Soil classification provides a poor indicator of carbon turnover rates in soil

Soil Biology & Biochemistry 43 (2011) 1688e1696

Contents lists available at ScienceDirect

Soil Biology & Biochemistry journal homepage: www.elsevier.com/locate/soilbio

Soil classification provides a poor indicator of carbon turnover rates in soil P. Simfukwe a, P.W. Hill a, B.A. Emmett b, D.L. Jones a, * a b

School of the Environment, Natural Resources & Geography, Bangor University, Gwynedd LL57 2UW, UK Centre for Ecology and Hydrology, Environment Centre Wales, Bangor, Gwynedd LL57 2UW, UK

a r t i c l e i n f o

a b s t r a c t

Article history: Received 3 December 2010 Received in revised form 9 April 2011 Accepted 24 April 2011 Available online 12 May 2011

Most soil surveys are based on soil geomorphic, physical and chemical properties, while many classifications are based on morphological properties in soil profile. Typically, microbial properties of the soil (e.g. biomass and functional diversity) or soil biological quality indicators (SBQIs) are not directly considered in soil taxonomic keys, yet soil classification schemes are often used to infer soil biological function relating to policy (e.g. soil pollution attenuation, climate change mitigation). To critically address this, our aim was to assess whether rates of carbon turnover in a diverse range of UK soils (n > 500) could effectively be described and sub-divided according to broadly defined soil groups by conventional soil classification schemes. Carbon turnover in each soil over a 90 d period was assessed by monitoring the mineralisation of either a labile (14C-labelled artificial root exudates) or more recalcitrant C source (14C-labelled plant leaves) in soil held at field capacity at 10
C. A double exponential first order kinetic model was then fitted to the mineralisation profile for each individual substrate and soil. ANOVA of the modelled rate constants and pool sizes revealed significant differences between soil groups; however, these differences were small regardless of substrate type. Principle component and cluster analysis further separated some soil groups; however, the definition of the class limits remained ambiguous. Exclusive reference values for each soil group could not be established since the model parameter ranges greatly overlapped. We conclude that conventional soil classification provides a poor predictor of C residence time in soil, at least over short time periods. We ascribe this lack of observed difference to the high degree of microbial functional redundancy in soil, the strong influence of environmental factors and the uncertainties inherent in the use of short term biological assays to represent pedogenic processes which have taken ca. 10,000 y to become manifest. Ó 2011 Elsevier Ltd. All rights reserved.

Keywords: Carbon cycling model Carbon sequestration Nutrient cycling Decomposition Residence time Soil organic matter Soil type

1. Introduction Conventional soil surveys and classifications are typically based on soil geomorphic, physical and chemical properties in which the microbial properties or soil biological quality indicators (SBQIs) are not directly considered. Yet, SBQIs can provide a greater indication of key processes operating in soils at the present time (Parr et al., 1992). In many instances it is this classification information that is required by policymakers when devising strategies for soil protection and evaluating ecosystem service provision. One example of this is the potential role that different soil types have to play in the sequestration of C within policy-relevant timescales (i.e. <30 years). This may be a feature more related to SQBIs rather than traditional soil classes which may reflect processes that dominated thousands or even millions of years ago.

* Corresponding author. Tel.: þ44 1248 382579; fax: þ44 1248 354997. E-mail address: [email protected] (D.L. Jones). 0038-0717/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.soilbio.2011.04.014

Approximately 80% of global, terrestrial biosphere C storage occurs in soil; however, the distribution of this C varies significantly between soil groups (IPCC, 2001). In some cases this C may be very old and may provide a false impression of C sequestration potential in the short term (Paul et al., 1997). Given current concerns about increasing concentrations of CO2 in the atmosphere and their potential effects on global climate, it is of the utmost importance that the factors controlling soil C storage in contrasting soils are understood. Empirically, the quantity of C stored in any soil is determined by the difference between rates of organic matter input and rates of organic matter loss. Most inputs of organic matter to soils are from plants, and most losses are due to decomposition of organic matter C by soil microbes and subsequent return to the atmosphere as respired CO2. To a first approximation, the rate of decomposition of organic matter in different soil types is determined by four factors: environmental conditions (e.g. climate, drainage and land management), the quality of the organic matter (e.g. C:N:P:polyphenol ratios), soil mineralogy (e.g. clay content and type) and existing C content

P. Simfukwe et al. / Soil Biology & Biochemistry 43 (2011) 1688e1696

(Chotte et al., 1998; Marschner et al., 2008; von Lützow et al., 2006). Organic matter inputs to most soils can be broadly characterised into two pools (van Hees et al., 2005). Pool 1 contains highly bioavailable, low molecular weight (MW) compounds (e.g. sugars, organic acids and amino acids) which have a turnover time in soil of hours, while Pool 2 contains more recalcitrant plant polymers (e.g. cellulose, lignin and some proteins) which break down over a daysto-months timescale. Low MW compounds are continually released to the soil through root exudation as well as entering soil via throughfall and when plant and microbial cells are lysed (Jones et al., 2004). Structural polymers are delivered at times of cell death (Jones et al., 2004; Nguyen, 2003). The decomposition of low MW compounds and that of more recalcitrant polymers has often been attributed to different taxa of soil microorganisms (Ekschmitt et al., 2008; McGill et al., 1981; Poll et al., 2008). Organic carbon storage in soil is highly sensitive to environmental change and its amount can vary significantly over decadal timescales. This contrasts strongly with the soil mineral fraction which may take millennia to significantly alter. Association of organic matter with soil minerals has often been linked to changes in its residence time in soils (Kögel-Knabner et al., 2008; Torn et al., 1997). In particular, longer C residence times have been attributed to organic matter protection by association with clay minerals (Saggar et al., 1996, 1999; von Lützow et al., 2006). Further, the capacity for protection of organic matter by association with minerals is dependent on the availability of mineral surfaces (Kögel-Knabner et al., 2008; von Lützow et al., 2006). Consequently, given similar mineralogy, younger soils with lower C contents may have a higher capacity for the protection of organic matter than soils with higher pre-existing C contents (i.e. those that are C saturated). Currently, the exact mechanisms controlling the residence time of organic matter in soil are not fully understood, and a variety of other factors such as soil pH and presence/availability of elements other than C can modify C residence times in individual soils (Kuzyakov et al., 2007; Paul et al., 2008; Rillig et al., 2007). Organic matter turnover rates in soil have frequently been estimated by the addition of isotopically-labelled substrates to soil and measuring mineralisation rates by capturing evolved CO2 (Boddy et al., 2007; Hill et al., 2008; Nguyen and Guckert, 2001). In this investigation we used this approach to examine the variation of substrate mineralisation across a broad range of major soil classes which are used at the national level for predicting greenhouse gas emissions. Environmental conditions and C substrate quantity and quality were controlled. Thus, we were able to directly investigate the effect of soil group on the turnover of organic matter. Specifically, we hypothesised that significant differences in mineralisation would occur between soil groups due to the differences in microbial community structure which occur across a wide range of physical (e.g. structure and mineralogy) and chemical (e.g. pH, NO 3 ) conditions in soil (Ekschmitt et al., 2008; Kögel-Knabner et al., 2008). Secondly, we also hypothesised that greater separation between the different soil groups would occur with a high MW complex source of organic matter in comparison to a low MW simple chemical form of organic matter addition.

