lable at ScienceDirect

Soil Biology & Biochemistry 42 (2010) 220e233

Contents lists avai

Soil Biology & Biochemistry

journal homepage: www.elsevier .com/locate/soi lb io

Determinants of soil organic matter chemistry in maritimetemperate forest ecosystems

Karen Vancampenhout a,b,*, Bruno De Vos c, Katinka Wouters a, Hans Van Calster c,Rudy Swennen a, Peter Buurman d, Jozef Deckers a

aDepartment of Earth and Environmental Sciences, K.U.Leuven, Celestijnenlaan 200E, B-3001 Leuven, BelgiumbDepartment of Biosciences and Technology, K.H.Kempen University College, Kleinhoefstraat 4, B-2440 Geel, BelgiumcResearch Institute for Nature and Forest, Gaverstraat 4, B-9500 Geraardsbergen, Belgiumd Earth System Science, Department of Environmental Sciences, Wageningen University, P.O. Box 47, NL 6700 AA Wageningen, The Netherlands

a r t i c l e i n f o

Article history:Received 23 February 2009Received in revised form17 September 2009Accepted 21 October 2009Available online 1 November 2009

Keywords:Organic carbonTemperate forestsPyrolysis-GC/MSFactor analysis

* Corresponding author at: Department of Earth aK.U.Leuven, Celestijnenlaan 200E, B-3001 Leuven, Befax: þ32 16 32 97 30.

E-mail address: karen.vancampenhout@ees.kuleuv

0038-0717/$ e see front matter � 2009 Elsevier Ltd.doi:10.1016/j.soilbio.2009.10.020

a b s t r a c t

While the influence of climate, vegetation, management and abiotic site factors on total carbon budgetsand turn-over is intensively assessed, the influences of these ecosystem properties on the chemicalcomplexity of soil organic matter (SOM) remains poorly understood. This study addresses the chemicalcomposition of NaOH-extracted SOM from maritime temperate forest sites in Flanders (Belgium) bypyrolysis-GC/MS. The studied forests were chosen based on dominant tree species (Pinus sylvestris, Fagussylvatica, Quercus robur and Populus spp.), soil texture and soil-moisture conditions. Differences inextractable-SOM pyrolysis products were correlated to site variables including dominant tree species,management of the woody biomass, site history, soil properties, total carbon stocks and indicators formicrobial activity. Despite of a typical high intercorrelation between these site variables, the influence ofthe dominant tree species is prominent. The extractable-SOM composition is strongly correlated to litterquality and available nutrients. In nutrient-poor forests with low litter quality, the decomposition ofrelatively recalcitrant compounds (i.e. short and mid-chain alkanes/alkenes and aromatic compounds)appears hampered, causing a relative accumulation of these compounds in the soil. However, if substratequality is favorable, no accumulations of recalcitrant compounds were observed, not even under highsoil-moisture conditions. Former heathland vegetation still had a profound influence on extractable-SOMchemistry of young pine forests after a minimum of 60 years.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction

As the soil is by far the largest terrestrial carbon reservoir, soilcarbon chemistry is of major importance in understanding carboncycles. Especially forest soils are of interest, as they hold 70% of allsoil C (Dixon et al., 1994; Huntington, 1995; Jandl et al., 2007).Carbon in soils is embedded in a complex mixture of chemicalcompounds, referred to as soil organic matter (SOM; Piccolo, 2001;K€ogel-Knabner, 2002; Kelleher and Simpson, 2006). The composi-tion of this mixture greatly influences decomposition processes andtherefore carbon cycling (Sollins et al., 1996; Dell'Abate et al., 2003;K€ogel-Knabner et al., 2006; Conant et al., 2008).

nd Environmental Sciences,lgium. Tel.: þ32 16 32 97 34;

en.be (K. Vancampenhout).

All rights reserved.

While the influence of climate, vegetation, management andabiotic site factors on total carbon budgets and turn-over is inten-sively assessed, high-resolution studies that elaborate on the influ-ences of these ecosystem properties on the chemical complexity ofSOM in natural ecosystems are much more scarce (Sollins et al.,1996; Gleixner et al., 1999; K€ogel-Knabner et al., 2005; von L€utzowet al., 2006; Buurman et al., 2009). So far, detailed studies addressingSOM composition have mainly focused on specific situations (e.g.Nierop et al., 2001a; Nierop et al., 2001b; de Alcantara et al., 2004;Buurman et al., 2007a; Schmitt et al., 2007; Kaal et al., 2008a,b;Verde et al., 2008; Ferreira et al., 2009; Santin et al., 2009). Moregeneral relations, linking the SOM chemistry to cross-ecosystemscale variations in climate, vegetation, management and abiotic sitefactors remain poorly understood. Soil type, vegetation cover andland use are reported to influence SOM chemistry, but these effectsremain inadequately documented (Senesi et al., 2003; Buurmanet al., 2009; Grandy et al., 2009). Vancampenhout et al. (2009)observedmajor differences in topsoil extractable-SOM composition

K. Vancampenhout et al. / Soil Biology & Biochemistry 42 (2010) 220e233 221

among major ecological biomes (tundra, taiga, steppe, coniferoustemperate forest, broadleaved temperate forest and tropical forest)yet stated that smaller scale variations, such as tree species, soilnutrients, sitehistoryandaltitude added to the effects of climate andlitter type (i.e. coniferous or broadleaved litter) and should beaddressed further.

This study therefore aims to establish a reconnaissance study inrelating environmental factors to differences in NaOH-extractable-SOM composition within a similar climatic context, i.e. maritimetemperate forests in Flanders (Belgium), by analytical pyrolysis-gaschromatography/mass spectrometry (pyrolysis-GC/MS). Differentforest types were considered dominated by common European treespecies (FAO, 2006a), i.e. Scots pine (Pinus sylvestris L.), Europeanbeech (Fagus sylvatica L.), Pedunculate oak (Quercus robur L.) andhybrid poplar (Populus spp.) and situated on different soil types.Current North and West European pine stands often result fromafforestation of former heathland, which affects SOM properties(von Oheimb et al., 2008). Hence, site history was also taken intoaccount. As the parameters controlling SOM chemistry remainpoorly understood, the study targets in-field conditions rather thanspecifically designed experiments. The study aims at (i) relativelycomparing differences in pyrolysis products of extractable-SOMcomposition between the chosen sites, (ii) relating these differ-ences to site factors including vegetation, management aspects,total carbon content, microbial activity proxies, litter layer char-acteristics and soil properties.

2. Materials and methods

2.1. Site selection and sampling

The region of Flanders has a temperate maritime climate, withan average yearly precipitation of 821mmand average temperatureof 9.7 �C (KMI, 2009). Soils aremainly formed onQuaternary aeolianor fluviatile deposits and textures range from mainly sandy orclayey in the north and north-west to mainly silty clay loam in thesouth. Under forest, the soilscape is dominated by Albeluvisols andLuvisols on heavy soil textures and by Podzol hydrosequences onsandy textures (FAO, 2006b). Important sylvicultural speciesinclude pine (P. sylvestris L. and Pinus nigraArn. var. calabrica (Loud.)Schneider), European beech (F. sylvatica L.), Pedunculate oak(Q. robur L.) and hybrid poplar (Populus spp.; Dumortier et al., 2007).

Nineteen forest sites were chosen based on (a) dominant treespecies and (b) soil properties assumed to primarily influence SOMdynamics (i.e. soil texture and soil moisture; Jenkinson and Cole-man, 2008). For sampling purposes, sites were grouped accordingto soil texture in two classes: i.e. ‘coarse’ (C; sandy loam andcoarser) and ‘fine’ (F; silt loam and finer) and further subdivided inhaving a ‘wet’ (W; highest groundwater level <50 cm deep) and‘dry’ (D; highest groundwater level >50 cm deep) soil-moistureregime. Two young pine forests on historical heathland anda recently forested heathland were also included in the dataset(Table 1). To elaborate upon the effects of prolonged anaerobicconditions, two samples were added taken on sites that are floodedby groundwater for long periods in winter, one birch-oak stand ona sandy former heathland (afforested since 1940 or earlier) and onealder-willow stand on sandy loam (afforested since 1775 or earlier).

All 21 samples were bulk samples of mineral soil, taken over theentire depth of the A-horizon. Samples were air dried and passedtrough a 2 mm sieve prior to laboratory analysis. The samples weregivena code reflecting thedominant tree speciesof the sampling site(i.e. Pi for pine, Fa for beech, Qu for oak, Po for poplar, Be for birch, Alfor alder and FH for Forested Heathland), the soil-texture class (C for‘coarse’ or F for ‘fine’) and the soil-moisture-regime class (D for dry,W for wet and WW for periodically flooded with groundwater).

2.2. Site variables and soil properties

Site variables included vegetation and standing biomassmanagement aspects, proxies for microbial biomass and activity,litter layer characteristics, total soil carbon and basic soil properties.

Vegetation and management aspects determined on site weredominant tree species, basal area, stem number, herb layer cover (asan indication for light reaching the forestfloor) andagedistributionofthe stand (even or uneven-aged). Forest age was based on compara-tive GIS analysis of maps (De Keersmaeker et al., 2001) and includesfour classes: (1) forested since 1775 or earlier, (2) forested since 1850or earlier, (3) forested since 1940 or earlier, (4) forested after 1940.

Total carbon was measured by dry combustion at 900 �C (Shi-madzu 5050A elemental analyzer) and nitrogen by the standardKjeldahl method (Gerhardt Vapodest 60 system). Total carbon andtotal nitrogen were expressed as mass percentages and as carbondensities (obtained by multiplying mass percentage of carbon bybulk density; De Vos, 2009). The hot water carbon (HWC) extract-able fraction (Ghani et al., 2003) and soil CO2 respiration (soda-limemethod, Edwards, 1982; Grogan, 1998) were evaluated as an indi-cation for the microbial carbon contribution to the SOM. Litter layercharacteristics included thickness and dry mass of the litter layerper m2. Humus index was determined based on Ponge andChevalier (2006), i.e. (1) eumull, (2) mesomull, (4) dysmull, (5)hemimoder, (6) eumoder, (7) dysmoder and extended with 2classes for (8) mor and (9) litter layer complexes with grasses.

Soil properties most commonly reported to affect total soilcarbon stocks are soil moisture, soil texture and acidity. Thereforesoil variables in this study included grain-size analysis, pH and soil-moisture regime. Grain-size was measured by laser diffractometry(Coulter LS 200) as percentage of clay (<8 mm; Konert and Van-denberghe, 1997), silt (8e50 mm) and sand (50 mme2 mm)according to the Belgian and USDA systems (Soil survey staff, 1975;Beuselinck et al., 1998; Buurman et al., 2001; Taubner et al., 2009).pHðH2OÞ was measured in a 1:5 w/w suspension, according to ISRICand FAO (2002). Soil-moisture regime was expressed as the depthof the winter groundwater table in cm below the soil surface, basedon redoximorphic soil features (Belgian soil classification system;Dudal, 1998): i.e. (b) >125 cm/90e125 cm, (c) 80e125 cm/60e90 cm, (d) 50e80 cm/40e60 cm, (e) 30e50 cm/20e40 cm, (f)0e30 cm/0e20 cm for the depth of the winter groundwater tableon loamy or clayey soils/sandy soils respectively.

