organic matter (Todd-Brown et al2012 Treseder et al 2012 Allison et al 2010 Lawrenceet al 2009 Bardgett et al 2008 Schimel and Weintraub2003) even if the magnitude of microbial control over soil Cdynamics in mineral soils remains poorly defined (Schimeland Schaeffer 2012) Microbially explicit approaches in re-cent models range in complexity from simple fungal to bac-terial ratios (Waring et al 2013) microbial guilds specializ-ing in different litter C substrates (Miki et al 2010 Moor-head and Sinsabaugh 2006) and complex community dy-namics (Allison 2012 Wallenstein and Hall 2012 Loreau2001) These models incorporate the complexity of micro-bial physiology and competitive interactions in litter decom-position dynamics and provide valuable insight into our un-derstanding of ldquoupstreamrdquo soil C inputs but focus less at-tention on the stabilization of microbial residues in mineralsoils Other recent work demonstrates that simple non-linearmicrobial models are feasible on the global scale and resultsin divergent responses compared with traditional soil biogeo-chemistry models (Wieder et al 2013) Again however thismicrobial modeling framework does not adequately capturehow microbial physiology and activity may facilitate the sta-bilization of SOM

Microbial attributes that regulate microbial residue inputsto SOM and their interactions with the soil matrix are ef-fectively absent in traditional soil biogeochemical models

and poorly accounted for in current microbially based mod-els Thus our chief motivation is to develop a process-basedmodeling framework to explore the potential role of micro-bial physiology and the stabilization of microbial biomassat the soilndashmineral interface as key drivers in the formationof SOM (Miltner et al 2012 Liang et al 2011 Bol et al2009 Grandy et al 2009) In this model litter and SOMturnover are governed by microbial biomass pools whichcorrespond to different microbial functional types Inherentphysiological differences between microbial functional typesprovide a basis to begin simulating below-ground biologicaland metabolic diversity and explore how the relative abun-dance of different microbial functional groups regulates bio-geochemical processes (Miki et al 2010) Microbial growthrates growth efficiency and turnover are subject to intrin-sic physiological constraints (Beardmore et al 2011 Mole-naar et al 2009 Dethlefsen and Schmidt 2007) but arealso sensitive to external forces such as resource chemistry(Manzoni et al 2012 Keiblinger et al 2010 Steinweg etal 2008 Rousk and Baringaringth 2007 Thiet et al 2006) suchthat both community composition and the soil environmentshould determine the optimization of physiological traits andtheir downstream influence on SOM dynamics

We use observational and theoretical insights to guideour incorporation of microbial physiological processes intopredictions of SOM stabilization Physiological differencesacross species have been linked to life-history strategies opti-mized for different resource environments (Resat et al 2012Beardmore et al 2011 Russell and Cook 1995) For in-stance in resource-rich environments fast-growing r strate-gists (copiotrophs) are typically characterized by a lowerMGE but higher growth rates relative to slower-growing Kstrategists (oligiotrophs Fierer et al 2012 Ramirez et al2012 Fierer et al 2007 Klappenbach et al 2000 Pianka1970) This physiology gives copiotrophs a competitive ad-vantage under resource-rich conditions such that they tendto dominate in these environments On an individual specieslevel MGE growth rates and turnover are expected to in-crease as resource quality increases However selection for acopiotroph-dominated community may drive up community-level growth rates and turnover at the expense of a lowerMGE The effect of this tradeoff on total biomass produc-tion and microbially derived inputs to physically protectedSOM is uncertain and remains a challenging often missingaspect of microbially based soil C models

In order to more rigorously evaluate dynamics betweenmicrobial physiology soil environmental conditions andSOM formation we introduce a new soil biogeochemistrymodel MIMICS (MIcrobial-MIneral Carbon Stabilization)MIMICS incorporates the relationships between microbialphysiology substrate chemical quality and the physical sta-bilization of SOM (Wang et al 2013 Goldfarb et al 2011Fontaine and Barot 2005) with the long-term aim of rep-resenting these interactions in global simulations using theCommunity Land Model (Lawrence et al 2011) Thus here



SOMpLITm MICr

Vmax Km

LITs MICK SOMc

τ

fi met

fi struc

fmet

fc

1-MGE

MGE

1-fc

I1-fmet

(1)