1689

UK, based on a stratified random sample of 1 km squares at gridpoints on a 15 km grid using the Institute of Terrestrial Ecology (ITE) Land Classification as the basis of the stratification (Scott, 2008). Fig. 1 shows the general location and distribution of samples across the UK. At each grid intersection, a 1 km2 sample area was selected. Within the 1 km2 sample area, 3 plots (5  5 m2) were randomly located and a single 15 cm long  4 cm diameter soil sample was collected from each of the plots. Topsoils were only selected for sampling to reflect standard practice in national monitoring schemes (Bellamy et al., 2005). The soil horizons sampled included H, O and A horizons with a, e, i, h, g, k and p sub-designations (FAO, 2006). Additional information about vegetation and soils were also collected from the same plots. The 1 km2 areas were stratified within the 45 major Land Classes of the UK. Across all land use categories, the dominant soil groupings (% of total) were: Brown soils (31%), Podzolic soils (15%), Surface water gley soils (18%), Peat soils (13%), Groundwater gley soils (12%), Lithomorphic soils (8%), and Pelosol soils (3%). The WRB equivalent categories for these soil groups are presented in Table 1. All the sites were characterised by a temperate climate with a NortheSouth mean annual temperature range of 7.5e10.6
C and EasteWest mean annual rainfall range from 650 to 1700 mm (Matthew, 2006). To normalise for soil moisture and ensure all soils were at field capacity, artificial rainfall (125 mM NaCl, 15.7 mM CaCl2, 1.3 mM

2. Material and methods 2.1. Soil sampling and preparation Soil samples were collected throughout the UK as part of the Centre for Ecology and Hydrology Countryside Survey (CS) 2007 (Emmett et al., 2010) with sites representing the main types of landscape and soil groups. To encompass all the major soil and land use types, a total of 524 soil samples were collected throughout the

Fig. 1. Map of the UK showing the soil sampling locations used in the study.

1690

P. Simfukwe et al. / Soil Biology & Biochemistry 43 (2011) 1688e1696

was chosen to reflect low MW root exudates and comprised a solution of 14C-glucose (50 mM), 14C-citrate (10 mM), 14C-fructose (5 mM), 14C-malate (5 mM), 14C-sucrose (5 mM) and 14C-succinate (2 mM) and possessed a specific activity of 8.4 Bq mmol1 C. The complex C substrate consisted of 14C-labelled shoots of Lolium perenne (L.) with a specific activity of 12.3 kBq g1. This plant was chosen as it is extremely common in the UK and is found growing in abundance on many of the soil groups. As L. perenne contains all the major chemical constituents found within plants (i.e. protein, cellulose, hemicelluloses, lignin, lipids etc.) we have used it as a model substrate. The 14C-enrichment of L. perenne plant material was performed by pulse labelling with 14CO2 at a constant specific activity and ambient CO2 concentration according to Hill et al. (2007). To characterise the 14C label in the plant material, a sequential chemical fractionation was performed according to Jones and Darrah (1994). Briefly, 50 mg of finely ground plant material was sequentially extracted in 8 ml deionised water for 30 min at 85
C, 8 ml 20% ethanol for 30 min at 80
C, 5 ml 0.3% HCl for 3 h at 95
C and 5 ml 1 M NaOH for 1 h at 95
C. After each extraction step, the sample was centrifuged (5000 g, 15 min), the supernatant removed and its 14C content determined using Optiphase 3Ò Scintillation fluid (PerkinElmer Corp., Waltham, MA) and a Wallac 1404 Liquid Scintillation Counter (PerkinElmer Corp., Waltham, MA). For each soil, a single replicate of 10 cm3 was placed into a sterile 50 cm3 polypropylene container. Either 0.5 ml of the 14C-labelled simple C substrate (artificial root exudates) or 100 mg of the 14 C-labelled complex C substrate (L. perenne shoots) was then added to the soil. A further 0.5 ml of distilled water was added to the soil receiving the complex C substrate to maintain the same moisture content in both treatments. A vial containing 1 M NaOH was then placed above the soil and the polypropylene containers hermetically sealed. The 14CO2 capture efficiency of the NaOH traps was >95%. The soils were then placed in the dark in a climatecontrolled room (10
C) and the NaOH traps exchanged after 0.5 h, 1 d, 7 d, 14 d, 28 d and 90 d. The 14CO2 in the NaOH traps was determined by liquid scintillation counting as described above.

Table 1 Comparable classification of the UK soil groups with those in the FAO World Reference Base Classification (WRB, 2006). Major UK soil group

World Reference Base

Brown Mainly Cambisols with some Luvisols, Acrisols Lithomorphic Leptosols and some Regosols Surface and groundwater gleys Mainly Gleysols, Planosols and some Fluvisols/Luvisols Podzolic Podzols Peat Histosols Pelosol Vertisols

CaSO4, 15.3 mM MgSO4, 12.3 mM H2SO4) was applied to each soil core (10
C) until the soils were fully wetted and 150 ml of leachate had been collected according to the protocol described by Emmett et al. (2008). This normalisation for soil moisture was undertaken to remove water availability as a variable in the C turnover studies. The soils were then incubated at 10
C for 28 d to equilibrate, after which the samples were broken up, mixed by hand, visible roots/ stones removed and the soils immediately used in the mineralisation experiments described below. Soil organic matter was determined on oven-dry soil (105
C, 24 h) by loss-on-ignition (375
C, 16 h). Total C and N were determined on air-dry soil using an Elementar Vario-EL analyzer (Elementaranalysensysteme GmbH, Hanau, Germany). Soil pH was determined in a 1:2.5 (w/v) soil:distilled water extract using fieldmoist soil. Available P was determined using the Olsen P method where air-dry soil was extracted with 0.5 M NaHCO3 (pH 8.5) and P in the extract determined colorimetrically using molybdate blue (Sparks, 1996). Exchangeable Ca and Al were determined by extracting field-moist soil with 1 M NH4Cl (1:5 w/v) according to Sparks (1996). Soil respiration was determined on intact cores by placing then in sealed chambers at 10
C and measuring the amount of CO2 accumulation in the headspace after 1 h using a Clarus 500 GC (Perkin Elmer Corp., Beverley MA). 2.2. Soil classification Soils were classified according to the England and Wales Soil Classification system (Avery, 1990). The system is hierarchical, defined at four successive categorical levels, with classes termed major soil groups, soil groups, soil subgroup and soil series. Soils were classified to one of the six major soil groups namely; Lithomorphic, Brown, Surface water Gley, Groundwater Gley, Podzolic, Peat and Pelosol soils. Table 1 shows the list and their equivalents in the FAO World Reference Base (WRB) soil classification scheme whilst the major properties for each soil group are presented in Table 2.