2.3. Extraction of SOM and pyrolysis-GC/MS

SOM was extracted by adding 50 ml NaOH (0.1 M) to 5 g of soilsample. The samples were shaken for 24 h under N2, centrifugedand decanted. Hence, only the NaOH-extractable fraction of theSOM is considered in this paper. To establish the proportion ofthe total carbon content extractable by 0.1 M NaOH for the soils inthis study, the carbon content of the NaOH-extract was likewisemeasured by dry combustion. This yield was expressed as NaOH-extracted C relative to the total C content of the sample (%C):

yieldð%CÞ ¼ CNaOH extracted=Ctotal

The NaOH-extract was acidified to pH 1e2 with HCl (1.5 M) and1 ml HF (48%) was added. The mixture was shaken for 48 h, dialyzed(cutoff diameter 12 000 Da) and freeze-dried (e.g. Nierop et al.,2001a, 2005; Buurman et al., 2007a,b; Kaal et al., 2008a,b). Thefraction thus obtained is further referred to as ‘extractable-SOM’ inthis paper (Vancampenhout et al., 2009). Freeze-dried samples werepyrolysed using a Curie-point pyrolyser (610 �C, 5 s), connected toa Carlo Erba GC (Milan, Italy) equipped with a fused silica column(Chrompack, 25 m, 0.25 mm i.d.) coated with CP-Sil5 using He as

Table 1Site and sample codes, grouped by dominant tree species, soil texture and soil-moisture regime of the sampling site. Soil type and site location are also indicated.

Site andsample code

Dominant tree species Soiltexturea

Soil-moistureregime

Soil typeb Sample site location Remarks

PoCW Populus euramericana‘Robusta’ or ‘Ghoy’

Sandy loam Wet Luvic Mollic Gleysol(Endoeutric)

Wijnendale, Ichtegem

PoCD Populus euramericana‘Robusta’

Sandy loam Dry Arenosol (Hypereutric) Hannecartbos,Oostduinkerke

A former ploughlayer is still visible

PoFW Pop trichocarapa � deltoides‘Beaupr�e’ or ‘Boelare’

Silty clay loam Wet Luvic Gleysol (Humic,Hypereutric, Epiclayic)

Oude Mombeek,St.Lambrechts-Herk

PoFW(2) Populus euramericana‘Robusta’

Silt loam Wet Epigleyic Luvisol(Hypereutric, Siltic)

Parikebos, Parike

PoFD Populus trichocarapa� deltoides ‘Beaupr�e’

Silt loam Dry Endogleyic Albeluvisol(Hypereutric, Siltic)

Balegem bos, Balegem

QuCW Quercus robur L. Sandy loam Wet Endogleyic Folic Umbrisol(Humic, Alumic, Hyperdystric)

Het Leen, Waarschoot

QuCD Quercus robur L. Sandy loam Dry Endostagnic Folic Umbrisol(Endoalbic, Hyperdystric)

Paddenpoelebos, Maldegem

QuFW Quercus robur L. Silt loam Wet Endogleyic Umbrisol(Humic, Alumic, Siltic)

Bos Ter Rijst, Schorisse Site contains remantsof Roman agriculture

QuFD Quercus robur L. Silt loam Dry Fragic Albeluvisol (Siltic) Zoni€en e Red Abby,Auderghem

Cultivated for ca. 100 yearsin the 17th century A.D.c

FaCW Fagus sylvatica L. Sandy loam Wet Endogleyic Folic Umbrisol(Brunic, Hyperdystric, Arenic)

Wijnendale, Ichtegem

FaCD Fagus sylvatica L. Sandy loam Dry Haplic Cambisol(Humic, Orthodystric)

Nieuwenhoven, Oostkamp

FaFW Fagus sylvatica L. Loam Wet Stagnic Cambisol (Epidystric) Drongengoed, UrselFaFD Fagus sylvatica L. Silt loam Dry Cutanic Fragic Albeluvisol

(Hyperalumic, Siltic)Zoni€en, Hoeilaart

PiC(W) Pinus sylvestris L. Loamy sand Wet Haplic Regosol(Hyperdystric, Arenic)

Kenisberg-Kruisberg, Schaffen

PiCD Pinus sylvestris L. Loamy sand Dry Umbric Podzol (Anthric) Gellikerheide, OpgrimbieYPiCW Pinus sylvestris L. Sand Wet Endogleyic Folic Podzol Het Kamp, Schilde Young pinus forest

(former heathland)YPiCD Pinus sylvestris L. Loamy sand Dry Folic Podzol Withoefse heide, Kalmthout Young pinus forest

(former heathland)FHCD Pinus sylvestris L.

(Forested Heathland)Sand Dry Haplic Podzol Withoefse heide, Kalmthout Heathland recently

converted to forest

Temporary flooded soilsBeCWW Betula pubescens,

Quercus robur L.Wet Umbric Gleysol

(Dystric, Endoarenic)Grootbroek-Bree,Molenbeersel

Temporally flooded

AlFWW Alnus glutinosa (L.)Gaertn.

Wet Thaptohistic Gleysol(Humic, Hypereutric)

Koolhembos, Puurs Temporally flooded

a Texture class is based on laser-diffractrometry (Taubner et al., 2009).b FAO, ISRIC and ISSS (2006).c Langohr (2001).

K. Vancampenhout et al. / Soil Biology & Biochemistry 42 (2010) 220e233222

a carrier gas (initial temperature: 40 �C, heating rate 7�C min�1, finaltemperature: 320 �C, hold time: 20 min), coupled to a Fisons MD(Ipswich, UK) 800 mass spectrometer (m/z 45e650, cycle time 1 s).

Pyrolysis-GC/MS implies a controlled thermal volatilization ofextracted SOM followed by gas chromatography separation andmass spectrometry identification. The resulting pyrogram enablessemi-quantification and relative comparison of the identifiedpyrolysed substances (Buurman et al., 2007a,b). Up to 300 differentchemical compounds were identified in each pyrolysed sample.Two hundred and six of these components were quantified basedon the two characteristic ions of each chemical compound (Table 2;Masslab 1.2.7; Fisons, Ipswich UK) according to de Alcantara et al.(2004) and Buurman et al. (2009). This method of relative semi-quantification cannot be used to reflect absolute C quantities, but itis particularly suitable for the comparison of complex spectra in Py-GC/MS studies (de Alcantara et al., 2004; Buurman et al., 2009). Therelative abundance RI of each of the 206 compounds in everysample was calculated as follows:

RAij ¼�Xij=

XXi

�*100

where Xij ¼ integrated surface of peak for component j in sample iand SXi is the sum of all 206 integrated surfaces for sample i. Only

compounds with average relative intensity > 0.1% were retained ina reduced dataset of 128 compounds (Table 2).

It should be noted that the compounds in Table 2 are pyrolysisproducts of the molecules present in the extracted SOM. Not allmolecule types are equally susceptible the pyrolysis-GC/MSprocedure described above. E.g. abundances of certain N-contain-ing groups or highly polar compounds such as fatty acids aretypically underestimated (Dignac et al., 2006). A comparison ofspectroscopic methods for assessing humic acid samples is given inBuurman et al. (2009). This study therefore does not presentabsolute data on total SOM composition, but aims at a relativecomparison of the chemical properties of the extractable SOM ofthe different sites described above, in relation to environmental sitecharacteristics.

2.4. Factor analysis and statistical analysis

Factor analysis based on principal components without rotationusing Statistica 6 (StatStoft, Tulsa) was performed on relativeabundances of the 128 selected pyrolysis compounds of the soilorganic matter. In order to relate the site variables to the results ofthis factor analysis, Spearman r correlation coefficients (SPSS 15.0;SPSS Inc. Chicago) were calculated between the factor scores

Table 2List of Pyrolysis-GC/MS-compounds found in the extractable SOM and retained in the reduced dataset, containing compound code, compound name, masses used forquantification and average retention time (RT).

Comp.code

Comp. name Masses RT (av.) Comp.code

Comp. name Masses RT (av.)

Soil lipids 7:1 C7 alkene 55 þ 69 3.53 19 C19 alkane 57 þ 71 25.488:1 C8 alkene 55 þ 69 4.94 20 C20 alkane 57 þ 71 26.859:1 C9 alkene 55 þ 69 6.91 21 C21 alkane 57 þ 71 28.1510:1 C10 alkene 55 þ 69 9.08 22 C22 alkane 57 þ 71 29.4111:1 C11 alkene 55 þ 69 11.28 23 C23 alkane 57 þ 71 30.6112:1 C12 alkene 55 þ 69 13.38 24 C24 alkane 57 þ 71 31.7513:1 C13 alkene 55 þ 59 15.39 25 C25 alkane 57 þ 71 32.8614:1 C14 alkene 55 þ 69 17.29 26 C26 alkane 57 þ 71 33.9215:1 C15 alkene 55 þ 69 19.09 27 C27 alkane 57 þ 71 34.9316:1 C16 alkene 55 þ 69 20.78 28 C28 alkane 57 þ 71 35.9218:1 C18 alkene 55 þ 69 23.91 29 C29 alkane 57 þ 71 36.8622:1 C22 alkene 55 þ 69 29.31 31 C31 alkane 57 þ 71 38.6425:1 C25 alkene 55 þ 69 32.72 33 C33 alkane 57 þ 72 40.369 C9 alkane 57 þ 71 7.14 F14 C14 fatty acid 60 þ 73 23.2510 C10 alkane 57 þ 71 9.33 F15 C15 fatty acid 60 þ 73 24.7311 C11 alkane 57 þ 71 11.51 F16 C16 fatty acid 60 þ 73 26.1412 C12 alkane 57 þ 71 13.71 F18 C18 fatty acid 60 þ 73 28.8113 C13 alkane 57 þ 71 15.6 Al Alcohol (C20) 55 þ 83 30.9414 C14 alkane 57 þ 71 17.44 K27 n-C27:0 methylketone 58 þ 59 36.9615 C15 alkane 57 þ 71 19.26 K29 n-C29:0 methylketone 58 þ 59 38.7916 C16 alkane 57 þ 71 20.93 Pr Prist-1-ene 56 þ 57 23.0217 C17 alkane 57 þ 71 22.54 T Squalene 69 þ 81 36.1618 C18 alkane 57 þ 71 24.05

Aromatics A1 Benzene 77 þ 78 3.08 A12 C3 benzene e comp. 105 þ 120 9.64A2 Toluene 91 þ 92 4.44 A13 Indene 115 þ 116 10.08A3 Ethylbenzene/xylene 91 þ 106 6.17 A14 Methylbenzene e comp. 115 þ 130 12.38A4 Dimethylbenzene/xylene 91 þ 106 6.34 A15 1-Methyl-1H-indene 115 þ 130 12.37A5 Styrene 103 þ 104 6.73 A16 Benzene compound 129 þ 142 12.48A6 Ethylbenzene (p-xylene) 91 þ 106 6.82 A17 1,1 Dimethyl-1H-indene 129 þ 144 14.58A7 Propylbenzene 91 þ 120 8.14 Pa1 Naphthalene 128 13.08A8 C3 benzene e comp. 105 þ 120 8.3 Pa2 x-Methyl naphtalene 141 þ 142 15.35A9 C3 benzene e comp. 105 þ 120 8.46 Pa3 x-Methyl naphtalene 141 þ 142 15.67A10 Ethylmethylbenzene e comp. 105 þ 120 8.7 Pa4 Trimethyl naphthalene 155 þ 170 20.69A11 C3 benzene e comp. 105 þ 120 9.03

Polysaccharides Ps1 2-Methylfuran 53 þ 82 2.58 Ps13 Methylbenzofuran 131 þ 132 11.35Ps2 Acetic acid 60 2.43 Ps14 Methylbenzofuran 131 þ 132 11.44Ps3 (2H)-furan-3-one 54 þ 84 4.59 Ps15 Alpha-D-glucopyranoside compound 57 þ 69 12.85Ps4 2-Furaldehyde 95 þ 96 4.91 Ps16 Alpha-D-glucopyranoside compound 57 þ 69 13.14Ps5 3-Furaldehyde 95 þ 96 5.28 Ps17 Sugar monomer 60 þ 71 13.38Ps6 Acetylfuran 109 þ 110 6.91 Ps18 Anhydrocyclofuranose 57 þ 73 13.98Ps7 5-Methyl-2-furaldehyde 109 þ 110 7.94 Ps19 1,4-Dideoxy-D-glycero-hex-1-

enopyranos-3-ulose87 þ 144 14.88

Ps8 3-Hydroxy-2-methyl-2-cyclopenten-1-one

55 þ 112 9.35 Ps20 Levogalactosan 60 þ 73 16.21

Ps9 Dianhydrorhamnose 113 þ 128 9.61 Ps21 Levomannosan 60 þ 73 17.45Ps10 2-Propan-2-one tetrahydrofuran 57 þ 72 10.57 Ps22 Levoglucosan 60 þ 73 18.16Ps11 Levoglucosenone 68 þ 98 10.93 Ps23 Dibenzofuran 139 þ 168 19.47Ps12 Maltol 71 þ 126 11.28

