storage and drivers of organic carbon in forest soils of southeast germany (bavaria) –...

11

Click here to load reader

Upload: ingrid

Post on 25-Dec-2016

217 views

Category:

Documents


3 download

TRANSCRIPT

Page 1: Storage and drivers of organic carbon in forest soils of southeast Germany (Bavaria) – Implications for carbon sequestration

Forest Ecology and Management 295 (2013) 162–172

Contents lists available at SciVerse ScienceDirect

Forest Ecology and Management

journal homepage: www.elsevier .com/ locate/ foreco

Storage and drivers of organic carbon in forest soils of southeast Germany (Bavaria)– Implications for carbon sequestration

Martin Wiesmeier a,⇑, Jörg Prietzel a, Frauke Barthold b, Peter Spörlein c, Uwe Geuß c, Edzard Hangen c,Arthur Reischl c, Bernd Schilling c, Margit von Lützow a, Ingrid Kögel-Knabner a,d

a Lehrstuhl für Bodenkunde, Department für Ökologie und Ökosystemmanagement, Wissenschaftszentrum Weihenstephan für Ernährung, Landnutzung und Umwelt,Technische Universität München, 85350 Freising-Weihenstephan, Germanyb Institute of Earth and Environmental Sciences, University of Potsdam, Karl-Liebknecht-Str. 24–25, 14476 Golm, Germanyc Bavarian Environment Agency (LfU), 95030 Hof, Germanyd Institute for Advanced Study, Technische Universität München, Lichtenbergstr. 2a, D-85748 Garching, Germany

a r t i c l e i n f o

Article history:Received 18 December 2012Received in revised form 15 January 2013Accepted 25 January 2013Available online 24 February 2013

Keywords:Tree species effectSoil organic matterClimate changeForest management

0378-1127/$ - see front matter � 2013 Elsevier B.V. Ahttp://dx.doi.org/10.1016/j.foreco.2013.01.025

⇑ Corresponding author. Tel.: +49 (0)8161 71 3679E-mail address: [email protected] (M. Wies

a b s t r a c t

Temperate forest soils of central Europe are regarded as important pools for soil organic carbon (SOC) andthought to have a high potential for carbon (C) sequestration. However, comprehensive data on total SOCstorage, particularly under different forest types, and its drivers is limited. In this study, we analyzed aforest data set of 596 completely sampled soil profiles down to the parent material or to a depth of1 m within Bavaria in southeast Germany in order to determine representative SOC stocks under differentforest types in central Europe and the impact of different environmental parameters. We calculated atotal median SOC stock of 9.8 kg m�2 which is considerably lower compared with many other inventorieswithin central Europe that used modelled instead of measured soil properties. Statistical analysesrevealed climate as controlling parameter for the storage of SOC with increasing stocks in cool, humidmountainous regions and a strong decrease in areas with higher temperatures. No significant differencesof total SOC storage were found between broadleaf, coniferous and mixed forests. However, coniferousforests stored around 35% of total SOC in the labile organic layer that is prone to human disturbance, for-est fires and rising temperatures. In contrast, mixed and broadleaf forests stored the major part of SOC inthe mineral soil. Moreover, these two forest types showed unchanged or even slightly increased mineralSOC stocks with higher temperatures, whereas SOC stocks in mineral soils under coniferous forest weredistinctly lower. We conclude that mixed and broadleaf forests are more advantageous for C sequestra-tion than coniferous forests. An intensified incorporation of broadleaf species in extent coniferous forestsof Bavaria would prevent substantial SOC losses as a result of rising temperatures in the course of climatechange.

� 2013 Elsevier B.V. All rights reserved.

1. Introduction

Forest ecosystems store more than 50% of total terrestrial car-bon (C) and are regarded to have a high potential for sequestrationof atmospheric CO2 (IPCC, 2000; Lorenz and Lal, 2010). In particulartemperate forests that cover only 8% of the global land surface playa key role for C sequestration that was estimated to be around 40%of total terrestrial C uptake with a hotspot in Europe (Nabuurset al., 1997; Martin et al., 2001; Goodale et al., 2002; Liski et al.,2002; Ciais et al., 2008; Lal, 2008; Wamelink et al., 2009; Tyrrellet al., 2012). Despite the importance of forests as major terrestrialC pool, there are large uncertainties regarding C storage in forestsoils which accounts for 60–70% of total forest C (IPCC, 2000; John-

ll rights reserved.

; fax: +49 (0)8161 71 4466.meier).

son and Curtis, 2001; Lorenz and Lal, 2010). Many authors criticizethe lack of forest soil data, particularly for deeper parts of the min-eral soil (e.g. Perruchoud et al., 1999; Baritz et al., 2010; Price et al.,2012; Tyrrell et al., 2012). Most studies that estimated the storageof soil organic carbon (SOC) focused on the organic layer and sur-face horizons in the upper 30 cm of the soil. However, tree rootsand thus input of organic matter (OM) extend to deep subsoil hori-zons down to a depth of 3 m that may contain more than 50% oftotal SOC stocks (Jobbagy and Jackson, 2000; Lorenz and Lal,2010; Rumpel and Kögel-Knabner, 2011). Further, there is a highspatial variability of forest SOC stocks and therefore, large numbersof samples are required to determine SOC stocks and assess differ-ences accurately (Lal, 2005; Schöning et al., 2006; Mäkipää et al.,2008; Spielvogel et al., 2009). Moreover, many forest SOC studiesare not based on measured soil properties but partly use modelledparameters for calculation of SOC stocks that could lead to a sys-tematic bias (Karjalainen et al., 2003; Lindner and Karjalainen,

Page 2: Storage and drivers of organic carbon in forest soils of southeast Germany (Bavaria) – Implications for carbon sequestration

M. Wiesmeier et al. / Forest Ecology and Management 295 (2013) 162–172 163

2007; Schrumpf et al., 2008). Also information about the mainenvironmental drivers for the storage of SOC in forest soils is lim-ited (Wamelink et al., 2009; Luyssaert et al., 2010).

In order to increase C stocks in forests, several managementpractices were discussed such as thinning, drainage, extending ofrotation period, fertilization, liming, site preparation, fire, stormand insect management, afforestation and reforestation, harvestmanagement and input of harvest residues (Burschel et al., 1993;Lal, 2005; Jandl et al., 2007; Nabuurs et al., 2008; Lorenz and Lal,2010; Carroll et al., 2012). However, one of the most promising ap-proaches to promote C sequestration in forests is a change in treespecies composition. Several studies investigated the storage ofSOC under different tree species and reported various effects(Augusto et al., 2002; De Vries et al., 2003; Hagen-Thorn et al.,2004; Ladegaard-Pedersen et al., 2005; Oostra et al., 2006; Schulpet al., 2008; Vesterdal et al., 2008). However, most of these studieswere again restricted to the organic layer and uppermost mineralhorizons and thus quantified only a certain proportion of totalSOC stocks. Therefore, more knowledge about forest type-specificSOC storage is needed before future composition of tree speciescan be recommended (Jandl et al., 2007; Vesterdal et al., 2012).

In this study we used a comprehensive data set of 596 completelysampled forest soil profiles down to the parent material or at least toa depth of 1 m within Bavaria in southeast Germany to gain insightinto the storage and driving factors of SOC. The data set consistedof 88 broadleaf, 331 coniferous and 177 mixed forest sites that weresampled for SOC, nitrogen (N), bulk density (BD), stone content (SC),pH and partly soil texture for each soil horizon. Our aims were to (1)determine total SOC stocks under different forest types, (2) revealthe main environmental parameters that control the storage ofSOC in forest soils and (3) derive information about C sequestrationin forest soils of Bavaria as affected by different forest types.

2. Materials and methods

2.1. Study area

Bavaria comprises an area of 70,550 km2 and is located in thesoutheast of Germany. The northwestern part of Bavaria is domi-

Fig. 1. Map of Bavaria with forest

nated by the southern German escarpment landscape that adjoinsin the east to low mountain ranges of the Bohemian Massif. South-wards a Molasse basin affiliates that ascends at the southern bor-der of Bavaria to the mountain range of the Alps. Elevationranges between 107 and 2962 m above sea level. Due to its loca-tion in central Europe, Bavaria has a sub-oceanic climate that ischaracterized by a transitional situation between a maritime cli-mate in the northwest and sub-continental influences in the east.Mean annual temperature and precipitation range from the escarp-ment landscape in the northwest to the Alps in the south between10 and 3 �C and 550 and 2500 mm, respectively. Around 35% of thearea of Bavaria is covered by forest (Fig. 1). Coniferous forests coveraround 35% of the total forest area (Schnell and Bauer, 2005) andare dominated by two species, Norway spruce (Picea abies) andScots pine (Pinus sylvestris). About 25% of the forest area is coveredby broadleaf forests which are dominated by European beech (Fa-gus sylvatica), pedunculate oak (Quercus robur) and sessile oak(Quercus petraea). In the remaining 40% of the forest area, mixedforests with coniferous and broadleaf species are found. Predomi-nant forest soil classes are soils with well-developed B horizons(Cambisols) and soils with water stagnation (Stagnosols, Albeluvi-sols, Planosols) according to the German soil systematic (AD-HOCAG Boden, 2005) and the equivalent Reference Soil Groups of theWRB system (IUSS Working Group WRB, 2006).

