Controls of bioavailability and biodegradability of dissolved organic matter in soils

kiểm soát dạng dễ tiêu và dạng dễ bị phân hủy của các hợp chất hữu cơ trong đất

Controls of bioavailability and biodegradability of

dissolved organic matter in soils

Bernd Marschnera,*, Karsten Kalbitzb a

Department Soil Science/Soil Ecology, Ruhr-University Bochum, D-44780 Bochum, Germany

b

Department of Soil Ecology, Bayreuth Institute for Terrestrial Ecosystem Research (BITO ¨ K),

University of Bayreuth, D-95440 Bayreuth, Germany Received 11 March 2002; accepted 9 December 2002

Abstract

In soils, dissolved organic matter (DOM) is probably the most bioavailable fraction of soilorganic matter, since all microbial uptake mechanisms require a water environment.Bioavailability describes the potential of microorganisms to interact with DOM It is aprerequisite for biodegradation and can be restricted, if DOM is present in small pores or withinsoil aggregates and therefore not accessible for microorganisms DOM biodegradation is defined

as the utilisation of organic compounds by soil microorganisms quantified by the disappearance

of DOM or O2 or by the evolution of CO2 The controlling factors for DOM biodegradabilitycan be divided into three groups, namely, intrinsic DOM quality parameters, soil and solutionparameters and external factors DOM characteristics that generally enhance its biodegradabilityare high contents of carbohydrates, organic acids and proteins for which the hydrophilic neutralfraction seems to be a good estimate In contrast, aromatic and hydrophobic structures that canalso be assessed by UV absorbance decrease DOM biodegradability, either due to theirrecalcitrance or due to inhibiting effects on enzyme activity Effects of solution parameters such

as Al, Fe, Ca and heavy metal concentrations on DOM biodegradability have been documented

in various studies, however with different, sometimes conflicting results Inhibitory effects ofmetals are generally attributed to toxicity of the organic complexes or the free metal ions Incontrast, the enhanced degradability observed in the presence of metal ions may be due toflocculation, as larger structures will provide better attachment for microbial colonies Asdegradation is dependent on microbial activity, the composition and density of the microbialpopulation used in the degradation studies also influence biodegradation Site-specific factors,such as vegetation, land use and seasonality of meteorological parameters control DOMcomposition and soil and soil solution properties and therefore also affect its biodegradability

0016-7061/02/$ - see front matter D 2002 Elsevier Science B.V All rights reserved.

doi:10.1016/S0016-7061(02)00362-2

* Corresponding author Tel.: +49-234-3222108; fax: +49-234-3214469.

E-mail address: bernd.marschner@ruhr-uni-bochum.de (B Marschner).

www.elsevier.com/locate/geoderma

The major obstacle for a better understanding of the controls of DOM biodegradability is thelack of a standardised methodology or at least systematic comparisons between the large number

of methods used to assess DOM biodegradability

D 2002 Elsevier Science B.V All rights reserved

Keywords: Bioavailability; Biodegradation; Dissolved organic matter; DOC

1 Introduction

In the past 10 years, much progress has been made in the understanding of dissolvedorganic matter (DOM) functions and dynamics in soils Today, it is commonly acknowl-edged that DOM can enhance the solubility and mobility of metals and organiccompounds (Blaser, 1994; Piccolo, 1994; Zsolnay, 1996; Marschner, 1999) and thuscontributes to pollutant transport or to micronutrient availability In the presence of DOM,weathering rates can be accelerated(Raulund-Rasmussen et al., 1998), and DOM plays acentral role during podsolisation (Lundstro¨m et al., 1995) Furthermore, DOM containsorganically bound nutrients such as N, P and S, and DOM dynamics will therefore alsoaffect their mobility and availability(Kalbitz et al., 2000; Kaiser et al., 2001a)

DOM is also a substrate for microorganisms In soils, DOM may be the most important

C source since soil microorganisms are basically aquatic and all microbial uptakemechanisms require a water environment(Metting, 1993) Furthermore, the soluble state

is presumably a prerequisite for the diffusion of substrates through microbial cellmembranes so that the degradation of solid phase organic matter or large molecules canonly occur after dissolution or hydrolysis by exoenzymes The initial phase of litterdecomposition is also strongly related to the amount of soluble compounds in the litter(Williams and Gray, 1974) This was also shown byMarschner and Noble (2000), where