Lignin monomers G1 4-Methoxyphenol (guaiacol) 109 þ 124 10.85 S1 2,6-Dimethoxyphenol (syringol) 139 þ 154 16.07G2 4-Methylguaiacol 123 þ 138 13.1 S2 4-Methylsyringol 153 þ 168 17.88G3 4-Ethylguaiacol 137 þ 152 14.87 S3 4-Vinylsyringol 165 þ 180 19.95G4 4-Vinylguaiacol 135 þ 150 15.51 S4 4-(Prop-1-enyl)syringol 181 þ 182 21.48G5 4-Methylguaiacol 107 þ 138 15.57 S5 4-(Prop-2-enyl)syringol trans 91 þ 194 22.18G6 4-(1-Propenyl) guaiacol 77 þ 164 16.36 S6 4-Acetylsyringol 181 þ 196 22.54G7 4-Formylguaicol 151 þ 152 16.89 S7 4-(Propan-2-one) syringol 167 þ 210 23.06G8 trans-4-(2-Propenyl)guaiacol 77 þ 164 18.06G9 4-Acetylguaiacol 151 þ 166 18.51G10 Guaiacylacetone 137 þ 180 19.94G11 Guaiacol-COOH vanillic acid 153 þ 168 20.91

Phenols Ph1 x-Methylphenol 107 þ 108 5.51 Ph6 3/4-Ethylphenol 107 þ 122 12.79Ph2 Phenol 66 þ 94 8.57 Ph7 4-Vinylphenol 91 þ 120 13.62Ph3 2-Methylphenol 107 þ 108 10.16 Ph8 Methoxytrimethylphenol 123 þ 166 19.36Ph4 3/4-Methylphenol 107 þ 108 10.6 Ph9 Tetramethylbutylphenol 107 þ 135 21.26Ph5 Dimethylphenol 107 þ 122 12.16

N-compounds N1 (1H)-Pyrrole, dimethyl 95 þ 96 3.56 N7 x-Dimethylpyridine 106 þ 107 7.57N2 Pyridine 52 þ 79 4.03 N8 Benzonitrile 76 þ 103 8.34N3 Pyrrole 67 4.1 N9 Indole 90 þ 117 15.02N4 x-Methylpyridine 66 þ 93 5.23 N10 1H-Indole-3-ethanamide 130 þ 131 16.83N5 x-Methyl-1H-pyrrole 80 þ 81 5.48 N11 Diketodipyrrole 93 þ 186 22.3N6 x-Methyl-1H-pyrrole, 80 þ 81 5.72 N12 Alkylamide 59 þ 72 31.2

K. Vancampenhout et al. / Soil Biology & Biochemistry 42 (2010) 220e233 223

Table 3NaOH-extraction yield, dominant tree species, properties of the standing biomass, litter layer characteristics, soil properties, soil carbon content and indications for microbialactivity for each of the selected study sites (codes as in Table 1).

YPiCD YPiCW PiCD PiC(W) FaCD FaCW FaFD FaFW QuCD QuCW QuFD QuFW PoCD PoCW PoFD PoFW PoFW(2)

NaOH-extr. yield (%C) 71 61 70 80 67 80 65 81 81 92 41 68 86 68 52 48 35

Domin. tree sp. Pine Pine Pine Pine Beech Beech Beech Beech Oak Oak Oak Oak Poplar Poplar Poplar Poplar PoplarBasal are (m2 ha�1) 41.2 26.2 33.2 37.7 27.1 53.3 35.5 28.1 43.6 41.5 33.8 53.2 n.a. 35.9 34.2 16 22.2Stem number 944 492 483 629 218 522 256 682 792 463 934 936 n.a. 405 189 109 600Herb cover (%) 16 100 100 32 64 6 12 84 12 100 0 100 85 84 100 100 66Forest age class 4 3 2 1 2 3 1 1 3 1 1 4 4 1 4 4 4Even-aged stand Yes Yes Yes No Yes Yes No No No No No No No No Yes Yes No

Humus index 8 9 9 8 5 8 8 5 6 7 7 4 5 8 1 1 2Thickness litter layer (cm) 9.3 11.7 10 1.5 7 9.7 4.7 10 5.8 10 6 1 2.2 n.a. 0.5 0.5 1Dry mass litter layer (kg m�1) 10.1 8.6 3.7 2.7 4.4 10.3 3.5 8.4 4.6 12.1 4.8 1.3 2.1 n.a. 0.3 0.2 0.7

Moisture class c d b c d d b e c d b e d e d f eClay (%) 2.9 2.7 6 7.8 9.1 9.9 19.8 19.6 8.9 8.5 16.8 18.6 6.2 14.7 14.3 33.4 18.6Silt (%) 11.4 9 16 10.4 25.8 27.5 72.5 35.9 15.9 27.8 72 64.8 22.6 27.2 72.2 53 75.9Sand (%) 85.7 88.4 77.9 81.8 65.1 62.7 7.6 44.5 75.2 63.7 11.2 16.7 71.2 58.1 13.5 13.7 5.4pH (H2O) 4 3.8 3.6 3.8 3.8 3.5 3.8 3.7 3.7 3.4 4.3 3.9 8 5.3 6.1 5.9 6.4N (%) 0.11 0.07 0.49 0.06 0.07 0.18 0.36 0.22 0.25 0.3 0.13 0.37 0.18 0.29 0.19 0.5 0.62N (Mg ha�1 cm�1) 0.02 0.01 0.06 0.01 0.01 0.02 0.04 0.03 0.08 0.04 0.02 0.05 0.02 0.04 0.02 0.06 0.06

C (%) 1.44 1.85 7.98 1.95 2.4 4.19 8.82 5.68 4.5 6.08 2.08 6.15 1.45 5.04 1.85 8.24 11.44C (Mg ha�1 cm�1) 0.22 0.27 0.92 0.27 0.34 0.55 0.97 0.70 0.51 0.74 0.28 0.74 0.13 0.63 0.24 0.92 1.14C/N ratio 13 26.5 16.2 34.5 32.8 22.8 24.5 25.4 17.6 20.5 16 16.5 8.1 17.2 9.7 16.4 18.3

CO2 resp. (mg m�2 h�1) 318 338.2 270.4 483.1 458 413.9 311.2 398 301.4 359.9 n.a. 449.7 n.a. 415.2 n.a. 423.9 521.1HWC (mg g�1) 1113.2 737.1 750.4 771.2 1058.7 1031.8 1179 677.2 1268.4 1112.3 n.a. 1455.1 n.a. 1433.8 n.a. 1435.7 1462.3HWC (% of TOC) 4.8 3.7 1.9 5.2 3.9 4.4 3.5 4 3.7 2.7 n.a. 2.5 n.a. 3.9 n.a. 2.3 2.5

K. Vancampenhout et al. / Soil Biology & Biochemistry 42 (2010) 220e233224

(first tree factors) and the values for ordinal and scale environ-mental variables. For class environmental variables, ANOVAF-values (SPSS 15.0; SPSS Inc. Chicago) were calculated, comparingthe mean factor scores for each class.

3. Results

3.1. Site characteristics

Results are summarized in Table 3 and Figs. 1 and 2. The specificecological requirements of each tree species and the effect of litterquality on both soil properties and soil microbial activity cause highinter-correlations among site characteristics, which is a typical

Fig. 1. Average total carbon density (black bars), average total nitrogen density (grey bars) anand Populus. Error bars denote the standard deviation.

feature of most temperate forests even when they are planted byman (Neirynck et al., 2000; Augusto et al., 2002; Hagen-Thorn et al.,2004; Jandl et al., 2007). Thickness and mass of the litter layer areobviously correlated, as well as sand, silt and clay content, total C,total N and C/N. In this dataset, strong correlations (p < 0.01) existbetween humus index and soil moisture and between soil textureand carbon content. For carbon and nitrogen contents results weresimilar when expressed as mass percentage or as density(Mg ha�1 cm�1).

On average, pine forests in this study have thin trees (high stemnumber relative to basal area) with a high herb cover. As expected,soils under pine are drier and sandier, low in N, low in C and havea high C/N ratio (Fig. 1, note the substantial standard deviations),

d average C/N ratio (white bars) of study sites under dominance of Pinus, Fagus, Quercus

Fig. 3. Factor analysis of the reduced dataset: factor loadings of the pyrolysiscompounds on the first three ordination axes. Compound codes as in Table 2. F1opposes aromatics and short and mid-chain aliphatics (relatively recalcitrant) to lignin(especially syringols) and N-compounds.

Fig. 2. Average thickness (white bars) and dry mass (hatched bars) of the litter layer,average CO2 respiration (black bars) and average hot water carbon content (HWC; greybars) of study sites dominated by Pinus, Fagus, Quercus and Populus. Error bars denotethe standard deviation.

K. Vancampenhout et al. / Soil Biology & Biochemistry 42 (2010) 220e233 225

thick and heavy litter layers, mor-type humus and low hot waterextractable carbon (HWC, Fig. 2). Average HWC increases uponincreasing litter quality (Cornelissen, 1996; Van Calster et al., 2007)and is significantly (p< 0.05) higher in samples taken under poplar(1444 � 15.9 mg g�1) as opposed to samples taken under pine orbeech (915 � 200 mg g�1). Also, a significant difference in HWCcontent (p < 0.05) was found between samples taken under pine(843 � 181 mg g�1) and under oak (1278 � 172 mg g�1). The averagesoil respiration is not significantly different among dominant treespecies.

On average, the beech stands are darker, have heavier adulttrees, thick and heavy litter layers and mor or moder type humus.The soils are low in N, have a high C/N ratio and lower microbialcontribution than oak and poplar stands. Oak stands in this studyon average are relatively dense (i.e. have the highest average basalarea and stem number), yet have a higher herbal cover. The litterlayer of these stands is less thick and consists of mull- to moder-type humus. Soils under oak on average contain more N, havea lower C/N and more microbial carbon than soils under beech.

Poplar stands considered in this study are typically very open, asreflected in the lower stem number and basal area and high herbcover of a usually grassy undergrowth. A significant difference inthickness of litter layer was observed between poplar (1.8� 0.8 cm)compared to pine and beech (8� 3.4 cm, p< 0.05, Fig. 2). The sametrend is observed concerning the dry mass of the litter layer,although not significant. Soils under poplar show highest soil-moisture conditions and clay content significantly varies amongsoil-moisture classes (p < 0.01). Average pH was significantlyhigher under poplar (6.3 � 1) vs. other tree species (3.8 � 0.2,p < 0.05). Average N content is lower and C/N higher for pine andbeech contrary to oak and poplar but standard deviations remainconsiderably high. C/N is significantly higher under beech(26.4� 4.4) as opposed to poplar (13.9� 4.7, p< 0.05). Noteworthyis the fact that C/N was very low in drier poplar stands but washigher with increasing soil moisture (Table 3). Microbial biomassand respiration are highest under poplar. For Quercus and Poplar, Cstocks appear higher in sites with high soil moisture.

3.2. General extractable-SOM chemistry

The proportion of total soil carbon extractable by 0.1 M NaOHvaried between 35 and 92% and is listed for each site in Table 3.Typically, this fraction varies considerably between soil types (Olk,2006) and lowest extraction yields were observed in soils with highclay contents (von L€utzow et al., 2007). Most undecomposed plantmaterial and polyvalent cation bridges in organo-mineral complexesare not dissolved by NaOH extraction (von L€utzow et al., 2007).