2.2. Selection of forest soil data

Available data from different soil surveys and permanent soilobservation sites in Bavaria provided by the Bavarian EnvironmentAgency (LfU) were screened to compile a representative data setfor forest soils. Only sampling locations were incorporated wheresoil profiles were sampled by horizon down to the parent materialor at least to 1 m depth. All soil horizons were analyzed for SOCand N concentration, BD, SC and pH. Minimum requirement forSOC and N analysis was a determination by dry combustion usinga CN elemental analyzer. Generally, only soil data was includedwhich was collected after 1990.

Overall, the selected sampling locations that fulfilled all require-ments amounted to 596 soil profiles (Fig. 1) with 88 locations under

areas and sampling locations.

Page 3: Storage and drivers of organic carbon in forest soils of southeast Germany (Bavaria) – Implications for carbon sequestration

164 M. Wiesmeier et al. / Forest Ecology and Management 295 (2013) 162–172

broadleaf forest, 177 locations under mixed forest and 331 locationsunder coniferous forest. The main part of the data constitutes a sam-pling campaign using a grid of 8 � 8 km within Bavaria that was con-ducted between 2000 and 2004 (Joneck et al., 2006). For each soilprofile a representative location was selected within a radius of500 m around the grid node in consideration of a homogeneous sam-pling area in terms of forest type, stand age of at least 70 years, relief,soil type and parent material. This approach should ensure that for-est sites with exceptional environmental conditions or which areinfluenced by recent land use changes are excluded from the study.Anthropogenic disturbances in the subsoil (e.g. deposits) were ex-cluded in a pre-exploratory survey using a soil auger. As the sam-pling of mineral and organic layers was conducted variouslywithin the growing season a systematic bias of seasonal-dependentsoil properties, particularly litterfall which is different between for-est types, can be excluded. For top- and subsoil horizons down to adepth of about 35 cm soil material was collected as composite sam-ple from eight sub-locations around the main soil profile in order tocover the small-scale heterogeneity of the soils. Below 35 cm sam-ples were taken from the main soil profile. A small proportion ofthe included soil profiles originate from permanent forest soil mon-itoring sites (Schubert, 2002) and regional soil surveys. With regardto the total forest area of Bavaria, a sampling density of one locationper 41 km2 was obtained.

2.3. Determination of SOC stocks and other soil properties

For the determination of BD, the mass of the oven-dry soil(105 �C) was divided by the volume of the soil cores (Hartge andHorn, 1989). For around 2% of organic and subsoil horizons, BDcould not be analyzed directly due to high amounts of stones inC horizons or low thickness of O horizons. In this case, BD was esti-mated based on BD values from adjacent sites with similar soiltype and parent material. Stone contents were estimated visuallyin the soil profiles according to AD-HOC AG Boden (2005). AllSOC and N concentrations were determined by dry combustionon CN elemental analyzer. Samples that contained CaCO3 wereheated to 500 �C for 4 h to remove organic carbon and the concen-tration of inorganic C of the residual material was determined bydry combustion. The SOC content was calculated by subtractingthe content of inorganic C from the total C concentration of the un-treated material. SOC and N stocks for each soil profile were calcu-lated using following equation:

EShz ¼Xhz

i

ECi � BDi � hi � 1� SCi

100

� �ð1Þ

where EShz is the total elemental stock (kg m�2) of all soil horizons hz,ECi is the elemental concentration (mg g�1) of horizon i, BDi is the bulkdensity (g cm�3) of horizon i, hi is the thickness (cm) of horizon i andSCi is the volumetric fraction of rock fragments >2 mm (%) of horizon i.

Soil texture was determined by wet sieving and sedimentationaccording to the method of Köhn (Gee and Bauder, 1986). Soil pHvalues were measured in 0.01 M CaCl2 at a soil/solution ratio of1/2.5 at room temperature. In order to compare soil horizon-spe-cific properties and to gain insight into the distribution of soilproperties with depth, soil data from single soil horizons wereaveraged (SOC and N concentration, BD, SC, pH) or added (SOCand N stock, thickness of soil horizon) to master horizons (L/O = or-ganic horizon; A = topsoil mineral horizon; B = subsoil mineralhorizon; C = parent material).

2.4. Environmental variables

In order to identify the parameters which control the storage ofSOC in Bavarian forest soils, a number of potential controlling vari-

ables were selected which describe the environmental conditionsat each sampling location. Based on a Digital Elevation Model(DEM) with a resolution of 25 m provided by the Bavarian Survey-ing and Mapping Authority (BVV), different topographic parame-ters were calculated (Wilson and Gallant, 2000) using ESRIsArcMap Version 10.0. As primary terrain attributes, elevation, slopeand curvature were determined. Contributing area (CA) and theTopographic Wetness Index (TWI) served as secondary parameters.The TWI was calculated using the following equation:

TWI ¼ lnSCAtan a

� �ð2Þ

where SCA is the specific contributing area and a is the slope. TheTWI is a topographic variable that indicates potential soil moistureconditions (Beven and Kirkby, 1978; Sorensen et al., 2006). To in-clude geology as a potential parameter influencing SOC stocks, theparent material (PM) was assigned for all sampling locations froma map with 35 PM classes (BAG500) provided by the Bavarian Envi-ronment Agency (LfU) (Geuß et al., 2011). As climatic variables, an-nual precipitation and mean annual temperature with a resolutionof 1 km recorded between 1981 and 2010 by the German WeatherService (DWD) were allocated.

2.5. Statistical analysis

Descriptive statistics were applied to describe the data setsincluding mean, standard deviation, median, minimum/maximumvalues, interquartile range and variance. Normal distribution wastested using the Kolmogoroff-Smirnoff test and the Shapiro–Wilktest. Due to the non-normal distributed nature of our data set aKruskal–Wallis one-way analysis of variance was used to test thesignificance of differences between forest types. As a post hoc test,the Wilcoxon–Mann–Whitney U-Test was calculated. For an anal-ysis of the variable importance for the storage of SOC, Spearman’srank correlation coefficients among all parameters were calculated.To include the nominally scaled parameters soil class and parentmaterial, the different classes of soil and parent material wereranked according to their potential to accumulate SOM. For soilclass, this was done on the basis of the mean depth of soil develop-ment, texture and wetness of the respective soil class. For parentmaterial, the degree of weatherability and clay contents in the par-ent material and the weathering product were considered. Theranked soil classes and parent materials were grouped into 5 and10 units, respectively. An incorporation of nominally scaled param-eters using dummy variables was dismissed due to the high num-ber of soil classes and parent materials. Due to strongintercorrelations between environmental parameters, principalcomponent analyses (PCAs) were conducted to extract the mainfactors controlling SOC storage. Stepwise multiple linear regressionmodels were calculated using the extracted factors. All statisticalcalculations were performed using the software IBM SPSS Statistics19.

3. Results

3.1. SOC and N stocks in forest soils of Bavaria

Basic soil characteristics and SOC and N stocks were calculatedfor master horizons of forest soils (Table 1). The organic layer (L/Ohorizons) and the A horizon showed similar amounts of SOC and N(2.4–2.5 and 0.1 kg m�2) despite considerably different SOC and Nconcentrations. Very high contents of SOC and N in the organiclayer were associated with a markedly low BD. In the A horizon,lower SOC and N concentrations were balanced by higher BD.The thickness and the pH of L/O and A horizons were comparable.