CO2release from a soil supplemented with different plant litters could largely be explained

by the disappearance of DOC(Fig 1) Similar results were obtained with soils incubated atdifferent temperatures(Marschner and Bredow, 2002).Cook and Allen (1992)also reportpositive relationships between initial DOC concentrations and CO2release during the first

5 weeks of a long-term incubation experiment However, at later stages, this relationship

no longer existed which they attributed to the depletion of degradable DOM compounds.Several other authors have found close correlations between DOM concentrations anddenitrification potentials or rates(Bijay-Singh et al., 1988; Isermann and Henjes, 1990; Pu

et al., 1999), thus indicating that the availability of biodegradable DOM may be aprerequisite for creating reducing condition in soils or in certain soil compartments(Zsolnay, 1996) On the other hand, Kalbitz et al (2003) found no evidence that DOMextracted from Oa, and A horizons is the most biodegradable fraction of soil organicmatter

DOM degradation is also an important process controlling DOM dynamics in soils.DOM inputs into the mineral soil generally greatly exceed DOM outputs with seepage.Until recently, this was mainly attributed to DOM retention through sorption (Guggen-berger et al., 1998) However, some newer calculations indicate that total C pools shouldthen be several orders of magnitude higher than generally observed in the field

B Marschner, K Kalbitz / Geoderma 113 (2003) 211–235 212

(Guggenberger and Kaiser, 2003; Moore, 1997) Therefore, the mineralisation rate ofDOM in subsoils must be much higher than thought previously(Qualls and Haines, 1992;Guggenberger et al., 1998), and thus, the biodegradation of DOM or of former DOMsorbed to mineral surfaces is the most likely explanation for the generally low DOM fluxestowards the groundwater reported byMichalzik et al (2001).

In their extensive review on DOM dynamics in soils,Kalbitz et al (2000)also point outthat the mechanisms and controls of DOM degradation in soils are still poorly understood

If the earlier stated assumption is correct, that the soluble state is a prerequisite for theuptake and degradation of organic matter by microorganisms, then DOM should play akey role in the stabilisation and destabilisation of soil organic matter, and thus, in Cdynamics and C pools of soils.Sollins et al (1996)present a conceptual model of SOMstabilisation and destabilisation for which they differentiate between three general sets ofcharacteristics affecting the stability of organic matter: recalcitrance, interactions andaccessibility For the conceptual understanding of mechanisms and controls, this approach

is very helpful However, one has to bear in mind that often, several mechanisms andprocesses interact to determine the stability or biodegradability of organic matter in soils

In this paper, factors and mechanisms are reviewed that control the microbialdegradation of DOM in soils The term ‘‘DOM’’ will be used for all organic substancessmaller than 0.45 Am that are suspended in aqueous solutions Strictly speaking, the termDOM can only be applied to organic matter in soil solutions extracted with lysimeters Inmost studies where soluble organic matter is obtained after extraction from the soil withbatch or percolation methods, this is not the truly dissolved phase, but the potentiallysoluble Zsolnay (1996) therefore suggests to clarify this by using the term ‘‘WEOM’’(water-extractable organic matter) Since this review deals with the degradability oforganic substances in the solution phase, a differentiation between the various methodswhereby this solution phase was obtained is not regarded as essential for the problem

Fig 1 Relationship between the change in water-extractable soluble organic compounds (DOC) and cumulative

CO 2 evolution in an Australian pasture topsoil during a 21-day incubation with different plant litter materials

(Marschner and Noble, 2000)

2 Bioavailability versus biodegradability

In pharmaceutical and toxicological studies with mammals, the term ‘‘bioavailability’’

is used to characterise the amount of a substance ingested and retained in the organism andthus becomes available for metabolic use It is therefore not a measure for the actualutilisation of this substance For organic molecules such as DOM compounds, this meansthat their uptake, i.e., bioavailability, must not necessarily result in their breakdown tosmaller entities or to complete mineralisation On the other hand, microorganisms excreteexoenzymes that promote extracellular degradation of compounds that are otherwise notbioavailable according to the above definition Therefore, in the context of DOM, the termbioavailability describes the potential of microorganisms to interact with these substances

