Organic Matter Degradation across Ecosystem Boundaries: The Need for a Unified Conceptualization

Opinion O ganic Ma e Deg ada ion ac oss Ecosys em Bounda ies: The Need o a Unified Concep ualiza ion Dolly N. Ko hawala, 1, *Anne M. Kelle man, 2 Nú ia Ca alán, 3,4 and La s J. T an ik 1 The global ca bon cycle connec s o ganic ma e (OM) pools in soil, eshwa e , and ma ine ecosys ems wi h he a mosphe e, he eby egula ing hei size and eac i i y. Due o he complexi y o biogeochemical p ocesses and his o ically compa men alized disciplines, ecosys em-specific concep ualiza ions o OM deg ada ion ha e eme ged independen ly o de elopmen s in o he ecosys ems. Recen discussions ega ding he ela i e impo ance o molecula composi ion and ecosys em p ope ies on OM deg ada ion ha e di e ged in opposing di ec- ions ac oss subdisciplines, lea ing ou unde s anding inconsis en . Ecosys em- dependen heo ies a e p oblema ic since p ope ies unique o an ecosys em may change in esponse o an h opogenic s esso s, including clima e change. The nex b eak h ough in ou unde s anding o OM deg ada ion equi es a shi in ocus owa ds de eloping a unified heo y o con ols on OM ac oss ecosys ems. Wha Con ols O ganic Ma e Pe sis ence and Reac i i y? The suscep ibili y o o ganic ma e (OM) o ei he pe sis and accumula e as a long- e m sink o ca bon, o cycle apidly and become mine alized (see Glossa y) in o a mosphe ic CO 2 , is a key ea u e o ca bon cycling ac oss he biogeosphe e. Ye , a seemingly simple ques ion emains un esol ed ac oss soil, eshwa e , and ma ine biogeochemis y: wha con ols he deg ada ion o OM? We know ha OM can be highly eac i e, and is deg aded wi hin minu es in some soil and eshwa e en i onmen s. Al e na i ely, i can also be highly pe sis en , las ing o millennia in soils, sedimen s, and he open ocean. Decades ago, he consensus would ha e been ha OM pe sis ence is eflec ed by i s molecula composi ion. In ime, flaws in his hinking we e e ealed, wi h he ecogni ion ha OM pe sis s a longe han can be explained by molecula composi ion alone [1]. Wi h his insigh and ecen echnological ad ances [2], he ela i e impo - ance o OM composi ion has shi ed in di e ging di ec ions o e ime (Box 1). Mos no ably, he soil and eshwa e lines o hinking a e de eloping in opposing di ec ions. The soil communi y came om a his o y o conside ing ha soil OM could p og essi ely become ‘ e ac o y’due o humi ica ion and selec i e p ese a ion [3–5], wi h molecula composi ion (e.g., lignin:N) being impo an o p edic ing deg ada ion a es [6,7]. The soil communi y now inc easingly ecognizes he impo ance o ecosys em p ope ies [8], wi h molecula composi ion being less ele an [9]. By con as , he eshwa e communi y fi s p oposed he i e con inuum concep du ing he 1980s [10], whe e shi s in he molecula composi ion o dissol ed o ganic ma e (DOM) wi h mo emen downs eam we e expec ed due o he p e e en ial use o subs a es. This heo y con- inues o be alida ed, e en wi h high- esolu ion me hods [11–13]. In he ma ine communi y, he e a e deba es abou he ela i e impo ance o molecula ly s able DOM [14] and en i onmen al con- s ain s [15–17] o unde s and why ma ine DOM has an a e age adioca bon age o 6000 yea s [18]. In ma ine sedimen s, he wide a iabili y in OM p ese a ion e ficiencies con inues o be Highligh s Concep ualiza ions o o ganic ma e (OM) deg ada ion a e di e ging wi hin and among he soil, eshwa e , and ma ine subdisciplines. The e is a need o s ep back om con- cep ualiza ions specific o ce ain eco- sys ems o holis ically unde s and wha con ols he deg ada ion o OM, ega d- less o he ecosys em. Ecosys em p ope ies a e changing ap- idly due o an h opogenic s esso s, and we encou age he de elopmen o con- cep ual and ea h sys em models o be mo e flexible and conside condi ions ha may cu en ly seem a ypical o a pa icula ecosys em. We emphasize ha he impo ance o molecula composi ion in egula ing OM deg ada ion should no be disca ded, bu a he we should unde s and when and why i appea s mo e o less ele an han ecosys em p ope ies. 1 Limnology/Depa men o Ecology and Gene ics, Uppsala Uni e si y, No by ägen18D, Uppsala 75236, Sweden 2 Na ional High Magne ic Field Labo a o y Geochemis y G oup and Depa men o Ea h, Ocean, and A mosphe ic Science, Flo ida S a e Uni e si y, Tallahassee, FL 32306, USA 3 Ca alan Ins i u e o Wa e Resea ch (ICRA), H2O building, Emili G ahi 101, 17003 Gi ona, Spain 4 Cu en a ilia ions: Uni ed S a es Geological Su ey (USGS), Wa e Mission A ea, 3215 Ma ine S ee , Rm E-127, Boulde , CO 80303, USA; and Labo a oi e des Sciences du Clima e de l’En i onnemen , LSCE, CEA, CNRS, UVSQ, 91191 Gi -Su -Y e e, F ance T ends in Ecology & E olu ion, Feb ua y 2021, Vol. 36, No. 2 h ps://doi.o g/10.1016/j. ee.2020.10.006 113 © 2020 The Au ho s. Published by Else ie L d. This is an open access a icle unde he CC BY license (h p://c ea i ecommons.o g/licenses/by/4.0/). T ends in Ecology & E olu ion OPEN ACCESS Box 1. Di e ging His o ical Concep ualiza ions o OM Deg ada ion in Soil, F eshwa e , and Ma ine Ecosys ems o e Time The Soil Pe spec i e: OM Pe sis ence as a Func ion o Ecosys em P ope ies His o ically, he molecula composi ion o OM (e.g., lignin:N, a oma ic con en , solubili y, molecula weigh , o p esence o humic subs ances) was hough o make soil OM pe sis [65–68]. Acco dingly, soil OM models assumed ha decay a es we e linked o molecula s uc u e [6,69]. These adi ionally held iewpoin s a e now ac i ely challenged [4,8,70], as e iewed elsewhe e [8](Figu e I). The long- e m p ese a ion o soil OM is inc easingly ecognized o be a unc ion o i s ecosys em p ope ies, such as empe a u e [71,72], biological inaccessibili y due o adso p ion o mine al su aces [8,60,73–75], and nu ien limi a ion [76], while molecula composi ion is conside ed less ele an [70]. A defining ea u e o soils is he p esence o mine al su aces, whe eby adso p ion and physical encapsula ion wi hin mic opo e spaces can e ain OM [75,77,78], and p o ec i om biological deg ada ion [3,4,70,79]. When s abilized, e en ‘labile’OM becomes p o ec ed om enzyma ic deg ada ion. In ac , soil OM wi h a adioca bon age o se e al housand yea s can be highly labile once des abilized back in o solu ion phase [48], and likewise o hawing pe ma os leacha es [80,81]. Acco dingly, i has been no ed ha adioca bon age pe se is no a good p edic o o long- e m pe sis ence o soil OM [50]. The F eshwa e Pe spec i e: DOM Pe sis ence due o Molecula Composi ion and Time In s a k con as o soil ecosys ems, eme ging eshwa e s udies a e ein o cing he long-held no ion ha molecula composi ion is ele an o DOM deg ada ion. DOM in eshwa e s ends o be highly eac i e (mean hal -li e o 2.5 yea s) wi h wa e esidence ime eme ging as a s ong p edic o o deg ada ion a es ac oss a ange o wa e bodies [39]. Thus, he e is an inhe en assump ion ha labile compounds a e selec i ely consumed fi s , and he emaining OM is inc easingly di ficul o deg ade [11,82–86]. Recen high- esolu ion app oaches ha e confi med ha , wi h ime, he e is a p e e en ial loss o oxidized, a oma ic compounds o high molecula weigh , whe eas educed alipha ic and N-con aining compounds a e esis an o deg ada ion o igh ly cycled, and, hus, pe sis en [13]. As wi h soil s udies, adioca bon age is a poo p edic o o long- e m pe sis ence in lakes and s eams [87,88]. Eme ging Ma ine Pe spec i e: Molecula Composi ion and Ecosys em Dependence Two main lines o hinking, which a e no exclusi e o each o he , p e ail in he ma ine communi y, as e iewed elsewhe e [42]. The o ma ion o e ac o y DOM du ing mic obial and ph oly ic deg ada ion, as well as he mogenesis, is suppo ed by de ailed molecula me hods [14,89] (Ma ine 1; Figu e I). Ano he heo y sugges s ha en i onmen al condi ions a e ele an [17,42]. One example is he dilu ion heo y [15,16], which sugges s ha highly dilu e condi ions o he deep sea (DOC: <42 μM C) es ic decomposi ion, a he han composi ional ecalci ance [16] (Ma ine 2; Figu e I). In ac , he ee ene gy eleased om highly labile subs a es such as glucose can be ou weighed by he high ene ge ic cos s o concen- a ing he subs a e wi hin he cell be o e me abolism [90]. Gi en ha DOM comp ises se e al housand indi idual compounds [91], he low abundance o any indi idual compound unde highly dilu e condi ions may limi specific ca abolic pa hways. Consequen ly, some DOM compounds may pe sis almos indefini ely in he deep ocean. Recen publica ions Olde publica ions Molecula composi ion domina es Ecosys em p ope ies domina e F eshwa e Ma ine 1 Soil [39] Ca alan e al. 2016 [13] Kelle man e al. 2015 [12] Sun e al. 1997 [10] Vanno e e al. 1980 [70] Lehmann and Klebe 2015 [8] Schmid e al. 2011 [16] A ie a e al. 2015 [15] Jannasch 1967 [89] Lech en eld e al. 2014 [91] He ko n e al. 2006 [1] Hedges e al. 2000 [93] Amon and Benne 1996 [94 ]Sollins e al. 1996 Ma ine 2 [11] Moshe e al. 2015 [17] Sex on e al. 2011 [95 ]K i zbe g e al. 2010 T ends T ends in in Ecology Ecology & E olu ion E olu ion Figu e I. Illus a ion o he P og ession in ou Concep ualiza ion o Dominan Con ols on O ganic Ma e (OM) Deg ada ion ac oss F eshwa e , Soil, and Ma ine Li e a u e. A ows unning nea he le sugges ha molecula composi ion is highly ele an o OM deg ada ion, while a ows unning o he igh sugges ha en i onmen al and biological con ols domina e. The eshwa e and soil li e a u e di e ge mos p ominen ly, while he e a e wo p e ailing heo ies in he ma ine li e a u e. This shows a lack o consensus ac oss, and wi hin, ecosys ems. Examples o li e a u e d i ing he di ec ion o a ows a e indica ed [1,8,10–13,15–17,39,70,89,91,93–95]. *Co espondence: dolly.ko [email protected] (D.N. Ko hawala). T ends in Ecology & E olu ion 114 T ends in Ecology & E olu ion, Feb ua y 2021, Vol. 36, No. 2 OPEN ACCESS deba ed, wi h con olling ac o s anging om molecula composi ion and age, o en i onmen al ac o s, such as associa ion wi h mine al pa icles [19–23]. Thus, ou undamen al concep ualiza- ion o con ols on OM deg ada ion is agmen ed and de eloping in an inconsis en manne wi hin and ac oss ecosys ems. Why Me ge Insigh s ac oss Ecosys ems? Soil, eshwa e , and ma ine ecosys ems a e significan ly di e en en i onmen s. Acco dingly, esea che s in hese a eas ha e de eloped specialized e minology and expe imen al app oaches, and ha e unique his o ies ha gene a ed cu en heo ies. Each ecosys em is unique in e ms o i s OM sou ces, decompose communi ies, ood-web dynamics, and ecosys em p ope ies. These di e ences inhe en ly imply ha he ela i e impo ance o he ac o s egula ing OM deg ada ion may ‘appea ’ o be unique o each ecosys em. Ye , ega dless o he ecosys em, OM is undamen ally a mix u e o deg ada ion by-p oduc s o igina ing om li ing biomass. In ac , aside om ex eme cases [24], OM o igina es om cellula ma e ial, including biomolecules (e.g., ca bohyd a es, p o eins, lipids, nucleic acids, and lignin). Acco dingly, we a gue he e ha he unde lying con ols o OM deg ada ion should be uni e sal, wi h he ela i e impo - ance o indi idual con ols enhanced o masked depending on local ecosys em p ope ies. We highligh he need o ques ion why OM deg ada ion ‘appea s’ o be egula ed di e en ly ac oss ecosys ems. Cu en ly, he li e a u e is ull o s imula ing discussions, and we ha e a unique oppo uni y o consolida e newly