Page 4: Storage and drivers of organic carbon in forest soils of southeast Germany (Bavaria) – Implications for carbon sequestration

M. Wiesmeier et al. / Forest Ecology and Management 295 (2013) 162–172 165

In the B horizon, higher amounts of SOC and N are stored (3.6 and0.3 kg m�2) despite lower concentrations due to a higher thicknessand BD. A slight increase of pH was observed. C horizons contrib-uted with 0.8 and 0.2 kg m�2 to SOC and N storage, respectively.They were characterized by low SOC and N concentrations, highBD, increased pH and higher stone contents. For soil texture, noconsiderable differences were determined between master hori-zons. For the entire soil profile down to the parent material or atleast to a depth of 1 m, forest soils of Bavaria stored 9.8 kg m�2

of SOC and 0.3 kg m�2 of N.Total SOC and N stocks were also calculated for different soil

classes according to the German soil systematic (AD-HOC AG Bod-en, 2005) and the related Reference Soil Groups of the WRB (IUSSWorking Group WRB, 2006) (Fig. 2). Lower SOC stocks than themedian value determined from the whole data set were detectedfor clay-rich soils (soil class D; Vertisols, Vertic Cambisols), soilsfrom limestone weathering residues (soil class C; Cambisols, Luvi-sols), Podzols (soil class P), soils with water stagnation (soil class S;Stagnosols, Albeluvisols, Planosols), groundwater soils (soil class G;Gleysols) and soils with clay migration (soil class L; Luvisols) rang-ing between 8.0 and 9.7 kg m�2. Only soils with well-developed Bhorizons (soil class B; Cambisols), soils with initial soil formation(soil class R; Leptosols, Regosols) and alluvial soils (soil class A;Fluvisols, Gleysols and Reference Soil Groups in combination withGleyic qualifier) showed higher amounts of 10.0–14.6 kg m�2 forSOC. Lower N stocks compared with the median value were foundfor soil classes D, P, S and B (0.2–0.3 kg m�2), whereas soil classes C,G, L, R and A showed higher amounts of 0.4–1.2 kg m�2.

Table 1Basic properties of forest soils according to master horizons (L/O = organic layer, A = topsoilpercentile in parentheses).

L/O A

SOC (mg g�1) 425.0 (377.0/460.0) 48.0 (32.4N (mg g�1) 16.5 (14.6/18.3) 2.9 (2.0/4C/N 25 (23/27) 18 (15/22BD (g cm�3) 0.13 (0.10/0.16) 0.91 (0.73SC (%) – 4 (0/10)T (cm) 5 (3/7) 6 (3/13)SOC (kg m�2) 2.5 (1.2/3.9) 2.4 (1.0/4N (kg m�2) 0.1 (0.0/0.2) 0.1 (0.0/0.Sand (%) – 44 (26/58Silt (%) – 39 (29/56Clay (%) – 17 (10/22pH (CaCl2) 3.5 (3.1/4.2) 3.4 (3.2/3

SOC, soil organic carbon; N, nitrogen; BD, bulk density; SC, stone content; T, thickness o

Fig. 2. Entire SOC and N stocks down to the parent material or to a depth of 1 m for diffaccording Reference Soil Groups of the WRB (IUSS Working Group WRB, 2006): D = clay-(Cambisols, Luvisols); P = Podzols; S = soils with water stagnation (Stagnosols, Albelu(Luvisols); B = soils with well-developed B horizons (Cambisols); R = soils with initial soilSoil Groups in combination with Gleyic qualifier) (Lines within the boxes give the meexcluding outliers, circles (1.5–3.0 interquartile range) and triangles (more than 3.0 inte

3.2. Pedogenetic, topographic and environmental controls on forestSOC stocks

To identify the parameters which control the accumulation ofSOC in forest soils, a correlation matrix including different pedoge-netic, topographic and environmental factors was calculated (Ta-ble 2). Total SOC stocks showed highly significant (P < 0.01)correlations with temperature, precipitation, elevation, contribut-ing area and parent material. However, all these parameters werecorrelated among themselves and with other parameters and can-not be regarded as independent determining factors. Thus, a prin-cipal component analysis (PCA) was carried out in order to extractthe main factors controlling SOC storage in forest soils (Table 3).The PCA extracted two factors that explained 50% of the total var-iance. Factor 1 was characterized by high loadings of the parame-ters elevation (0.95), temperature (�0.92) and precipitation (0.92).Factor 2 was driven by the TWI (0.76) and slope (�0.61). A multiplelinear regression model that included these factors revealed factor1 as most important factor (Beta value of 0.51) for total SOC storagein forest soils followed by factor 2 (Beta value of 0.09). The regres-sion model explained 34% of the variance of total SOC stocks.

To gain insight into the driving factors for SOC storage over thesoil profile depth, correlations between SOC stocks from masterhorizons and pedogenetic, topographic and environmental param-eters were calculated (Table 4). SOC stocks in the organic layerwere strongly correlated (P < 0.01) with pH, temperature, slopeand parent material. For A horizon stocks, highly significant corre-lations were found with clay content, pH, temperature, precipita-

, B = subsoil, C = parent material) in Bavaria (median values with the 25th and the 75th

B C

/70.7) 7.5 (4.7/12.1) 2.7 (1.6/4.6).4) 1.3 (0.8/1.7) 0.8 (0.3/1.4)) 11 (7/16) 6 (1/11)/1.10) 1.36 (1.23/1.46) 1.50 (1.37/1.64)

8 (1/25) 30 (3/62)51 (34/73) 38 (25/55)

.2) 3.6 (2.4/5.7) 0.8 (0.4/1.5)3) 0.3 (0.2/0.5) 0.2 (0.1/0.5)) 42 (26/56) 48 (29/68)) 37 (27/47) 33 (24/43)) 21 (13/32) 19 (9/34).9) 4.1 (3.9/4.4) 4.7 (4.1/7.5)

f the horizon.

erent soil classes of the German soil systematic (AD-HOC AG Boden, 2005) and therich soils (Vertisols, Vertic Cambisols); C = soils from limestone weathering residuesvisols, Planosols); G = groundwater soils (Gleysols); L = soils with clay migrationformation (Leptosols, Regosols); A = alluvial soils (Fluvisols, Gleysols and Reference

dian, boxes the 25th and 75th percentile, whiskers the lowest and highest valuesrquartile ranges) represent outliers and extremes).

Page 5: Storage and drivers of organic carbon in forest soils of southeast Germany (Bavaria) – Implications for carbon sequestration

Table 2Correlation matrix of total SOC stocks and different pedogenetic, topographic and environmental parameters.

SOC Temp. Prec. Elev. Exp. Slope Curv. CA TWI Soil C.

Temp. �0.315**

Prec. 0.402** �0.659**

Elev. 0.402** �0.837** 0.789**

Exp. �0.029 0.012 �0.055 �0.028Slope 0.104* �0.395** 0.385** 0.307** �0.120**

Curv. �0.020 �0.010 0.014 0.034 0.006 0.031CA 0.113** �0.077 0.041 0.007 �0.014 0.147** �0.536**

TWI �0.102* 0.395** �0.378** �0.312** 0.127** �0.987** �0.051 �0.107**

Soil C. 0.078 �0.240** 0.186** 0.170** 0.006 0.287** 0.083* �0.017 �0.278**

PM 0.213** �0.293** 0.280** 0.360** �0.100* 0.077 �0.011 0.077 �0.082* 0.006

SOC, soil organic carbon stock (kg m�2); Temp., mean annual temperature (�C); Prec., annual precipitation (mm); Elev., elevation (m a.s.l.); Exp., exposition; Curv., curvature;CA, contributing area; TWI, Topographic Wetness Index; Soil C., soil class according to AD-HOC AG Boden (2005); PM, parent material.

* Level of significance: p 6 0.05.** Level of significance: p 6 0.01.

Table 3Rotated component matrix derived from a principal component analysis of variablescontrolling forest SOC stocks.

Factor 1 Factor 2

Elevation 0.95 �0.02Temperature �0.92 0.06Precipitation 0.92 �0.09Parent material 0.38 0.16TWI �0.39 0.76Slope 0.60 �0.61CA 0.11 0.47Curvature �0.17 �0.45Soil class 0.21 �0.35Exposition 0.03 0.32

Table 4Correlation of SOC stocks from master horizons (L/O = organic layer, A = topsoil,B = subsoil, C = parent material) with different pedogenetic, topographic and envi-ronmental parameters.

