Essay: Plants play an important role in ecosystem processes

Plants play an important role in ecosystem processes (Aerts & Chapin 2000; Lambers et al. 2000; Díaz et al. 2004). In the last decade, there has been an ongoing effort to understand the role of plants in ecosystem functioning and to predict how plants could respond to natural or man-induced environmental changes (Díaz & Cabido 1997; de Bello et al. 2010). Several plant traits were proposed to summarize the major dimensions of variation in plant ecological strategies and to provide insights on the main opportunities and selective forces that shaped life history, architecture, resource allocation, and physiology of plants (Grime 1977; Westoby 1998; Lambers et al. 2000; Westoby et al. 2002; Bär Lamas et al. 2016). Most of these traits converge across a wide range of environments and were successfully used to identify plant functional groups across broad regional environmental gradients (Díaz et al. 2004; Cornwell et al. 2008). Among traits, plant height, specific leaf area, and structural plant defenses were identified as meaningful traits converging across environmental gradients (Fonseca et al. 2000; Westoby et al. 2002; Moreno et al. 2010; May et al. 2013). However, meaningful plant traits related to ecosystem processes across narrow environmental gradients were scarcely explored. Particularly, trait variation across aridity gradients in desert ecosystems could be relevant in relation to changes in ecosystem processes such as primary production, litter decomposition and nutrient release and cycling. In these ecosystems, low and highly variable precipitation exerts a strong selective pressure on morphological and physiological traits allowing plants to tolerate or avoid extremely dry conditions (Noy Meir 1973; Reich et al. 2003). Plant trait variation across aridity gradients may be related to species replacement and/or plant phenotypic plasticity having different effects on the patterns of relevant ecosystem processes such as primary production, litter decomposition, and nutrient release and cycling (Lambers et al. 2000; Wright et al. 2004; Golodets et al. 2009; Moreno et al. 2010; Moreno & Bertiller 2015). In this context, the selection of an adequate set of meaningful plant traits related to key ecosystem processes may increase our ability to predict vegetation functional responses to natural or man-induced environmental change across arid ecosystems (Woodward & Cramer 1996; Díaz & Cabido 1997; de Bello et al. 2010).
Many studies identified plant functional groups using easily measurable life history traits summarizing fundamental plant trade-offs (Shaver & Chapin 1980; Westoby 1998; Cornwell et al. 2008; Zhao et al. 2014; Bär Lamas et al. 2016). However, other studies showed that relevant ecosystem processes such as leaf litter decomposition and nutrient release or immobilization could be better explained by specific traits such as leaf chemistry (Cornelissen et al. 1999; Wardle 2002). Litter decomposition has an important impact on nutrient cycling and C fluxes from the land to the atmosphere (Swift et al. 1979; Lambers et al. 2008). This process is mainly controlled by climatic variations across biomes (Berg et al. 1993, 2003; Parton et al. 2007; Lambers et al. 2008) and primarily by intrinsic litter traits at local scales (Aerts 1996, 1997; Killingbeck & Whitford 1996; Carrera et al. 2000, 2009; Wright et al. 2004; Vargas et al. 2006). In general, attributes of litter chemistry such as C/N or secondary compounds/N ratios were described as good proxies of litter decomposition, N mineralization, and N immobilization rates at local scales (Adams & Attiwill 1986). Low-productive species characteristic of arid ecosystems such as evergreen shrubs commonly produce long lasting tissues with large amounts of secondary compounds (Aerts & Chapin 2000; Hättenschwiler & Vitousek 2000) acting as defenses against herbivores, pathogens, water stress (Grime et al. 1996; Aerts & Chapin 2000; Wardle et al. 2002) and UV-B radiation (Meijkamp et al. 1999). Complex C compounds like phenolics or other secondary metabolites could also retard litter mass decay and nutrient release during decomposition process (Hättenschwiler & Vitousek 2000; Niinemets & Tamm 2005).
Few studies were focused to identify the main plant traits linked to key processes, such as litter mass decay or C and nutrient cycling within ecosystems (e.g. Cornwell et al. 2008) or biomes (e.g. Aguiar et al. 1996; Chapin et al. 1996). The selection of these related synthetic traits may help to assemble plant species in functional groups in order to predict eventual changes in ecosystem functional properties under changing environments at local scales (Chapin et al. 2009; Pietsh et al. 2014). In this sense, we selected dominant species across an aridity gradient and grouped them by plant growth forms and by different combinations of a set of morphological and chemical traits commonly used to group plant species in several biogeographic regions (e.g. Grime 1977; Westoby 1998; Bär Lamas et al. 2016). First, we explored whether plant morphological and chemical traits and rates of decomposition processes change across a narrow aridity gradient. Then, we asked whether different plant species groupings based on either growth forms or different combinations of morpho-chemical traits reflect contrasting patterns of senesced leaf mass decay and N release or immobilization during leaf decomposition, and which are the meaningful traits synthesizing the role of plants species in decomposition processes (leaf mass decay and N release/immobilization).