acqui ed knowledge de i ed om one ecosys em o help explain ou lie s in adjacen ecosys ems. Ul ima ely, we encou age mo ing owa ds he de elopmen o a mo e holis ic pe spec i e o OM deg ada ion ha is ecosys em independen , and c i ical o he de elopmen o ea h sys em models, pa icula ly hose linking e es ial and aqua ic ecosys ems [25]. Ideally, ou concep ual unde s anding o OM deg ada ion should be obus and flexible o changing en i onmen al condi ions. Gi en ha ecosys em p ope ies a e highly dynamic, he a e o DOM has impo an implica ions o ecosys em unc ioning, including whe he i is a ca bon sou ce o sink. Fac o s such as pe ma os hawing, inc eased p ima y p oduc ion, shi ing decompose communi y composi ion, inc eased suspended sedimen loads o aqua ic sys ems [26,27], eu ophica ion, ocean acidifica ion [28], land-use change [29,30], and ex eme clima e e en s [31,32] can shi OM deg ada ion a es. These changes can a e se ecosys em bounda ies. Thus, a majo challenge o ea h sys em models is he inco po a ion o eliable es ima es o OM deg ada ion a es ac oss all ecosys ems, pa icula ly unde changing en i onmen al condi ions. E en small shi s in biosphe e OM deg ada ion a es can ha e la ge, nonlinea consequences on he elease o he g eenhouse gases, CO 2 and CH 4 [33]. Thus, he e we aise awa eness o di e ging concep ualiza ions o OM deg ada ion ac oss ecosys ems, iden i y ea u es o eme ging heo ies ha could be con ibu ing o di e ging iewpoin s, and encou age u u e esea ch e o s o de elop amo eunified unde s anding o OM deg ada ion. The Complex In e play be ween Molecula Composi ion and Ecosys em P ope ies In ecen yea s, he e has been a shi away om ecognizing he ole o OM composi ion in he soil sciences, wi h emphasis on disc edi ing he humifica ion heo y [9]. In ac , ‘omi ing any emphasis on subs a e composi ion’has been sugges ed [9]. By con as , he p e ailing eshwa- e pe spec i e and one ma ine pe spec i e is ha molecula composi ion emains highly ele an o p edic ing OM deg ada ion [13,14]. Howe e , i is impo an o ecognize ha ecosys em p ope ies and molecula composi ion a e no mu ually exclusi e and, a a gi en place and ime, one may ‘appea ’mo e ele an han ano he . Chemical he modynamics s a es ha , unde s anda d condi ions, he molecula composi ion o a compound de e mines i mine aliza ion is Glossa y Ac i a ion ene gy (E a ): minimum amoun o ene gy equi ed o a eac ion o p oceed. Ea is an ene gy ba ie ha a eac ion mus o e come while ansi ioning om ini ial o final he modynamic s a e, and egula es he a e o he eac ion. Chemical he modynamics: field o s udy dedica ed o defining he c i e ia esponsible o p edic ing whe he a eac ion is easible and will p oceed spon aneously. The a o abili y o a eac ion o p oceed he modynamically does no de e mine i s a e. Dilu ion heo y: also known as he molecula di e si y hypo hesis; sugges s ha low concen a ions, pa icula ly o a complex mix u e, can limi he abili y o an o ganism o me abolize a subs a e. Ecosys em p ope ies: include he ex insic en i onmen al condi ions ha could cons ain he deg ada ion o OM, including abio ic and bio ic ac o s. Examples include empe a u e, pH, oxygen a ailabili y, nu ien s, adso p ion o mine als, and decompose communi y composi ion. Func ional g oups: specificg oupso a oms o bonds wi hin an o ganic compound ha a e esponsible o i s unc ional cha ac e is ics, including how i is likely o eac . Ca boxyl and phenolic g oups a e examples. Gibbs ee ene gy (ΔG): calcula ed as hedi e encein eeene gybe ween eac an s and p oduc s o a gi en eac ion a cons an empe a u e and p essu e. When ΔGis nega i e, he eac ion is spon aneous; when ΔG eaches ze o, he eac ion has eached chemical equilib ium. Humi ica ion: b eakdown and ans o ma ion o o ganic ma e ial in o a complex mix u e o humus ha is molecula ly dis inc om s a ing ma e ials and can pe sis in soils. Isome ic di e si y: ange o s uc u al con o ma ions o molecules wi h he same molecula o mula and mass; he numbe o possible isome ic s uc u es inc eases wi h molecula mass and could eflec unc ional di e si y in e ms o suscep ibili y o deg ada ion. Mine alized/mine aliza ion: desc ibes a se o biological and physicochemical eac ions deg ading OM in o ino ganic compounds (e.g., ca bon dioxide and me hane). Molecula composi ion: gene ic e m desc ibing he chemical composi ion o OM, including i s chemodi e si y, op ical T ends in Ecology & E olu ion OPEN ACCESS T ends in Ecology & E olu ion, Feb ua y 2021, Vol. 36, No. 2 115 a o able. The he modynamic a o abili y, o Gibbs ee ene gy (ΔG), conside s bond ene gies, he chemical s uc u e o OM, unc ional g oups, and he oxida ion s a e o ca bon [34]. When ΔGis nega i e, he eac ion p oceeds spon aneously, and ene gy is eleased. Howe e , in he case o DOM, ΔGis gene ally posi i e [34] and, hus, equi es coupling o he educ ion o a e minal elec on accep o o o ma ion o a educed ca bon p oduc [34]. Howe e , he mody- namics is no conce ned wi h he me abolic pa hway o kine ics o he eac ion. Ra he , a kine ic ene gy ba ie , called he ac i a ion ene gy (E a ), mus be o e come, and his egula es he a e o eac ions. Thus, en i onmen al condi ions a e ele an in cons aining he a e o he eac ion and e en ‘labile’OM can pe sis in ce ain scena ios. I condi ions change, such ha a specific enzyme becomes a ailable o example, he ac i a ion ene gy is educed allowing he eac ion o p oceed apidly. Thus, in some scena ios, a eac ion can be he modynamically a o able bu kine ically un a o able. In such a scena io, e en labile DOM can pe sis . Al e na i ely, a non he modynamically a o able eac ion can occu i he ac i a ion ene gy is