SOC (kg m�2) L/O A B C

Clay – 0.428** 0.054 0.063pH �0.693** 0.453** 0.097* 0.231**

Temp. �0.169** �0.239** �0.197** �0.073Prec. �0.035 0.364** 0.282** 0.124*

Elev. 0.123** 0.277** 0.256** 0.127*

Exp. 0.034 �0.082* �0.018 �0.003Slope �0.119** 0.145** 0.053 0.134**

Curv. 0.005 0.002 �0.031 0.062CA �0.014 0.094* 0.045 0.037TWI 0.110** �0.140** �0.044 �0.129*

Soil C. 0.023 0.053 �0.036 0.151**

PM 0.006 0.187** 0.120** �0.029

Temp., mean annual temperature (�C); Prec., annual precipitation (mm); Elev.,elevation (m a.s.l.); Exp., exposition; Curv., curvature; CA, contributing area; TWI,Topographic Wetness Index; Soil C., soil class according to AD-HOC AG Boden

166 M. Wiesmeier et al. / Forest Ecology and Management 295 (2013) 162–172

tion, elevation, slope and parent material. For subsoil horizons (Band C), strong correlations were only found with temperature, pre-cipitation and elevation.

(2005); PM, parent material.* Level of significance: p 6 0.05.

** Level of significance: p 6 0.01.

3.3. Storage of SOC and N and basic soil characteristics of differentforest types

Total SOC and N stocks were calculated for broadleaf, mixed andconiferous forests (Fig. 3). Generally, no significant differences(P < 0.05) were found between forest types with SOC and N stocksranging between 9.4–9.9 kg m�2 and 0.2–0.4 kg m�2, respectively.Slightly lower stocks were determined for coniferous forest soilsbut differences were not significant. In order to reveal the distribu-tion of SOM with depth, SOC and N stocks as well as other soilproperties were calculated for master horizons of soils under dif-ferent forest types (Fig. 4). Basic differences regarding the depthdistribution of SOM between the different forest types were foundin the organic layer and the A horizon. Broadleaf forests exhibitvery shallow organic layers (2 cm) that store relatively lowamounts of SOC and N (0.6 and 0.02 kg m�2). In mixed and conifer-ous forests, the thickness of the organic layer was 2–3 times higherand consequently much higher amounts of SOC (1.8 and3.2 kg m�2) and N (0.08 and 0.12 kg m�2) are stored. The oppositetrend was found for A horizons. The amount of SOC and N in A hori-zons continuously decreased from broadleaf forests (4.1 and0.21 kg m�2) to mixed forests (2.6 and 0.13 kg m�2) and to conifer-ous forests (1.4 and 0.07 kg m�2). No considerable differences werefound for B and C horizons of different forest types. In B horizons,the highest amounts of SOC were stored (3.4–4.2 kg m�2) due tohigh horizon thicknesses (44–52 cm). Much lower amounts ofSOC were determined for C horizons (0.8–1.1 kg m�2). In contrast,the amount of N increased with depth. In B horizons, N stocks ran-ged between 0.25 and 0.34 kg m�2 and rose to 0.40–0.49 kg m�2 in

the C horizon apart from coniferous forests with only 0.15 kg m�2.Generally, no considerable differences between forest types weredetermined for SOC and N concentrations that decreased withdepth and for BD values, which increased with depth.

For a further characterization of soils under different foresttypes, soil texture, stone contents, pH values and C/N ratios werecalculated for master horizons (Fig. 5). In general, soils underbroadleaf forests contained considerably higher contents of siltand clay in all horizons compared with mixed and coniferous forestsoils. The latter were characterized by high contents of sand in allhorizons that were almost twice as high compared with broadleafforests. For mixed and coniferous forest soils, contents of clay ran-ged between 12% and 18% over the total soil profile without appre-ciable differences with depth. In contrast, broadleaf forests showeda marked increase of clay contents from the A horizon (18%) to thesubsoil (28–34%). Silt contents generally decreased with depth andranged between 29% and 38% for mixed and coniferous forests and39–54% for broadleaf forests. Contents of sand remained constantover the soil profile for broadleaf forests (24–28%) and increasedwith depth for mixed (44–52%) and coniferous forests (45–60%).For stone contents, major differences were only detected in the Chorizon with considerably lower amounts in broadleaf forests(15%) compared with mixed (48%) and coniferous forests (32%). Ba-sic differences between forest types were found regarding pH val-ues. Broadleaf forests showed generally higher pH values thanmixed and coniferous forest, particularly in the organic layer with

Page 6: Storage and drivers of organic carbon in forest soils of southeast Germany (Bavaria) – Implications for carbon sequestration

Fig. 3. Total SOC and N stocks down to the parent material or to a depth of 1 m fordifferent forest types (B = broadleaf forest, M = mixed forest, C = coniferous forest)(Lines within the boxes give the median, boxes the 25th and 75th percentile,whiskers the lowest and highest values excluding outliers, circles (1.5–3.0interquartile range) and triangles (more than 3.0 interquartile ranges) representoutliers and extremes).

M. Wiesmeier et al. / Forest Ecology and Management 295 (2013) 162–172 167

pH 5.0 compared to 3.9 and 3.2 for mixed and coniferous forests,respectively. In A horizons, pH values slightly decreased whereasa strong increase was observed down to the C horizon both forbroadleaf forests (pH 7.2) as well as mixed and coniferous forests(pH 5.5 and 4.3). Ratios of C to N constantly decreased with depthfor all forest types and were slightly smaller in L/O horizons frombroadleaf forests compared with coniferous forests.

4. Discussion

4.1. Total storage of SOC in Bavarian forest soils

The analysis of a comprehensive data set of 596 forest soil pro-files within Bavaria revealed a median SOC value of 9.8 kg m�2. A

Fig. 4. SOC and N concentration, bulk density (BD), thickness and SOC and N stocks for mawithin the boxes give the median, boxes the 25th and 75th percentile, whiskers the lowtriangles (more than 3.0 interquartile ranges) represent outliers and extremes).

high proportion was stored in the organic layer (25%) that can berelated to a relatively low incorporation of OM into the mineral soilby the soil fauna due to low pH values and poor litter quality indi-cated by high C/N ratios in L/O horizons (Table 1). Our results arewithin the range of general SOC estimations down to a depth of1 m for temperate forest soils of 6.0–13.9 kg m�2 (IPCC, 2000; Lal,2005; Tyrrell et al., 2012) and European forest soils of 8.1–13.7 kg m�2 (Nabuurs et al., 1997; Goodale et al., 2002; De Vrieset al., 2003). Jobbagy and Jackson (2000) reported higher SOCstocks of 17.4 kg m�2 for temperate deciduous forests but this esti-mation was based on only 60 soil profiles.

Forest SOC inventories from adjacent regions and countries incentral Europe with comparable environmental conditions andtree species allow a direct comparison with our estimations. ForDenmark, considerably higher SOC stocks of 12.5–16.9 kg m�2

were calculated down to a depth of 1 m or for whole soil profiles(Nabuurs and Mohren, 1993; Krogh et al., 2003; Vejre et al.,2003). The authors related these relatively high estimations to apartial lack of bulk density data and to an inclusion of poorlydrained forest locations that accumulate high amounts of SOC.Higher forest SOC stocks of 14.8–15.5 kg m�2 for a depth of 1 mwere also reported for Belgium, but similarly, stock calculationswere partly based on estimated soil properties for bulk densityand stone content (Lettens et al., 2005). The authors state that thiscould lead to an overestimation of SOC stocks. A calculation of for-est SOC stocks with our data set using widely applied pedotransferfunctions (PTFs) for the estimation of bulk densities resulted in anoverestimation of up to 50% (Wiesmeier et al., 2012b). Thus, soilparameters which are necessary for the determination of SOCstocks should be generally measured and PTFs seem to be only use-ful if they are calibrated individually under different environmen-tal conditions and for all sampling depths (De Vos et al., 2005;Schrumpf et al., 2008). However, SOC stock estimations for forestsoils of Switzerland that also partly used PTFs for bulk density were

ster horizons of soils under broadleaf (B), mixed (M) and coniferous (C) forest (Linesest and highest values excluding outliers, circles (1.5–3.0 interquartile range) and

Page 7: Storage and drivers of organic carbon in forest soils of southeast Germany (Bavaria) – Implications for carbon sequestration

Fig. 5. Contents of sand, silt and clay, stone contents (SC), pH values and C/N ratios for master horizons of soils under broadleaf (B), mixed (M) and coniferous (C) forest (Lineswithin the boxes give the median, boxes the 25th and 75th percentile, whiskers the lowest and highest values excluding outliers, circles (1.5–3.0 interquartile range) andtriangles (more than 3.0 interquartile ranges) represent outliers and extremes).