As a measure for the actual utilisation of organic compounds by soil microorganisms,the term ‘‘biodegradability’’ is chosen In a strict sense, this still encompasses twoalternative or sequential processes:

(1) microbial uptake or breakdown of the original compounds which are then used for thebiosynthesis of microbial cell materials

(2) complete mineralisation to obtain energy and inorganic nutrients

Depending on the analytical tools used to monitor the degradation process, these twoprocesses are considered to different degrees If microbial utilization of DOM isdetermined by the increase in microbial biomass, then only the assimilated organic carbon(AOC) is considered (Escobar and Randall, 2001), while the mineralized fraction isneglected If DOC disappearance is used as a measure for biodegradation, it is not possible

to differentiate between microbial incorporation and mineralisation

Another aspect that points to the complexity of this issue is illustrated in the study byAmon and Benner (1996)where low-molecular DOM (< 1000 Da) was less degradablethan high-molecular DOM However, bacterial growth efficiency was much higher withthe low-molecular DOM fraction, thus indicating that this seemingly less-degradablefraction contained more compounds needed for bacterial biomass production

AOC is a measure for the ability of water to support heterotrophic growth(Escobar andRandall, 2001) Therefore, AOC is mainly an important parameter for waterworks Itrepresents only a small portion of the entire biodegradable DOM and will not be furtherdiscussed in this paper mainly dealing with soil DOM We will focus on biodegradability

of terrestrial DOM, i.e., the utilisation of organic compounds by soil microorganismsquantified by the disappearance of DOM or O2or by the evolution of CO2

3 Methods for the determination of DOM biodegradability

As indicated in the previous section, no generally accepted standard method for thedetermination of DOM biodegradability exists so far A method published by theInternational Organization for Standardization determines the biochemical oxygen demand(BOD) of aqueous media, which is based on a 5-day incubation, using solid sewage sludge

as an inoculum(ISO 10707, 1994) However, this method is mainly used to determine the

B Marschner, K Kalbitz / Geoderma 113 (2003) 211–235 214

BOD in effluents from sewage treatment plants that have to meet certain BOD levels inmany countries.

The methods used by soil scientists, groundwater hydrologists, limnologists or ographers to quantify the biodegradable organic carbon are quite diverse(Table 1) Moststudies are conducted in cultures where any of the following parameters may vary:– type of incubation (batch culture, flow-through bioreactor)

ocean-– type and size of incubation vessel

As a matter of fact, in this review, no two studies performed in different laboratories usedthe same set of parameters for the determination of DOM biodegradability in their batchexperiments This means that the reported results cannot be compared with each other,which is a major obstacle for scientific discussions and progress

Of all the parameters listed above, two seem to be most crucial for the quantification ofDOM biodegradability: duration of incubation and measure for biodegradation

Many long-term incubations (>10 days) showed that DOM generally consists at least of

a rapidly degradable fraction (fast BDOM or labile DOM), a fraction that is degraded moreslowly, and the recalcitrant fraction that remains in solution even after very longincubation periods (up to 180 days) Little is known about the nature of the compounds

in these different DOM pools, but it is generally assumed that the labile DOM consistsmainly of simple carbohydrate monomers (i.e., glucose, fructose), low-molecular organicacids (i.e., citric, oxalic, succinic acid), amino acids, amino sugars and low-molecular-weight proteins(Lynch, 1982; Qualls and Haines, 1992; Guggenberger et al., 1994; Ku¨seland Drake, 1999; Kaiser et al., 2001b; Koivula and Ha¨nninen, 2001) These compoundscan directly be utilised by a large number of different organisms and therefore do notrequire a special set of enzymes(Lynch, 1982)

