Scholarly article on topic 'Effects of drought-induced forest die-off on litter decomposition'

Effects of drought-induced forest die-off on litter decomposition Academic research paper on "Biological sciences"

Share paper
Academic journal
Plant Soil
OECD Field of science

Academic research paper on topic "Effects of drought-induced forest die-off on litter decomposition"

Plant Soil (2016) 402:91-101 DOI 10.1007/s11104-015-2762-4



Effects of drought-induced forest die-off on litter decomposition

Josep Barba • Francisco Lloret • Jorge Curiel Yuste

Received: 23 March 2015 / Accepted: 30 November 2015 /Published online: 11 December 2015 © Springer International Publishing Switzerland 2015


Aims Drought-induced forest die-off and subsequent species replacement may modify environmental conditions and eventually affect litter decomposition. We aimed to disentangle the effects of tree species and die-off state on litter decomposition in a mixed forest where Pinus sylvestris populations experiencing severe drought-induced die-off are being replaced by Quercus ilex.

Methods Litter bags with leaves and fine roots from both species were placed under canopies representing three habitats of the die-off and replacement process (healthy and dead P. sylvestris and healthy Q. ilex). Mass was assessed over 3 years.

Results Species-specific chemistry of litter (C:N ratio) had a direct effect on mass loss, but also indirect effects, attributed to the decomposer microbial community

Responsible Editor: Klaus Butterbach-Bahl.

Electronic supplementary material The online version of this article (doi:10.1007/s11104-015-2762-4) contains supplementary material, which is available to authorized users.

J. Barba (*) • F. Lloret

CREAF, Cerdanyola del Vallès, 08193 Barcelona, Catalonia, Spain

e-mail: J. Barba • F. Lloret

UnivAutónoma Barcelona (UAB), Cerdanyola del Vallès, 08193 Barcelona, Catalonia, Spain

J. C. Yuste

Museo Nacional de Ciencias Naturales (MNCN), CSIC, Madrid 28006, Spain

associated with a given habitat-species. In their respective original habitats, oak leaves decomposed 44 % faster than pine needles, whereas oak roots decomposed 46 % slower than pine roots.

Conclusions Forest die-off and species replacement affected litter decomposition. This effect can have great implications in forest functioning, particularly if drought-induced die-off worsens in the next decades, according with the trend observed in the studied system.

Keywords Litter decomposition. Forest die-off. Home field advantage. Carbon cycle. Mediterranean forest


The aerobic decomposition of dead organic matter is one of the main sources of CO2 emission from terrestrial ecosystems, consequently playing a critical role in their carbon (C) and nutrient balances at both local (Santa Regina 2001; Bonanomi et al. 2010) and global scales (Prentice et al. 2001; Canadell et al. 2007; Stocker et al. 2013). Therefore, in order to improve current predictions of ecosystem responses to climate change, it is crucial to understand the drivers controlling litter decomposition dynamics (Cao and Woodward 1998). Since the first half of the twentieth century, temperature, moisture and vegetation have been described as the main drivers of litter decomposition (Waksman and Gerretsen 1931). The key role of temperature and moisture is based on the well-known fact that the enzyme kinetics involved in microbial decomposition are very

sensitive to both water and temperature (Davidson and Janssens 2006).

Furthermore, vegetation controls organic C decomposition in different ways. Firstly, vegetation determines decomposition via species-specific litter quality, because the differences in chemical litter composition between plant species imply different litter degradabili-ty, and hence different rates of decomposition. This influence of litter quality on litter decomposition has been described from local (Saura-Mas et al. 2012; Wang et al. 2014) to regional scales (Melillo et al. 1982; Vivanco and Austin 2006; Cornwell et al. 2008). Accordingly, different indicators of litter quality such as C:N ratio, nutrient (N and/or P) content, and the content of some structural molecules (e.g., lignin or holocellulose), have been correlated with litter decom-posability (Gallardo and Merino 1993; Couteaux et al. 1995; Gholz et al. 2000; Vivanco and Austin 2006; Bonanomi et al. 2010). More specifically, litter's initial C:N ratio has been identified as one of the best chemicals predictors of litter decomposition (Melillo et al. 1982; Parton et al. 2007; Berg and McClaugherty 2008; Bonanomi et al. 2010). Moreover, for a given species, different organs (i.e., leaves, fine roots and twigs) present a different chemical composition and, therefore, different rates of decomposition (Vivanco and Austin 2006; Freschet et al. 2013; Wang et al. 2014). Secondly, vegetation has the ability to modify environmental conditions, such as temperature and moisture (Binkley and Giardina 1998; Yuan et al.

2012), thus indirectly determining decomposition by affecting enzyme kinetics (Cornwell et al. 2008; Freschet et al. 2012) or photodegradation rates associated with exposure to radiation (Austin and Vivanco 2006). Thirdly, vegetation can influence litter decomposition via its co-evolution with the soil decomposer community (Vivanco and Austin 2008; Ayres et al. 2009a), resulting in specific tree-species soil communities (Grayston and Prescott 2005; Waldrop and Firestone 2006; Curiel Yuste et al. 2012) with different functional diversity (Waldrop and Zak 2004; Wallenstein et al.

2013). This co-evolution between tree species and their soil communities is reflected by the capacity of specific microbial community to decompose more efficiently the litter of the plant species from which is derived (Austin et al. 2014). This effect, called home-field-advantage (HFA) (Ayres et al. 2009a, b; Austin et al. 2014), is widespread in forest ecosystems, enhancing litter decomposition by 8 % on average (Ayres et al. 2009b).