2.4. Mineralisation kinetics A double first order kinetic model was fitted to the experimental data for each substrate in each individual soil using SigmaPlot v10.0 using a least squares minimization routine (SPSS Inc., Chicago, IL) where

Y ¼ ½a1  expðk1 tÞ þ ½a2  expðk2 tÞ

(1)

14

and where Y represents the amount of C remaining in the soil, a1 and a2 describe the size of the two organic matter pools in the model at time 0, k1 and k2 are the exponential coefficients describing the rate of turnover of pools a1 and a2 respectively, and t is time after substrate addition.

2.3. Mineralisation of substrates A simple or complex 14C-isotopically labelled C substrate was used to estimate mineralisation rates in soil. The simple C substrate

Table 2 Properties of the major soil groups. Values are expressed on a dry weight basis and represent mean  SD. ** and *** indicate significant differences between soil groups at the P < 0.01 and P < 0.001 levels respectively. Superscript letters indicate significant between soil groups at the P < 0.05 level. Soil group

Soil pH

Lithomorphic Brown Groundwater gley Surface water gley Pelosol Podzolic Peat ANOVA

6.0 6.5 6.6 5.9 6.8 4.9 4.6 ***

      

1.4a 1.2a 1.2a 1.3a 1.4a 0.9b 0.6b

Soil organic matter (%) 35.1 10.0 10.9 24.9 7.3 28.8 75.9 ***

      

32.8b 12.5d 13.5d 30.0bc 3.3cd 27.8b 27.9a

C-to-N ratio 16.2 12.2 13.4 14.2 10.7 17.9 24.1 ***

      

6.4bc 4.1e 4.4cde 5.5cd 0.8de 7.0b 8.5a

Olsen P (mg kg1) 41 32 32 25 28 21 44 **

      

69ab 27abc 34abc 28bc 22abc 20c 51a

Bulk density (g cm3) 0.56 1.04 0.98 0.78 1.10 0.57 0.20 ***

      

0.38c 0.33a 0.36a 0.43b 0.21a 0.33c 0.21d

Exchangeable Ca (mmol kg1) 42  39  42  54  62  20  46  ***

39a 27a 31a 49a 40a 29b 50a

Exchangeable Al (mmol kg1) 3.4 2.1 2.3 6.3 1.3 7.1 9.6 **

      

5.6ab 4.8b 5.4ab 14.6ab 4.8ab 8.9ab 38.0a

Soil respiration (mg C cm3 h1) 20.8 8.6 13.2 14.9 10.2 17.9 43.1 ***

      

33.0b 11.0c 24.1bc 19.2bc 26.4bc 32.9bc 37.1a

Soil respiration (g C g SOC1 h1) 6.4 7.3 8.1 6.6 4.7 5.1 4.5 ***

      

5.6ab 5.6a 6.4a 5.0ab 3.4a 3.9b 3.8b

P. Simfukwe et al. / Soil Biology & Biochemistry 43 (2011) 1688e1696

For the simple C substrate, pool a1 is attributable to the rapid use of substrate in catabolic processes leading to loss of 14CO2 in respiration, while pool a2 is attributable to the slower turnover of C incorporated into the microbial biomass via anabolic processes (Boddy et al., 2007, 2008). For the complex C substrate, pool a1 is attributable to the rapid use of labile C (e.g. simple sugars, proteins, amino acids), while pool a2 is attributable to the slower turnover of both the C incorporated into the microbial biomass via anabolic processes and the plant structural C (e.g. cellulose, hemicelluloses and lignin) (Ingwersen et al., 2008). The half-life (HL1) of the substrate pool (a1) was calculated as follows:

HL1 ¼ lnð2Þ=k1

(2)

When 14C-labelled substrates are transformed by microbial processes, a proportion of the 14C remains in the soil and so may enter and re-enter the biomass repeatedly (Kouno et al., 2001; Boddy et al., 2007). Consequently, due to the uncertainty of connectivity between pools a1 and a2 we did not calculate the halflife for pool a2. The stabilisation of organic matter C in soils has been linked to the soil mineralogy, and especially to clay and silt content (Paul et al., 2008; Six et al., 2002; Stewart et al., 2009). As an index of the stabilisation of C in soils we calculated the Biophysical Quotient over the 90 d incubation period (BQ; Saggar et al., 1994, 1999) where

BQ ¼

14

C  respired=14 C  residual in soil

1691

activity (as measured by basal CO2 efflux) were apparent between soils when expressed on a volumetric basis, however, when these were normalised for soil organic matter these differences were far less apparent. Differences were also apparent in chemistry between the soil groups with significant differences apparent in available P, Ca and Al and the soil’s Ca:Al ratio. Overall, clear differences were seen between all soil groups with respect to a particular soil property. The only exception to this was between the brown and groundwater gley soil groups which were not significantly different in all the properties measured here (P > 0.05; Table 1). Overall, the greatest differences in soil properties were seen between the peat and podzolic soils in comparison to the other soil groupings.

3.2. Distribution of

14

C in plant material

Of the total 14C contained in the plant material and subsequently added to soil 32.9  1.5% was extractable by water, 4.2  0.2% by ethanol, 16.8  0.6% by HCl, 27.5  0.4% by NaOH and 18.5  2.2% was insoluble residue. These components approximately correspond to readily decomposable or neutral-detergent soluble C (water and ethanol soluble), cellulose and hemicellulose (HCl soluble) and lignin (NaOH soluble and insoluble-humus) fractions of organic matter respectively (Domisch et al., 1998; Ekschmitt et al., 2008; Moorhead and Sinsabaugh, 2006).

(3)

2.5. Statistical analysis Kinetic parameters describing mineralisation rates for the individual soils within each of the seven soil groups were compared using a one way ANOVA using SPSS v14.0 (SPSS Inc., Chicago, IL). This ANOVA was done for each substrate independently. Post hoc multiple comparisons (pairwise) tests were made using Gabriel test where homogeneity of variance was assumed and Games-Howell procedure where unequal variance was assumed to identify significant differences among specific group pairs. We accepted P  0.05 as an indication of statistical significance. Differences in the major soil properties between the groups were tested by ANOVA with Tukey pair-wise comparison (threshold P < 0.05) using Minitab v16 (Minitab Inc., State College, PA). A principal component analysis (PCA) and a cluster analysis were carried out using Minitab v15 (Minitab Inc.) to explore the interrelationships between soil groups and the kinetic model parameters. For the cluster analysis, the average linkage method and a squared Euclidean distance measure were used with the similarity level measured on the vertical axis. The variables were standardised to minimize the effect of scale differences since the variables were in different units. 3. Results 3.1. Properties of the soil groups As expected, there were significant differences in the major chemical, biological and physical properties between the 7 soil groups (Table 2). Many of these were related to significant differences in soil organic matter (SOM) content (e.g. bulk density, total N, C-to-N ratio). In addition to those presented in Table 2, significant differences between soil groups were also apparent for moisture content at field capacity, soil solution DOC and soluble humic substances which were also related to SOM content (data not presented). Significant differences in total soil microbial

3.3. Substrate mineralisation Following addition of the 14C-labelled substrates (both simple C and complex substrates) to the soil, there was an initial rapid phase of 14CO2 evolution followed by a secondary slower phase of evolution (Fig. 2). The double exponential decay equation gave a good fit to the biphasic experimental data for both substrate forms (mean  SEM; r2 > 0.9984  0.0002 and 0.9994  0.0001 for simple and complex C substrates respectively; n ¼ 524; Fig. 2). The exponential decay coefficients and half-lives (HL) describing the mineralisation of both substrate are presented in Table 3. The amount of 14C recovered and the calculated biophysical quotient (BQ) in the seven soil major soil groups are shown in Table 4.