low. Since OM is a complex mix u e o compounds, each wi h i s own ΔGand E a ,i isuse ul o conside a con inuum o he modynamic and kine ic ac o s a play. Acco dingly, he ne e ec migh be ha one ‘appea s’mo e ele an han ano he a a pa icula place and ime. Fo ins ance, i compounds a e app oaching he modynamic con ol a he han kine ic con ol, he o e all eac ion a e o he OM pool migh eflec ene gy di e ences be ween eac an s and p oduc s. In his case, he molecula composi ion can appea o be impo an o he ‘appa en ’ a e o he bulk DOM eac ion [scena io (i) in Figu e 1]. In his case, mo e bioa ailable and pho olabile compounds migh be deg aded fi s , ollowed by less biolabile and pho olabile compounds. When en i onmen al cons ain s a e imposed (such as limi a ions o he decom- pose communi y), he ac i a ion ene gy inc eases, esul ing in slowe kine ics [scena io (ii) in Figu e 1]. Wi hin any pool o OM, he e can be compe ing ac o s a play. Fo ins ance, he quali y o plan li e in soils could con ibu e o OM pools spanning ac oss a ange o kine ics om scena io (i) o (iii) depending on how hey o m associa ions wi h mine al su aces [35]. Tempe a- u e is ele an , being imbedded in he calcula ion o ΔG, and being ele an o kine ics, such as o enzyma ic deg ada ion [36,37]. Likewise, a ange o o he en i onmen al ac o s, such as pH, nu ien s, oxygen a ailabili y, soil mois u e, he absence o p esence o sola adia ion, and adso p ion o mine al su aces, can shi kine ics. Unde highly cons ained en i onmen al cha ac e is ics, elemen al composi ion, unc ional g oup composi ion, o isome ic di e si y. Ne deg ada ion a es Mass o OM pool (iii) En i onmen al cons ains high (ii) In e media e (i) Molecula composi ion can appea ele an OM eac i eOM pe sis s HighLow OM a ailabili y T ends T ends in in Ecology Ecology & E olu ion E olu ion Figu e 1. Ti le. Scena ios depic ing pools o o ganic ca bon ac oss a con inuum o deg ada ion a es and masses unde condi ions ha a e (i) no cons ained by en i onmen al and biological condi ions and highly eac i e, esul ing in a small mass o s o ed ca bon, such as in eshwa e s; (ii) an in e media e scena io whe eby some en i onmen al cons ain s slow deg ada ion a es (such as nu ien limi a ion), esul ing in a la ge pool size; and (iii) a scena io unde highly cons ained condi ions whe eby o ganic ma e (OM) is di ficul o access and deg ada ion a es a e educed subs an ially esul ing in an accumula ion o o ganic ma e (e.g., sedimen , mine al soils, o pea lands). The e e sible a ows indica e ha condi ions a e fluid ac oss a con inuum o scena ios and dynamic, wi h he po en ial o be in e changeable. Fo ins ance, i en i onmen al and biological cons ain s placed on scena ios (ii) and (iii) a e li ed, deg ada ion a es can shi o scena io (i) whe eby he decompose communi y is op imized, and deg ada ion a es inc ease. In his concep ualiza ion, we assume ha p oduc ion a es a e cons an ac oss he con inuum. T ends in Ecology & E olu ion 116 T ends in Ecology & E olu ion, Feb ua y 2021, Vol. 36, No. 2 OPEN ACCESS condi ions, such as anoxia in pea lands o pe ma os soils, he kine ic ba ie inc eases such ha OM can pe sis o cen u ies o millennia [scena io (iii) in Figu e 1]. This scena io is pa icula ly ele an whe e OM is physically sepa a ed o p o ec ed om enzyma ic o sola pho ochemical deg ada ion. As deg ada ion a es slow, he size o he OM pool can be expec ed o inc ease [ om (i) o (iii) in Figu e 1]. A key poin o ecognize is ha a single pool o OM can encoun e a ull ange o cons ain s as i a els wi hin o be ween ecosys ems, and in e changeably appea o be egula ed mo e, o less, by molecula composi ion [38] ela i e o ecosys em p ope ies. Fo ins ance, OM ozen in pe ma os [scena io (iii)] can become biological a ailable upon hawing [scena io (i)], and once again be bu ied in lake sedimen upon o ming associa ions wi h pa icu- la e ma e ial [scena io (iii); Figu e 1]. Likewise, DOM in a soil en i onmen can be adso bed o mine als and p o ec ed om decomposi ion [scena io (iii)] and subsequen ly deso bed back in o solu ion phase [scena io (i), Figu e 1]. The Rela ionship be ween Residence Time and Reac i i y In cases whe e he molecula composi ion is ound o egula e deg ada ion, as in eshwa e ecosys ems [13,39,40], he e is a s ong nega i e ela ionship be ween deg ada ion a e and wa e esidence ime [39](Figu e 2A). This nega i e ela ionship be ween esidence ime and deg- ada ion a es has also been obse ed o ma ine sedimen s [20,41], bu a a longe imescales. DOM in eshwa e ecosys ems is la gely accessible o biological and pho odeg ada ion, lying concep ually nea scena io (i) in Figu e 1. Howe e , e en in eshwa e s, a sizable ac ion o he ela ionship be ween wa e esidence ime and deg ada ion a es emains unexplained (59%) [39], sugges ing ha en i onmen al cons ain s con ibu e o he unexplained a iabili y in Log esidence ime Log OM eac i i y (k) Log esidence ime Log OM eac i i y (k) (A) Residence ime− eac i i y ela ionship (B) Dis up ed esidence ime− eac i i y ela ionship Pe iod o inaccessibili y T ends T ends in in Ecology Ecology & E olu ion E olu ion Figu e 2. Ti le. Illus a ion o (A) he ela ionship be ween esidence ime and eac i i y (e.g., [39]) wi h minimal en i onmen al cons ain s, and (B) unde a scena io whe e deg ada ion is dis up ed by ecosys em p ope ies making o ganic ma e (OM) inaccessible o biological deg ada ion o an ex ended pe iod o ime (pink-b oken line), be o e becoming accessible once again. Using adioca bon da ing o p edic eac i i y in scena io (B) would likely mis ep esen he amoun o ime OM was exposed o deg ada ion p ocesses. T ends in Ecology & E olu