However, how soil communities are able to efficiently decompose different substrates and how differences in litter degradability could influence the correct interpretation of HFA is still under debate (Freschet et al. 2012; Keiser et al. 2014).

All these factors highlight the complexity of the controls of litter decomposition dynamics, which drives the C sink capacity of soils from terrestrial ecosystems, and the paucity of our knowledge of above-belowground interactions. Another major source of uncertainty, for instance, is that most research has focused on the decomposition patterns of above-ground litter (needles and/or leaves), whereas the decomposition of fine roots, which accounts for at least half of the litter produced by vegetation (Montero et al. 2005; Clemmensen et al. 2013), and for most of the C incorporated in soil in the long-term (Clemmensen et al.

2013) has been only marginally studied. In this regard, it is important to understand how climate change-induced shifts in vegetation health (Lloret et al. 2012) may alter above-belowground interactions and hence rates of organic matter decomposition and nutrient turnover. Drought- and heat-induced tree die-off and mortality have been reported over the last few decades around the world (Allen et al. 2010), particularly in South Europe and the Mediterranean Basin (Lloret et al. 2004; Della-Marta et al. 2007; Briffa et al. 2009; Carnicer et al. 2011). Changes towards ecosystems with a more limited supply of water (Giorgi and Lionello 2008; Mariotti 2010) have been associated with the decline of keystone species that have their southern limit of distribution in the Mediterranean Basin (Lenoir et al. 2010; Vayreda et al. 2013; Carnicer et al. 2014). This is the case with Pinus sylvestris L., which, in some areas of the Iberian peninsula, is being replaced by other species such as Quercus ilex L., which are better adapted to drought (Vila-Cabrera et al. 2013; Carnicer et al.

2014). Whereas many studies have speculated about the possible significant impact on C and nutrients dynamics in vegetation shifts induced by climate change (Cornwell et al. 2008; Ayres et al. 2009a, b; Ball et al. 2009; Mclaren and Turkington 2010; Freschet et al.

2013), really few studies have been directly designed to estimate how climate-change-induced secondary succession may affect forest C dynamics and the capacity of terrestrial ecosystems to sequester C (Diaz-Pines et al.


The aim of this study is to assess how changes in litter quality associated with climate-change-induced

vegetation shifts affect litter decomposition rates. We addressed the climate-driven forest succession from P sylvestris to Q. ilex occurring in the Prades Mountains (NE Iberian Peninsula). This information on litter decomposition in P. sylvestris forests is particularly relevant for regional assessment of the C and nutrient balance because this widely distributed species is experiencing severe die-off episodes in different regions (Martinez-Vilalta and Pinol 2002; Bigler et al. 2006). Leaf (senescent) and fine-root (fresh) litter bags from both species were placed beneath healthy P. sylvestris, dead P. sylvestris and Q. ilex canopies in a fully crossed factorial design, and decomposition and C and N content over 3 years were measured. Specifically, we tested if drought-induced forest succession modifies litter decomposition through changes in litter quality (litter effects), changes in the soil environment (habitat effects) and their interaction.

Materials and methods

Our experiment was performed in a mixed forest on the northwest-facing hillside in Titilar Valley, Prades Mountains (NE Iberian Peninsula; 41°13'N, 0°55'E; 1015 m asl). The climate is Mediterranean, with a mean annual temperature of 11.3 °C and mean annual precipitation of 664 mm (period 1951-2010) (Ninyerola et al. 2007a, b). The experimental area was located on a 35° hill slope, on metamorphic schist substrate that outcrops onto a large part of the study site. Soils are xerochrepts with clay loam texture and high gravel content (46 % volume). Organic horizons cover most of the soil and outcrops with variable thickness. For more information about the studied area, see Hereter and Sánchez (1999) and Barba et al. (2013). The mixed forest, which has not been managed for the last 30 years (Here§ et al. 2012), is mainly composed of Pinus sylvestris L. (Scots pine) (54 % of the forest basal area and mean diameter at breast height [DBH] of 0.32 m) and Quercus ilex L. ssp ilex (Holm oak) (41 % of the total BA and DBH of 0.15 m). The study area has been affected by several drought events since the 1990s, particularly the P. sylvestris population (Martínez-Vilalta and Piñol 2002), producing an average mortality of 12 % of standing trees and mean crown defoliation of 52 % (Vila-Cabrera et al. 2013). This situation, coupled with contrasted recruitment rates between both species (low rates in P sylvestris and high rates in Q. ilex) (Vila-

Cabrera et al. 2013), is leading to a progressive replacement of P. sylvestris by Q. ilex as the dominant over-storey species.

The decomposition experiment considered three types of trees, representing different stages of the ongoing forest succession (Healthy P sylvestris [HPs], Dead P sylvestris [DPs] and Q. ilex [Qi]). We established two meters around trees as the respective rhizosphere-influence area on soil environment (hereafter, habitat). Five replicates (hereafter, microsites) of each of these three habitat types were selected on a 1-ha study site. Microsites of the three habitat types were spatially randomized since die-off pattern was diffused. Selected dead pines had died in the nineties as consequence of serve drought events (Martinez-Vilalta and Pinol 2002).