The slowly degradable or relatively stable DOM fraction probably contains charides (i.e., breakdown products of cellulose, hemicellulose) and other plant or micro-bially derived compounds or degradation products that require special tools fordegradation These enzymes are probably only produced when labile substrates are nolonger available or they are limited to few organisms that therefore do not need to competefor the labile substrates (k- vs R-strategists according toPaul and Clark, 1996) Even theso-called recalcitrant DOM fraction is not fully nondegradable as it would otherwiseaccumulate in subsoils to a much higher degree than observed(Moore, 1997) However,

polysac-Table 1

Summary of methods used in different studies for the determination of the biodegradability of DOM or DOM fractions

Zsolnay and Steindl, 1991d;

Qualls and Haines, 1992 ;

Boissier and Fontvieille, 1993 ;

Raymond and Bauer, 2001

inoculum + nutrients

Zsolnay, 1996;

Marschner and Bredow, 2002

Andersson and Nilsson, 2001; Ogawa et al., 2001

Hongve, 1999d;

Hongve et al., 2000 d

;

Søndergaard et al., 2000

Amon and Benner,

Amon et al., 2001 d

Jandl and Sletten, 1999; Jandl and Sollins, 1997; Moran et al., 2000

Søndergaard and Worm, 2001

Blaser, 1989

Merckx et al., 2001;

Marschner and Noble, 2000; Marschner and Bredow, 2002

this fraction must consist of structures that are not easily cleaved by enzymes, such aslignin degradation products or compounds strongly altered through preceding degradationsteps(Joergensen, 1998).

As pointed out by Qualls and Haines (1992)and Kalbitz et al (2003), soil solutionscontain very different amounts of these fractions, and consequently, the kinetics of thedegradation process will be very different Quantification of the contribution of DOM

to the stable C pool in the mineral subsoil requires a quantification of rapidly andslowly degradable DOM fractions and their mean residence times Knowledge aboutthe size of the biodegradable DOM fraction is not sufficient Kalbitz et al (2003)reported two DOM solutions with a similar portion of biodegradable DOC, whereas thedecomposition constants of the rapidly and slowly degradable fraction differed to agreat extent

The other aspect that can strongly influence the result of a biodegradation assay is howDOM degradation is quantified If the change in DOC concentration is used, then samplestaken from the incubation solution will need to be filtered at 0.45 Am Any DOCtransformed into microbial biomass or other particulate carbon resulting from coagulationand precipitation will be largely retained on the filter and its disappearance will thereforefalsely be interpreted as degradation On the other hand, if samples are left unfiltered andTOC is used as a measure, microbially incorporated DOC will be regarded as notdegraded According toSøndergaard et al (2000), microbial biomass may be in the range

of 10% of TOC, so that the error made with TOC or DOC measurements may be small.However, the error could be much higher if all particulate C is considered (C which doesnot pass through a 0.45-Am filter) Thus, Kalbitz et al (2003) reported a DOC mine-ralisation after 90 days of only 9% from CO2data, although the DOC content of thissample declined by 50% However, a problem with TOC analysis is that an adequate andreproducible sampling of the suspension and a complete combustion of the particlescannot be guaranteed

The other measure for DOM mineralisation is CO2efflux from the samples(Table 1).However, even with this method, errors can occur due to CO2dissolution in water (3.4 g

CO2/l under atmospheric pressure in distilled water) if CO2is not trapped and depletedfrom the atmosphere of the incubation vessel and when the solution was not in equilibriumwith atmospheric CO2initially

Another approach for the determination of DOM degradability is using ‘‘bioreactors’’filled with glass beads that are colonized by microorganisms to form so-called biofilms ontheir surfaces (Yano et al., 1998; Søndergaard et al., 2000) DOM solutions are passedthrough such flow-through reactors and DOM degradation is determined from thedifference in DOC concentrations between in- and out-flow, usually with residence times

of 10 – 24 h In a comparative study, Søndergaard et al (2000) showed that thedegradability of DOM determined with such a system is closely related to DOMdegradability in batch cultures after 135 – 151 days (r2= 0.73) and reaches about 90% ofthe batch values This high efficiency of the bioreactor can be explained by the relativelyhigh microbial density compared to batch cultures which allows more intensive microbialinteractions with DOM and its degradation products within the biofilm Pinney et al.(2000)describe another type of bioreactor where they used biologically active sand in abatch vessel