ion OPEN ACCESS T ends in Ecology & E olu ion, Feb ua y 2021, Vol. 36, No. 2 117 deg ada ion a es. In a complimen a y s udy o whole-lake dissol ed o ganic ca bon (DOC) budge s, eac i i y was s ongly ela ed o wa e esidence ime, and nu ien s a us helped p edic whe he he lake was a ca bon sou ce o sink [40]. In he ocean, ini ial deg ada ion a es o algal DOM a e also apid, wi h esidence ime likewise being in e sely linked o eac i i y [42]. In ac , mos ma ine DOM is gene a ed om p ima y p oduc i i y and consumed wi hin weeks, wi h only ≈1% le a e a decade [42]. Only an ex emely small ac ion o he DOM pool accumula es, ul ima ely pe sis ing o millennia. In soil en i onmen s, when he oppo uni y o be physically p o ec ed by mine al su aces is emo ed, soil DOM can ha e simila deg ada ion a es (<0.001–0.30 d –1 )[43] o i s eshwa e coun e pa (0.001–0.50 d –1 )[44,45]. In iguingly, he molecula composi ion o soil DOM can help p edic how i unc ions; o example, whe he associa ions wi h mine al su aces will o m [43,46,47]. (See Figu e 3.) When Time S ands S ill: Disconnec ing Ch onological Time ( 14 C) om Reac i i y In he subsoil en i onmen , soil OM wi h a adioca bon age o se e al housand yea s has been ound o be highly ‘labile’once des abilized in o solu ion [48], as o shale deg ada ion [49]. This obse a ion was p e iously in e p e ed as a eason o disc edi he conside a ion ha esidence ime is a p edic o o OM eac i i y [50]. In consolida ing his appa en disc epancy, i appea s ha ,when OM is biologically inaccessible, he ela ionship be ween esidence ime and eac i i y may be dis up ed (Figu e 2B). When he oppo uni y o enzyma ic deg ada ion, o example, is elimina ed, he ime– eac i i y clock is e ec i ely paused. Ye , ch onological ime con inues, and he ime– eac i i y ela ionship esumes once he cons ain is li ed. The ch onological age o OM is eflec ed by adioca bon da ing o 14 C; a p oxy o he ime since he o iginal plan issue fixed a mosphe ic CO 2 , and con inues i espec i e o deg ada ion. Thus, when OM is biologically inaccessible due o physical and chemical ac o s, such as occlu- sion in mine al po e spaces [51], ex emely dilu e condi ions [15,16], being ozen in pe ma os [52], o anoxia [53], he adioca bon da e is i ele an o i s eac i i y o deg ada ion a e. Recogni ion ha a esidence ime– eac i i y clock ope a es independen ly om ch onological ime is a ele an conside a ion and helps consolida e he ac ha OM can pe sis in some en i- onmen s o a long ime. We emphasize ha i would be e oneous o dis ega d he ole o ime in p edic ing OM eac i i y and s ess ha e o should a he be placed on unde s anding when and why he esidence ime– eac i i y clock is paused. No bac e ial g ow h Bac e ial g ow h Dissol ed o ganic ca bon concen a ion (µM C) Molecula eac i i y 0 50 100 150 200 Labile OM Re ac o y OM (2) (1) Deep sea h eshold concen a ion T ends T ends in in Ecology Ecology & E olu ion E olu ion Figu e 3. Ti le. The cu en concep ualiza ion o he dilu ion heo y (a ow 1) sugges s ha bac e ial g ow h is limi ed below a h eshold dissol ed o ganic ca bon concen a ion ound in he deep sea (black-b oken line, 42 uM C [16]), e en i he molecula composi ion o he o ganic ma e (OM) emains labile wi h p og essi e decomposi ion. An al e na i e heo y in eshwa e is ha he molecula composi ion o OM becomes p og essi ely less eac i e wi h deg ada ion (a ow 2). I emains unclea whe he he dilu ion heo y holds in ecosys ems ou side o he deep sea and whe e he minimum h eshold concen a ion o e ec i e bac e ial g ow h lies o eshwa e ecosys ems, such as glacial mel wa e s. Figu e inspi ed by [92]. T ends in Ecology & E olu ion 118 T ends in Ecology & E olu ion, Feb ua y 2021, Vol. 36, No. 2 OPEN ACCESS The Need o Tes Theo ies ac oss Ecosys em Bounda ies As ou unde s anding o OM deg ada ion dynamics p og esses, i becomes clea ha some concep s a e inconsis en ac oss subdisciplines, and ca e ully designed c oss-ecosys em s udies a e equi ed. Ideally, eme ging s udies would include OM om se e al sou ces and di e - en en i onmen al condi ions and challenge p e iously p oposed hypo heses by es ing hei s eng hs and limi a ions ac oss ecosys em bounda ies. The dilu ion heo y is one example o a heo y ha could benefi om being es ed ou side he ma ine en i onmen . The dilu ion hypo hesis exp esses ha , a concen a ions ypical o he deep sea (42 μMC),DOM compounds a e in ac labile, bu indi idual subs a e concen a ions lie below he h eshold o ene ge ic equi emen s o p oka yo es [16]. When DOM om he deep sea was concen a ed up o a ac o o 10, A ie a e al. [35] ound enhanced bac e ial ac i i y. This led o he conclusion ha concen a ion was limi ing bac e ial u iliza ion o DOM. Howe e , compa able eshwa e s udies ound ha , unde inc easingly dilu e condi ions (down o 40 μM C), p oka yo ic ac i i y was no cons ained [38,54]. In addi ion, highly eac i e OM has been ound in glacial mel wa e s a concen a ions simila o, o a lowe han, he deep sea (11 μMC)[55,56]. Thus, i is possible ha he h eshold concen a ion ele an o he deep sea (42 μM C) does no apply ac oss ecosys ems and is composi ion dependen . One c oss-ecosys em s udy es ing he dilu ion heo y ound ha DOM om deep ocean wa e s exhibi s highe isome ic di e si y han lake DOM. High isome ic di e si y g ea ly educes he concen a ion o any indi idual compound down o picomola concen a ions, and he eby a p io i suppo s he dilu ion heo y. Howe e , he eshwa e lake sample had a highe numbe o chemical o mulas assigned, sugges ing a wide di e si y o molecules bu a highe concen a- ions indi idually [57,58]. Thus, i is i al o ecognize ha he h eshold concen a ion may