To assess litter mass loss over time we use the litter bags method. Despite the suitability of this method for separating the effects of litter decomposition from microbially-stabilized plant-derived tissue or loses of fragmented tissues through the mesh pores is under debate (Cotrufo et al. 2013), is the most used method to study litter decomposition (Cotrufo et al. 2009), specially indicated for field experiments allowing a large number of replicates.

Freshly senescent needles, leaves and living fine roots (diameter thinner than 2 mm) from P. sylvestris and Q. ilex were collected from the same study area and oven-dried at 60 °C for 24 h. Litter bags (0.5 mm nylon mesh and size 7.5x8.5 cm) were filled with a known dry-weight amount of litter (0.5-1 g) (Ps needles, Qi leaves, Ps roots and Qi roots). Mesh size was large enough to allow microbial and fungi activity as well as small access by arthropods, but small enough to avoid major losses of the smallest litter portions (Killham 1994). Six litter bags containing each litter type were placed on each microsite, with a total of 360 litter bags (3 habitat types X 5 microsite replicates X 4 litter types X 6 litter-type replicates). A square metal fence (1 x 1 m) was installed on each microsite and litter bags were placed inside to avoid disturbances from wild boars during the experiment. Leaf litter bags were placed on the surface and fine-root litter bags were buried at a depth of 5-10 cm. We did not remove either the organic horizons underneath the litter bags at the beginning of the experiment or the litterfall during the experiment - as commonly done (i.e., Vivanco and Austin 2008) - since we wanted to mimic the natural conditions of the decomposition process as accurate as possible. Litter bags were placed in the field in July 2011 and collected 0.16,

0.5, 1, 1.5,2 and 3 years after bags bury and oven-dried at 60 °C for 24 h. The remaining litter was dry cleaned with a brush and weighed. Each individual sample was ground and analysed for total carbon [%C] content by CHNS organic microanalysis, using combustion coupled with gas chromatography (EUROVECTOR, EA3011, Milano, Italy). Similarly, initial litter quality (%N and %C) was assessed for three samples of each litter type.

To control the possible effects of the microsites' environmental differences on decomposition process, soil water content (SWC) and soil temperature were measured every 2 weeks from January 2012 to July 2013 at each microsite. SWC was measured by time domain reflectometry (TDR) (Tektronix 1502C, Beaverton, Oregon, USA). One TDR probe 15 cm long was permanently installed in the upper soil on each microsite. In order to correct the SWC measurements for stone content, gravimetrical SWC was regressed against TDR measurements (for more information, see Poyatos et al. (2013). Soil temperature was measured at 10 cm, using a thermometer (OMEGA, HH806AU, Stamford, USA). Additional information about soil properties such as pH, N availability, SOM content, soil bacterial community composition at the different habitats could be found in Curiel Yuste etal. (2012).

To assess the mass loss for litter and habitat types, we used a general linear model (GLM) coupled with an exponential decay equation (expressed as ln (Mt / M0), where Mt is the remaining dry mass of each sample on the sampled date and M0 is the initial dry mass) for each litter type (fixed factor) and habitat type (fixed factor) (Saura-Mas et al. 2012), and we included time as an additional variable in the model. Since the ln (Mt / M0) divided by time has been defined as the decomposition constant (k) (Olson 1963), the modelled slope for each pair of combinations (4 litter types X 3 habitat types) represents the decomposition constant k of each combination. As all the litterbags were collected on the same microsites throughout the experiment, microsite was also included in the model as a random factor. The model also contained the interactions between litter type, habitat type and time. Since this interaction was significant, the effect of litter was analysed separately in an additional model that included microsite and habitat type as random factors and time. Similarly, a model for habitat type was built, including microsite and litter type as random factors and time.

Moreover, the remaining mass (%) was analysed by GLM models. The temporal pattern of remaining mass was analysed separately for litter and habitat types in different models, which also included microsite as a random factor and time. These models also tested differences between litter and habitat types, respectively, for each collection event. Overall differences between litter or habitat types for the whole period of time were analysed with similar GLM models, but considering time as a random factor. A log-odd transformation was applied to achieve normal distribution in the % of remaining mass (i.e., log[x/(1-x)]).

To test the possible effect of initial litter quality on the decomposition process, linear regression was fitted with the mean (± SE) of the k values obtained in each habitat type (n=3) and mean (± SE) C:N values obtained in three samples analysed at the start of the experiment.

All the analyses were carried out using R 3.0.3. (R Foundation for Statistical Computing, Vienna, Austria). The mixed-effects models were performed using the R packages nlme and lme4 (Pinheiro et al. 2009; Bates etal. 2014).


Litter decomposition rates

The decomposition rate of the different litter types varied between habitats, as supported by a significant interaction between litter and habitat effects (Table 1, adjusted R2 of the model was 0.75). All litter types were decomposed with a similar k in HPs habitat, but there were differences in the DPs and Qi habitats (Fig. 1). In both the DPs and Qi habitats, Qi leaves and Ps roots showed consistently higher decomposition rates than Ps needles and Qi roots. In the Qi habitat, Qi leaves showed the highest decomposition rate, followed by Ps roots, and Qi roots showed the lowest decomposition rates, whereas Ps needles showed intermediate rates between the Ps roots and Qi roots. The Ps needle k was higher in HPs than in DPs habitats and showed intermediate values in the Qi habitat. Similarly, Qi leaves were decomposed faster in Qi habitats than in Ps habitats. The k of Ps roots did not show any significant differences in any of the habitats, and Qi roots showed higher k in both the Ps habitats (HPs and DPs) than in the Qi habitat. Overall, decomposition rates varied across the different litter origins: in Ps litter, the decomposition