However, these flow-through bioreactors initially require long equilibration times (up

to 6 months) and continuous maintenance to achieve reproducible degradation rates(McDowell, personal communication) Other problems of these reactors are the release ofDOM into the solution and the adsorption of DOM onto the biofilms Furthermore, itseems impossible to determine the pool sizes and the residence times of a rapidly and aslowly degradable fraction

The other listed parameters will also affect the result of DOM biodegradation ments because these are mainly discussed under soil and solution properties (Section 5.2).Here are some examples Shaking of DOM solutions during the incubation could hinderthe development of hyphae which could result in an underestimation of the biodegradation

measure-by fungi Addition of nutrients will at least accelerate DOM biodegradation resulting inhigher degradation rates in comparison to incubation without addition of nutrients(Schmerwitz, 2001) Furthermore, an enhanced coagulation and precipitation due toincreased ionic strengths (Kalbitz et al., 2000) is imaginable Finally, data about thedensity of added microorganisms are scarce in published studies on DOM biodegradation,ranging from 1000 – 2000(Miettinen et al., 1999) to 104 CFU (Volk et al., 2000), and0.48  105cells ml 1(Buffam et al., 2001)in the inoculated sample Mostly, it is reportedthat 1% (v/v) of inoculum was added

4 Controls of DOM bioavailability

The bioavailability of DOM is reduced if the possibilities of microorganisms to interactwith DOM are restricted These may be physical restrictions, such as inaccesibility ofDOM in very small pores or chemical restrictions, such as DOM sorption to solid surfaces.4.1 Pore size

DOM in small pores is not accessible for microorganisms, i.e., in pores with diametersbelow 0.2 Am (Zsolnay, 1997) This pore size class contains water that is not plantavailable and hardly participates in transport processes Consequently, DOM in these poreswill only become bioavailable through diffusion into larger water-filled pores Althoughenzymes excreted by microorganism may enter these pores, this is also limited todiffusion, as well as the movement of the breakdown products out of the pores In someclayey soils, up to the 50% (v/v) of the total pore volume is in this size class and DOMcould be preserved there from microbial breakdown However, to date, little is knownabout the amounts and quality of DOM in different pore size classes, because the soilwater in the smallest pore sizes cannot be directly extracted for analysis

Zsolnay and Steinweg (2000) have attempted to overcome this problem by using astepwise extraction technique to obtain DOM fractions from different pore size classes In

a first step, the undisturbed samples are percolated to obtain the so-called mobile DOM.Soil solution from the mesopores (0.2 – 6 Am) is then extracted by centrifugation, and theremaining DOM is extracted in a batch-shake procedure, with a mild salt solution For thethree soils examined bySteinweg (2002), DOM in the percolates was always the leastbiodegradable, thus supporting the assumption that this DOM pool should be depleted first

B Marschner, K Kalbitz / Geoderma 113 (2003) 211–235 218

due to its high bioavailability In the other two fractions, DOM degradability was similarand much higher, thus indicating some physical protection in less-accessible pores Since

71 – 82% of DOM was obtained in the batch extracts, DOM conservation in the smallestpores may be of major importance in soils

4.2 Soil aggregation

A similar mechanism of restricted bioavailability of DOM in small pores may occurwithin aggregates since various studies have shown that the disruption of aggregatesstimulates microbial activity (i.e., Elliott, 1986; Ladd et al., 1993) and that aggregatescontain more young and less-altered plant-derived organic matter than the bulk soil(Skjemstad et al., 1990; Puget et al., 1995; Six et al., 2000) However, no studies havebeen encountered where DOM from within aggregates was compared to bulk soil DOM interms of its biodegradability