be a iable in o he en i onmen s. Whe he hisisdue oDOMcomposi iono en i onmen al cons ain s needs o be answe ed h ough a common e o . Likewise, i would be use ul o de elop a be e unde s anding o how e ec i e a pa icula enzyme, o se o enzymes, migh be a clea ing he nume ous s uc u al isome s ep esen ed wi hin a single mass o cha ge peak in a mass spec um. Concluding Rema ks Resol ing disc epancies, such as when and why a pa icula heo y holds, is ele an o p edic ing u u e global ca bon fluxes. This is pa icula ly impo an whe e sligh shi s in he eac i i y o la ge pools o o ganic ca bon can ha e subs an ial consequences o he flux o he g eenhouse gases, CO 2 and CH 4 . Fo ins ance, i is use ul o unde s and when and why esidence ime and eac i i y a e connec ed and disconnec ed, and when and why molecula composi ion appea s o be mo e ele an han ecosys em p ope ies. C oss-ecosys em s udies a e c i ical o making he nex se o impo an b eak h oughs in ou unde s anding o con ols on OM deg ada ion [39,59,60]. C oss-ecosys em s udies may be expe imen ally challenging a imes [61], ye , ha e p e iously p o en o gi ea deepe and mo e holis ic unde s anding in he fields o nu ien limi a ion and unde s anding o he ole o ex acellula enzymes [62,63]. E en ually, hese mo e obus heo ies can be inco po a ed in o ea h sys em models. Al hough we acknowledge ha ecosys em p ope ies a y widely ac oss ecosys ems, we a gue ha undamen al con ols on OM deg ada ion a e uni e sal. This boundless p ope y [64], whe eby ca bon exchanges ac oss ecosys ems indisc imina ely, o ces us o mo e igo ously challenge he alidi y o concep s ha ha e di e ged in adjacen ecosys ems. To uni y his agmen ed unde s anding, we should ecognize ha he impo ance o molecula composi ion can appea o be ‘masked’, because ecosys em p ope ies a e possibly limi ing deg ada ion, and al e na i ely, whe e molecula composi ion appea s o egula e deg ada ion, ecosys em Ou s anding Ques ions Fo hose wo king, o planning on wo king, ac oss ecosys em bounda ies, how do we o e come he se o expe imen al challenges o wo king wi h OM ac oss highly con as ing ecosys- ems,suchasma ixe ec s? Wha is he bes measu e o OM eac- i i y ha can be used in a consis en manne ac oss soil, eshwa e , and ma ine subdisciplines? Since he eac- i i y o OM is highly a iable and can be measu ed as adioca bon age, u no e ime, hal -li es, and deg ada ion a es, a common desc ip o o OM eac i i y is equi ed o e ec i ely compa e eac i - i y ac oss ecosys ems. How can we accu a ely measu e he dis ibu ion o OM deg ada ion a es ac oss soil, eshwa e , and ma ine ecosys ems? How do ea h sys em models accommoda eaflexibleconcep ualiza ion o con olsonOMdeg ada ionandinclude a uni e sal measu e o molecula composi ion? How can we mos e ec i ely educe he unce ain y associa ed wi h deg ada ion a es en e ed in o ea h sys em models? In mo ing o wa d, how do we bes design empi ical s udies ocused on uni ying con ols on OM eac i i y ha c oss ecosys em bounda ies? T ends in Ecology & E olu ion OPEN ACCESS T ends in Ecology & E olu ion, Feb ua y 2021, Vol. 36, No. 2 119 p ope ies could be p o iding he oppo uni y o mo e apid deg ada ion. Since he same pool o OM migh be exposed o ei he ex eme ( as o slow kine ics), ou concep ualiza ion should be flexible enough o include bo h molecula composi ion and ecosys em p ope ies as ele an con ols, and swi ch back and o h as needed. Ra he han complacen ly accep ing di e ging ‘ecosys em specific’ iewpoin s, we sugges chal- lenging and es ing exis ing heo ies ac oss ecosys em bounda ies o e eal mo e uni e sal mechanisms behind OM p ese a ion and eac i i y. We gi e se e al examples o ques ions ha emain o be add essed (see Ou s anding Ques ions). To add ess hese, we encou age collabo a i e esea ch om scien is s ac oss ecosys ems, in e ac ions be ween expe imen alis s and modele s and e o s o use mo e accessible e minology. Ul ima ely, we p opose s i ing owa ds a concep ualiza ion o OM deg ada ion ha is independen o he ecosys em and flexible o changing ecosys em p ope ies. Acknowledgmen s This wo k was suppo ed by unding om he Swedish Na ional Science Founda ion (D.N.K., s a ing g an 2016-04108; and L.J.T., 2018-04524), he Swedish Resea ch Council o En i onmen , Ag icul u al Sciences, and Spa ial Planning (D.N.K., 201800778; and L.J.T., 219-2009-1692), he Knu and Alice Wallenbe g Founda ion (L.J.T., KAW 2013.0091), an ea ly ca ee esea ch awa d om King Ca l XVI Gus a o Sweden (D.N.K.), a Ma ie Sklodowska Cu ie Ac ion o he Eu opean Commission o N.C. (CHROME - 839709) and a g an o A.M.K. om he Malméns Founda ion. Re e ences 1. Hedges, J.I. e al. (2000) The molecula ly-uncha ac e ized componen o nonli ing o ganic ma e in na u al en i onmen s. O g. Geochem. 31, 945–958 2. Mo an, M.A. e al. (2016) Deciphe ing ocean ca bon in a changing wo ld. P oc. Na l. Acad. Sci. U. S. A. 113, 3143–3151 3. Klebe , M. e al. (2011) Old and s able soil o ganic ma e is no necessa ily chemically ecalci an : implica ions o modeling concep s and empe a u e sensi i i y. Glob. Change Biol. 17, 1097–1107 4. on Lu zow, M. e al. (2006) S abiliza ion o o ganic ma e in empe a e soils: mechanisms and hei ele ance unde di e en soil condi ions - a e iew. Eu . J. Soil Sci. 57, 426–445 5. Hayes, M.H.B. and Swi , R.S. (2020) Vindica ion o humic subs ances as a key componen o o ganic ma e in soil and wa e . Ad . Ag on. 1–37 6. Pa on, W.J. e al. (1987) Analysis o ac o s con olling soil o ganic ma e le els in g ea plains g asslands. Soil Sci. Soc. Am. J. 1173–1179 7. Melillo, J.M. e al. (1982) Ni ogen and lignin con ol o ha dwood lea li e decomposi ion dynamics. Ecology 63, 621–626 8. Schmid , M.W.I. e al. (2011) Pe sis ence o soil o ganic ma e as an ecosys em p ope y. Na u e 478, 49–56 9. Klebe , M. (2010) Wha is ecalci an soil o ganic ma e ? En i on. Chem. 7, 320–332 10. Vanno e, R.L. e al. (1980) Ri e con inuum concep . Can. J. Fish. Aqua . Sci. 37, 130–137 11. Moshe , J.J. e al. (2015) Longi udinal shi s in dissol ed o ganic ma e chemogeog aphy and chemodi e si y wi hin headwa e s eams: a i e con inuum ep ise. Biogeochemis y 124, 371–385 12. Sun, L. e al. (1997) Use o elemen al composi ion o p edic bioa ailabili y o dissol ed o ganic ma e in a Geo gia i e . Limnol. Oceanog . 