Table 1 Results of the general linear model (GLM) testing for the effects of litter and microsite, and their interaction, on the decomposition constant (Yr-1)

Variables df F-value p-value

Litter 3 7.701 <0.001

Habitat 6 1.753 0.175

Litter x Habitat 6 2.244 0.039

rates were higher in needles (0.14±0.01 year 1) than in roots (0.17±0.01 year ), while the opposite trend was observed for Qi leaves (0.19±0.02year )androot(0.13 ±0.01 year ) (GLM with litter type as predictor and microsite and habitat type as random factors and time, ^<0.05) (Fig. 2a). Furthermore, when considering all three habitats together, litter composition (k) almost significantly correlated (R2=0.76; ^=0.081) with the initial litter quality (C:N ratio) (Fig. 3, Table 2). When the four litter types were considered together, no differences appeared in decomposition rates between habitat types (Fig. 2b) (GLM with habitat type as predictor and microsite and litter type as random factors, p>0.05). No differences were found in the measured environmental conditions between habitat types (GLM with soil temperature or soil water content as independent variables, habitat type as predictor and microsite as random factor; ^=0.94 for soil temperature and p=0.16 for soil water content). Additionally, seasonal patterns of soil temperature and SWC did not show differences among the different habitats (Supplementary Material, Figure 1).

Since HPs and Qi represented the two forest succes-sional extremes, the Ps needles and roots decomposition

rates beneath HPs were compared with the Qi leaves and roots decomposition rates beneath Qi. Ps needles in the HPs habitat showed on average of 44 % lower k than Qi leaves in the Qi habitat (p=0.046). Ps roots in the HPs habitat showed on average 46 % higher k than Qi roots in the Qi habitat (p=0.046).

Mass remaining over time

The mass remaining over time decreased for all litter types (Fig. 4a) and habitats (Fig. 4b). Its temporal evolution varied between litter types (F=16.58, p<0.001) but no significant differences were found between habitat types for the whole time period (F=1.56, p=0.250, GLMs with litter types and habitat types as predictors and microsite and time as random factors). Qi roots maintained the highest remaining mass throughout the studied period, while Qi leaves presented an accelerated biomass loss in comparison to the other litter types around 1.5 years after starting. The biomass loss of the two types of Ps litter remained quite similar over the 3 years, with a tendency towards an increase in roots after 1.5 years.


Here, we show that drought-induced secondary succession from P sylvestris to Q. ilex may substantially alter patterns of litter decomposition and N dynamics in Mediterranean forests. This alteration may be due to both differences in litter quality between these two tree

Fig. 1 Decomposition constants (k) of Pinus sylvestris and Quercus ilex litter (leaves and fine roots) across three habitats obtained by the GLM. The lower case indicates significant differences in k between litter types within each habitat and the capital letters indicate significant differences in k between habitats within each litter type (p<0.05)

Ps Qi Ps Qi Ps Qi Ps Gt I':. Qi ps Qt

needles leaves roots roots needles leaves roots roots needles leaves roots roots


010 0.05 0.00

Ps rteedfes Qi leaves Ps roots Qf roots


Fig. 2 Effects of litter type (a) and habitat (b) on the decomposition constant (k) (mean ± SE) obtained by the GLM. The lower case indicates significant differences in k between litter types (a)

0.05 -

0.00 -I--1---1---1--

HPs DPs Qi


and habitats (b) (p<0.05). Significant differences between litter types and habitat types were obtained from GLMs that considered only the respective variables as fixed factors (see text)

species and the differential capacities of the microbial communities associated with the habitats to decompose the different litter types.

Litter quality exerted a major control over litter decomposition, which is something that has already been widely observed both at local (Gallardo and Merino 1993; Bonanomi et al. 2010; Aponte et al. 2012) and regional scales (Couteaux et al. 1995; Vivanco and Austin 2006; Cornwell et al. 2008). In our study, initial litter C:N ratio appeared as a good predictor of litter quality since it correlated quite well with the decomposition rate constant (k) of the litter types, independently of habitat. However, other chemical controls on litter decomposition may also be important - for instance,

60 70 so

C:N ratio

Fig. 3 Relationship between the initial C:N ratio of litter (leaves and fine roots from Pinus sylvestris and Quercus ilex) and decomposition constant (k) (mean ± SE), calculated as the mean of k considering the three habitat types

initial lignin content, which is usually negatively correlated with decomposition constant (Cornwell et al. 2008). In fact, lignin content in Ps needles has been reported to be higher than in Qi leaves (Kattge et al. 2011; Mediavilla et al. 2011), in agreement with our observations of the lower decomposition rates of Ps needles. Other physical controls could also underlie the differences in litter decomposition observed (Cornwell et al. 2008). Qi leaves show higher area/volume ratio than Ps needles (Kattge et al. 2011), enhancing microbial accessibility and consequently decomposition rates. Leaf litter usually decomposes better than fine-root litter (Gholz et al. 2000; Vivanco and Austin 2006; Freschet et al. 2013) due to its better quality (i.e., low C:N) (Bird and Torn 2006; Wang et al. 2010), as we observed in Q. ilex, but not for P sylvestris, although the latter's needles showed higher C:N than fine roots.