3.4. Dependence of substrate mineralisation on substrate type Overall, the recovery of 14CO2 was greater and faster (P  0.05) from soils to which the simple substrate was added in comparison to those to which the complex plant substrate was added. By day 7, the recovery of 14CO2 from the simple C substrate amended soils represented 29e32% of the total 14C added. An additional 15e17% was recovered over the remaining 83 d. In contrast, when the more complex plant substrate was added to soils, only 8e12% was recovered as 14CO2 in 7 d and another 29e33% over the remaining 83 d. The rate of mineralisation in the soils amended with the simple C substrate decreased sharply at about day 7, compared to the plant amended soils where the decrease in mineralisation was more gradual. The half-lives calculated from k1 for the complex plant C substrate were approximately 7e10-fold greater (P < 0.001) than the half-lives calculated for the soils amended with the simple C substrate (Table 3). In contrast, the k2 rate constant describing the mineralisation of pool a2 was 6e39% greater for the more complex C substrate (yielding shorter half-times if calculated) than those for the simple C substrate (P < 0.001).

1692

P. Simfukwe et al. / Soil Biology & Biochemistry 43 (2011) 1688e1696

A

Simple C substrate

Brown Groundwater gley Lithomorphic Peat Pelosol Podzolic Surface water gley

90

80

70

60

50

B

100

Complex C substrate

90

80

70

group 2 containing the Peats and group 3 containing the Lithomorphic and Pelosol soils. 3.6. Dependence of mineralisation of the complex C substrate on soil group Microbial allocation of the 14C derived from the labelled plant material to the rapidly-respired pool (a1) was significantly different among the soil groups. Of most significance was the 25e38% lower allocation to pool a1 in the Peat soils in comparison to the other soils (P < 0.001) with the exception of the Podzols (Table 3). In addition, allocation of 14C to pool a1 was 24 and 26% lower (P < 0.05) for the Podzols than for the Brown and Pelosol soils respectively. There were no significant differences (P > 0.05) among the soil groups with respect to the half-time of pool a1 (HL1) with the shortest half-life being only ca. 15.5% shorter than the longest. The exponential rate constant describing the slower phase of 14C loss from the plant material (k2) showed significant differences (P < 0.001) among the soil groups, although these differences were very small. Overall, the k2 values were 15e18% larger for the Brown and Pelosol soils than for the Lithomorphic, Podzolic, Peat or Surface water gley soils. The biophysical quotient (BQ) was also significantly different (P < 0.001) among soil groups and followed the series: Brown ¼ Pelosol > Groundwater gley > Surface water gley ¼ Lithomorphic > Podzolic > Peat. There was no significant relationship (P > 0.05) between the rate of C cycling through the pools a1 and a2, and no correlation was observed in the relationships of k1 and k2 values between and within the two substrate types. 3.7. Multivariate analyses

60

14

C remaining in soil (% of the total

14

C added)

14

C remaining in soil (% of the total

14

C added)

100

50 0

30

60

90

Time (days) Fig. 2. The amount of 14C remaining in different soil groups after the addition of a simple 14C-labelled substrate (root exudates; Panel A) and a more chemically complex 14C-labelled substrate (plant leaf material; Panel B). The curves represent fits of a double first order decay model to the presented data. These curves are not those used to calculate curve parameters presented and discussed as these were calculated for each soil independently. The legend is the same for both panels. Values are means  SEM.

3.5. Dependence of mineralisation of the simple C substrate on soil group The amount of 14C from the simple C substrate allocated to the rapidly-respired pool (a1) was not significantly different (P > 0.05) among the seven soil groups (Table 3). The substrate 14C allocation to pool a1 in the Brown soils (highest allocation) was only 13% more than in Pelosols (lowest allocation). Although the half-time of this pool (HL1), derived from exponential coefficient k1, showed significant differences between soil groups (P < 0.05), only the Lithomorphic soils proved different from the rest. The exponential coefficient for the slower phase of mineralisation (k2) separated the soils into three groups (P < 0.001), the first being the Brown soils which showed a faster mineralisation rate than the second group which contained the Groundwater gleys, and which were faster than the third which contained the Lithomorphic, Peat, Pelosol, Podzolic and Surface water gley soils. The biophysical quotient (BQ) also revealed significant differences between soil groups with three groups apparent (P < 0.001), namely, group 1 containing the Browns Groundwater gley, Podzolic and Surface water gley soils,

Principal component analysis (PCA; Fig. 3) and cluster analysis tree diagram (dendrogram, Fig. 4) were used to assess the relationships among the kinetic model parameters and soil groups. The PCA biplot yielded a clear separation of some soil groups with respect to several principal components. The first two principle component accounted for 76% (47 and 29% respectively) of the total variance observed in the variables, to explain differences in the soil group. The first axis represents a gradient from variables on Brown and Groundwater gley soils (on the right) separated from the Lithomorphic, Surface water gley and Peat soils (on the left) with the Pelosol and Podzolic soils (in the middle) being intermediate. The axis 1 is a substrate half-life gradient with HL2 for the simple C substrate and HL1 for the plant substrate scoring high for soil groups on the left and HL1 for the simple C substrate and k2 for the simple C substrate for soil groups on the right (compare Fig. 3 and Table 3). The second axis separated the Pelosols (on top) from the Peats and Podzolic soils (on the bottom) with all other soils being intermediate. However, the PCA suffers from the horseshoe effect and therefore we cannot easily tell whether the Pelosols are at one end of a secondary gradient, or if its position at the end of axis 2 is merely a distortion. The direction of the variable arrows indicates the greatest change in magnitude of the variable, whereas its length may be related to the rate of change (Ramette, 2007). Angles between variable arrows reflect their correlations, e.g. putative interactions between variables (Ramette, 2007). The dendrogram (Fig. 4) from the cluster analysis was used to demonstrate the relationship amongst the soils. Upon examination of the similarity and distance levels, the final partitioning was set to identify three soil groups which were distinctly different from each other but also share common characteristics within themselves. The groups formed were: group 1, comprising Brown, Groundwater gley and Podzolic soils; group 2, comprising Lithomorphic, Peat and Surface water gley soils, and group 3, comprising Pelosols.