42, 714–721 13. Kelle man, A.M. e al. (2015) Pe sis ance o dissol ed o ganic ma e in lakes ela ed o i s molecula cha ac e is ics. Na . Geosci. 8, 454–457 14. He ko n, N. e al. (2006) Cha ac e iza ion o a majo e ac o y componen o ma ine dissol ed o ganic ma e . Geochim. Cosmochim. Ac a 70, 2990–3010 15. Jannasch, H.W. (1967) G ow h o ma ine bac e ia a limi ing con- cen a ions o o ganic ca bon in seawa e . Limnol. Oceanog . 12, 264–272 16. A ie a, J.M. e al. (2015) Dilu ion limi s dissol ed o ganic ca bon u iliza ion in he deep ocean. Science 348, 331–333 17. Sex on, P.F. e al. (2011) Eocene global wa ming e en s d i en by en ila ion o oceanic dissol ed o ganic ca bon. Na u e 471, 349–352 18. Hansell, D.A. e al. (2009) Dissol ed o ganic ma e in he ocean: a con o e sy s imula es new insigh s. Oceanog aphy 22, 202–211 19. Hedges, J.I. and Keil, R.G. (1995) Sedimen a y o ganic ma e p ese a ion - an assessmen and specula i e syn hesis. Ma . Chem. 49, 81–115 20. Middelbu g, J.J. (1989) A simple a e model o o ganic ma e decomposi ion in ma ine sedimen s. Geochim. Cosmochim. Ac a 53, 1577–1581 21. Blai , N.E. and Alle , R.C. (2012) The a e o e es ial o ganic ca bonin hema ineen i onmen .Annu. Re . Ma . Sci. 4, 401–423 22. Bu dige, D.J. (2007) P ese a ion o o ganic ma e in ma ine sedimen s: Con ols, mechanisms, and an imbalance in sedi- men o ganic ca bon budge s? Chem. Re . 107, 467–485 23. A nd , S. e al. (2013) Quan i ying he deg ada ion o o ganic ma e in ma ine sedimen s: A e iew and syn hesis. Ea h-Sci. Re . 123, 53–86 24. McCollom, T.M. (2013) Mille -U ey and beyond: wha ha e we lea ned abou p ebio ic o ganic syn hesis eac ions in he pas 60 yea s? Annu. Re . Ea h Plane . Sci. 41, 207–229 25. Bow ing, S.P.K. e al. (2019) ORCHIDEE MICT-LEAK ( 5459), a global model o he p oduc ion, anspo , and ans o ma ion o dissol ed o ganic ca bon om A c ic pe ma os egions - Pa 1: a ionale, model desc ip ion, and simula ion p o ocol. Geosci. Model De . 12, 3503–3521 26. Walling, D.E. and Fang, D. (2003) Recen ends in he suspended sedimen loads o he wo ld's i e s. Glob. Plane . Chang. 39, 111–126 27. Sy i ski, J.P.M. e al. (2005) Impac o humans on he flux o e es ial sedimen o he global coas al ocean. Science 308, 376–380 28. K oeke , K.J. e al. (2013) Impac s o ocean acidifica ion on ma ine o ganisms: quan i ying sensi i i ies and in e ac ion wi h wa ming. Glob. Change Biol. 19, 1884–1896 29. Pos , W.M. and Kwon, K.C. (2000) Soil ca bon seques a ion and land-use change: p ocesses and po en ial. Glob. Change Biol. 6, 317–327 30. Guo, L.B. and Gi o d, R.M. (2002) Soil ca bon s ocks and land use change: a me a analysis. Glob. Change Biol. 8, 345–360 T ends in Ecology & E olu ion 120 T ends in Ecology & E olu ion, Feb ua y 2021, Vol. 36, No. 2 OPEN ACCESS 31. Ciais, P. e al. (2005) Eu ope-wide educ ion in p ima y p oduc- i i y caused by he hea and d ough in 2003. Na u e 437, 529–533 32. Schimel, D.S. e al. (1994) Clima ic, edaphic, and bio ic con ols o e s o age and u no e o ca bon in soils. Glob. Biogeochem. Cycle 8, 279–293 33. Ciais, P. e al. (2013) Ca bon and o he biogeochemical cycles. In Clima e Change 2013: The Physical Science Basis. Con ibu ion o Wo king G oup I o he Fi h Assessmen Repo o he In e go e nmen al Panel on Clima e Change (S ocke , T.F. e al., eds), pp. 465–570, Camb idge Uni e si y P ess 34. LaRowe, D.E. and Van Cappellen, P. (2011) Deg ada ion o na u al o ganic ma e : a he modynamic analysis. Geochim. Cosmochim. Ac a 75, 2030–2042 35. Co u o, M.F. e al. (2013) The Mic obial E ficiency-Ma ix S abiliza ion (MEMS) amewo k in eg a es plan li e decomposi- ion wi h soil o ganic ma e s abiliza ion: do labile plan inpu s o m s able soil o ganic ma e ? Glob. Change Biol. 19, 988–995 36. A nos i, C. (2004) Speed bumps and ba icades in he ca bon cycle: subs a e s uc u al e ec s on ca bon cycling. Ma . Chem. 92, 263–273 37. Sinsabaugh, R.L. (1994) Enzyma ic analysis o mic obial pa e ns and p ocess. Biol. Fe il. Soils 17, 69–74 38. Koehle , B. e al. (2012) Reac i i y con inuum o dissol ed o ganic ca bon decomposi ion in lake wa e . J. Geophys. Res.-Biogeosci. 117, 1–14 39. Ca alán, N. e al. (2016) O ganic ca bon decomposi ion a es con olled by wa e e en ion ime ac oss inland wa e s. Na . Geosci. 9, 501–504 40. E ans, C. e al. (2017) Va iabili y in o ganic ca bon eac i i y ac oss lake esidence ime and ophic g adien s. Na . Geosci. 10, 832–835 41. Boud eau, B.P. e al. (2008) Commen on "Physical model o he decay and p ese a ion o ma ine o ganic ca bon". Science (New Yo k, N.Y.) 319, 1616 au ho eply 1616 42. Di ma , T. (2015) Reasons behind he long- e m s abili y o dis- sol ed o ganic ma e . In Biogeochemis y o Ma ine Dissol ed O ganic Ma e (Hansell, D.A. and Ca lson, C.A., eds), pp. 369–388, Else ie 43. Kalbi z, K. e al. (2003) Biodeg ada ion o soil-de i ed dissol ed o ganic ma e as ela ed o i s p ope ies. Geode ma 113, 273–291 44. Hanson, P.C. e al. (2011) Fa e o alloch honous dissol ed o ganic ca bon in lakes: a quan i a i e app oach. PLoS ONE 6, 1–12 45. Polla d, P.C. (2013) In si u apid measu es o o al espi a ion a e cap u e he supe labile DOC bac e ial subs a es o eshwa e . Limnol. Oceanog -Me h. 11, 584–593 46. Kaise , K. and Kalbi z, K. (2012) Cycling downwa ds - dissol ed o ganic ma e in soils. Soil Biol. Biochem. 