168 M. Wiesmeier et al. / Forest Ecology and Management 295 (2013) 162–172

only slightly higher (9.8 kg m�2 down to C horizon without organichorizons, Perruchoud et al., 2000; 11.9 kg m�2 down to 1 m depth,Bolliger et al., 2008) compared to our results.

For Germany and Bavaria, forest SOC data is available from na-tional forest inventories with sampling grids of 8 � 8 km con-ducted between 1987 and 2008 (BZE I and II). The calculation forGerman forest soils revealed higher SOC stocks of 10.8 kg m�2

down to a depth of 90 cm (extrapolated to 100 cm:12.1 kg m�2)(Wolff and Riek, 1996; Baritz, 1998; Baritz and Strich, 2000). ForBavarian forests, also higher SOC stocks of 11.1–11.7 kg m�2 werecalculated (Schubert, 2010; Hangen and Schubert, 2011). Higherforest stocks determined in these studies can be related to theinclusion of forested peat locations with OC-rich Histosols whichwere excluded in this study. Due to a possible overestimation offorest SOC stocks in several studies that included modelled soilparameters, our estimation of the total amount of SOC in forestsoils of 9.8–11.7 kg m�2 (depending on the inclusion of forestedHistosols) is considerably lower than results from other countriesof central Europe. Besides modelling of soil parameters, a high spa-tial variability of SOC stocks in forest soils within decimeters tometers probably contributes to overestimations of forest SOC stor-age (Schöning et al., 2006; Mäkipää et al., 2008; Price et al., 2012).Particularly sampling close to the trunks resulted in twice as highSOC stocks compared with a sampling design that included severalsatellite locations at greater distances to the trunk as it was appliedin this study (J. Prietzel, personal communication).

4.2. Driving factors of SOC storage in forest soils

The combined approach of regression, principal component andmultiple linear regression analyses clearly showed that the factorclimate, indicated by high loadings of the climate-related parame-ters temperature, precipitation and elevation, mainly controlledthe total storage of SOC in forest soils of Bavaria (Tables 2 and 3).The climate-effect on the storage of SOC has two components. First,

the amount of precipitation controls the above- and belowgroundnet primary productivity of trees and thus the input of OM into thesoil. High amounts of precipitation also lead to acidification of for-est soils that is associated with a lower decomposition of SOM(Meier and Leuschner, 2010). Further, humid climatic conditionsare associated with a profound weathering of the parent materialand thus with the formation of OC-stabilizing minerals. A calcula-tion of SOC stocks stratified to precipitation classes within Bavariashowed generally higher stocks with higher precipitation (Fig. 6).Low amounts of SOC (8.3 kg m�2) were found for regions with pre-cipitation below 800 mm in lowlands of northern Bavaria that con-tinuously increased to 13.1 kg m�2 within the high precipitationregime at the boarder of the Alps (1200–1400 mm) (Fig. 7). Onlyin regions with very high precipitation (>1400 mm) in elevatedareas of the Alps, slightly lower SOC stocks were determined, prob-ably due to unfavourable conditions for SOC accumulation in termsof a short growing season and erosion. A positive correlation be-tween precipitation and overall SOC stocks was also found in a glo-bal analysis of SOC stocks down to a depth of 3 m (Jobbagy andJackson, 2000) and in numerous forest SOC studies (e.g. Paulet al., 2002; Callesen et al., 2003; Baritz et al., 2010; Meier andLeuschner, 2010).

The temperature is the second climate-related componentinfluencing SOC storage. Generally, the microbial decompositionof SOM is temperature-dependent as its complex molecular attri-butes have a high intrinsic temperature sensitivity (Davidson andJanssens, 2006; von Lützow and Kögel-Knabner, 2009; Conantet al., 2011). Although this relationship is governed by multipleconstraints (stability of SOM, substrate availability, physiology ofthe soil microflora and physicochemical controls as pH, water, oxy-gen and nutrients), numerous in situ observations and laboratoryexperiments revealed decreasing amounts of SOM in forest soilsdue to increased soil respiration with rising temperatures (e.g. Job-bagy and Jackson, 2000; Lorenz and Lal, 2010; Meier and Leusch-ner, 2010; Tyrrell et al., 2012; Vesterdal et al., 2012). A clear

Page 8: Storage and drivers of organic carbon in forest soils of southeast Germany (Bavaria) – Implications for carbon sequestration

Fig. 6. Relationship between total SOC stocks and mean annual temperature as well as mean annual precipitation (Lines within the boxes give the median, boxes the 25th and75th percentile, whiskers the lowest and highest values excluding outliers, circles (1.5–3.0 interquartile range) and triangles (more than 3.0 interquartile ranges) representoutliers and extremes).

Fig. 7. Mean annual temperature and annual precipitation in Bavaria determined between 1981 and 2010.

M. Wiesmeier et al. / Forest Ecology and Management 295 (2013) 162–172 169

decrease of SOC stocks with increasing temperatures was alsofound for forest soils of Bavaria (Fig. 6). Forest SOC stocks continu-ously decreased from cool, alpine regions of the Alps (3–4 �C) with25.1 kg m�2 down to 8.8 kg m�2 in the lowlands of northwest Ba-varia (8–9 �C) (Fig. 7). Only for the regions with highest tempera-tures (9–10 �C) slightly higher amounts of SOC were found that isprobably related to very fertile soils in this part of Bavaria. Thestrong temperature-dependence of forest SOC highlights the riskof future SOC losses with rising temperatures in the course of cli-mate change.

Besides climate as the dominant driver for the storage of SOC inforest soils, the factor soil moisture, indicated by the TWI and anegative correlation with slope, was detected (Tables 2 and 3).Large TWI values are usually found in lower positions of a land-scape with large contributing areas and indicate increased likeli-hood of saturated conditions (Sorensen et al., 2006; Grabs et al.,2009) due to topographic position. The reduced mineralization ofOM in groundwater soils in these positions is responsible for astrong accumulation of SOC stocks in forest soils of Bavaria as itwas also indicated by high amounts of SOC in alluvial soils

(Fig. 2). In a related study, we detected TWI as most importantparameter controlling SOC and N storage in agricultural soils of Ba-varia (Wiesmeier et al., submitted for publication).

An analysis of the correlation of SOC stocks of master horizonswith pedogenetic, topographic and environmental factors furtherclarified the variable importance and allowed an incorporation ofhorizon-related parameters (Table 4). The temperature-depen-dence of SOC storage was confirmed for all horizons but seemedto be slightly stronger in the subsoil. This is in line with resultsof Schubert (2010) who found stronger associations of forest SOCstocks with temperature, precipitation and elevation with increas-ing soil depth. The authors suppose that older SOM in the subsoilmirrors the continuous impact of climate on SOC storage over sev-eral centuries. However, this phenomenon could probably also beexplained by the OC-depletion effect of land use that is more pro-nounced in the topsoil than in the subsoil and masks the climate-effect. In contrast, the organic layer and the A horizon are stronglyinfluenced by the tree species composition through litterfall asindicated by strong correlations with pH values. Higher pH valuesof broadleaf litter are associated with a higher bioturbative incor-

Page 9: Storage and drivers of organic carbon in forest soils of southeast Germany (Bavaria) – Implications for carbon sequestration

Fig. 8. SOC stocks in the organic layers and mineral soils under different foresttypes according to classes of mean annual temperature (mean values and standarddeviation, horizontal lines within the bars indicate median values).

170 M. Wiesmeier et al. / Forest Ecology and Management 295 (2013) 162–172

poration of OM into the A horizon whereas in coniferous forests,acidic litter accumulates at the soil surface (Figs. 4 and 5). A closerelationship between litter quality (indicated by C/N ratios, pH andlignin contents) and SOC accumulation was determined in numer-ous studies (e.g. Augusto et al., 2002; Hagen-Thorn et al., 2004;Oostra et al., 2006; Prescott, 2010; Vesterdal et al., 2012). More-over, a strong correlation was found between clay contents andSOC stocks in the A horizon. The stabilizing effect of clay mineralswith high surface areas on SOM is well known and was regarded asdecisive for SOC accumulation in forest soils (Paul et al., 2002; Lal,2005; Baritz et al., 2010; Price et al., 2012). Remarkably, this rela-tionship was only visible in the A horizon but not in subsoil hori-zons. Thus, the total storage of SOC in forest soils of Bavaria doesnot depend on soil texture as it was also indicated by relativelylow SOC stocks of soil classes that are characterized by high claycontents (Fig. 2).