The pyrolysis products retained in the reduced dataset are listedin Table 2. Pyrolysis compounds were grouped according to theirprobable origin and chemical similarity. They include a range ofalkenes (n:1) and alkanes (n); pristene (Pr); four fatty acids (Fn);two methylketones (Kn); an alcohol (Al); a terpene (squalene, T;Tegelaar et al., 1989b; van Bergen et al., 1997; Buurman et al., 2005,2007a,b); aromatic compounds (A) containing benzene and indenerelated structures, toluene and styrene; four polyaromaticcompounds (Pa; Saiz-Jimenez, 1994; Nierop et al., 2001b); (poly)saccharide-derived pyrolysis products (Ps) including furans,

K. Vancampenhout et al. / Soil Biology & Biochemistry 42 (2010) 220e233226

furaldehydes and monomeric sugars (Moldoveanu, 1998; Pageet al., 2002), lignin compounds derived from plant lignins of theguaiacol (G) and the syringol (S) types (Hedges and Mann, 1979;Saiz-Jimenez and de Leeuw, 1986b; Moldoveanu, 1998; K€ogel-Knabner, 2002; Killops and Killops, 2005); nine phenol products(Ph) with various substituents (Ralph and Hatfield, 1991; Moldo-veanu, 1998; Nierop et al., 1999); and twelve nitrogen-containingcompounds (N; Schulten and Schnitzer, 1997; van Bergen et al.,1998a; Hatcher et al., 2001).

3.3. Factor analysis

The first three factors explain 37.7%, 17.6% and 9.3% respectively,of the observed variation. Factor loadings of the variables (i.e. thepyrolysis compounds) and scores of the cases (i.e. the samples) areshown in Figs. 3 and 4.

In the F1eF2 space (Fig. 3), short and mid-chain alkanes plottogether with most alkenes, methylketones and the C20 alcoholand have strongly negative scores on F1. The C15 alkane and alkeneplot outside this cluster. The long-chain alkanes (C28, C29 and C31)at the top of the diagram plot close to levoglucosan (Ps 22). Within

Fig. 4. Factor analysis of pyrolysis compounds (reduced dataset): factor scores of thesamples on the first three ordination axes (codes as in Table 1). A gradient in dominanttree species can be observed along F1.

the alkanes, most samples have a clear dominance of the C15member and to a lesser extent the C12 member (Fig. 5aec). Underpine, the dominance of the C15 alkane seems to increase in olderpine forests. Also samples under oak with low soil moisture havea clear dominance of C15 alkane, especially QuFD. This dominanceis not prominent in Fagus samples, where C12 seems moreimportant. Drier poplar stands also show C15 dominance. TheC15:1 alkene shows a clear dominance in the young forests, i.e. sitesQuFD, YPiCD, YPiCW, FHCWand PoCD. The C27 alkene, is importantin site FaFD (Zoni€en). In Fig. 3a, this compound has low positiveloadings on both F1 and F2 and as a result hardly differentiatessamples.

The heavier aromatics and the polyaromatics Pa1 and Pa3e4plot together with the short andmid-chain alkane cluster describedabove, except for A16 (benzene compound) and A17 (dimethy-lindene) found in the upper right quadrant. The lighter aromaticsA1e8 and Pa2 have more negative loadings on F2 and includetoluene. Most polysaccharide-derived compounds plot in the upperright quadrant, with levoglucosan (Ps22, found in the alkane/alkene cluster) as a notable exception and the benzofurans(Ps13e14 and 23), which have low loadings on F1 but stronglynegative ones on F2. N-containing compounds have high loadingson F1, but weak negative ones on F2 and plot at far right, with theexception of benzonitrile (N8), which plots with the benzofuransand alkylamide (N12), which plots with levoglucosan (Ps22) at thetop of the diagram. The other levosugars (Ps20 and Ps21) havemuch lower positive loadings on F2.

Syringols, guaiacols and phenols are found in the lower rightquadrant of the F1eF2 space (Fig. 3). Especially syringols and mostphenols (Ph2e7) have high positive loadings on the first axis.

Factor 3 explains only 9.3% of all variation. Strongly positivescores are found for guaiacols (Fig. 4.3b) and strongly negative onesfor some aromatics (i.e. A2 e toluene, A5e styrene), 2-methylfuran(Ps1), benzonitrile (N8) and some N-compounds, notably N2(pyridine).

In the factor scores diagram of Fig. 4, a clear gradient in domi-nant tree species can be observed along F1, ranging from forestedheathland at the far left, through young pine forest, old pine forest,beech forest, oak forest to poplar forest at the far right. Young pinestands and forested heathland have the most negative scores.Samples taken under beech (except for FaFD) have slightly negativescores on F1, while samples taken under oak have slightly positivescores. The second factor F2 differentiates between sample QuFDon the one hand (highest positive score) and samples taken underoak on coarse textured soils and beech (FaFD, FaFW, FaCD, FaCW,QuCW and QuCD, highest negative scores) on the other. The F1eF3diagram (Fig. 4) differentiates mainly between the sample takenunder forested heathland and the other samples, but the separationof tree species remains intact. Most poplar stands have negativescores as well.

3.4. Correlation of extractable-SOM chemistry to site characteristics

F1 is significantly correlated with tree species, litter layer char-acteristics, soil properties and HWC (Table 4). Pine stands on dry,sandy soils with considerable litter accumulation in the litter layerand mor-type humus are correlated with negative values on F1.These stands have low soil pH, low soil N content and low HWCcontent. Broadleaved stands with mull humus types on rich soilswith high microbial contribution (HWC) are positively correlatedwith F1.

F2 and F3 in general are poorly correlated to the measuredenvironmental factors. F2 is negatively correlated with the mass ofthe litter layer alone. Finally, F3 is correlated with dominant treespecies, humus index, soil moisture, pH and respiration. Poplar

Fig. 5. Alkane distribution in (a) pine stands, (b) beech and oak stands, (c) Poplar stands and (d) temporary flooded sites.

K. Vancampenhout et al. / Soil Biology & Biochemistry 42 (2010) 220e233 227

stands with high soil moisture and high CO2 respiration have mostnegative scores on this axis.

4. Discussion

4.1. Interpretation of pyrolysis compounds

As pyrolysis-GC/MS generates data on pyrolysis products of themolecules present in the extracted SOM, comparison of thesecompounds to literature is necessary to verify their origin (Buur-man et al., 2009).

Short and mid-chain aliphatic compounds in pyrograms of soilsmainly originate from aliphatic biopolyesters in the SOM, derivedfrom plant lipid precursors (such as suberan and cutan) or micro-bial lipids. According to Buurman et al. (2006, 2007b), alkanes andalkenes with chain length 12e28 which cluster together in factoranalysis should be considered relatively recalcitrant degradationproducts of vegetation, rather than microbial polymers. Long-chainalkanes (C27, C29, C30, C31) on the other hand originate fromrelatively undecomposed vegetal precursors (plant waxes) and areeasily broken down in shorter chains (Eglinton and Hamilton, 1967;Saiz-Jimenez and de Leeuw, 1987; Walton, 1990; K€ogel-Knabneret al., 1992a,b; Saiz-Jimenez, 1995; Tegelaar et al., 1995; Dove et al.,1996; van Bergen et al., 1997; Moldoveanu, 1998; Nierop, 1998; vanBergen et al., 1998b; Nierop and Buurman, 1999; Nierop et al.,2001b; Killops and Killops, 2005; Buurman et al., 2005, 2007b,2009; Hajje and Jaffe, 2006; Vancampenhout et al., 2009; Mikuttaet al., 2006). A dominance of the C15, C12 and C14 homologuesimilar to the one observed in this studywas described by Buurmanand Roscoe (2008) and Ferreira et al. (2009). The origin of this effectmay be microbial but remains uncertain.

Fatty acids are usually underestimated in pyrolysis-GC/MS dueto their high polarity and decarboxylation during pyrolysis(Saiz-Jimenez, 1994). Nevertheless, all pyrograms show a domi-nance of F16, typical for plant-derived humic material (Buurmanet al., 2009).

Aromatic pyrolysis products originate from SOM compoundsderived from proteins (toluene), tannins and other (poly)phenolsincluding black carbon (Saiz-Jimenez and de Leeuw, 1986; Ralphand Hatfield, 1991; Saiz-Jimenez, 1994; Maie et al., 2003). Cycliza-tion of decarboxylated fatty acids during pyrolysis can also yieldalkylbenzenes (Buurman et al., 2009). As for the short and mid-chain aliphatics, aromatics are considered relatively recalcitrant to(further) microbial decay (Saiz-Jimenez and de Leeuw, 1987; vonL€utzow et al., 2006; Lorenz et al., 2007).

Polysaccharide-derived pyrolys-products include furans, fur-aldehydes, benzofuranes and monomeric sugar molecules, whichrelate to SOM molecules with both vegetal and microbial precur-sors. Ps9, Ps11 and Ps15-22 (and particularly the levosugars) arecommonly related to biopolymers with cellulose-type precursors,referring to relatively weakly-decomposed, plant-derived SOM.Smaller polysaccharide pyrolysis products are mostly associatedwith microbial material (Ps1, Ps2, Ps4-6). Benzofurans are likewiseassociated with more strongly decomposed SOM material, denot-ing charring or microbial influence (Saiz-Jimenez and de Leeuw,1987; K€ogel et al., 1988; Ralph and Hatfield, 1991; Sollins et al.,1996; Moldoveanu, 1998; Hatcher et al., 2001; Page et al., 2002;Helfrich et al., 2006; Vancampenhout et al., 2009).

Syringols and guaiacols are lignin-derived compounds. SOMfound under coniferous vegetation typically shows a clear domi-nance of lignin compounds related to guaiacol (coniferyl derived),while broadleaved forests have SOM with both guaiacol and

Table 4Spearman r correlation coefficients (r; calculated between the factor scores ofsamples on the first three factors and values for ordinal and scale environmentalvariables) and ANOVA F-values (calculated for class environmental variables,comparing means between each variable class and the factor scores), relating sitevariables to the results of the factor analysis in Fig. 4.

Factor 1 Factor 2 Factor 3

F r F r F r

Forest managementDominant tree species 16.56** 1.67 5.17*Basal area �0.08 �0.23 0.44Stem number �0.34 0.24 0.37Cover herb layer 0.25 0.15 �0.13Forest age class 2.50 0.20 0.82Even-aged or uneven-aged 3.29 0.11 1.01

Litter layerHumus index L0.69** �0.08 0.49*Thickness of litter layer L0.76** �0.42 0.35Dry mass of litter layer L0.69** L0.50* 0.31

Soil propertiesMoisture class 0.55* �0.13 L0.62**Clay percentage 0.65** 0.02 �0.35Silt percentage 0.69** 0.00 �0.31Sand percentage L0.73** �0.04 0.39pH (H2O) 0.59* 0.44 L0.56*N (%) 0.67** 0.04 0.07N (Mg/ha) 0.55* �0.02 0.12

Soil carbonC (%) 0.47 �0.07 �0.05C (Mg/ha) 0.36 �0.08 0.01C/N ratio �0.41 �0.35 0.07

Microbial activityCO2 respiration 0.35 0.15 L0.73**HWC 0.77** 0.24 �0.27HWC/total C ratio �0.49 �0.33 �0.05

**Correlation is significant at 0.01 level.*Correlation is significant at 0.05 level.

Fig. 6. Ratios of the C3-lignin compounds (C3G and C3S) to unsubstituted guaiacol(G1) and syringol (S1), indicative for lignin degradation status. Black bars denote theratio for guiaiacol-components; grey bars denote the ratio for syringol compounds forthe samples indicated on the X-axis. Samples taken under coniferous vegetation do notcontain syringol-components.

K. Vancampenhout et al. / Soil Biology & Biochemistry 42 (2010) 220e233228

syringol (sinapyl derived) compounds (Hedges and Mann, 1979;Killops and Killops, 2005). Althoughmost literaturementions fasterside chain degradation in syringols than in guaiacols (Nierop et al.,2001b; Chefetz et al., 2002), this does not appear to be reflected inthe present study: Fig. 6 clearly shows higher C3S/S1 than C3G/G1ratios, where C3G and C3S are the lignins with intact side chains asopposed to S1 and G1, which are devoid of a side chain in the paraposition (Zech et al., 1992).