The lack of any significant differences in the abiotic environmental variables between habitats points to the role of habitat specificities of soil decomposer communities as major controllers of the observed differences between habitats in decomposition rates (Curiel Yuste et al. 2012; Keiser et al. 2014). This statement is partially supported by the habitat-specific soil bacterial community found at the same study site (Curiel Yuste et al. 2012). It has been hypothesized elsewhere that soil decomposer communities associated with the distinct stages of forest succession exhibit a specific capacity to decompose litter of varying quality (Freschet et al. 2012; Keiser et al. 2014). In our case, we observed that soil decomposer communities under HPs habitats were

Table 2 Initial litter quality on the study site

Ps needles Qi leaves Ps roots Qi roots

C (%) 52.51 a±0.50 50.79 b±0.11 48.62 c±0.17 48.20 c±0.31

N (%) 0.67 b±0.02 0.93 a±0.01 0.69 b±0.01 0.65 b±0.01

C:N 80.02 a±2.61 54.90 c±0.35 70.16 b±0.96 74.45 ab±1.08

The different letters indicate differences from litter from the post-hoc analysis (Tukey test) (P<0.05)

able to decompose litter of differing quality to similar rates, whereas soil decomposer communities under DPs and Qi were more selective and significantly capable of decomposing better litter of higher quality. Thus, these

functional differences along the drought-induced suc-cessional gradient endorse the existence of a concomitant microbial succession in such habitats, as reported for soil bacteria communities in the studied forest

Fig. 4 Mass remaining (mean ± SE) for each litter type (a) and habitat type (b) at 6 collection times (0.16, 0.5, 1, 1.5, 2 and 3 years, n=5). The lower case indicates significant differences in remaining mass between litter types (panel a) within collection times. The capital letters indicate significant differences in remaining mass significant over time within litter types (panel a) and within habitat types (panel b). No significant differences were found in remaining mass between microsites within collection times (b). Comparisons were made with GLMs, with microsite as a random factor (p<0.05). Statistical analyses were performed with remaining mass log-odd transformed to achieve normality

(Curiel Yuste et al. 2012) and in other ecosystems (Waldrop and Firestone 2006; Wickings et al. 2012; Keiser et al. 2014). Despite information of fungal diversity and composition associated at the different stages of pines die-off and Holm oak replacement at the study site is not available, soil fungal communities are usually even more tree-species specific than bacterial communities (Urbanová et al. 2015), reinforcing the idea that every habitat has a specific soil microbial community (both bacterial and fungal).

Therefore, litter decomposition rates in forest subjected to drought-induced species replacement would be modified not only by alterations in litter quantity and quality, but also by changes in the decomposer community associated with tree replacement.

However, despite the differences in litter decomposition between microbial communities seem quite clear, some uncertainties on the bacterial vs. fungal relative contribution remained still unsolved, especially in a drying context. While soils of Scots pines forests have been found more dominated by fungus than by bacteria compared to Quercus forests in a similar forest succession type (Díaz-Pinés et al. 2014), fungal communities have been reported to be more drought-resistant than bacterial communities (Wilkinson et al. 2002; Curiel Yuste et al. 2011; Flores-Rentería et al. 2015; Tardy et al. 2015) and thus, fungal relative proportion increases in drier conditions (Jensen et al. 2003). How these changes in bacterial-fungal communities due to drought-induced shifts may affect ecosystem functioning such as litter decomposition is still unclear and deserves further studies.

Our results only partially supported the HFA hypothesis, which proposes higher litter mass loss under the species producing a given litter type (at home) than under other species (away). Qi leaves were more efficiently decomposed under "home" (Qi) than under "away" habitats (HPs or DPs; 36 % difference, on average). Soil decomposer community under Qi seems, therefore, more specialized in the decomposition of higher quality organic matter (lower C:N ratio), such as Qi leaves. However, the decomposer community in the HPs habitat seemed to be generalist, and thus able to decompose litter of very different quality at similar rates (Fig. 1) refuting a potential local adaptation of the decomposer communities, as suggested by HFA hypothesis. . Moreover, Ps roots showed similar decomposition rates across habitats, and for Qi roots, the decomposition rates were actually higher in HPs than

in Qi habitats. Overall, these results do not support the HFA hypothesis, whereby the HFA should be, according to theory, more pronounced in more recalcitrant litter (Stricklandetal. 2009,2013; MilcuandManning2011). Our results do concur, however, with 25 % of the experiments that performed reciprocal litter transplants between tree species without observing any stimulation of decomposition at home (Ayres et al. 2009b).

The comparison of litter decomposition between the two forest succession extremes can help to predict future trends in the C dynamics of these forests. While Qi leaves decompose faster than Ps needles beneath their respective species, for fine roots the opposite trend was observed: rates of fine-root decomposition in the HPs habitat were higher than the rates of Qi fine-root decomposition in the Qi habitat, suggesting that secondary succession may produce a substantial decoupling of above- and belowground trends in organic matter decomposition. Thus, P. sylvestris replacement by Q. ilex would imply a faster decomposition of superficial leaf-derived soil layers but a slower decomposition of root-derived material, which is generally the major contributor to soil organic matter (Clemmensen et al. 2013). Nevertheless, the amount of litter produced above and belowground by the different species was not measured in this study, and the final C balance would ultimately depend on both, the decomposition rates and the contribution of the above- and belowground biomass of the different species to the C pool (Berg 2000; Bardgett etal. 2013).


