N-driven changes in a plant community affect leaf-litter traits and may delay organic matter decomposition in a Mediterranean maquis

Abstract

Organic matter (OM) decomposition is typically controlled by climate, soil properties, litter quality and soil microorganisms. Availability of nitrogen (N) also influences decomposition, but its effects on decomposition are controversial and most studies have only addressed decomposition of individual plant species grown under high N availability. We experimentally manipulated the dose of available N in a Mediterranean Basin maquis in south-western Europe, with low ambient N deposition (5.2 kg N ha⁻¹ yr⁻¹) and low soil N content (0.1%). N availability was modified by the addition of 40 and 80 kg N ha⁻¹ yr⁻¹ as NH₄NO₃. Control plots were not fertilized. After 2.5 years of N additions, we accounted for the integrated effects of N enrichment on litter decomposability taking into consideration the N-driven changes in the whole plant community (changes in plant species composition and litter quality). We collected soil from the no N addition treatment (control) and three types of leaf-litter (from three N addition treatments – 0, 40 and 80 kg N ha⁻¹ yr⁻¹) from the N-manipulation field experiment and performed a microcosms controlled decomposition study. Distinct leaf-litter traits were quantified (N and lignin concentration and C/N and lignin/N ratios) and related with decomposition and soil microbial biomass and activity. The leaf-litter consisted mostly of leaves from summer semi-deciduous shrubs, but relative to the control (no N addition), the treatment receiving 80 kg N ha⁻¹ yr⁻¹ had twice the amount of evergreen sclerophyll leaf-litter and higher lignin and N concentrations giving lower C/N and lignin/N ratios. As a result, OM decomposition in the microcosms containing 80 kg N ha⁻¹ yr⁻¹ litter was slower (with concomitant reduction in soil microbial biomass and activity) than in those containing 40 kg N ha⁻¹ yr⁻¹ litter. At the ecosystem level, N-driven changes in plant community altered leaf-litter traits (e.g. increased litter lignin and N content and decreased lignin/N ratio), which were powerful determinants of litter decomposition rates under controlled conditions. The results suggest that increasing N availability in this nutrient poor Mediterranean maquis may select species with litter traits that could delay decomposition and increase soil OM accumulation.

Introduction

The balance between vegetation inputs and organic matter (OM) decomposition determines the size of the soil OM pools (Baer et al., 2010) and controls nutrient cycling in terrestrial ecosystems (Knorr et al., 2005). In turn, the rate of litter decomposition is controlled by climate, soil properties, litter composition (Fioretto et al., 1998, 2001; Alarcón-Gutiérrez et al., 2008; Austin and Ballaré, 2010) and soil microbial community (Fioretto et al., 2001). However, nitrogen (N) availability can also influence decomposition and nutrient-cycling dynamics (Schimel and Bennett, 2004; Knorr et al., 2005; Liu et al., 2010), with potential consequences for OM decomposition and accumulation.

Increased N availability can change plant community litter traits directly and/or indirectly. Depending on the plant species and/or community, N enrichment can directly increase litter N concentration and decrease the C/N ratio of whole plant communities (e.g. Californian coastal sage scrub and chaparral – Vourlitis et al., 2009) and/or individual plant species (e.g. Cistus ladanifer from Mediterranean Basin maquis – Dias et al., 2012), making litter more readily decomposable. N enrichment is a powerful indirect driver of plant diversity changes (Sala et al., 2000), with alterations in plant traits (e.g. species-specific N and lignin concentrations) strongly influencing litter inputs and decomposition rates (Cornwell et al., 2008).

Knorr et al. (2005) conducted a meta-analysis of empirical studies to examine the effects of N enrichment on litter decomposition, concluding that N enrichment could increase, decrease or have no effect on litter decomposition depending on fertilization rate, site-specific ambient N-deposition level, and litter quality. However, none of the analysed studies had considered the effect of increased N availability on local plant biodiversity and consequently on leaf-litter traits.

In this work we intended to assess the effects of increased N availability on leaf-litter decomposability at the plant community level, in particular, in the semi-natural Mediterranean Basin ecosystems, where N deposition is predicted to increase three fold by 2050 (Galloway et al., 2004; Phoenix et al., 2006). The dynamics of litter decomposition need to be understood in order to inform management of these biodiversity hotspots (Phoenix et al., 2006). Recently, Incerti et al. (2011) developed a process-based model of litter decomposition for Mediterranean ecosystems, but did not account for the effect of the initial N content of litter. The few studies made on the effects of increased N on decomposition of Mediterranean litter (Sirulnik et al., 2007; Alarcón-Gutiérrez et al., 2008; Kazakou et al., 2009) were not conclusive (Ochoa-Hueso et al., 2011) and did not consider N-driven changes at the whole community level. In order to elucidate the integrated effects of N-driven changes on plant litter on decomposition, the present study was performed in a N-poor Mediterranean ecosystem very responsive to N availability (Dias et al., 2011a, 2012), whose vegetation may be grouped into two main plant functional types: summer semi-deciduous and evergreen-sclerophylls. Each group has been characterized on the basis of its phenology (Correia, 1988), water relations, carbon exchange properties (Werner et al., 1999), soil surface characteristics (Cruz et al., 2008), N use (Dias et al., 2011b) and strategies for regeneration after fire (Trabaud, 1981; Keeley, 1986; Clemente et al., 2005).

Our objective was to focus on the biotic processes of decomposition (Austin and Ballaré, 2010), so the leaf-litter had to be crushed (1 mm). Under field conditions this would cause its loss due to wind and/or rain, thus overestimating decomposition, so we setup a litter decomposition experiment under controlled conditions (microcosms). Acknowledging the importance of drying-rewetting events, especially in Mediterranean ecosystems (Fierer and Schimel, 2002), and that in maquis soils decomposition peaks in autumn (Rutigliano et al., 2009; Simões et al., 2009), we mimicked litter decomposition after a long dry period similar to the Mediterranean summer. To exclude the effect of N-driven changes in soil microbial community, we used only one type of soil (Control) and three types of leaf-litter (from three N addition treatments – 0, 40 and 80 kg ammonium nitrate-N ha⁻¹ yr⁻¹). Soil and litters were collected from an ongoing N-manipulation field study in a Mediterranean Basin maquis where increased N concentration (and decreased C/N ratio) of the dominant plant species (Dias et al., 2012) and changes in plant community (Dias et al., 2011a) had already been observed. Our working hypothesis was based on the differences between leaf and litter traits of summer semi-deciduous and evergreen sclerophylls (Schlesinger and Mavis, 1981; Correia and Catarino, 1994; Fioretto et al., 2005): the N-driven changes in the relative contribution of each group to whole community litter may result in alterations of whole community litter traits (e.g. N and lignin content, C/N and lignin/N ratios) and may not be proportional to the availability of N, partially explaining the controversial results concerning the effect of N availability on decomposition rates.