P. Simfukwe et al. / Soil Biology & Biochemistry 43 (2011) 1688e1696

1693

Table 3 Coefficients for the first order decomposition model describing the turnover of a simple C substrate and a more complex C substrate in a range of soil groups. The pool size and the mineralisation rate constant for the fast and slow phases of the kinetics model are represented by a1 and a2, and k1 and k2 respectively. The half-times for the respective pools were defined by 0.693/k. Values represent means  SEM. NS indicates no significant difference between soil groups (P > 0.05) while *, ** and *** indicate significant differences between soil groups at the P < 0.05, P < 0.01 and P < 0.001 levels respectively. Superscript letters indicate significant between soil groups at the P < 0.05 level. Soil group

Pool a1 (% of total

k1 (d1) 14

Half-life for pool a1 (h)

C)

Simple C substrate (root exudates) Brown 32.5  0.5 Surface water gley 29.2  0.8 Groundwater gley 31.1  1.2 Lithomorphic 29.9  1.2 Peat 30.0  0.8 Pelosol 28.2  1.2 Podzolic 31.3  0.8 ANOVA NS

0.74 0.81 0.80 0.96 0.78 0.81 0.76 *

Complex C substrate (plant shoots) Brown 19.8  0.4a Surface water gley 18.9  0.7a Groundwater gley 18.7  0.8a Lithomorphic 16.9  1.2a Peat 12.6  0.7b Pelosol 20.4  1.5a Podzolic 15.0  0.7c ANOVA ***

0.097 0.093 0.108 0.099 0.097 0.095 0.102 NS

      

0.03a 0.04b 0.04b 0.05c 0.03b 0.10b 0.03a

      

0.002 0.004 0.004 0.006 0.005 0.005 0.004

4. Discussion 4.1. Dependence of substrate mineralisation on soil group and implications for soil C sequestration The rate of soil organic matter (SOM) mineralisation is controlled by many biotic and abiotic factors of which, temperature, moisture content and substrate quality are typically deemed to be the most important (Cavigelli et al., 2005; Kuzyakov et al., 2007). In our study, all these factors were kept constant across the experiments to permit direct comparison of C turnover in our contrasting soil groups. This implies that observed treatment differences in soil 14 CO2 evolution patterns were wholly dependent upon the intrinsic properties of the soil (e.g. available nutrients, microbial community structure, exchangeable cations etc). Contrary to expectation, especially given the known differences in the physico-chemical properties and below-ground biodiversity in UK soils (Table 2 Bardgett et al., 2001), the effect of soil group on organic matter decomposition rate was relatively minor. Here we showed that the release of 14CO2 from the soils with the lowest mineralisation rates

1.15 1.00 1.06 0.82 1.00 1.02 1.10 * 7.7 8.3 7.0 7.8 8.2 7.5 7.7 NS

      

      

0.05a 0.05a 0.07a 0.05b 0.05a 0.12a 0.07a

0.2 0.3 0.3 0.5 0.4 0.4 0.3

Pool a2 (% of total

k2 (d1) 14

C)

67.0 70.3 68.3 69.7 69.5 71.5 68.2 NS

      

0.1 0.8 1.2 1.2 0.8 1.2 0.8

0.0035 0.0031 0.0032 0.0027 0.0028 0.0028 0.0031 ***

      

0.0001a 0.0001c 0.0002b 0.0001c 0.0001c 0.0002c 0.0001c

80.2 81.0 81.2 83.0 87.3 79.6 84.9 ***

      

0.4a 0.7a 0.8a 1.2a 0.7b 1.5a 0.7c

0.0039 0.0033 0.0036 0.0034 0.0034 0.0039 0.0034 ***

      

0.0001a 0.0001c 0.0001b 0.0001c 0.0001c 0.0003a 0.0001c

was only 12 and 19% (simple and complex C respectively) less than that from the soils with the overall highest mineralisation rates. Given such a wide range of soil groups from a similarly wide range of ecosystems were investigated, it is surprising that the capacity of soil microbes to take up and utilize the added organic matter varied so little. However, the results are relatively consistent with the small differences in basal respiration observed between the soil groups when CO2 efflux was expressed on an SOM basis (Table 2). Consequently, the predicted capacity for C storage in soils was also largely independent of soil group under the conditions tested here. Our results are in accordance with Ananyeva et al. (2008) who also found no major differences in microbial efficiency (qCO2) in soil types from different climatic regions across European Russia. The lack of soil group difference in low MW substrate use (e.g. amino acids, organic acids) has also been noted in studies that have

Table 4 Total 14C recovered (as 14CO2) and the calculated biophysical quotient (BQ) in the seven soil groups for the simple C substrate (root exudates) and the complex C substrate (plant leaves). BQ is the ratio of the respired 14C and residual 14C remaining in the soil after 90 d. Values represent means  SEM. Superscript letters indicates significant between soil groups. **, *** indicate significant differences between soil groups at the P < 0.01 and P < 0.001 level respectively. Soil group

Brown Surface water gley Groundwater gley Lithomorphic Peat Pelosol Podzolic ANOVA

Simple C substrate

Complex C substrate

Respired C (% total 14C added)

BQ

Respired C (% total 14C added)

BQ

50.2  0.7a 45.8  1.0b

1.03  0.03a 0.94  0.05a

43.5  0.4a 39.6  0.6b

0.78  0.01a 0.67  0.02c

48.9  1.7a

0.95  0.06a

40.9  0.6b

0.70  0.02b

45.8 45.8 44.1 47.9 **

   

1.7ab 1.0b 1.8b 1.1ab

0.81 0.86 0.81 0.97 ***

   

0.04b 0.03c 0.05b 0.04a

39.0 35.6 43.8 37.4 ***

   

0.9b 0.6c 1.6ab 0.6bc

0.65 0.56 0.83 0.61 ***

   

0.02c 0.01d 0.03a 0.02e

Fig. 3. A PCA biplot of mineralisation model kinetic parameters and the seven major soil groups investigated in this study. The parameters are represented by lines and the abbreviation of the parameter and the soils by dots and soil names abbreviations. The suffix or prefix _re and _p represent simple (root exudates) and complex (plant leaf materials) carbon substrates respectively. HL1 and HL2 indicates half-lives for the substrate 2-pool model, a1 and a2 represent the size of these pools and and k1 and k2 the rate constants for these pools respectively. BQ indicates the biophysical quotient and 90 the amount or substrate mineralised after 90 d. Further details of the 2-pool model and its coefficients are provided in the Materials and methods section. SWG and GWG indicate surface water and groundwater gley soils respectively.