Phenols in pyrolysis can have a variety of origins includinglignins, carbohydrates, proteins or polyphenols (Ralph and Hatfield,1991; Moldoveanu, 1998; van Bergen et al., 1998b; Lobe et al.,2002). 4-Vinylphenol (Ph7) may be a pyrolysis product of grass-lignin (Saiz-Jimenez and de Leeuw, 1986; Tegelaar et al., 1989a;Ralph and Hatfield, 1991).

Little is known about the origin of N-containing compounds inSOM. They may originate from vegetal or microbial precursors, ormay be formed by chemical interactions (Schulten and Schnitzer,1997; van Bergen et al., 1998a). Abundance of N-compunds in SOMin non-fertilised soils is linked to highmicrobial influence (Schultenet al., 1992; Gleixner et al., 1999; Nierop et al., 2001b; Grandy et al.,2009). Benzonitrile (N8), plotting with the benzofurans in thefactor analysis, may denote charring or microbial influence (Buur-man and Roscoe, submitted for publication; Vancampenhout et al.,2008, 2009). Buurman et al. (2007a) associated 3-methyl(1H)pyrrole (N6), pyridine (N2), pyrrole (N3) and benzonitrile (N8)with relatively degraded SOM and high microbial input, while3/4-methylpyridine (N4), indole (N9) and alkylamides (N12)were associatedwith relatively fresh SOM. The position of the latterin the F1eF2 factor space, close to levoglucosan (Ps22) and

the long-chain aliphatics, may therefore confirm the findings ofBuurman et al. (2007a). On the other hand, alkylamides and levo-glucosan are sometimes attributed to microbial organic matter(Moldoveanu, 1998; van Bergen et al., 1998a; Buurman and Roscoe,submitted for publication), which may explain why the otherlevosugars (Ps20 and Ps21) have much lower positive loadings onF2 than levoglucosan.

4.2. Interpretation of the factor analysis

Considering the position of major compounds in the F1eF2factor space and the origin of the pyrolysis compounds discussedabove, Fig. 3 can be interpreted as follows: samples with stronglynegative scores on the first factor F1 contain (i) a relative accu-mulation of aromatics and short and mid-chain aliphatics, whichare fairly recalcitrant and indicative of degraded plant material;(ii) a relative absence of lignin (especially syringols indicativefor broadleaved vegetation) and (iii) a relative absence of Ncompounds. Strongly positive loadings on this axis on the otherhand denote (i) the abundance of N-compounds, most likely linkedto high microbial influence, and (ii) abundance of lignins and mostof the polysaccharide compounds. The high loading of 4-vinyl-phenol (Ph7) on F1 may relate to grass-lignin.

Strongly negative loadings on F2 in combination with negativeloadings on F1 indicate a relative abundance of possible markers ofcharring (benzofurans, benzonitrille, polyaromatics), while thelower right quadrant of the F1eF2 space represents a high abun-dance of lignin fragments, partly in combination with N-com-pounds. The top centre of the F1eF2 space in Fig. 4.3 representscompounds both indicative for relatively fresh (Ph8, alkanes 28e31and possibly Ps22 and N12) and microbial (Ps3, alkane 15 andpossibly Ps22 and N12) material, suggesting an admixture of both.

Strongly positive scores on factor F3 are found for guaiacols(Fig. 4.3b, related to angiosperm-derived lignin) Strongly negativeones are found in the lower left quadrant for some aromatics

K. Vancampenhout et al. / Soil Biology & Biochemistry 42 (2010) 220e233 229

(i.e. A2 e toluene, A5 e styrene), 2-methylfuran (Ps1) and benzoni-trile (N8), all indicative for degraded material, while smaller poly-saccharide- andN-compounds (notably pyridine (N2)) dominate thelower right quadrant, suggesting a high microbial contribution.

Factor scores in Fig. 4 emphasise a strong effect of dominant treespecies on extractable-SOM chemistry (F1). Former heathlandstands are still clearly distinguisable. Pine and former heathlandcontain high relative amounts of degraded, fairly recalcitrantaromatics and short and mid-chain aliphatics, while N-compoundsprevail under poplar, suggesting high microbial influence. Asconiferous vegetation does not have syringol-type lignin precur-sors, syringol compounds have strongly positive scores on F1.4-Vinylphenol (Ph7) is abundant under poplar as well, confirmingits relation to the typical grassy undergrowth under open poplarstands.

The second factor F2 stresses the deviating composition ofsample QuFD (containing an admixture of both relatively fresh anddecomposed material) compared to samples taken under oak onlight textured soils and under beech, which are richer in lignins. Inthe F1eF3 diagram (Fig. 4b), the positions of the sample takenunder forested heathland (high in strongly degraded material) andthose taken under poplar stands with high soil-moisture regimes(high in smaller N-components) differentiate from the others.

4.3. Influence of environmental factors

Despite the high intercorrelation between the environmentalfactors, the influence of dominant tree species on F1 is striking(Fig. 4). Moreover, species are arranged along F1 according to theirlitter quality (ranging from pine over beech, oak and finally poplar;Cornelissen, 1996; Aerts and Chapin, 2000; Van Calster et al., 2007;De Deyn et al., 2008), indicating that input quality is an importantdeterminant of the extractable-SOM composition in maritimetemperate forests. Both aboveground leaf litter as belowgroundroot input may contribute to this effect (Pollierer et al., 2007; Millar,1974), but cannot be differentiated in this study. Muys (1995),Hobbie et al. (2006) and Vancampenhout et al. (2009) evidence theimportance of litter quality for SOM dynamics and many authorsdiscuss the effect of overstory species on the soil's biologicalcommunity (Bauzon et al., 1969; Tyler, 1992; Mardulyn et al., 1993;Baldock et al., 1997; H€attenschwiler et al., 2005; Derry et al., 1998;Gholz et al., 2000; Priha et al., 2001; Bardgett et al., 2005; Mayer,2008). Besides the choice of dominant tree species, standingbiomass management options apparently had no significant influ-ence on extractable-SOM chemistry, which is in line with the weakeffect of silvicultural practices on the humified fractions intemperate beech forests reported by Hedde et al. (2008). SampleFaFD had a somewhat better litter quality than most beech forestsdue to the presence of a calcareous subsoil (Langohr and Sanders,1985; Langohr, 1986; Van Hemelrijck et al., 2004), which may causeits deviating extractable-SOM composition in Fig. 4.

An important accumulation of mid-chain alkenes/alkanes andaromatics is observed in the extractable SOM under pine forests.This accumulation coincides with a strong degradation of lignin-and cellulose-derived compounds (Fig. 3) and associated toecosystems with poor litter quality, moder-type humus with thickand heavy litter layers, having low soil pH, low soil N content andlow HWC content (Table 4). Several authors evidence the negativeeffect on decomposition rate of poor input quality on one hand(Anderson, 1991; Cornelissen, 1996; Baldock et al., 2004; Hameret al., 2004; Polyakova and Billor, 2007; Barbier et al., 2008) and lowlevels of nitrogen and low pH on the other (van Bergen et al., 1998b;von L€utzow et al., 2008). Moreover, the accumulation of recalci-trants (aromatics and short and mid-chain aliphatics) appearsa common feature in soils where not all conditions for efficient

decay are met (due to substrate limitations, occlusion, acidity, toxicions, complexation, sorption, low fertility or oxygen deprivation;Sollins et al., 1996; Baldock et al., 1997; Helfrich et al., 2006;Buurman and Roscoe, submitted for publication; Ferreira et al.,2009; K€ogel-Knabner et al., 2008a,b; Marschner et al., 2008) Underthese conditions, stoichiometric constraints hamper the decom-position of recalcitrant compounds in favor of more easilydecomposable material, such as lignins and cellulose (Sollins et al.,1996; Kiem and K€ogel-Knabner, 2003; Hessen et al., 2004; Crosset al., 2007; Marschner et al., 2008; Vancampenhout et al., 2009)hence explaining the accumulation evidenced in Fig. 3 and itscorrelation to humus type, litter layer characteristics and microbialcontribution (HWC). Fig. 6 likewise indicates stronger side chaindegradation in guaiacols under pine. Soils under forested heathlandand pine moreover have very low relative amounts of N-compounds (Figs. 3 and 4).

Oppositely, no accumulation of recalcitrants is present in theextractable SOM under poplar, which coincides with high abun-dance of N compounds, polysaccharides and syringols (Figs. 3 and 4).Although N-compounds may have different origins, enrichmentof N-compounds in unfertilized soils is generally linked to highmicrobial influence by microbial re-synthesis (van Bergen et al.,1998a; von L€utzow et al., 2006; Ekschmitt et al., 2008; Marschneret al., 2008) especially in combination with large contents of poly-saccharide products (except levosugars; Ferreira et al., 2009). This isfurther evidenced by the high positive correlation of F1 to microbialproxies (HWC) and to ecosystems with high litter quality, mull-typehumus and thin and light litter layers (Table 4). Kuzyakov et al.(2000), Hamer et al. (2004) and Marschner et al. (2008) mentionpositive priming in soil decomposition processes, where the pres-ence of energy rich substrates favors the decomposition of recalci-trants. Remarkably, the soil's nutrient status shows strongercorrelation to the decomposition of recalcitrants than oxygendeprivation (Table 4). This is further illustrated in the extractable-SOM chemistry of BeCWW (on oligothropic and acid soil) andAlFWW (on rich alluvial soil). The extract obtained from BeCWWwas extremely rich in aromatics and aliphatic biopolymers. Espe-cially the long alkanes C23, C25e29 and C31 and the clear odd-over-even dominance in the longer alkanes (Fig. 5d) indicate that theoriginal plant-derived alkane chemistry remains almost intact.Extractable SOM originating from AlFWW on the other hand waschemically very similar to that of the non-flooded poplar standsdescribed in the factor analysis.

No general relation of extractable-SOM chemistry with timeunder forest was found, yet a clear effect of age was observed inpine forests (Fig. 4). Forested heathland and the younger pineforests (forested since 1940 or longer) havemore negative scores onF1, as compared to PiC(W) and PiCD (permanently forested at leastsince 1775 and 1850, respectively). This indicates that the formerheathland vegetation still has a detectable influence on theextractable-SOM chemistry under pine forests established for 60years or more (i.e. an abundance of relatively recalcitrant aromaticsand mid-chain/short aliphatics). Under poplar, beech and oakhowever, the period under forest does not appear to have mucheffect on extractable-SOM chemistry. Generally, chemical recalci-trance is assumed to be mainly important only during the firstdecade of SOM decomposition (von L€utzow et al., 2006, 2008). Thisstudy however indicates that e for the extractable fraction-chem-ical recalcitrant compounds can persist after land-use changes fortime scales much larger than a decade if conditions are oligotrophicand substrate quality is poor.

In the extractable SOM from forested heathland and young pinestands, the prominence of smaller aromatics, polyaromatics andaliphatics could be related to the regular incidence of fires inheathland (Vandewiele, 2001; Knicker et al., 2006) as both FHCD

K. Vancampenhout et al. / Soil Biology & Biochemistry 42 (2010) 220e233230

and YPiCD show a particularly high abundance of compoundsrelated to charring (benzofurans (Ps13, 14 and Ps23) and benzoni-trile (N8, Fig. 3)).

No correlations could be established between the extractable-SOM composition and total carbon contents and carbon densities ofthe study sites. Hence, quantitative total carbon measurements donot address the chemical quality of the extractable-SOM fraction.Remarkably, the C/N ratio shows no correlation to any of the axesdespite its well-known effects on turn-over. This lack of correlationcan however be attributed to high C/N ratios found both under pineand beech on one hand, and poplar stands with high soil-moistureregimes on the other (Table 3).