4.3 Sorption

In the presence of mineral solid phases, the mineralisation of plant-derived drates or simple organic compounds such as glucose and citrate can be greatly reduced,especially if charged molecules like citrate or oxalate interact with charged minerals such

carbohy-as clay minerals or goethite(Jones and Edwards, 1998; Miltner and Zech, 1998; Stro¨m

et al., 2001) The observed reduced biodegradability of soil organic matter throughsorption to mineral surfaces is considered to be one or the most important stabilisationprocesses, and it is extensively reviewed by Sollins et al (1996) and Kaiser andGuggenberger (2000) However, the mechanisms of this sorption process are still aspoorly understood, as the reasons why sorbed materials may be less degradable Incontrast,Guggenberger and Kaiser (2003)estimated a mean residence time of the sorbedorganic carbon of about 4 – 30 years and thus challenged the commonly assumedsorptive stabilization of DOM They hypothesised that natural soil surfaces are covered

by biofilms with a high affinity for DOM, so that the observed ‘‘sorption’’ may indeedenhance bioavailability and subsequent biodegradation Only sorption onto purelymineral surfaces would thus result in an effective stabilization of DOM (Guggenbergerand Kaiser, 2003) On the other hand, sorption is generally not irreversible, so thatsorbed materials may return to the solution phase and thus become bioavailable again.Desorption will depend on the nature of the sorbate and sorbent and is affected bysolution composition Kaiser and Guggenberger (2000)have shown that the hydrophilicDOM fraction is much more easily desorbed than the hydrophobic DOM fraction, which

is also less biodegradable (see below)

4.4 Drought

DOM availability for microorganisms is reduced when soils become dry, since thisgreatly limits microbial activity and decreases diffusive transport processes towards theremaining moist sites of activity The strong stimulation of microbial activity afterrewetting dry soils is therefore often attributed to the accumulation of easily degradable

substances such as cellular materials from the desiccated organisms in the dry soil(Lundquist et al., 1999; Zsolnay et al., 1999; Merckx et al., 2001).

5 Factors controlling DOM biodegradability

The biodegradability of DOM is controlled by numerous factors that can be dividedinto three categories The first set of factors are intrinsic DOM characteristics that aredetermined by molecular structure, functional group content or size of the molecules Thesecond set of factors consists of soil properties that can influence the degradation process,such as nutrient availability, microbial community structure and the presence of toxicsubstances or other soil solution constituents At the third level, external factors such as thetemperature and rainfall regime and the associated vegetational cycles will induce aseasonal variability of both DOM inputs and microbial activity which can affect intrinsicDOM quality parameters and soil and soil solution properties

5.1 Intrinsic DOM quality parameters

5.1.1 Molecular size

Considering the uptake mechanisms of microorganisms, one could expect that smallerDOM molecules or units should be ingested and degraded preferentially Evidence for thiswas found in one of our studies where the biodegradability of DOM in ultrafiltrates of thesize class < 1000 Da was three- to fourfold higher than in the size class < 10,000 Da or in thebulk DOM solution(Table 2) However, this was only true for DOM extracted from soilsamples that were collected in early spring In summer, biodegradability of DOM was muchlower, with no differentiation between size classes The reason for this is probably thedepletion of degradable compounds by the activated microorganisms during late spring andsummer On the other hand, the preferential degradation of small compounds in the springsample may not be a size effect but due to chemical characteristics.Kaiser et al (2001b)showed for a forest soil that easily degradable carbohydrates, amino sugars and proteinsaccumulate during winter and these compounds would largely appear in the small size class.For aquatic DOM, Amon and Benner (1994, 1996) found opposite results In theirsamples obtained from the Gulf of Mexico and the Amazon River and nearby coastalocean waters, they determined a much higher C mineralisation from larger DOM sizefaction (>1000 Da) compared to smaller DOM Since most DOM in the Amazon was inthe larger size fraction and marine DOM consisted mainly of the small size fraction, they

Table 2

Effect of sampling date on the biodegradability of total DOM and DOM in two size fractions (ultrafiltration) from solutions obtained from percolating undisturbed soil samples from an arable field with 0.01 M CaCl 2 (DOC after 5-day incubation at 20 jC in % of initial DOC)

Values in rows followed by the same letter are not significantly different ( p < 0.05, Duncan test).