P. Simfukwe et al. / Soil Biology & Biochemistry 43 (2011) 1688e1696

Similarity

1694

Soil types Fig. 4. Cluster analysis tree diagram (dendrogram) showing three different soil groups at about <50% similarity level in the kinetic model parameter means. SWG and GWG indicate surface water and groundwater gley soils respectively.

included soils from different geographical locations sampled both at a local and global scale (Jones et al., 2005, 2009). Indeed, from Jones (1999) we note that there appears to be greater effect of soil depth (topsoil versus subsoil) on low MW substrate mineralisation in comparison to topsoils collected from different geographical locations. This finding also has important implications for future national monitoring schemes where we recommend that C turnover in subsoil samples should also be examined. This is particularly pertinent considering recent reports that subsoils can represent a major store of C which can be highly dynamic (Lorenz and Lal, 2005; Don et al., 2011). As sorption of substrates to the soil’s solid phase can significantly reduce substrate bioavailability (van Hees et al., 2005), we had expected our seven soil groups to differentially influence mineralisation, particularly as our simulated root exudate mixture contained strongly sorbing organic anions (e.g. citrate). However, this was not apparent, possibly because the largest differences in sorption between soil groups occur in sub-surface horizons and possibly because rapid microbial uptake can largely negate the impact of solid phase sorption (Fischer et al., 2010). Our mineralisation assays focused on topsoils. Thus, we acknowledge that it is possible that greater differences in C assimilation rates between soils than we have found occur deeper in the soil profile. Nevertheless, the topsoil receives the bulk of plant C inputs and most of the microbial biomass is located there. Consequently, as, the gross quality of root exudates varies little between plant species (i.e. they all contain appreciable quantities of easily assimilated sugars, amino acids and organic acids; Jones et al., 2004), and the soil microbial community is strongly C limited, overall differences in assimilation rate between soils are probably small. A constraint of our study was that the mineralisation assays were only performed for a relatively short period (3 months) relative to the time taken for pedogenesis in the UK (ca. 10,000 y since the last glacial period; Avery, 1990). Based on an average soil C density of 120e200 t C ha1 and a pedogenic period of active C accumulation of 10,000 y for UK soils (Howard et al., 1995), this equates to a theoretical linear C accumulation rate of 0.01e0.02 t C ha1 y1 (ca. 5e10 mg C kg1 y1). However, it is well established that C accumulation in soils is non-linear due to the progressive reduction in the potential for chemical and physical SOM protection and consequently most soils in the UK have been at quasi steady with regard to C accumulation for hundreds, if not thousands of years (King et al., 2005; Hopkins et al., 2009). Therefore the rates of net C accumulation in soil can be expected to

be extremely small (<104 t C ha1 y1), particularly in comparison to annual above- and below-ground C inputs for typical vegetation types within our study area (ca. 1 to 10 t C ha1 y1; Tipping et al., 2010). Therefore in retrospect it may not be surprising that few differences were seen between the soil groups. Our results therefore confirm the long held view that plant residue quality, moisture and temperature are the key regulators of C storage, factors that were normalised in our laboratory incubation experiments. Our experiments do, however, suggest that the composition of the soil microbial community is relatively unimportant in SOM decomposition at least in the well established land uses present in the UK. This goes against recent reviews that have suggested that we need to focus further research on understanding microbial community structure if we are to accurately model soil C dynamics (Drenovsky et al., 2008; McGuire and Treseder, 2010). Our results, alongside others, suggest that significant functional redundancy exists in the microbial population and that this would not represent a sensitive modelling parameter (Fitter et al., 2005). Despite the similarities in mineralisation between soil groups, we did observe some small soil group-dependent differences. Most notably, the microbial community in the peat soils tended to allocate the lowest proportion of both C substrates to the rapidlyrespired pool a1, and also had relatively low k2 values for both substrates. Thus, as would be expected from the large existing C stores (IPCC, 2001), peat soils had the lowest decomposition rates and, given equal input rates, would have the greatest capacity for soil C sequestration, regardless of substrate type. This tendency for peat soils to allocate more C to the slow-respired pool a2 and low decomposition rates was more apparent with the plant substrate. We ascribe this to low rates of enzyme activity due to the low O2 content (which inhibits phenol oxidase) and the abundance of humic substances in soil solution (which binds to enzymes causing steric hindrance; Brannon and Sommers, 1985). Conversely, the relatively fertile Brown soils had a relatively high C allocation to pool a1, high k2 values and high BQ values for both substrates, leading to a lower C sequestration rate per unit of added C than most other soils. High capacity to store new C has often been attributed to young, low C soils such as Lithomorphic soils where sites for organic matter protection are unoccupied i.e. high C saturation deficit (Paul et al., 2008; Stewart et al., 2009; von Lützow et al., 2006). Despite having a relatively short HL1 for the labile substrate showing an active microbial biomass, Lithomorphic soils had smaller k2 values for both substrates than Brown soils, showing a longer residence time for most of the added C regardless of the form of C. The clay rich Pelosols also appeared to have a relatively long residence time for C. For the simple C substrate they had relatively high allocation to pool a2, low k2 and low BQ values. In contrast, the capacity of Pelosols to sequester C derived from plant litter appeared to be relatively low and similar to that of Brown soils. This suggests that protection of microbes, and microbially modified C by clay particles was significant, whilst the decomposition rate of unmodified plant material was not significantly altered by clay content.

4.2. Impact of soil group on microbial C use efficiency Our results also suggest that the flow of C through the soil microbial community may be broadly similar between soil groups. Although large differences in C use efficiency occur for different substrates (Boddy et al., 2007; Fischer et al., 2010), our results suggest that the C use efficiency is remarkably similar across soil groups being on average 0.70  0.01 mmol biomassC mmol substrate-C1 (root exudate C). Again, this supports previous global surveys showing similarities in C use efficiency for

P. Simfukwe et al. / Soil Biology & Biochemistry 43 (2011) 1688e1696

soil microbial communities when supplied with amino acid-C (Jones et al., 2009). For the root exudate mixture, it is known that depletion of pool a1 corresponds to loss of substrate from the soil solution and that these simple compounds are relatively short lived in soil in an intact form (<24 h; Jones, 1999). Therefore, it can be assumed that the majority of the C remaining in the soil was immobilized in the soil microbial biomass (i.e. pool a2; Boddy et al., 2008; Fischer et al., 2010). The results presented here suggest very little difference in the rates of C turnover within the soil microbial community (k2 values presented in Table 3). Again this supports a meta-analysis of soil microbial biomass by Gonzalez-Quiñones et al. (2011) which suggests that the size of the microbial biomass is also stable in topsoils relative to the amount of organic matter inputs. 4.3. Interrelationships among variables The PCA and the cluster analysis revealed some distinct groups of soil groups with respect to the first and second principal components used. Although some soil groups were clearly distinct from others (e.g. Peats versus Brown), other soils were not (e.g. Podzolics versus Surface water gleys) making it difficult to separate them (Figs. 3 and 4). Cutting the dendrogram at 50% similarity level yielded three distinct soil groups (Fig. 4) with the Browns, Groundwater gley, Podzolics forming group 1; the Lithomorphics, Peats and Surface water gleys in group 2 and the Pelosols forming group 3. Elevating the final partitioning to a higher similarity level, could result in splitting Browns from the Groundwater gleys and Podzolics in Group 1 making four overall soil groups. 5. Conclusions Using a large number of soils sampled over a wide geographical scale we set out to test whether different soil groups had different intrinsic capacities to mineralise and sequester C under laboratory conditions. Although the rate of decomposition of two contrasting C substrates varied between individual soil samples, few systematic soil group effects could be identified. Overall, most mineralisation model parameters proved remarkably similar across soil groups, suggesting a lack of difference in C sequestration potential between soil groups. Our results suggest that soil groups defined at the major group level are not of sufficient resolution to allow predictions of variation in C turnover at a national scale when assayed in the laboratory using standard protocols. We ascribe this lack of observed difference to the high degree of microbial functional redundancy in soil, the strong influence of environmental factors and the uncertainties inherent in the use of short term biological assays to represent pedogenic processes which have taken ca. 10,000 y to become manifest. Acknowledgements We would like to thank Robert Mills, Robert Griffiths, David Cooper and Ed Rowe at the Centre for Ecology and Hydrology for help with the experiments and data handling. This work was funded by the Natural Environment Research Council (NERC) and a consortium of UK government departments and agencies headed by the UK Department for Environment, Food and Rural Affairs (Contract no. CR0360). References Ananyeva, N.D., Susyan, A.E., Chernova, V.O., Wirth, S., 2008. Microbial respiration activities of soils from different climatic regions of European Russia. European Journal of Soil Biology 44, 147e157.