52, 29–32 47. Ko hawala, D.N. e al. (2012) Selec i e adso p ion o dissol ed o ganic ma e o mine al soils. Geode ma 189-190, 334–342 48. Ca aco, N. e al. (2010) Millennial-aged o ganic ca bon subsidies o a mode n i e ood web. Ecology 91, 2385–2393 49. Pe sch, S.T. e al. (2001) C-14-dead li ing biomass: e idence o mic obial assimila ion o ancien o ganic ca bon du ing sha e wea he ing. Science 292, 1127–1131 50. Ma in-Spio a, E. e al. (2014) Pa adigm shi s in soil o ganic ma e esea ch a ec in e p e a ions o aqua ic ca bon cycling: anscending disciplina y and ecosys em bounda ies. Biogeochemis y 117, 279–297 51. Ekschmi , K. e al. (2008) Soil-ca bon p ese a ion h ough habi a cons ain s and biological limi a ions on decompose ac i i y. J. Plan Nu . Soil Sci. 171, 27–35 52. Schuu , E.A.G. e al. (2015) Clima e change and he pe ma os ca bon eedback. Na u e 520, 171–179 53. Bas iken, D. e al. (2004) Deg ada ion o dissol ed o ganic ma e in oxic and anoxic lake wa e . Limnol. Oceanog . 49, 109–116 54. Eile , A. e al. (2003) He e o ophic bac e ial g ow h e ficiency and communi y s uc u e a di e en na u al o ganic ca bon concen a ions. Appl. En i on. Mic obiol. 69, 3701–3709 55. Hood, E. e al. (2009) Glacie s as a sou ce o ancien and labile o ganic ma e o he ma ine en i onmen . Na u e 462, 1044–1047 56. Singe , G.A. e al. (2012) Biogeochemically di e se o ganic ma e in Alpine glacie s and i s downs eam a e. Na . Geosci. 5, 710–714 57. Za k, M. e al. (2017) Molecula p ope ies o deep-sea dissol ed o ganic ma e a e p edic able by he cen al limi heo em: e idence om andem FT-ICR-MS. Ma . Chem. 191, 9–15 58. Za k, M. and Di ma , T. (2018) Uni e sal molecula s uc u es in na u al dissol ed o ganic ma e . Na . Commun. 9, 1–8 59. Hedges, J.I. e al. (1997) Wha happens o e es ial o ganic ma e in he ocean? O g. Geochem. 27, 195–212 60. Hemingway, J.D. e al. (2019) Mine al p o ec ion egula es long- e m global p ese a ion o na u al o ganic ca bon. Na u e 570, 228–231 61. Hedges, J.I. and Oades, J.M. (1997) Compa a i e o ganic geo- chemis ies o soils and ma ine sedimen s. O g. Geochem. 27, 319–361 62. A nos i, C. e al. (2014) Ex acellula enzymes in e es ial, esh- wa e , and ma ine en i onmen s: pe spec i es on sys em a i- abili y and common esea ch needs. Biogeochemis y 117, 5–21 63. Else , J.J. e al. (2007) Global analysis o ni ogen and phospho- us limi a ion o p ima y p oduce s in eshwa e , ma ine and e es ial ecosys ems. Ecol. Le . 10, 1135–1142 64. Ba in, T.J. e al. (2009) The boundless ca bon cycle. Na . Geosci. 2, 598–600 65. Abe , J.D. e al. (1990) P edic ing long- e m pa e ns o mass- loss, ni ogen dynamics, and soil o ganic ma e o ma ion om ini ial fine li e chemis y in empe a e o es ecosys ems. Can. J. Bo .-Re . Can. Bo . 68, 2201–2208 66. Alexande , M. (1981) Biodeg ada ion o chemicals o en i on- men al conce n. Science 211, 132–138 67. Schul en, H.R. and Schni ze , M. (1997) Chemical model s uc- u es o soil o ganic ma e and soils. Soil Sci. 162, 115–130 68. Piccolo, A. (2001) The sup amolecula s uc u e o humic subs ances. Soil Sci. 166, 810–832 69. Jenkinson, D.S. (1990) The u no e o o ganic-ca bon and ni ogen in soil. Philos. T ans. R. Soc. B-Biol. Sci. 329, 361–368 70. Lehmann, J. and Klebe , M. (2015) The con en ious na u e o soil o ganic ma e . Na u e 528, 60–68 71. Åg en, G.I. and Bosa a, E. (2002) Reconcilling di e ences in p edic ions o empe a u e esponse o soil o ganic ma e . Soil Biol. Biochem. 34, 129–132 72. Da idson, E. and Janssens, I. (2006) Tempe a u e sensi i i y o soil ca bon decomposi ion and eedbacks o clima e change. Na u e 440, 165–173 73. Kalbi z, K. e al. (2005) S abiliza ion o dissol ed o ganic ma e by so p ion o he mine al soil. Soil Biol. Biochem. 37, 1319–1331 74. Kaise , K. and Guggenbe ge , G. (2000) The ole o DOM so p- ion o mine al su aces in he p ese a ion o o ganic ma e in soils. O g. Geochem. 31, 711–725 75. To n, M.S. e al. (1997) Mine al con ol o soil o ganic ca bon s o age and u no e . Na u e 389, 170–173 76. Vi ousek, P.M. and Fa ing on, H. (1997) Nu ien limi a ion and soil de elopmen : expe imen al es o a biogeochemical heo y. Biogeochemis y 37, 63–75 77. Kaise , K. and Guggenbe ge , G. (2003) Mine al su aces and soil o ganic ma e . Eu . J. Soil Sci. 54, 219–236 78. Ko hawala, D.N. e al. (2009) Soil p ope ies con olling he ad- so p ion o dissol ed o ganic ca bon o mine al soils. Soil Sci. Soc. Am. J. 73, 1831–1842 79. Sollins, P. e al. (2006) O ganic C and N s abiliza ion in a o es soil: e idence om sequen ial densi y ac iona ion. Soil Biol. Biochem. 38, 3313–3324 80. Wang, Y.H. e al. (2018) Selec i e leaching o dissol ed o ganic ma e om alpine pe ma os soils on he Qinghai-Tibe an Pla eau. J. Geophys. Res.-Biogeosci. 123, 1005–1016 81. D ake, T.W. e al. (2018) The epheme al signa u e o pe ma os ca bon in an A c ic flu ial ne wo k. J. Geophys. Res.-Biogeosci. 123, 1475–1485 82. C eed, I.F. e al. (2015) The i e as a chemos a : esh pe spec i es on dissol ed o ganic ma e flowing down he i e con inuum. Can. J.Fish.Aqua .Sci.72, 1272–1285 83. Cu is, P.J. and Schindle , D.W. (1997) Hyd ologic con ol o dis- sol ed o ganic ma e in low-o de P ecamb ian Shield Lakes. Biogeochemis y 36, 125–138 T ends in Ecology & E olu ion OPEN ACCESS T ends in Ecology & E olu ion, Feb ua y 2021, Vol. 36, No. 2 121