Factor F2 shows little correlation to the environmental factors inTable 4. It is strongly influenced by the deviating composition ofsample QuFD, compared to the other oak stands. It has an unusuallyhigh contribution of alkanes C15, C28-31, high relative contributionof Ps22 and N12 and unusually low in lignins and N-compounds(vide supra, Figs. 3 and 4). The sampling site was part of Zoni€en-forest from1775 onward, but was intensivelymanaged for 100e150years in the 17th century by the nearby abbey. This history stillaffects the site's understory vegetation andC/N ratio today (Langohr,2001). Spiegelberger et al. (2006) mention a similar long-lastingeffect (more than 70 years) of previous management on vegetation,and describe lasting effects on the soil microbial biomass. Whetheror not the QuFD site history is responsible for its deviating extract-able-SOM composition cannot be concluded in this study. Finally,oak stands on sandy soils are more similar to the beech group inFig. 4. In these forests, secondary tree species include beech andlarch (Larix spp.), while oak stands on heavier textured soils weremixedwith beech, hornbeam (Carpinus betulus L.) and alder (DeVos,1998).

F3 differentiates the composition of the forested heathlandsample (FHCD) and samples from poplar stands with high soilmoisture (PoXW, strongly negative scores). The latter explains theaxis' strong correlation to soil-moisture content. The correlation toCO2 respiration is most probably due to temporarily drying of theseotherwisewet soils during themeasuring period (August), inducinga sudden increase in respiration (De Deyn et al., 2008; Knohl et al.,2008). Notably, some aromatics (i.e. A2 e toluene, A5 e styrene),benzonitrile (N8) and 2-methylfuran (Ps1) are high in the forestedheathland, emphasising again the degraded status of the extractableSOMunder formerheathland,while somesmallerN-compoundsarehigh in wet poplar stands (pyridine compounds N2 and N7).

5. Conclusion

The results of this study indicate that, if climatic conditions arecomparable, vegetation is a major factor determining the extract-able-SOM composition in maritime temperate forests. Theseobservations further elaborate the relations found betweenecosystems and extractable-SOM chemistry on a global scale byVancampenhout et al. (2009). Nevertheless, soil nutrient status waslikewise important and small-scale site factors such as site history,soil texture and soil-moisture gain importance. Apart from treespecies, management choices regarding the standing biomass didnot significantly affect extractable-SOM chemistry.

Furthermore, aliphatics and aromatics accumulate in theextractable SOM of ecosystems with poor soils and low litterquality. Their decomposition seems hampered in such conditions,whereas all carbon sources seem effectively used if conditions fordecomposition are favorable (i.e. eutrophic ecosystems with highlitter quality, which show no accumulation of recalcitrants yet highamounts of N-compounds). Correlations between extractable-SOMchemistry, litter layer properties and hot water carbon contentfurther support this hypothesis.

The long-lasting influence of former heathland on the accu-mulation of recalcitrants in the extractable-SOM compositionindicates that chemical recalcitrance is relevant in certain ecosys-tems for several decades.

Acknowledgements

This study was funded by K.U.Leuven and supported by an FWOgrand (Research Foundatione Flanders). The authors wish to thankProf. R. Langohr, Prof. R. Merckx, Prof. P. Jacobs, Prof. E. Smolders,L. Vancampenhout, Dr. A. Temme and M. Broekhuijsen for theirsupport and wish to acknowledge the work of E.J. Velthorst formaking the pyrograms.

References

Aerts, R., Chapin, F.S., 2000. The Mineral Nutrition of Wild Plants Revisited:a Re-evaluation of Processes and Patterns. In: Advances in Ecological Research,vol. 30. Academic Press Inc, San Diego, pp. 1e67.

Anderson, J.M., 1991. The effects of climate change on decomposition processes ingrassland and coniferous forests. Ecological Applications 1, 326e347.

Augusto, L., Ranger, J., Binkley, D., Rothe, A., 2002. Impact of several common treespecies of European temperate forests on soil fertility. Annals of Forest Science59 (3), 233e253.

Baldock, J.A., Oades, J.M., Nelson, P.N., Skene, T.M., Golchin, A., Clarke, P., 1997.Assessing the extent of decomposition of natural organic materials using solid-state C-13 NMR spectroscopy. Australian Journal of Soil Research 35 (5),1061e1083.

Baldock, J.A., Masiello, C.A., Gelinas, Y., Hedges, J.I., 2004. Cycling and composition oforganic matter in terrestrial and marine ecosystems. Marine Chemistry 92,39e64.

Barbier, S., Gosselin, F., Balandier, P., 2008. Influence of tree species on understoryvegetation diversity and mechanisms involved e a critical review for temperateand boreal forests. Forest Ecology and Management 254, 1e15.

Bardgett, R.D., Bowman, W.D., Kaufmann, R., Schmidt, S.K., 2005. A temporalapproach to linking aboveground and belowground ecology. Trends in Ecology& Evolution 20, 634e641.

Bauzon, D., Van den Driessche, R., Dommergues, Y., 1969. Litter effect: I. Differentialin situ influence of forest litters on some biological soil characteristics. Oeco-logia Plantarum 4 (2), 99e102.

Beuselinck, L., Govers, G., Poesen, J., Degraer, G., Froyen, L., 1998. Grain-size analysisby laser diffractometry: comparison with the sieve-pipette method. Catena 32,193e208.

Buurman, P., Pape, T., Reijneveld, J.A., de Jong, F., van Gelder, E., 2001. Laser-diffraction and pipette-method grain sizing of Dutch sediments: correlationsfor fine fractions of marine, fluvial and loess samples. Netherlands Journal ofGeosciences 80 (2), 49e57.

Buurman, P., van Bergen, P.F., Jongmans, A.G., Meijer, E.L., Duran, B., van Lagen, B.,2005. Spatial and temporal variation in podzol organic matter studied bypyrolysis-gas chromatography/mass spectrometry and micromorphology.European Journal of Soil Science 56 (2), 253e270.

Buurman, P., Nierop, K.G.J., Pombal, X.P., Cortizas, A.M., 2006. Molecular chemistryby pyrolysis-GC/MS of selected samples of the Penido Vello peat deposit,Galicia, NW Spain. In: Martini, P., Martinez Cortizas, A., Chesworth, W. (Eds.),Peatlands-Evolution and Records of Environmental and Climate Changes.Elsevier, Amsterdam (Chapter 10).

Buurman, P., Peterse, F., Martin, G.A., 2007a. Soil organic matter chemistry inallophanic soils: a pyrolysis-GC/MS study of a Costa Rican Andosol catena.European Journal of Soil Science 58, 1330e1347.

Buurman, P., Schellekens, J., Fritze, H., Nierop, K.G.J., 2007b. Selective depletion oforganic matter in mottled podzol horizons. Soil Biology & Biochemistry 39,607e621.

Buurman, P., Nierop, K.G.J., Kaal, J., Senesi, N., 2009. Analytical pyrolysis and ther-mally assisted hydrolysis and methylation of EUROSOIL humic acid samples ea key to their source. Geoderma 150, 10e22.

Buurman, P., Roscoe, R. Different chemistry of free light, occluded light, andextractable SOM fractions in soils of cerrado, tilled and untilled fields, MinasGerais, Brazil e a pyrolysis-GC/MS study. European Journal of Soil Science,Submitted for publication.

Chefetz, B., Salloum, M.J., Deshmukh, A.P., Hatcher, P.G., 2002. Structural compo-nents of humic acids as determined by chemical modifications and carbon-13NMR, pyrolysis-, and thermochemolysis-gas chromatography/mass spectrom-etry. Soil Science Society of America Journal 66 (4), 1159e1171.

Conant, R.T., Drijber, R.A., Haddix, M.L., Parton, W.J., Paul, E.A., Plante, A.F., Six, J.,Steinweg, J.M., 2008. Sensitivity of organic matter decomposition to warmingvaries with its quality. Global Change Biology 14 (4), 868e877.

Cornelissen, J.H.C., 1996. An experimental comparison of leaf decomposition rates ina wide range of temperate plant species and types. Journal of Ecology 84 (4),573e582.

K. Vancampenhout et al. / Soil Biology & Biochemistry 42 (2010) 220e233 231

Cross, W.F., Wallace, J.B., Rosemond, A.D., 2007. Nutrient enrichment reducesconstraints on material flows in a detritus-based food web. Ecology 88 (10),2563e2575.

de Alcantara, F.A., Buurman, P., Curi, N., Neto, A.E.F., van Lagen, B., Meijer, E.L., 2004.Changes in soil organic matter composition after introduction of riparianvegetation on shores of hydroelectric reservoirs (Southeast of Brazil). SoilBiology & Biochemistry 36, 1497e1508.

De Deyn, G.B., Cornelissen, J.H.C., Bardgett, R.D., 2008. Plant functional traits andsoil carbon sequestration in contrasting biomes. Ecology Letters 11 (5),516e531.

De Keersmaeker, L., Rogiers, N., Lauriks, R., De Vos, B., 2001. GIS Data with Affor-estation on Historical Maps. Report VLINA C97/06. Project ‘Ecosystem VisionForest in Flanders’, issued by the Flemish Government and the Flemish ImpulsProgramme for Nature Development, Brussels (in Dutch).

De Vos, B., 1998. Chemical Element Analysis of the Forest Floor in Macro-inverte-brate Soil Fauna Plots. Report IBW Bd R:98.005.. Institute for Forestry and GameManagement, Ministry of the Flemish Government, Geraardsbergen.

De Vos, B., 2009. Uncertainties of forest soil carbon Stock assessment in Flanders.Dissertationes de Agricultura, University of Leuven, Leuven.

Dell'Abate, M.T., Benedetti, A., Brookes, P.C., 2003. Hyphenated techniques ofthermal analysis for characterisation of soil humic substances. Journal ofSeparation Science 26 (5), 433e440.

Derry, A.M., Staddon, W.J., Trevors, J.T., 1998. Functional diversity and communitystructure of microorganisms in uncontaminated and creosote-contaminatedsoils as determined by sole-carbon-source-utilization. World Journal ofMicrobiological Biotechnology 14, 571e578.

Dignac, M.F., Houot, S., Derenne, S., 2006. How the polarity of the separationcolumn may influence the characterization of compost organic matter bypyrolysis-GC/MS. Journal of Analytical and Applied Pyrolysis 75, 128e139.

Dixon, R.K., Brown, S., Houghton, R.A., Solomon, A.M., Trexler, M.C., Wisniewski, J.,1994. Carbon pools and flux of global forest ecosystems. Science 263 (5144),185e190.

Dove, H., Mayes, R.W., Freer, M., 1996. Effects of species, plant part, and plant age onthe n-alkane concentrations in the cuticular wax of pasture plants. AustralianJournal of Agricultural Research 47 (8), 1333e1347.

Dudal, R., 1998. The soil map of Belgium in a European framework. Pedologie-Themata. Journal of the Belgian Soil Science Society 5, 13e19.

Dumortier, M., De Bruyn, L., Hens, M., Peymen, J., Schneiders, A., Van Daele, T., VanReeth, W. (Eds.), 2007. Natuurrapport 2007. Toestand van de natuur in Vlaan-deren: cijfers voor het beleid, nr. 4. Mededeling van het Instituut voor Natuur-en Bosonderzoek, Brussel (in Dutch).

Edwards, N.T., 1982. The use of soda-lime for measuring respiration rates interrestrial systems. Pedobiologica 23, 321e330.

Eglinton, G., Hamilton, R.J., 1967. Leaf epicuticular waxes. Science 156, 1322.Ekschmitt, K., Kandeler, E., Poll, C., Brune, A., Buscot, F., Friedrich, M., Gleixner, G.,

Hartmann, A., Kastner, M., Marhan, S., Miltner, A., Scheu, S., Wolters, V., 2008.Soil-carbon preservation through habitat constraints and biological limitationson decomposer activity. Journal of Plant Nutrition and Soil Science e ZeitschriftFur Pflanzenernahrung und Bodenkunde 171 (1), 27e35.