In this study, we are the first, to our knowledge, to use a real-case scenario of the effects of die-off-driven forest succession on litter decomposition and N dynamics. Our study shows that besides changes in soil temperature and moisture availability, climate change-driven succession from P sylvestris to Q. ilex is responsible for modifying both leaf and fine-root decomposition through changes in the chemical nature of the litter and the relationship between above-ground vegetation species and the belowground environment, including local decomposer communities. In this particular study, drought-induced replacement of Scots pines by Holm oaks seems to provoke significant changes, firstly in the chemical composition of litter and secondly in the ability of different microbial communities to decompose

organic matter. The result should be a net increase in the decomposition rates of the "above-ground" litter (moving from recalcitrant Ps needles to more palatable Qi leaves) but a net reduction in the rates of "belowground" litter decay, largely due to the lower capacity of the microbial communities under colonizer Holm-oaks to decompose root material as compared with those under pines. Our results also suggest, therefore, that in order to correctly predict the effects of climate change effects on C dynamics on forests, models should closely examine changes in both the chemical composition and functioning of the decomposer communities associated with drought-induced secondary successions.

Acknowledgments The authors thank I. Azcoitia, G. Barba, M. Gol and C. Recasens for help in fieldwork and sample processing, and J. Martinez-Vilalta for his valuable comments. We thank the anonymous referees for all their constructive comments and advice. This study was supported by the Spanish Government projects CGL2009-08101, CGL2010-16373, CGL2012-32965 and CGL2013-42271-P, by the Government of Catalonia grants (2009-SGR-00247 and 2014-SGR-453) and by the TRY initiative on plant traits ( JB was supported by an FPI scholarship (BES-2010-036558) from the Spanish Ministry of Economy and Competitiveness. JCY acknowledges the support of the "Ramon y Cajal" programme.


Allen CD, Macalady AK, Chenchouni H et al (2010) A global overview of drought and heat-induced tree mortality reveals emerging climate change risks for forests. For Ecol Manag 259:660-684

Aponte C, García LV, Marañón T (2012) Tree species effect on litter decomposition and nutrient release in Mediterranean oak forests changes overtime. Ecosystems 15:1204-1218 Austin AT, Vivanco L (2006) Plant litter decomposition in a semiarid ecosystem controlled by photodegradation. Nature 442: 555-558

Austin AT, Vivanco L, González-Arzac A, Pérez LI (2014) There's no place like home? An exploration of the mechanisms behind plant litter-decomposer affinity in terrestrial ecosystems. New Phytol 204:207-214 Ayres E, Steltzer H, Berg S, Wall DH (2009a) Soil biota accelerate decomposition in high-elevation forests by specializing in the breakdown of litter produced by the plant species above them. J Ecol 97:901-912 Ayres E, Steltzer H, Simmons BL et al (2009b) Home-field advantage accelerates leaf litter decomposition in forests. Soil Biol Biochem 41:606-610 Ball BA, Bradford MA, Coleman DC, Hunter MD (2009) Linkages between below and aboveground communities: Decomposer responses to simulated tree species loss are largely additive. Soil Biol Biochem 41:1155-1163

Barba J, Curiel Yuste J, Martinez-Vilalta J, Lloret F (2013) Drought-induced tree species replacement is reflected in the spatial variability of soil respiration in a mixed Mediterranean forest. For Ecol Manag 306:79-87 Bardgett RD, Manning P, Morrien E, De Vries FT (2013) Hierarchical responses of plant-soil interactions to climate change: consequences for the global carbon cycle. J Ecol 101:334-343

Bates D, Maechler M, Bolker D, Walker S (2014) lme4: linear

mixed-effects models using Eigen and S4 Berg B (2000) Litter decomposition and organic matter turnover in

northern forest soils. For Ecol Manag 133:13-22 Berg B, McClaugherty C (2008) Plant litter: decomposition, humus formation, carbon sequestration. Springer-Verlag Bigler C, Braker OU, Bugmann H et al (2006) Drought as an inciting mortality factor in scots pine stands of the Valais, Switzerland. Ecosystems 9:330-343 Binkley D, Giardina C (1998) Why do tree species affect soils? The warp and woof of tree-soil interactions. Biogeochemistry 42:89-106

Bird JA, Torn MS (2006) Fine roots vs. needles: a comparison of 13C and 15N dynamics in a ponderosa pine forest soil. Biogeochemistry 79:361-382 Bonanomi G, Incerti G, Antignani Vet al (2010) Decomposition and nutrient dynamics in mixed litter of Mediterranean species. Plant Soil 331:481-496 Briffa KR, van der Schrier G, Jones PD (2009) Wet and dry summers in europe since 1750: evidence of increasing drought. Int J Climatol 29:1894-1905 Canadell JG, Le Quere C, Raupach MR et al (2007) Contributions to accelerating atmospheric CO2 growth from economic activity, carbon intensity, and efficiency of natural sinks. Proc Natl Acad Sci U S A 104:18866-18870 Cao MK, Woodward FI (1998) Dynamic responses of terrestrial ecosystem carbon cycling to global climate change. Nature 393:249-252