B Marschner, K Kalbitz / Geoderma 113 (2003) 211–235 220

concluded that the larger molecules are more bioreactive not due to their size, but becausethey are fresher, i.e., less degraded than the smaller compounds Therefore, again size isonly a secondary attribute and the primary factor controlling DOM biodegradability would

be structural characteristics

5.1.2 Chemical structure and spectroscopic properties

Carbohydrates and amino acids are highly decomposable in soils(Haider, 1992)and areutilized preferentially by microorganisms during degradation of different compounds inDOM solutions(Volk et al., 1997; Amon et al., 2001; Kalbitz et al., 2003) However,Volk

et al (1997) stated that the often used classification of carbohydrates as labile DOMcomponents should be seen with caution, as carbohydrates can also be bound to stableDOM compounds

Compounds with alkyl or aromatic structural units generally accumulate during thedecomposition of soil organic matter(Baldock et al., 1992; Ko¨gel-Knabner et al., 1992;Baldock and Preston, 1995; Huang et al., 1999)and have thus been associated with a lowbiodegradability.Boissier and Fontvieille (1993)found that phenols and polyphenols wereclosely related to the amount of nondegradable DOM in incubation experiments Similarly,Wershaw and Kennedy (1998)observed a relative increase in aromatic structures duringlitter decomposition Kalbitz et al (2003) showed that the biodegradability of DOMextracted from forest litter layers was negatively correlated to its content in aromaticstructures determined with1H-NMR

Spectroscopic properties are commonly determined for DOM characterisation.Traina et

al (1990)andChin et al (1994)have shown that the specific UV absorbance of humic andfulvic acids between 250 and 280 nm is closely correlated to their content in aromaticstructures

Since aromatic structures are generally quite recalcitrant, one would expect a negativerelationship between the specific UV absorbance and the biodegradability of DOM Thisindeed was observed in one of our studies (Fig 2a, Jo¨demann unpublished results) with alinear correlation coefficient of r = 0.69 for 28 solutions obtained from an arable soil thathad been stored fresh, air-dried or frozen and then extracted with 1 mM CaCl2solutionwith a percolation procedure with either undisturbed samples or after homogenization Inthe same samples, the decrease in DOC concentration after biodegradation was highlysignificantly correlated (r = 0.85) with an increase in specific UV absorbance (Fig 2b),thus indicating that UV-inactive substances were degraded preferentially Other authorsalso reported close correlations between DOM degradability and specific UV absorbance(Gilbert, 1988; Zoungrana et al., 1998; Pinney et al., 2000; Kalbitz et al., 2003)and someeven found nonlinear relationships, where biodegradability increases exponentially withdecreasing UV absorbance

However, specific UV absorbance of DOM is not always a reliable predictor forbiodegradability.Marschner and Bredow (2002)show that the biodegradability of DOMfrom soil samples incubated at different temperatures varied greatly from 8% to 61% but wasnot related to specific UV absorbance of either total DOM or of its size fractions If oneaccepts the assumption that UV absorbance is a measure for the recalcitrant aromaticstructures, then these results clearly show that the non-aromatic compounds also greatlydiffer in biodegradability A low biodegradability of aliphatic compounds may be due to

binding to aromatic structures (i.e., lignocellulose) or to a high degree of polymerisation oroxidation(Guggenberger et al., 1994), but this cannot be assessed with simple spectroscopicmethods.

More recently, fluorescence spectroscopy has been used successfully to obtaininformation about the biodegradability of DOM (Glatzel et al., 2003; Kalbitz et al.,2003)using the assumption that more condensed aromatic structures with a red-shiftedfluorescence are less biodegradable than structures with a low degree of condensation andconjugation Zsolnay et al (1999) showed that a humification index calculated fromfluorescence data can help to differentiate between microbial cell lysis products and morehumified DOM.Parlanti et al (2000)stressed the usefulness of fluorescence spectroscopy

as an indicator for biological activity and humification in coastal waters

Fig 2 Relationship between DOC degradation and specific UV absorbance of DOC extracted with 0.01 M CaCl 2

solution from differently treated top soil samples from an arable field Sample treatments included drying and freezing prior to extraction (a) Relationship between initial UV absorbance and degradation of DOC after 5 days

of incubation (b) Relationship between the change in UV absorbance during incubation and the degradation of DOC.

B Marschner, K Kalbitz / Geoderma 113 (2003) 211–235 222