1695

Avery, B.W., 1990. Soils of the British Isles. Cambridge University Press, Cambridge. Bardgett, R.D., Jones, A.C., Jones, D.L., Kemmitt, S.J., Cook, R., Hobbs, P.J., 2001. Soil microbial community patterns related to the history and intensity of grazing in sub-montane ecosystems. Soil Biology & Biochemistry 33, 1653e1664. Bellamy, P.H., Loveland, P.J., Bradley, R.I., Lark, R.M., Kirk, G.J.D., 2005. Carbon losses from all soils across England and Wales 1978-2003. Nature 437, 245e248. Boddy, E., Hill, P.W., Farrar, J., Jones, D.L., 2007. Fast turnover low molecular weight components of the dissolved organic carbon pool of temperate grassland field soil. Soil Biology & Biochemistry 39, 827e835. Boddy, E., Roberts, P., Hill, P.W., Farrar, J., Jones, D.L., 2008. Turnover of low molecular weight dissolved organic C (DOC) and microbial C exhibit different temperature sensitivities in Arctic tundra soils. Soil Biology & Biochemistry 40, 1557e1566. Brannon, C.A., Sommers, L.E., 1985. Stability and mineralization of organic phosphorus incorporated into model humic polymers. Soil Biology & Biochemistry 17, 221e227. Cavigelli, A.M., Lengnick, L.L., Buyer, S.J., Fravel, D.C., Handoo, Z., McCarty, G., Millner, P., Sikora, L., Wright, S., Vinyard, B., Rabenhorst, M., 2005. Landscape level variation in soil resources and microbial properties in a no-till corn field. Applied Soil Ecology 29, 99e123. Chotte, J.L., Ladd, J.N., Amato, M., 1998. Sites of microbial assimilation, and turnover of soluble and particulate 14C-labeled substrates decomposing in a clay soil. Soil Biology & Biochemistry 30, 205e218. Domisch, T., Finer, L., Karsisto, M., Laiho, R., Laine, J., 1998. Relocation of carbon from decaying litter in drained peat soils. Soil Biology & Biochemistry 30, 1529e1536. Don, A., Schumacher, J., Freibauer, A., 2011. Impact of tropical land-use change on soil organic carbon stocks e a meta-analysis. Global Change Biology 17, 1658e1670. Drenovsky, R.E., Feris, K.P., Batten, K.M., Hristova, K., 2008. New and current microbiological tools for ecosystem ecologists: towards a goal of linking structure and function. American Midland Naturalist 160, 140e159. Ekschmitt, K., Kandeler, E., Poll, C., Brune, A., Buscot, F., Friedrich, M., Gleixner, G., Hartmann, A., Kästner, M., Marhan, S., Miltner, A., Scheu, S., Wolters, V., 2008. Soil-carbon preservation through habitat constraints and biological limitations on decomposer activity. Journal of Plant Nutrition and Soil Science 171, 27e35. Emmett, B.A., Frogbrook, Z.L., Chamberlain, P.M., Griffiths, R., Pickup, R., Poskitt, J., Reynolds, B., Rowe, E., Rowland, P., Spurgeon, D., Wilson, J., Wood, C.M., 2008. Countryside Survey. Soils Manual. NERC/Centre for Ecology & Hydrology, CS Technical Report No.3/07, CEH Project Number: C03259, 180 pp. Emmett, B.A., Reynolds, B., Chamberlain, P.M., Rowe, E., Spurgeon, D., Brittain, S.A., Frogbrook, Z., Hughes, S., Lawlor, A.J., Poskitt, J., Potter, E., Robinson, D.A., Scott, A., Wood, C., Woods, C., 2010. Countryside Survey. Soils Report from 2007. CS Technical Report No. 9/07. Centre for Ecology and Hydrology, Wallingford, UK. FAO, 2006. Guidelines for Soil Description, fourth ed. Food and Agriculture Organization of the United Nations, Rome, Italy. Fischer, H., Ingwersen, J., Kuzyakov, Y., 2010. Microbial uptake of low-molecularweight organic substances out-competes sorption in soil. European Journal of Soil Science 61, 504e513. Fitter, A.H., Gilligan, C.A., Hollingworth, K., Kleczkowski, A., Twyman, R.M., Pitchford, J.W., 2005. Biodiversity and ecosystem function in soil. Functional Ecology 19, 369e377. Gonzalez-Quiñones, V., Stockdale, E.A., Banning, N., Hoyle, F.C., Sawada, Y., Wherrett, A., Jones, D.L., Murphy, D.V., 2011. Soil microbial biomass e Interpretation and consideration for soil monitoring. Australian Journal of Soil Research, in press. van Hees, P.A.W., Jones, D.L., Finlay, R., 2005. The carbon we do not see - the impact of low molecular weight compounds on carbon dynamics and respiration in forest soils: a review. Soil Biology & Biochemistry 37, 1e13. Hill, P.W., Marshall, C., Williams, G.G., Blum, H., Harmens, H., Jones, D.L., Farrar, J.F., 2007. The fate of photosynthetically-fixed carbon in Lolium perenne L. grassland as modified by elevated CO2, nitrogen and sward management. New Phytologist 173, 766e777. Hill, P.W., Farrar, J.F., Jones, D.L., 2008. Decoupling of microbial glucose uptake and mineralization in soil. Soil Biology & Biochemistry 40, 616e624. Hopkins, D.W., Waite, I.S., McNicol, J.W., Poulton, P.R., Macdonald, A.J., O’Donnell, A.G., 2009. Soil organic carbon contents in long-term experimental grassland plots in the UK (palace leas and park grass) have not changed consistently in recent decades. Global Change Biology 15, 1739e1754. Howard, P.J.A., Loveland, P.J., Bradley, R.I., Dry, F.T., Howard, D.M., Howard, D.C., 1995. The carbon content of soil and its geographical-distribution in Great Britain. Soil Use and Management 11, 9e15. IPCC, 2001. Climate Change: The Scientific Basis. Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge. Ingwersen, J., Poll, C., Streck, T., Kandeler, E., 2008. Micro-scale modelling of carbon turnover driven by microbial succession at a biogeochemical interface. Soil Biology & Biochemistry 40, 864e878. Jones, D.L., 1999. Amino acid biodegradation and its potential effects on organic nitrogen capture by plants. Soil Biology & Biochemistry 31, 613e622. Jones, D.L., Darrah, P.R., 1994. Simple method for 14C labeling root material for use in root decomposition studies. Communications in Soil Science and Plant Analysis 25, 2737e2743. Jones, D.L., Hodge, A., Kuzyakov, Y., 2004. Plant and mycorrhizal regulation of rhizodeposition. New Phytologist 163, 459e480.