4.3. The impact of forest type on SOC storage – implications for carbonsequestration

A comparison of total SOC stocks under broadleaf, coniferousand mixed forests revealed no significant tree species effect onSOC storage (Fig. 3). This finding is contradictory to several studiesthat determined variable amounts of SOC under different tree spe-cies. Generally, a beneficial effect of coniferous species, particularlyNorway spruce and Scots pine, for SOC sequestration was reportedin many studies due to an accumulation of acidic litter in the or-ganic layer (e.g. Borken and Beese, 2005; Ladegaard-Pedersenet al., 2005; Oostra et al., 2006; Schulp et al., 2008). However, alarge part of those studies failed to incorporate also the subsoilfor the calculation of SOC stocks. For example, Matos et al.(2010) concluded that admixture of Sessile oak (Quercus petraea)in pure stands of Scots pine leads to reduced SOC stocks in the top-soil with increasing age of oaks. However, only the first 20 cm ofthe soil were investigated, what probably led to a considerableunderestimation of SOC in mixed stands. Particularly in the oldadmixtures (124 year old oaks), easily decomposable OM fromoak was likely incorporated deeper than 20 cm by bioturbationand there was probably also an high input of OM into the subsoildue to leaching of dissolved organic carbon (DOC) and deeper root-ing systems. Vesterdal et al. (2008) showed that tree species withlow SOC stocks in the organic layer had generally higher amountsin the underlying mineral soil. They conclude that this oppositetrend may offset the differences in the organic layer and thus thereare no significant differences between tree species if whole soilprofiles are regarded.

However, from a simple comparison of total SOC stocks underdifferent forest types sampled at various locations, no conclusionsshould be drawn regarding carbon sequestration. On the one hand,there is an unequal distribution of broadleaf, coniferous and mixedforests within Bavaria that might distort the results. Moreover, alarge part of total forest SOC stocks is present as litter in the organ-ic layer and as particulate organic matter (POM) in the mineral soilthat constitute a labile SOC pool (von Lützow et al., 2008). As thisunprotected pool is more prone to human disturbance, forest firesand rising temperatures (Jandl et al., 2007; Vesterdal et al., 2008;von Lützow and Kögel-Knabner, 2009; Meier and Leuschner,2010) the contribution of labile SOC, particularly in the organiclayer, to total stocks under different forest types is crucial for man-agement of carbon sequestration. Generally, changes of speciescomposition that result in translocation of SOM from the forestfloor into the mineral horizon are regarded as beneficial for sus-tainable C sequestration in forests (Prietzel, 2004; Jandl et al.,2007; Vesterdal et al., 2008, 2012). Therefore, forest type-specificstocks separated into SOC in the organic layer and the mineral soilwere calculated for temperature classes within Bavaria (Fig. 8). The

stratification according to temperature seems to be sufficient as itwas identified as the main driver of SOC storage. A masking effectof differences between forest types in terms of soil texture, stonecontent and total soil depth can be excluded as these parametersdid not affect SOC storage substantially (Figs. 4 and 5). For all foresttypes a general increase of SOC stocks with decreasing tempera-tures was detected. However, clear differences were determinedin terms of the distribution of SOC in the organic layer and the min-eral soil. In a wide temperature range of 6–10 �C, coniferous foreststored considerably more C in the organic layer compared withmixed and broadleaf forests. In cool, elevated regions below 6 �Cconiferous stored substantial amounts of SOC in the mineral soilbut as only a few locations under mixed forest were sampled inthis area, no comparison can be drawn. Although the total amountsof SOC are similar under all forest types, coniferous forest soilsseem to be less suitable in terms of long-term carbon sequestrationas a large part of SOC is stored in the form of a labile pool. Mixedand broadleaf forests are more advantageous for C sequestrationas a high proportion of SOC is stabilized in the mineral soil. Theseconclusions are based on the assumption that leaching of DOC,which was not detected in this study, is of minor importance andis not significantly different between forest types. Thus, an incor-poration of broadleaf species in coniferous forests, as it was pro-posed by Prescott (2010), would be beneficial for long-term Csequestration as enhanced activity of soil macrofauna, inducedby a higher litter quality (Fig. 5), leads to an incorporation of OMin the mineral soil in form of stable soil aggregates. Establishmentof mixed forests as an effective way to promote C sequestrationwas also proposed by Jandl et al. (2007), Hangen and Schubert(2011) and Prietzel and Bachmann (2012).

Besides a general decrease of SOC of all forest types withincreasing temperatures, mixed and broadleaf forests tended toshow unchanged or even slightly increased mineral SOC stockswith higher temperatures initiated at a temperature >7 �C. Thisdemonstrates that the actual temperature sensitivity of pools sta-bilized by interactions with mineral surfaces and metal ions canbe very low due to the low substrate availability (high ‘cancellingeffect’) (von Lützow & Kögel-Knabner 2009). Also SOC stored in or-ganic layers seems to be unaffected by higher temperatures inmixed and broadleaf forests. Moreover, in the warmest regions ofBavaria (>9 �C), even the total amount of SOC under coniferous for-ests strongly declines and is considerably lower compared tomixed and broadleaf forests. In the light of increasing temperaturesin the course of climate change projected for central Europe andBavaria in the next decades (LfU, 2008), mixed and broadleaf for-ests seem to be much more resistant towards SOC losses thanconiferous forests. Moreover, mixed forests are also less suscepti-

Page 10: Storage and drivers of organic carbon in forest soils of southeast Germany (Bavaria) – Implications for carbon sequestration

M. Wiesmeier et al. / Forest Ecology and Management 295 (2013) 162–172 171

ble to pests and abiotic disturbances (Jandl et al., 2007; Prietzel andBachmann, 2012). The maintenance of expanded coniferous forestareas could lead to a notable decrease of SOC and turn forest soilsinto C sources instead of sinks.

5. Conclusions

Total SOC stocks in forest soils of Bavaria were calculated to be9.8 kg m�2, what is considerably lower than other estimations incentral Europe. The application of PTFs as well as sampling closeto trunks is probably responsible for overestimations in severalSOC inventories. We recommend a sampling design that includesseveral locations with various distances from trunks and measure-ment of all soil parameters down to the parent material for anaccurate determination of forest SOC stocks. The storage of SOCin forest soils was mainly driven by climate with higher stocks un-der cool, humid regimes and lower stocks in warmer, drier regions.This highlights the importance of mountainous forest soils in theAlps and low mountain ranges of Bavaria as a major C pool. Strongcorrelations with climate were superimposed in the topsoil by claycontent and tree species composition indicated by pH values. Nosignificant differences of entire SOC storage were detected be-tween broadleaf, mixed and coniferous forests. However, conifer-ous forests accumulated around 35% of total SOC in theunprotected organic layer that is prone to human disturbance, for-est fires and rising temperatures. In contrast, mixed and particu-larly broadleaf forests stored significantly higher amounts of SOCin the mineral soil. Moreover, mineral SOC stocks of mixed andbroadleaf forests remained unchanged or were even slightly higherin regions with higher temperatures, demonstrating their low ac-tual temperature sensitivity, whereas coniferous mineral SOCstocks were continuously lower. Thus, we conclude that an incor-poration of broadleaf species in expanded coniferous forests of Ba-varia associated with a translocation of SOC from the forest floorinto the mineral soil would promote C sequestration in the long-term, particularly under increased temperatures.

Acknowledgements

We thank Alfred Schubert from the Bavarian State Institute forForestry for providing forest soil data. Peter Schad is acknowledgedfor the allocation of soil classes to the WRB system. We are gratefulto the the Bavarian State Ministry of the Environment and PublicHealth for funding the project ‘‘Der Humuskörper bayerischerBöden im Klimawandel – Auswirkungen und Potentiale’’.

References

AD-HOC AG Boden, 2005. Bodenkundliche Kartieranleitung. Bundesanstalt fürGeowissenschaften und Rohstoffe (Ed.). E. SchweizerbartscheVerlagsbuchhandlung, Stuttgart.

Augusto, L., Ranger, J., Binkley, D., Rothe, A., 2002. Impact of several common treespecies of European temperate forests on soil fertility. Ann. For. Sci. 59, 233–253.

Baritz, R., 1998. Kohlenstoffvorräte der Waldböden Deutschlands.Bundesforschungsanstalt für Forst – und Holzwirtschaft, Eberswalde.

Baritz, R., Strich, S., 2000. Forests and the national greenhouse gas inventory ofGermany. Biotechnol., Agronomy, Soc. Environ. 4, 267–271.

Baritz, R., Seufert, G., Montanarella, L., Van Ranst, E., 2010. Carbon concentrationsand stocks in forest soils of Europe. For. Ecol. Manage. 260, 262–277.