(2)

(6)

(7)

(5)

(8)

(4)

(10)

(3)

(9)



2 Methods






dCs

dt= MIC times

Vmaxtimes Cs

Km + Cs (1)














entering MICr

10 2 6 2c ndash


2 2 2 2d ndash




entering MICr



































0 2 4 6 8 10

020

4060

80100

time (y)

Mas

s re

mai

ning

()

(a)

0 2 4 6 8 10

time (y)

(b)


3 Results




SO

M c

hang

e (g

C m

minus2)

-3000-2500-1000

-500

0

(a)

year

Tota

l C c

hang

e (g

C m

minus2)

0 5 10 15 20 500

-4500

-2500

-1500

-500

(b)





























4 Discussion

















































Edited by S Zaehle

References
































































































































































SOMpLITm MICr

Vmax Km

LITs MICK SOMc

τ

fi met

fi struc

fmet

fc

1-MGE

MGE

1-fc

I1-fmet

(1)

(2)

(6)

(7)

(5)

(8)

(4)

(10)

(3)

(9)



2 Methods






dCs

dt= MIC times

Vmaxtimes Cs

Km + Cs (1)














entering MICr

10 2 6 2c ndash


2 2 2 2d ndash




entering MICr



































0 2 4 6 8 10

020

4060

80100

time (y)

Mas

s re

mai

ning

()

(a)

0 2 4 6 8 10

time (y)

(b)


3 Results




SO

M c

hang

e (g

C m

minus2)

-3000-2500-1000

-500

0

(a)

year

Tota

l C c

hang

e (g

C m

minus2)

0 5 10 15 20 500

-4500

-2500

-1500

-500

(b)





























4 Discussion

















































Edited by S Zaehle

References







































































































































































entering MICr

10 2 6 2c ndash


2 2 2 2d ndash




entering MICr



































0 2 4 6 8 10

020

4060

80100

time (y)

Mas

s re

mai

ning

()

(a)

0 2 4 6 8 10

time (y)

(b)


3 Results




SO

M c

hang

e (g

C m

minus2)

-3000-2500-1000

-500

0

(a)

year

Tota

l C c

hang

e (g

C m

minus2)

0 5 10 15 20 500

-4500

-2500

-1500

-500

(b)





























4 Discussion

















































Edited by S Zaehle

References




















































































































































































0 2 4 6 8 10

020

4060

80100

time (y)

Mas

s re

mai

ning

()

(a)

0 2 4 6 8 10

time (y)

(b)


3 Results




SO

M c

hang

e (g

C m

minus2)

-3000-2500-1000

-500

0

(a)

year

Tota

l C c

hang

e (g

C m

minus2)

0 5 10 15 20 500

-4500

-2500

-1500

-500

(b)





























4 Discussion

















































Edited by S Zaehle

References












































































































































































0 2 4 6 8 10

020

4060

80100

time (y)

Mas

s re

mai

ning

()

(a)

0 2 4 6 8 10

time (y)

(b)


3 Results




SO

M c

hang

e (g

C m

minus2)

-3000-2500-1000

-500

0

(a)

year

Tota

l C c

hang

e (g

C m

minus2)

0 5 10 15 20 500

-4500

-2500

-1500

-500

(b)





























4 Discussion

















































Edited by S Zaehle

References
































































































































































0 2 4 6 8 10

020

4060

80100

time (y)

Mas

s re

mai

ning

()

(a)

0 2 4 6 8 10

time (y)

(b)


3 Results




SO

M c

hang

e (g

C m

minus2)

-3000-2500-1000

-500

0

(a)

year

Tota

l C c

hang

e (g

C m

minus2)

0 5 10 15 20 500

-4500

-2500

-1500

-500

(b)





























4 Discussion

















































Edited by S Zaehle

References






















































































































































































4 Discussion

















































Edited by S Zaehle

References










































































































































































4 Discussion

















































Edited by S Zaehle

References


































































































































































4 Discussion

















































Edited by S Zaehle

References










































































































































































































Edited by S Zaehle

References






























































































































































































Edited by S Zaehle

References












































































































































































Edited by S Zaehle

References



































































































































































Edited by S Zaehle

References
























































































































































































































































































































































































































































































































































































Documents

Integrating microbial physiology and physio-chemical principles in