1696

P. Simfukwe et al. / Soil Biology & Biochemistry 43 (2011) 1688e1696

Jones, D.L., Kemmitt, S.J., Wright, D., Cuttle, S.P., Bol, R., Edwards, A.C., 2005. Rapid intrinsic rates of amino acid biodegradation in soils are unaffected by agricultural management strategy. Soil Biology & Biochemistry 37, 1267e1275. Jones, D.L., Kielland, K., Sinclair, F.L., Dahlgren, R.A., Newsham, K.K., Farrar, J.F., Murphy, D.V., 2009. Soil organic nitrogen mineralization across a global latitudinal gradient. Global Biogeochemical Cycles 23, GB1016. Kögel-Knabner, I., Guggenberger, G., Kleber, M., Kandeler, E., Kalbitz, K., Scheu, S., Eusterhues, K., Leinweber, P., 2008. Organo-mineral associations in temperate soils: integrating biology, mineralogy, and organic matter chemistry. Journal of Plant Nutrition and Soil Science 171, 61e82. King, J.A., Bradley, R.I., Harrison, R., 2005. Current trends of soil organic carbon in English arable soils. Soil Use and Management 21, 189e195. Kouno, K., Wu, J., Brookes, C.P., 2001. Turnover of biomass C and P in soil following incorporation of glucose or ryegrass. Soil Biology & Biochemistry 34, 617e622. Kuzyakov, Y., Hill, P.W., Jones, D.L., 2007. Root exudate components change litter decomposition in a simulated rhizosphere depending on temperature. Plant and Soil 290, 293e305. Lorenz, K., Lal, R., 2005. The depth distribution of soil organic carbon in relation to land use and management and the potential of carbon sequestration in subsoil horizons. Advances in Agronomy 88, 35e66. von Lützow, M., Kögel-Knabner, I., Ekschmitt, K., Matzner, E., Guggenberger, G., Marschner, B., Flessa, H., 2006. Stabilization of organic matter in temperate soils: mechanisms and their relevance under different soil conditions-review. European Journal of Soil Science 57, 426e445. 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., Rethenmeyer, J., Schäffer, A., Schmidt, M.W.I., Schwark, L., Wiessenberg, G.L.B., 2008. How relevant is recalcitrance for the stabilization of organic matter in soils? Journal of Plant Nutrition and Soil Science 171, 91e110. Matthew, P., 2006. Climate Memorandum No 21. A Spatial Analysis of Trends in the UK Climate Since 1914 Using Gridded Datasets. Version 1.1. Scientific Consultancy, Met Office, Exeter. McGill, W.B., Hunt, H.W., Woodmansee, R.G., Reuss, J.O., 1981. PHOENIX, a model of the dynamics of carbon and nitrogen in grassland soil. In: Clark, F.E., Roswall, T. (Eds.), Terrestrial Nitrogen Cycles. Processes, Ecosystem Strategies and Management Impacts. Blackwell Publishing, Malden, USA, pp. 49e115. McGuire, K.L., Treseder, K.K., 2010. Microbial communities and their relevance for ecosystem models: decomposition as a case study. Soil Biology & Biochemistry 42, 529e535. Moorhead, D.L., Sinsabaugh, R.L., 2006. A theoretical model of litter decay and microbial interaction, by the Ecological Society of America. Ecological Monographs 76, 151e174. Nguyen, C., 2003. Rhizodeposition of organic C by plants: mechanisms and controls. Agronomie 23, 375e396.

Nguyen, C., Guckert, A., 2001. Short-term utilisation of 14C-[U] glucose by soil microorganisms in relation to carbon availability. Soil Biology & Biochemistry 33, 53e60. Parr, J.F., Papendick, R.I., Hornick, S.B., Meyer, M.E., 1992. Soil quality: attributes and relationship to alternative and sustainable agriculture. American Journal of Alternative Agriculture 7, 5e11. Paul, E.A., Follett, R.F., Leavitt, S.W., Halvorson, A., Peterson, G.A., Lyon, D.J., 1997. Radiocarbon dating for determination of soil organic matter pool sizes and dynamics. Soil Science Society of America Journal 61, 1058e1067. Paul, S., Flessa, H., Veldkamp, E., Lopez-Ulloa, M., 2008. Stabilization of recent soil carbon in the humid tropics following land use changes: evidence from aggregate fractionation and stable isotope analysis. Biogeochemistry 87, 247e263. Poll, C., Marhan, S., Ingwersen, J., Kandeler, E., 2008. Dynamics of litter carbon turnover and microbial abundance in a rye detritusphere. Soil Biology & Biochemistry 40 (6), 1306e1321. Ramette, A., 2007. Multivariate analyses in microbial ecology. FEMS Microbiology Ecology 62 (2), 142e160. Rillig, M.C., Caldwell, B.A., Wösten, H.A.B., Sollins, P., 2007. Role of proteins in soil carbon and nitrogen storage: controls on persistence. Biogeochemistry 85, 25e44. Saggar, S., Tate, K.R., Feltham, C.W., Childs, C.W., Parshotam, A., 1994. Carbon turnover in a range of allophanic soils amended with 14C-labelled glucose. Soil Biology & Biochemistry 26, 1263e1271. Saggar, S., Parshotam, A., Sparling, G.P., Feltham, C.W., Hart, P.B.S., 1996. 14C-labelled ryegrass turnover and residence times in soils varying in clay content mineralogy. Soil Biology & Biochemistry 28, 1677e1686. Saggar, S., Parshotam, A., Hedley, C., Salt, G., 1999. 14C-labelled glucose in New Zealand soils. Soil Biology & Biochemistry 31, 2025e2037. Scott, W.A., 2008. Countryside Survey Statistical Report. Countryside Survey Technical Report No.4/07. Centre for Ecology & Hydrology, Wallingford, UK. Six, J., Conant, R.T., Paul, E.A., Paustian, K., 2002. Stabilization mechanisms of soil organic matter: implications for C-saturation of soils. Plant and Soil 241, 155e176. Sparks, D.L., 1996. Methods of Soil Analysis. Part 3. Chemical Methods. Soil Science Society of America, Madison, WI. Stewart, C.E., Paustian, K., Conant, R.T., Plante, A.F., Six, J., 2009. Soil carbon saturation: implications for measurable carbon pool dynamics in long-term incubations. Soil Biology & Biochemistry 41, 357e366. Tipping, E., Chamberlain, P.M., Bryant, C.L., Buckingham, S., 2010. Soil organic matter turnover in British deciduous woodlands, quantified with radiocarbon. Geoderma 155, 10e18. Torn, M.S., Trumbore, S.E., Chadwick, O.A., Vitousek, P.M., Hendricks, D.M., 1997. Mineral control of soil organic carbon storage and turnover. Nature 389, 170e173. WRB, 2006. World Reference Base for Soil Resources 2006. Report 103. Food and Agriculture Organization of the United Nations, Rome, Italy.