Beven, K.J., Kirkby, M.J., 1978. A physically based, variable contributing area modelof basin hydrology. Hydrol. Sci. Bull. 24, 43–69.

Bolliger, J., Hagedorn, F., Leifeld, J., Bohl, J., Zimmermann, S., Soliva, R., Kienast, F.,2008. Effects of land-use change on carbon stocks in Switzerland. Ecosystems11, 895–907.

Borken, W., Beese, F., 2005. Soil respiration in pure and mixed stands of Europeanbeech and Norway spruce following removal of organic horizons. Can. J. For.Res. – Rev. Can. Rech. For. 35, 2756–2764.

Burschel, P., Kursten, E., Larson, B.C., Weber, M., 1993. Present role of Germanforests and forestry in the national carbon budget and options to its increase.Water Air Soil Pollut. 70, 325–340.

Callesen, I., Liski, J., Raulund-Rasmussen, K., Olsson, M.T., Tau-Strand, L., Vesterdal,L., Westman, C.J., 2003. Soil carbon stores in Nordic well-drained forest soils –relationships with climate and texture class. Glob. Change Biol. 9, 358–370.

Carroll, M., Milakovsky, B., Finkral, A., Evans, A., Ashton, M.S., 2012. Managingcarbon sequestration and storage in temperate and boreal forests. In: Ashton,M.S., Tyrrell, M.L., Spalding, D., Gentry, B. (Eds.), Managing Forest Carbon in aChanging Climate. Springer, New York, pp. 205–226.

Ciais, P., Schelhaas, M.J., Zaehle, S., Piao, S.L., Cescatti, A., Liski, J., Luyssaert, S., Le-Maire, G., Schulze, E.D., Bouriaud, O., Freibauer, A., Valentini, R., Nabuurs, G.J.,2008. Carbon accumulation in European forests. Nat. Geosci. 1, 425–429.

Conant, R.T., Ryan, M.G., Agren, G.I., Birge, H.E., Davidson, E.A., Eliasson, P.E., Evans,S.E., Frey, S.D., Giardina, C.P., Hopkins, F.M., Hyvonen, R., Kirschbaum, M.U.F.,Lavallee, J.M., Leifeld, J., Parton, W.J., Steinweg, J.M., Wallenstein, M.D.,Wetterstedt, J.A.M., Bradford, M.A., 2011. Temperature and soil organic matterdecomposition rates – synthesis of current knowledge and a way forward. Glob.Change Biol. 17, 3392–3404.

Davidson, E.A., Janssens, I.A., 2006. Temperature sensitivity of soil carbondecomposition and feedbacks to climate change. Nature 440, 165–173.

De Vos, B., Van Meirvenne, M., Quataert, P., Deckers, J., Muys, B., 2005. Predictivequality of pedotransfer functions for estimating bulk density of forest soils. SoilSci. Soc. Am. J. 69, 500–510.

De Vries, W., Reinds, G.J., Posch, M., Sanz, M.J., Krause, G.H.M., Calatayud, V., Renaud,J.P., Dupouey, J.L., Sterba, H., Vel, E.M., Dobbertin, M., Gundersen, P., Voogd,J.C.H., 2003. Intensive Monitoring of Forest Ecosystems in Europe – TechnicalReport 2003. UN/ECE, Brussels, Geneva.

Gee, G.W., Bauder, J.W., 1986. Particle-size analysis. In: Page, A.L. (Ed.), Methods ofSoil Analysis, Part 1, Physical and Mineralogical Methods. Soil Science Society ofAmerica, Wisconsin, pp. 383–411.

Geuß, U., Hangen, E., Martin, W., 2011. Hintergrundwerte von anorganischen undorganischen Schadstoffen in Böden Bayerns. Bayerisches Landesamt fürUmwelt, Hof.

Goodale, C.L., Apps, M.J., Birdsey, R.A., Field, C.B., Heath, L.S., Houghton, R.A.,Jenkins, J.C., Kohlmaier, G.H., Kurz, W., Liu, S.R., Nabuurs, G.J., Nilsson, S.,Shvidenko, A.Z., 2002. Forest carbon sinks in the Northern Hemisphere. Ecol.Appl. 12, 891–899.

Grabs, T., Seibert, J., Bishop, K., Laudon, H., 2009. Modeling spatial patterns ofsaturated areas: a comparison of the topographic wetness index and a dynamicdistributed model. J. Hydrol. 373, 15–23.

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. For. Ecol. Manage. 195, 373–384.

Hangen, E., Schubert, A., 2011. Vorrat organischen Kohlenstoffs in WaldbödenBayerns. Bodenschutz 3, 80–83.

Hartge, K.H., Horn, R., 1989. Die physikalische Untersuchung von Böden. EnkeVerlag, Stuttgart.

IPCC, 2000. Land Use, Land-Use Change, and Forestry. Intergovernmental Panel onClimate Change (IPCC), Geneva.

IUSS Working Group WRB, 2006. World Reference Base for Soil Resources 2006.World Soil Resources Reports No. 103, FAO, Rome.

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, 253–268.

Jobbagy, E.G., Jackson, R.B., 2000. The vertical distribution of soil organic carbon andits relation to climate and vegetation. Ecol. Appl. 10, 423–436.

Johnson, D.W., Curtis, P.S., 2001. Effects of forest management on soil C and Nstorage: meta analysis. For. Ecol. Manage. 140, 227–238.

Joneck, M., Hangen, E., Martin, W., Spörlein, P., Foullois, N., Außendorf, M., Reischl,A., Wittenbecker, M., 2006. Wissenschaftliche Grundlagen für den Vollzug derBodenschutzgesetze in Bayern (GRABEN) – ein Projekt stellt sich vor.Bodenschutz 2 (2006), 32–38.

Karjalainen, T., Pussinen, A., Liski, J., Nabuurs, G.J., Eggers, T., Lapvetelainen, T.,Kaipainen, T., 2003. Scenario analysis of the impacts of forest management andclimate change on the European forest sector carbon budget. Forest Policy Econ.5, 141–155.

Krogh, L., Noergaard, A., Hermansen, M., Greve, M.H., Balstroem, T., Breuning-Madsen, H., 2003. Preliminary estimates of contemporary soil organic carbonstocks in Denmark using multiple datasets and four scaling-up methods. Agric.Ecosyst. Environ. 96, 19–28.

Ladegaard-Pedersen, P., Elberling, B., Vesterdal, L., 2005. Soil carbon stocks,mineralization rates, and CO2 effluxes under 10 tree species on contrastingsoil types. Can. J. For. Res. – Rev. Can. Rech. For. 35, 1277–1284.

Lal, R., 2005. Forest soils and carbon sequestration. For. Ecol. Manage. 220, 242–258.Lal, R., 2008. Soil carbon stocks under present and future climate with specific

reference to European ecoregions. Nutr. Cycl. Agroecosyst. 81, 113–127.Lettens, S., Van Orshovena, J., van Wesemael, B., De Vos, B., Muys, B., 2005. Stocks

and fluxes of soil organic carbon for landscape units in Belgium derived fromheterogeneous data sets for 1990 and 2000. Geoderma 127, 11–23.

LfU, 2008. Bayerns Klima im Wandel – erkennen und handeln. BayerischesLandesamt für Umwelt (LfU), Augsburg.

Lindner, M., Karjalainen, T., 2007. Carbon inventory methods and carbon mitigationpotentials of forests in Europe: a short review of recent progress. Eur. J. For. Res.126, 149–156.

Liski, J., Perruchoud, D., Karjalainen, T., 2002. Increasing carbon stocks in the forestsoils of western Europe. For. Ecol. Manage. 169, 159–175.

Lorenz, K., Lal, R., 2010. Carbon Sequestration in Forest Ecosystems. Springer, NewYork.

Page 11: Storage and drivers of organic carbon in forest soils of southeast Germany (Bavaria) – Implications for carbon sequestration

172 M. Wiesmeier et al. / Forest Ecology and Management 295 (2013) 162–172

Luyssaert, S., Ciais, P., Piao, S.L., Schulze, E.D., Jung, M., Zaehle, S., Schelhaas, M.J.,Reichstein, M., Churkina, G., Papale, D., Abril, G., Beer, C., Grace, J., Loustau, D.,Matteucci, G., Magnani, F., Nabuurs, G.J., Verbeeck, H., Sulkava, M., van der Werf,G.R., Janssens, I.A., 2010. The European carbon balance. Part 3: forests. Glob.Change Biol. 16, 1429–1450.