Carnicer J, Coll M, Ninyerola M et al (2011) Widespread crown condition decline, food web disruption, and amplified tree mortality with increased climate change-type drought. Proc Natl Acad Sci U S A 108:1474-1478 Carnicer J, Coll M, Pons X et al (2014) Large-scale recruitment limitation in Mediterranean pines : the role of Quercus ilex and forest successional advance as key regional drivers. Glob Ecol Biogeogr 23:371-384 Clemmensen KE, Bahr A, Ovaskainen O et al (2013) Roots and associated fungi drive long-term carbon sequestration in boreal forest. Science 339:1615-1618 Cornwell WK, Cornelissen JHC, Amatangelo K et al (2008) Plant species traits are the predominant control on litter decomposition rates within biomes worldwide. Ecol Lett 11:10651071

Cotrufo MF, Del Galdo I, Piermatteo D (2009) Litter decomposition: concepts, methods and future perspectives. In: Kutsch WL, Bahn M, Heinemeyer A (eds) Soil carbon dynamics. An integrated methodology. Cambridge University Press, New York, pp 76-90

Cotrufo MF, Wallenstein MD, Boot CM et al (2013) The Microbial Efficiency-Matrix Stabilization (MEMS) framework integrates plant litter decomposition with soil organic matter stabilization: do labile plant inputs form stable soil organic matter? Glob Chang Biol 19:988-995

Couteaux MM, Bottner P, Berg B (1995) Litter decomposition

climate and litter quality. Trends Ecol Evol 10:63-66 Curiel Yuste J, Peñuelas J, Estiarte M et al (2011) Drought-resistant fungi control soil organic matter decomposition and its response to temperature. Glob Chang Biol 17:14751486

Curiel Yuste J, Barba J, Fernandez-Gonzalez AJ et al (2012) Changes in soil bacterial community triggered by drought-induced gap succession preceded changes in soil C stocks and quality. Ecol Evol 2:3016-3031 Davidson EA, Janssens IA (2006) Temperature sensitivity of soil carbon decomposition and feedbacks to climate change. Nature 440:165-173 Della-Marta PM, Haylock MR, Luterbacher J, Wanner H (2007) Doubled length of western European summer heat waves since 1880

Díaz-Pinés E, Schindlbacher A, Godino M et al (2014) Effects of tree species composition on the CO2 and N2O efflux of a Mediterranean mountain forest soil. Plant Soil 384:243-257 Flores-Rentería D, Curiel Yuste J, Rincón A et al (2015) Habitat fragmentation can modulate drought effects on the plant-soil-microbial system in Mediterranean Holm oak (Quercus ilex) forests. Microb Ecol 69:798-812 Freschet GT, Aerts R, Cornelissen JHC (2012) Multiple mechanisms for trait effects on litter decomposition: moving beyond home-field advantage with a new hypothesis. J Ecol 100:619-630

Freschet GT, Cornwell WK, Wardle DA et al (2013) Linking litter decomposition of above- and below-ground organs to plant-soil feedbacks worldwide. J Ecol 101:943-952 Gallardo A, Merino J (1993) Leaf decomposition in two Mediterranean ecosystems of southwest Spain: influence of substrate quality. Ecology 74:152-161 Gholz HL, Wedin DA, Smitherman SM et al (2000) Long-term dynamics of pine and hardwood litter in contrasting environments: toward a global model of decomposition. Glob Chang Biol 6:751-765

Giorgi F, Lionello P (2008) Climate change projections for the

Mediterranean region. Glob Planet Chang 63:90-104 Grayston SJ, Prescott CE (2005) Microbial communities in forest floors under four tree species in coastal British Columbia. Soil Biol Biochem 37:1157-1167 Here? AM, Martínez-Vilalta J, Claramunt López B (2012) Growth patterns in relation to drought-induced mortality at two Scots pine (Pinus sylvestris L.) sites in NE Iberian Peninsula. Trees 26:621-630

Hereter A, Sánchez JR (1999) Experimental areas of Prades and Montseny. In: Roda F, Retana J, Gracia CA, Bellot J (eds) Ecology of Mediterranean evergreen oak forests. Springer, Berlin, pp 15-27 Jensen KD, Beier C, Michelsen A, Emmett BA (2003) Effects of experimental drought on microbial processes in two temperate heathlands at contrasting water conditions. Appl Soil Ecol 24:165-176

Kattge J, Díaz S, Lavorel S et al (2011) TRY- a global database of

plant traits. Glob Chang Biol 17:2905-2935 Keiser AD, Keiser DA, Strickland MS, Bradford MA (2014) Disentangling the mechanisms underlying functional differences among decomposer communities. J Ecol 102:603-609 Killham K (1994) The soil biota. In: Killham K (ed) Soil ecology. Cambridge University Press, p 196

Lenoir J, Gégout JC, Dupouey JL et al (2010) Forest plant community changes during 1989-2007 in response to climate warming in the Jura Mountains (France and Switzerland). J Veg Sci 21:949-964 Lloret F, Siscart D, Dalmases C (2004) Canopy recovery after drought dieback in holm-oak Mediterranean forests of Catalonia (NE Spain). Glob Chang Biol 10:2092-2099 Lloret F, Escudero A, Iriondo JM et al (2012) Extreme climatic events and vegetation: the role of stabilizing processes. Glob Chang Biol 18:797-805 Mariotti A (2010) Recent changes in the Mediterranean water cycle: a pathway toward long-term regional hydroclimatic change? J Clim 23:1513-1525 Martínez-Vilalta J, Piñol J (2002) Drought-induced mortality and hydraulic architecture in pine populations of the NE Iberian Peninsula. For Ecol Manag 161:247-256 Mclaren JR, Turkington R (2010) Plant functional group identity differentially affects leaf and root decomposition. Glob Chang Biol 16:3075-3084 Mediavilla S, González-Zurdo P, García-Ciudad A, Escudero A (2011) Morphological and chemical leaf composition of Mediterranean evergreen tree species according to leaf age. Trees Struct Funct 25:669-677 Melillo JM, Aber J, Muratore J (1982) Nitrogen and lignin control of hardwood leaf litter decomposition dynamics. Ecology 3: 621-626