FAO, 2006a. Global Forest Resources Assessment 2005. FAO, Rome.FAO, 2006b. Guidelines for Soil Description, fourth ed. FAO, Rome.FAO, ISRIC, ISSS, 2006. World Reference Base for Soil Resources 2006. World Soil

Resources Reports 103. FAO, Rome.Ferreira, F.P., Vidal-Torrado, P., Buurman, P., Macias, F., Otero, X.L., Boluda, R., 2009.

Pyrolysis-gas chromatography/mass spectrometry of soil organic matterextracted from a Brazilian mangrove and Spanish salt marshes. Soil ScienceSociety of America Journal 73, 841e851.

Ghani, A., Dexter, M., Perrott, K.W., 2003. Hot-water extractable carbon in soils:a sensitive measurement for determining impacts of fertilisation, grazing andcultivation. Soil Biology and Biochemistry 35 (9), 1231e1243.

Gholz, H.L., Wedin, D.A., Smitherman, S.M., Harmon, M.E., Parton, W.J., 2000. Long-term dynamics of pine and hardwood litter in contrasting environments:toward a global model of decomposition. Global Change Biology 6 (7),751e765.

Gleixner, G., Bol, R., Balesdent, J., 1999. Molecular insight into soil carbon turnover.Rapid Communications in Mass Spectrometry 13 (13), 1278e1283.

Grandy, A.S., Strickland, M.S., Lauber, C.L., Bradford, M.A., Fierer, N., 2009. Theinfluence of microbial communities, management, and soil texture on soilorganic matter chemistry. Geoderma 150, 278e286.

Grogan, P., 1998. CO2 flux measurement using soda lime: correction for waterformed during CO2 adsorption. Ecology 79 (4), 1467e1468.

Hagen-Thorn, A., Callesen, I., Armolaitis, K., Nihlgard, B., 2004. The impact of sixEuropean tree species on the chemistry of mineral topsoil in forest plantationson former agricultural land. Forest Ecology and Management 195 (3), 373e384.

Hajje, N., Jaffe, R., 2006. Molecular characterization of Cladium peat from the FloridaEverglades: biomarker associations with humic fractions. Hydrobiologia 569,99e112.

Hamer, U., Marschner, B., Brodowski, S., Amelung, W., 2004. Interactive priming ofblack carbon and glucose mineralisation. Organic Geochemistry 35 (7),823e830.

Hatcher, P.G., Dria, K.J., Kim, S., Frazier, S.W., 2001. Modern analytical studies ofhumic substances. Soil Science 166 (11), 770e794.

H€attenschwiler, S., Tiunov, A.V., Scheu, S., 2005. Biodiversity and litter decompo-sition interrestrial ecosystems. Annual Review of Ecology Evolution andSystematics 36, 191e218.

Hedde, M., Aubert, M., Decaens, T., Bureau, F., 2008. Dynamics of soil carbon ina beechwood chronosequence forest. Forest Ecology and Management 255 (1),193e202.

Hedges, J.I., Mann, D.C., 1979. Characterization of plant-tissues by their ligninoxidation-products. Geochimica et Cosmochimica Acta 43 (11), 1803e1807.

Helfrich, M., Ludwig, B., Buurman, P., Flessa, H., 2006. Effect of land use on thecomposition of soil organic matter in density and aggregate fractions asrevealed by solid-state C-13 NMR spectroscopy. Geoderma 136 (1e2), 331e341.

Hessen, D.O., Agren, G.I., Anderson, T.R., Elser, J.J., De Ruiter, P.C., 2004. Carbon,sequestration in ecosystems: the role of stoichiometry. Ecology 85 (5),1179e1192.

Hobbie, S.E., Reich, P.B., Oleksyn, J., Ogdahl, M., Zytkowiak, R., Hale, C.,Karolewski, P., 2006. Tree species effects on decomposition and forest floordynamics in a common garden. Ecology 87 (9), 2288e2297.

Huntington, T.G., 1995. Carbon sequestration in an aggrading forest ecosystem inthe Southeastern USA. Soil Science Society of America Journal 59 (5),1459e1467.

ISRIC, FAO, 2002. In: Van Reeuwijk, L.P. (Ed.), Procedures for Soil Analysis, sixth ed.ISRIC, Wageningen.

Jandl, R., Lindner, M., Vesterdal, L., Bauwens, B., Baritz, R., Hagedorn, F., Johnson, D.W., Minkkinen, K., Byrne, K.A., 2007. How strongly can forest managementinfluence soil carbon sequestration? Geoderma 137 (3e4), 253e268.

Jenkinson, D.S., Coleman, K., 2008. The turnover of organic carbon in subsoils. Part2. Modelling carbon turnover. European Journal of Soil Science 59 (2),400e413.

Kaal, J., Brodowski, S., Baldock, J.A., Nierop, K.G.J., Cortizas, A.M., 2008a. Charac-terisation of aged black carbon using pyrolysis-GC/MS, thermally assistedhydrolysis and methylation (THM), direct and cross-polarisation C-13 nuclearmagnetic resonance (DP/CP NMR) and the benzenepolycarboxylic acid (BPCA)method. Organic Geochemistry 39 (10), 1415e1426.

Kaal, J., Martinez-Cortizas, A., Nierop, K.G.J., Buurman, P., 2008b. A detailed pyrol-ysis-GC/MS analysis of a black carbon-rich acidic colluvial soil (Atlantic ranker)from NW Spain. Applied Geochemistry 23 (8), 2395e2405.

Kelleher, B.P., Simpson, A.J., 2006. Humic substances in soils: are they reallychemically distinct? Environmental Science and Technology 40 (15),4605e4611.

Kiem, R., K€ogel-Knabner, I., 2003. Contribution of lignin and polysaccharides to therefractory carbon pool in C-depleted arable soils. Soil Biology and Biochemistry35 (1), 101e118.

Killops, S.D., Killops, V., 2005. Introduction to Organic Geochemistry. BlackwellPublishing, Malden.

KMI, 2009. Database of the Royal Meteorological Institute. www.kmi.be (consulted22.01.2009).

Knicker, H., Almendros, G., Gonzalez-Vila, F.J., Gonzalez-Perez, J.A., Polvillo, O., 2006.Characteristic alterations of quantity and quality of soil organic matter causedby forest fires in continental Mediterranean ecosystems: a solid-state C-13 NMRstudy. European Journal of Soil Science 57, 558e569.

Knohl, A., Soe, A.R.B., Kutsch, W.L., Gockede, M., Buchmann, N., 2008. Representa-tive estimates of soil and ecosystem respiration in an old beech forest. Plant andSoil 302 (1e2), 189e202.

K€ogel, I., Hempfling, R., Zech, W., Hatcher, P.G., Schulten, H.R., 1988. Chemical-composition of the organic-matter in forest soils. 1. Forest litter. Soil Science146, 124e136.

K€ogel-Knabner, I., 2002. The macromolecular organic composition of plant andmicrobial residues as inputs to soil organic matter. Soil Biology and Biochem-istry 34 (2), 139e162.

K€ogel-Knabner, I., De Leeuw, J.W., Hatcher, P.G., 1992a. Nature and distribution ofalkyl carbon in forest soil profiles e implications for the origin and humifi-cation of aliphatic biomacromolecules. Science of the Total Environment 118,175e185.

K€ogel-Knabner, I., Hatcher, P.G., Tegelaar, E.W., De Leeuw, J.W., 1992b. Aliphaticcomponents of forest soil organic-matter as determined by solid-state C-13NMR and analytical pyrolysis. Science of the Total Environment 113 (1e2),89e106.

K€ogel-Knabner, I., von L€utzow, M., Guggenberger, G., Flessa, H., Marschner, B.,Matzner, E., Ekschmitt, K., 2005. Mechanisms and regulation of organic matterstabilisation in soils. Geoderma 128 (1e2), 1e2.

K€ogel-Knabner, I., Chenu, C., Kandeler, E., Piccolo, A., 2006. Biological and physi-cochemical processes and control of soil organic matter stabilization andturnover. European Journal of Soil Science 57 (4), 425.

K€ogel-Knabner, I., Ekschmitt, K., Flessa, H., Guggenberger, G., Matzner, E.,Marschner, B., von Luetzow, M., 2008a. An integrative approach of organicmatter stabilization in temperate soils: linking chemistry, physics, and biology.Journal of Plant Nutrition and Soil Science e Zeitschrift fur Pflanzenernahrungund Bodenkunde 171 (1), 5e13.

K€ogel-Knabner, I., Guggenberger, G., Kleber, M., Kandeler, E., Kalbitz, K., Scheu, S.,Eusterhues, K., Leinweber, P., 2008b. Organo-mineral associations in temperatesoils: integrating biology, mineralogy, and organic matter chemistry. Journal ofPlant Nutrition and Soil Science e Zeitschrift fur Pflanzenernahrung und Bod-enkunde 171 (1), 61e82.

Konert, M., Vandenberghe, J., 1997. Comparison of laser grain size analysis withpipette and sieve analysis: a solution for the underestimation of the clay frac-tion. Sedimentology 44, 523e535.

Kuzyakov, Y., Friedel, J.K., Stahr, K., 2000. Review of mechanisms and quantificationof priming effects. Soil Biology and Biochemistry 32 (11e12), 1485e1498.

K. Vancampenhout et al. / Soil Biology & Biochemistry 42 (2010) 220e233232

Langohr, R., 1986. Une fore

Ï

t aux pieds de ‘limon’, sol et relief en Fo

Ï

ret de Soignes:des t�emoins uniques. R�eserves Naturelles, 3.

Langohr, R., 2001. L'anthropisation du paysage p�edologique agricole de la Belgiquedepuis le N�eolithique ancien e apports de l'arch�eop�edologie. Etude et Gestiondes Sols 8 (2), 215e229.

Langohr, R., Sanders, J., 1985. The Belgian Loess Belt in the Last 20 000 Years:Evolution of Soils and Relief in the Zonien Forest. In: Soils and QuaternaryLandscape Evolution. John Wiley and Sons Ltd., pp 359e371.

Lobe, I., Du Preez, C.C., Amelung, W., 2002. Influence of prolonged arable croppingon lignin compounds in sandy soils of the South African highveld. EuropeanJournal of Soil Science 53 (4), 553e562.

Lorenz, K., Lal, R., Preston, C.M., Nierop, K.G.J., 2007. Strengthening the soil organiccarbon pool by increasing contributions from recalcitrant aliphatic bio(macro)molecules. Geoderma 142 (1e2), 1e10.

Maie, N., Behrens, A., Knicker, H., K€ogel-Knabner, I., 2003. Changes in the structureand protein binding ability of condensed tannins during decomposition of freshneedles and leaves. Soil Biology & Biochemistry 35, 577e589.

Mardulyn, P., Godden, B., Echezarreta, P.A., Penninckx, M., Gruber, W., Herbauts, J.,1993. Changes in humus microbiological activity-induced by the substitution ofthe natural beech forest by Norway spruce in the Belgian Ardennes. ForestEcology and Management 59 (1e2), 15e27.

Marschner, B., Brodowski, S., Dreves, A., Gleixner, G., Gude, A., Grootes, P.M.,Hamer, U., Heim, A., Jandl, G., Ji, R., Kaiser, K., Kalbitz, K., Kramer, C.,Leinweber, P., Rethemeyer, J., Schaeffer, A., Schmidt, M.W.I., Schwark, L.,Wiesenberg, G.L.B., 2008. How relevant is recalcitrance for the stabilization oforganic matter in soils? Journal of Plant Nutrition and Soil Science e Zeitschriftfur Pflanzenernahrung und Bodenkunde 171 (1), 91e110.

Mayer, P.M., 2008. Ecosystem and decomposer effects on litter dynamics along anold field to old-growth forest successnional gradient. Acta Oecologica e Inter-national Journal of Ecology 33 (2), 222e230.

Mikutta, R., Kleber, M., Torn, M.S., Jahn, R., 2006. Stabilization of soil organic matter:association with minerals or chemical recalcitrance? Biogeochemistry 77 (1),25e56.