Mäkipää, R., Lehtonen, A., Peltoniemi, M., 2008. Monitoring carbon stock changes inEuropean forests using forest inventory data. In: Dolman, A.J., Valentini, R.,Freibauer, A. (Eds.), The Continental-Scale Greenhouse Gas Balance of Europe.Springer, New York, pp. 191–214.

Martin, P.H., Nabuurs, G.J., Aubinet, M., Karjalainen, T., Vine, E.L., Kinsman, J., Heath,L.S., 2001. Carbon sinks in temperate forests. Annu. Rev. Energy Environ. 26,435–465.

Matos, E.S., Freese, D., Slazak, A., Bachmann, U., Veste, M., Hüttl, R.F., 2010. Organic-carbon and nitrogen stocks and organic-carbon fractions in soil under mixedpine and oak forest stands of different ages in NE Germany. J. Plant Nutr. SoilSci. 173, 654–661.

Meier, I.C., Leuschner, C., 2010. Variation of soil and biomass carbon pools in beechforests across a precipitation gradient. Glob. Change Biol. 16, 1035–1045.

Nabuurs, G.J., Mohren, G.M.J., 1993. Carbon in Dutch forest ecosystems. Neth. J.Agric. Sci. 41, 309–326.

Nabuurs, G.J., Paivinen, R., Sikkema, R., Mohren, G.M.J., 1997. The role of Europeanforests in the global carbon cycle – a review. Biomass Bioenergy 13, 345–358.

Nabuurs, G.J., Thurig, E., Heidema, N., Armolaitis, K., Biber, P., Cienciala, E.,Kaufmann, E., Makipaa, R., Nilsen, P., Petritsch, R., Pristova, T., Rock, J.,Schelhaas, M.J., Sievanen, R., Somogyi, Z., Vallet, P., 2008. Hotspots of theEuropean forests carbon cycle. For. Ecol. Manage. 256, 194–200.

Oostra, S., Majdi, H., Olsson, M., 2006. Impact of tree species on soil carbon stocksand soil acidity in southern Sweden. Scand. J. Forest Res. 21, 364–371.

Paul, K.I., Polglase, P.J., Nyakuengama, J.G., Khanna, P.K., 2002. Change in soil carbonfollowing afforestation. For. Ecol. Manage. 168, 241–257.

Perruchoud, D., Kienast, F., Kaufmann, E., Bräker, O.U., 1999. 20th century carbonbudget of forest soils in the Alps. Ecosystems 2, 320–337.

Perruchoud, D., Walthert, L., Zimmermann, S., Lüscher, P., 2000. Contemporarycarbon stocks of mineral forest soils in the Swiss Alps. Biogeochemistry 50,111–136.

Prescott, C.E., 2010. Litter decomposition: what controls it and how can we alter itto sequester more carbon in forest soils? Biogeochemistry 101, 133–149.

Price, S.P., Bradford, M.A., Ashton, M.S., 2012. Characterizing organic carbon stocksand flows in forest soils. In: Ashton, M.S., Tyrrell, M.L., Spalding, D., Gentry, B.(Eds.), Managing Forest Carbon in a Changing Climate. Springer, New York, pp.7–30.

Prietzel, J., 2004. Humus changes after introduction of beech and oak into Scots-pine monocultures. J. Plant Nutr. Soil Sci. 167, 428–438.

Prietzel, J., Bachmann, S., 2012. Changes in soil organic C and N stocks after foresttransformation from Norway spruce and Scots pine into Douglas fir, Douglas fir/spruce, or European beech stands at different sites in Southern Germany. For.Ecol. Manage. 269, 134–148.

Rumpel, C., Kögel-Knabner, I., 2011. Deep soil organic matter – a key but poorlyunderstood component of terrestrial C cycle. Plant Soil 338, 143–158.

Schnell, A., Bauer, A., 2005. Die zweite Bundeswaldinventur 2002: Ergebnisse fürBayern. Bayerische Landesanstalt für Wald und Forstwirtschaft (LWF), Freising.

Schöning, I., Totsche, K.U., Kögel-Knabner, I., 2006. Small scale spatial variability oforganic carbon stocks in litter and solum of a forested Luvisol. Geoderma 136,631–642.

Schrumpf, M., Schumacher, J., Schöning, I., Schulze, E.D., 2008. Monitoring carbonstock changes in European soils: process understanding and samplingstrategies. In: Dolman, A.J., Valentini, R., Freibauer, A. (Eds.), The Continental-Scale Greenhouse Gas Balance of Europe. Springer, New York, pp. 153–189.

Schubert, A., 2002. Bayerische Waldboden-Dauerbeobachtungsflächen –Bodenuntersuchungen. Wissenschaftszentrum Weihenstephan für Ernährung,Landnutzung und Umwelt der Technischen Universität München undBayerische Landesanstalt für Wald und Forstwirtschaft, Freising.

Schubert, A., 2010. Organisch gebundener Kohlenstoff im Waldboden. LWFWaldforschung aktuell 78, 11–14.

Schulp, C.J.E., Nabulars, G.J., Verburg, P.H., de Waal, R.W., 2008. Effect of tree specieson carbon stocks in forest floor and mineral soil and implications for soil carboninventories. For. Ecol. Manage. 256, 482–490.

Sorensen, R., Zinko, U., Seibert, J., 2006. On the calculation of the topographicwetness index: evaluation of different methods based on field observations.Hydrol. Earth Syst. Sci. 10, 101–112.

Spielvogel, S., Prietzel, J., Auerswald, K., Kögel-Knabner, I., 2009. Site-specific spatialpatterns of soil organic carbon stocks in different landscape units of a high-elevation forest including a site with forest dieback. Geoderma 152, 218–230.

Tyrrell, M.L., Ross, J., Kelty, M., 2012. Carbon dynamics in the temperate forest. In:Ashton, M.S., Tyrrell, M.L., Spalding, D., Gentry, B. (Eds.), Managing ForestCarbon in a Changing Climate. Springer, New York, pp. 77–107.

Vejre, H., Callesen, I., Vesterdal, L., Raulund-Rasmussen, K., 2003. Carbon andnitrogen in Danish forest soils – contents and distribution determined by soilorder. Soil Sci. Soc. Am. J. 67, 335–343.

Vesterdal, L., Schmidt, I.K., Callesen, I., Nilsson, L.O., Gundersen, P., 2008. Carbon andnitrogen in forest floor and mineral soil under six common European treespecies. For. Ecol. Manage. 255, 35–48.

Vesterdal, L., Elberling, B., Christiansen, J.R., Callesen, I., Schmidt, I.K., 2012. Soilrespiration and rates of soil carbon turnover differ among six commonEuropean tree species. For. Ecol. Manage. 264, 185–196.

von Lützow, M., Kögel-Knabner, I., 2009. Temperature sensitivity of soil organicmatter decomposition-what do we know? Biol. Fert. Soils 46, 1–15.

von Lützow, M., Kögel-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 aconceptual model. J. Plant Nutr. Soil Sci. 171, 111–124.

Wamelink, G.W.W., Wieggers, H.J.J., Reinds, G.J., Kros, J., Mol-Dijkstra, J.P., van Oijen,M., de Vries, W., 2009. Modelling impacts of changes in carbon dioxideconcentration, climate and nitrogen deposition on carbon sequestration byEuropean forests and forest soils. For. Ecol. Manage. 258, 1794–1805.

Wiesmeier, M., Hübner, R., Barthold, F.K., Spörlein, P., Geuß, U., Hangen, E., Reischl,A., Schilling, B., von Lützow, M., Kögel-Knabner, I., submitted for publication.Amount, distribution and driving factors of organic carbon and nitrogen inagricultural soils of southeast Germany (Bavaria).

Wiesmeier, M., Spörlein, P., Geuss, U., Hangen, E., Haug, S., Reischl, A., Schilling, B.,von Lützow, M., Kögel-Knabner, I., 2012. Soil organic carbon stocks in southeastGermany (Bavaria) as affected by land use, soil type and sampling depth. Glob.Change Biol. 18, 2233–2245.

Wilson, J., Gallant, J., 2000. Terrain Analysis: Principles and Applications. John Wiley& Sons, Inc., New York.

Wolff, B., Riek, W., 1996. Deutscher Waldbodenbericht 1996 – Ergebnisse derbundesweiten Bodenzustandserhebung im Wald von 1987–1993 (BZE).Bundesministerium für Ernährung, Landwirtschaft und Forsten, Bonn.