Milcu A, Manning P (2011) All size classes of soil fauna and litter quality control the acceleration of litter decay in its home environment. Oikos 120:1366-1370 Montero G, Ruiz-Peinado R, Muñoz M (2005) Producción de biomasa y fijación de CO2 por los bosques españoles. Monografías del INIA. Serie Forestal no 13., Madrid Ninyerola M, Pons X, Roure JM (2007a) Objective air temperature mapping for the Iberian Peninsula using spatial interpolation and GIS. Int J Climatol 27:1231-1242 Ninyerola M, Pons X, Roure JM (2007b) Monthly precipitation mapping of the Iberian Peninsula using spatial interpolation tools implemented in a geographic information system. Theor Appl Climatol 89:195-209 Olson JS (1963) Energy storage and the balance of producers and

decomposers in ecological systems. Ecology 44:322-331 Parton W, Silver WL, Burke IC et al (2007) Global-scale similarities in nitrogen release patterns during long-term decomposition. Science 315:361-364 Pinheiro J, Bates D, DepRoy S (2009) Linear and nonlinear mixed

effects models. R package version 3.1-96 Poyatos R, Aguadé D, Galiano L et al (2013) Drought-induced defoliation and long periods of near-zero gas exchange play a key role in accentuating metabolic decline of Scots pine. New Phytol 200:388-401 Prentice I, Farquhar G, Fasham M (2001) The carbon cycle and atmospheric carbon dioxide. In: Pitelka L, Rojas AR (eds) Climate change 2001: the scientific basis. GRID-Arendal in 2003., pp 183-237 Santa Regina I (2001) Litter fall, decomposition and nutrient release in three semi-arid forests of the Duero basin, Spain. Forestry 74:347-358 Saura-Mas S, Estiarte M, Peñuelas J, Lloret F (2012) Effects of climate change on leaf litter decomposition across post-fire plant regenerative groups. Environ Exp Bot 77:274-282

Stocker TF, Qin D, Plattner GK, et al. (eds) (2013) IPCC, 2013: climate change 2013: the physical science basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change Strickland MS, Lauber C, Fierer N, Bradford MA (2009) Testing the functional significance of microbial community composition. Ecology 90:441-451 Strickland MS, McCulley RL, Bradford MA (2013) The effect of a quorum-quenching enzyme on leaf litter decomposition. Soil Biol Biochem 64:65-67 Tardy V, Chabbi A, Charrier X et al (2015) Land use history shifts in situ fungal and bacterial successions following wheat straw input into the soil. PLoS One 10:e0130672 Urbanova M, Snajdr J, Baldrian P (2015) Composition of fungal and bacterial communities in forest litter and soil is largely determined by dominant trees. Soil Biol Biochem 84:53-64 Vayreda J, Gracia M, Martinez-Vilalta J, Retana J (2013) Patterns and drivers of regeneration of tree species in forests of peninsular Spain. J Biogeogr 40:1252-1265 Vila-Cabrera A, Martinez-Vilalta J, Galiano L, Retana J (2013) Patterns of forest decline and regeneration across Scots pine populations. Ecosystems 16:323-335 Vivanco L, Austin AT (2006) Intrinsic effects of species on leaf litter and root decomposition: a comparison of temperate grasses from North and South America. Oecologia 150:97107

Vivanco L, Austin AT (2008) Tree species identity alters forest litter decomposition through long-term plant and soil interactions in Patagonia, Argentina. J Ecol 96:727-736

Waksman SA, Gerretsen FC (1931) Influence of temperature and moisture upon the nature and extent of decomposition of plant residues by microorganisms. Ecology 12:33-60 Waldrop MP, Firestone MK (2006) Response of microbial community composition and function to soil climate change. Microb Ecol 52:716-724 Waldrop MP, Zak DR (2004) Divergent responses of soil micro-bial activity and soil C storage to atmospheric N deposition suggests decomposer communities are not functionally redundant. Ecol Soc Am Annu Meet Abstr 89:527 Wallenstein MD, Haddix ML, Ayres E et al (2013) Litter chemistry changes more rapidly when decomposed at home but converges during decomposition-transformation. Soil Biol Biochem 57:311-319 Wang H, Liu S, Mo J (2010) Correlation between leaf litter and fine root decomposition among subtropical tree species. Plant Soil 335:289-298 Wang W, Zhang X, Tao N et al (2014) Effects of litter types, microsite and root diameters on litter decomposition in Pinus sylvestris plantations of northern China. Plant Soil 374:677-688 Wickings K, Grandy AS, Reed SC, Cleveland CC (2012) The origin of litter chemical complexity during decomposition. Ecol Lett 15:1180-1188 Wilkinson SC, Anderson JM, Scardelis SP et al (2002) PLFA profiles of microbial communities in decomposing conifer litters subject to moisture stress. Soil Biol Biochem 34:189-200 Yuan Z, Gazol A, Wang X et al (2012) What happens below the canopy? Direct and indirect influences of the dominant species on forest vertical layers. Oikos 121:1145-1153