Millar, C.S., 1974. Decomposition of coniferous leaf litter. In: Dickinson, C.H.,Pugh, G.J.F. (Eds.), Biology of Plant Litter Decomposition. Academic Press,London, pp. 105e128.

Moldoveanu, S.C., 1998. Analytical Pyrolysis of Natural Organic Polymers. Elsevier,Amsterdam, 496 pp.

Muys, B., 1995. The influence of tree species on humus quality and nutrient avail-ability on a regional scale (Flanders, Belgium). Plant and Soil 168e169,649e660.

Neirynck, J., Mirtcheva, S., Sioen, G., Lust, N., 2000. Impact of Tilia platyphyllos Scop.,Fraxinus excelsior L., Acer pseudoplatanus L., Quercus robur L. and Fagus sylvaticaL. on earthworm biomass and physico-chemical properties of a loamy topsoil.Forest Ecology and Management 133 (3), 275e286.

Nierop, K.G.J., 1998. Origin of aliphatic compounds in a forest soil. OrganicGeochemistry 29 (4), 1009e1016.

Nierop, K.G.J., Buurman, P., 1999. Water-soluble organic matter in incipient podzols:accumulation in B horizons or in fibres? European Journal of Soil Science 50,701e711.

Nierop, K.G.J., Buurman, P., de Leeuw, J.W., 1999. Effect of vegetation on chemicalcomposition of H horizons in incipient podzols as characterized by C-13 NMRand pyrolysis-GC/MS. Geoderma 90 (1e2), 111e129.

Nierop, K.G.J., Pulleman, M.M., Marinissen, J.C.Y., 2001a. Management inducedorganic matter differentiation in grassland and arable soil: a study usingpyrolysis techniques. Soil Biology and Biochemistry 33 (6), 755e764.

Nierop, K.G.J., van Lagen, B., Buurman, P., 2001b. Composition of plant tissues andsoil organic matter in the first stages of a vegetation succession. Geoderma 100(1e2), 1e24.

Nierop, K.G.J., van Bergen, P.F., Buurman, P., van Lagen, B., 2005. NaOH andNa4P2O7 extractable organic matter in two allophanic volcanic ash soils ofthe Azores Islands e a pyrolysis GC/MS study. Geoderma 127 (1e2),36e51.

Olk, D.C., 2006. A chemical fractionation for structure-function relations of soilorganic matter in nutrient cycling. Soil Science Society of America Journal 70,1013e1022.

Page, D.W., van Leeuwen, J.A., Spark, K.M., Mulcahy, D.E., 2002. Pyrolysis charac-terisation of plant, humus and soil extracts from Australian catchments. Journalof Analytical and Applied Pyrolysis 65 (2), 269e285.

Piccolo, A., 2001. The supra-molecular structure of humic substances. Soil Science166, 810e832.

Pollierer, M.M., Langel, R., Korner, C., Maraun, M., Scheu, S., 2007. The under-estimated importance of belowground carbon input for forest soil animal foodwebs. Ecology Letters 10 (8), 729e736.

Polyakova, O., Billor, N., 2007. Impact of deciduous tree species on litterfall quality,decomposition rates and nutrient circulation in pine stands. Forest Ecology andManagement 253 (1e3), 11e18.

Ponge, J.F., Chevalier, R., 2006. Humus Index as an indicator of forest stand and soilproperties. Forest Ecology and Management 233 (1), 165e175.

Priha, O., Grayston, S.J., Hiukka, R., Pennanen, T., Smolander, A., 2001. Microbialcommunity structure and characteristics of the organic matter in soils underPinus sylvestris, Picea abies and Betula pendula at two forest sites. Biology andFertility of Soils 33 (1), 17e24.

Ralph, J., Hatfield, R.D., 1991. Pyrolysis-GCeMS characterization of forage materials.Journal of Agricultural and Food Chemistry 39 (8), 1426e1437.

Saiz-Jimenez, C., 1994. Analytical pyrolysis of humic substances e pitfalls, limita-tions, and possible solutions. Environmental Science and Technology 28 (11),1773e1780.

Saiz-Jimenez, C., 1995. Reactivity of the aliphatic humic moiety in analyticalpyrolysis. Organic Geochemistry 23, 955e961.

Saiz-Jimenez, C., de Leeuw, J.W., 1986. Lignin pyrolysis products e their structuresand their significance as biomarkers. Organic Geochemistry 10 (4e6), 869e876.

Saiz-Jimenez, C., de Leeuw, J.W., 1987. Chemical-structure of a soil humic-acid asrevealed by analytical pyrolysis. Journal of Analytical and Applied Pyrolysis 11,367e376.

Santin, C., Gonzalez-Perez, M., Otero, X.L., Alvarez, M.A., Macias, F., 2009. Humicsubstances in estuarine soils colonized by Spartina maritima. Estuarine Coastaland Shelf Science 81, 481e490.

Schmitt, A., Glaser, B., Borken, W., Matzner, E., 2007. Repeated freeze-thaw cylceschanged organic matter quality in a temperate forest soil. In: InternationalSymposium on Soil Processes under Extreme Meteorological Conditions,February 25e28, Bayreuth, Germany.

Schulten, H.R., Leinweber, P., Reuter, G., 1992. Initial formation of soil organic-matter from grass residues in a long-term experiment. Biology and Fertility ofSoils 14, 237e245.

Schulten, H.R., Schnitzer, M., 1997. The chemistry of soil organic nitrogen: a review.Biology and Fertility of Soils 26 (1), 1e15.

Senesi, N., D'Orazio, V., Ricca, G., 2003. Humic acids in the first generation ofEUROSOILS. Geoderma 116, 325e344.

Soil Survey Staff, 1975. Soil Taxonomy: A Basic System of Soil Classification forMaking and Interpreting Soil Surveys. USDA-SCS. Agric. Handb. 436. U.S. Gov.Print. Office, Washington, DC.

Sollins, P., Homann, P., Caldwell, B.A., 1996. Stabilization and destabilization of soilorganic matter: mechanisms and controls. Geoderma 74 (1e2), 65e105.

Spiegelberger, T., Hegg, O., Matthies, D., Hedlund, K., Schaffner, U., 2006. Long-termeffects of short-term perturbation in a subalpine grassland. Ecology 87,1939e1944.

Taubner, H., Roth, B., Tippk€otter, R., 2009. Determination of soil texture: comparisonof the sedimentation method and the laser-diffraction analysis. Journal of PlantNutrition and Soil Science 172, 161e171.

Tegelaar, E.W., Deleeuw, J.W., Holloway, P.J., 1989a. Some mechanisms of flashpyrolysis of naturally-occurring higher-plant polyesters. Journal of Analyticaland Applied Pyrolysis 15, 289e295.

Tegelaar, E.W., Deleeuw, J.W., Saizjimenez, C., 1989b. Possible origin of aliphaticmoieties in humic substances. Science of the Total Environment 81e2, 1e17.

Tegelaar, E.W., Hollman, G., Vandervegt, P., Deleeuw, J.W., Holloway, P.J., 1995.Chemical characterization of the periderm tissue of some angiosperm species erecognition of an insoluble, nonhydrolyzable, aliphatic biomacromolecule(Suberan). Organic Geochemistry 23 (3), 239e251.

Tyler, G., 1992. Tree species affinity of decomposer and ectomycorrhizal macrofungiin beech (Fagus sylvatica L.), oak (Quercus robur L.) and hornbeam (Carpinusbetulus L.) forests. Forest Ecology and Management 47 (1e4), 269e284.

van Bergen, P.F., Bull, I.D., Poulton, P.R., Evershed, R.P., 1997. Organic geochemicalstudies of soils from the Rothamsted classical experiments: 1. Total lipidextracts, solvent insoluble residues and humic acids from Broadbalk wilderness.Organic Geochemistry 26 (1e2), 117e135.

van Bergen, P.F., Flannery, M.B., Poulton, P.R., Evershed, R.P., 1998a. OrganicGeochemical Studies of Soils from Rothamsted Experimental Station: IIINitrogen-containing Organic Matter in Soil from Geescroft Wilderness,Nitrogen-Containing Macromolecules in the Bio- and Geosphere. In: ACSSymposium Series. American Chemical Society, Washington, pp. 321e338.

van Bergen, P.F., Nott, C.J., Bull, I.D., Poulton, P.R., Evershed, R.P., 1998b. Organicgeochemical studies of soils from the Rothamsted Classical Experiments e IV.Preliminary results from a study of the effect of soil pH on organic matter decay.Organic Geochemistry 29 (5e7), 1779e1795.

Van Calster, H., Baeten, L., De Schrijver, A., De Keersmaeker, L., Rogister, J.E.,Verheyen, K., Hermy, M., 2007. Management driven changes (1967e2005) insoil acidity and the understorey plant community following conversion ofa coppice-with-standards forest. Forest Ecology and Management 241 (1e3),258e271.

Vancampenhout, K., Wouters, K., Caus, A., Buurman, P., Swennen, R., Deckers, J.,2008. Fingerprinting of soil organic matter as a proxy for assessing climate andvegetation changes in last interglacial palaeosols (Veldwezelt, Belgium).Quaternary Research 69, 145e162.

Vancampenhout, K., Wouters, K., De Vos, B., Buurman, P., Swennen, R., Deckers, J.,2009. Differences in chemical composition of soil organic matter in naturalecosystems from different climatic regions e a pyrolysis-GC/MS study. SoilBiology and Biochemistry 41, 568e579.

Vandewiele, S., 2001. Spontane evolutie van de bosstructuur en effect van brand opde bosontwikkeling in de Withoefse Heide. Unpublished M.Sc. thesis, GhentUniversity, Ghent (in Dutch).

Van Hemelrijck, I., Muys, B., De Vos, B., 2004. Hoge bomen vangen veel wind:verklaringen voor windworp van beuk in het Zoni€enwoud. Bosrevue 8 (1e4) (inDutch).

Verde, J.R., Buurman, P., Martinez-Cortizas, A., Macias, F., Arbestain, M.C., 2008.NaOH-extractable organic matter of andic soils from Galicia (NW Spain) underdifferent land use regimes: a pyrolysis GC/MS study. European Journal of SoilScience 59, 1096e1110.

von L€utzow, M., K€ogel-Knabner, I., Ekschmitt, K., Matzner, E., Guggenberger, G.,Marschner, B., Flessa, H., 2006. Stabilization of organic matter in temperate

K. Vancampenhout et al. / Soil Biology & Biochemistry 42 (2010) 220e233 233

soils: mechanisms and their relevance under different soil conditions e

a review. European Journal of Soil Science 57 (4), 426e445.von L€utzow, M., K€ogel-Knabner, I., Ekschmitt, K., Flessa, H., Guggenberger, G.,

Matzner, E., Marschner, B., 2007. SOM fractionation methods: relevance tofunctional pools and to stabilization mechanisms. Soil Biology & Biochemistry39, 2183e2207.

von L€utzow, M., K€ogel-Knabner, I., Ludwig, B., Matzner, E., Flessa, H., Ekschmitt, K.,Guggenberger, G., Marschner, B., Kalbitz, K., 2008. Stabilization mechanisms oforganic matter in four temperate soils: development and application of

a conceptual model. Journal of Plant Nutrition and Soil Science e Zeitschrift furPflanzenernahrung und Bodenkunde 171 (1), 111e124.

von Oheimb, G., Hardtle, W., Naumann, P.S., Westphal, C., Assmann, T., Meyer, H.,2008. Long-term effects of historical heathland farming on soil properties offorest ecosystems. Forest Ecology and Management 255 (5e6), 1984e1993.

Walton, T.J., 1990. Waxes, cutin and suberin. In: Harwood, J.L., Bowyer, J.R. (Eds.),Methods in Plant Biochemistry. Academic Press, London, pp. 105e158.

Zech,W., Ziegler, F., K€ogel-Knabner, I., Haumaier, L.,1992.Humic substances distributionand transformation in forest soils. Science of the Total Environment 118, 155e174.