The role of physical disturbance for litter decomposition and nutrient cycling in coastal sand dunes

Highlights

• Physical disturbance increases ecosystem functioning in stressful costal sand dunes.

• Dune geomorphology impacts ecosystem functioning when affecting sand deposition.

• Plant community composition changes in response to increasing ecosystem functioning.

Abstract

Disturbance increases ecosystem functioning in productive habitats but its effect in stressful conditions is less documented, although this is crucial for understanding the resilience of disturbed systems to natural and anthropogenic disturbances. Our goal is to assess the influence of physical disturbance for ecosystem functioning in coastal sand dunes. We set up an experimental design, including two treatments in four blocks, in a four km-long dune site from South West France. We simulated physical disturbance from marine and wind origin, digging Experimental Notches (EN), in the incipient (West EN treatment) and established foredunes (East EN treatment), respectively and compared the effects of EN to controls along transects including 13 positions from the beach to the transition dune behind ENs. We sampled litter decomposition rate, elevation variation, wind abrasion, sand grain size and vegetation composition. We also used drones to quantify sand deposition sheets during severe winter storms. Litter decomposition rate was the highest where sand accumulated the most, at the ecotone between the established foredune and transition dune and in the East EN treatment. This increase of ecosystem functioning was correlated to wind patterns. However, there was also a strong alongshore variability, with important sand deposition sheets occurring in some blocks depending on dune geomorphology. Vegetation composition was mainly influenced by shoreline distance, but also by the block and EN treatment, with a strong interaction between these three effects. We conclude that physical disturbance increase ecosystem functioning in the stressful conditions of the Atlantic sand dunes, only when sand accumulates, whereas excessive disturbances enhancing sand erosion are not favorable for ecosystem functioning.

Introduction

The role of disturbance for community structure and ecosystem functioning has fascinated ecologists for decades (Connell, 1978; Grime, 1973; Huston, 1979; MacArthur and Wilson, 1963; Paine and Levin, 1981; Pickett and White, 1985). Grime (Grime, 1973) proposed to separate disturbance from environmental stress to straightforwardly assessing plant species functional strategies, community richness and composition, and ecosystem functioning in natural and managed ecosystems. Disturbance can be defined as a decrease in plant biomass due to either biotic or abiotic factors (herbivory, trampling, ploughing, fire, flooding, soil erosion, sediment accumulation) and stress as a decrease in plant growth and community productivity due to mainly four environmental constraints, shade, drought, oligotrophy and sub-optimal temperatures (Grime, 1973; Grime, 1974). This distinction is crucial to understand the role of disturbance for ecosystem functioning since stress is always related to a decrease in community productivity, whereas disturbance can show different relationships with productivity.

Concerning plant species richness, Huston (Huston, 1979) proposed that physical disturbance decreases species richness in stressful environments but promotes it in favorable ones due to decreasing competition. Thus, Grime (Grime, 1973), Huston (Huston, 1979) (see (Huston, 2014) for a synthesis) proposed the occurrence of a unimodal (i.e., highest diversity at mid-position along the gradient) relationship between plant species richness and stress, disturbance, and community biomass. This hypothesis has been supported by a number of empirical studies, for example Ross et al. (Ross et al., 2019) and Maalouf et al. (Maalouf et al., 2012) for productive herbaceous communities and stressful calcareous grasslands, respectively. However, this relationship has been recently debated (Adler et al., 2011; Fraser et al., 2015).

Ecosystem functioning refers to the ecological processes that control the fluxes of energy, nutrients and organic matter through an environment (e.g., decomposition, primary production and nutrient cycling) (Cardinale et al., 2012). It is also widely acknowledged that the absence of disturbance in low stress conditions leads to a progressive decline in ecosystem functioning. For example, Wardle et al. (Wardle et al., 2004) studying the long-term evolution of six forest ecosystems from tropical, temperate, and boreal biomes, showed that litter decomposition, biomass, and decomposer microbe activity decreased over time in absence of major disturbance, mainly due to increasing phosphorus limitation. Fortunel et al. (Fortunel et al., 2009) also found an overall decrease in ecosystem functioning, measured by litter quality and decomposition, with decreasing disturbance for herbaceous systems from ten European sites exhibiting moderate to low stress conditions. Finally, Peltzer et al. (Peltzer et al., 2010) reviewed the findings from studies of long-term chronosequences assessing nutrient cycling for systems spanning the boreal, temperate, and subtropical zones. Although they also found an overall decrease in ecosystem functioning through time, they stressed the occurrence of exceptions in arid climates with no ecosystem retrogression or slower depletion in nutrients. The depletion of nutrients through time is highly dependent on climate drainage (Delgado-Baquerizo et al., 2013; Huston, 2012), with increasing phosphorus limitation and decreasing productivity during succession in wet climates only. This explains why nutrient cycling is improved by disturbance in wet climates.

However, if disturbance (e.g., wildfires, grazing) is essential to maintain ecosystem functioning in low stress conditions, its role in stressful systems is less known, with contrasting results, either increases or decreases of ecosystem functioning, depending on disturbance type (Deng et al., 2013; Dimitrakopoulos et al., 2006; Wardle et al., 1997). Additionally, even in wet climates, changing and intensifying disturbance regimes has been shown to drive positive feedback loops between vegetation composition, structure, and ecosystem function that could reduce ecosystem resilience and trigger ecosystem collapse (Bowd et al., 2019; Lindenmayer et al., 2016). Such effects are even more likely to occur in stressful systems with lower resilience to natural and anthropogenic disturbances (Kéfi et al., 2007). Thus, the role of disturbance for ecosystem functioning is not clearly established in stressful systems and its occurrence might be more dependent on the frequency and intensity of disturbances than in wet climates. The resilience of ecosystems to natural and anthropogenic disturbances is highly dependent on the ability of plant species to recolonize disturbed habitats and, thus, to plant growth and ecosystem functioning. If disturbance is decreasing ecosystem functioning and, thus, plant growth and community biomass in stressful systems, this may induce a collapse of ecological communities and ecosystem services (Kéfi et al., 2007; Michalet, 2006). This will support the CSR model of Grime (Grime, 1974) that consider there is no viable strategy for plant in highly stressed and disturbed systems. Therefore, increasing our knowledge on the role of physical disturbance for ecosystem functioning in stressful systems is key for managing and restoring disturbed stressful ecosystems, such as arid ecosystems or coastal sand dunes.

We choose to assess in coastal dunes the role of physical disturbance for ecosystem functioning in stressful environments since community structure and ecosystem functioning are known to be mainly driven by these two environmental constraints in these systems (Forey et al., 2008; Maun, 1998). Coastal dunes are generally occurring along coasts having sufficient sand supply, prevailing onshore winds and the presence of vegetation, or other obstructions, to trap the sand transported by the wind (Nordstrom, 2015). They are subjected to wave action, likely to induce drastic coastal erosion during storms (Castelle et al., 2015; Guisado-Pintado and Jackson, 2018). Additionally, coastal dunes are highly influenced by wind processes, transporting sand from the beach to the dune (Hesp and Walker, 2013), and the presence of plants limiting wind erosion by trapping sand with their aerial parts (Zarnetske et al., 2012). Interactions between these physical and biotic processes can lead to the formation of blowouts in coastal dunes (Hesp, 2002), promoting sediment transfers from the beach to the back dune by wind acceleration in the blowout (Hesp and Walker, 2013).

Plant communities in costal dunes are subjected to a large range of environmental stresses such as saline spray, swash inundation (principally in the beach and incipient foredune), drought, nutrient deficiency (Hesp, 1991; Hesp and Martinez, 2007; Martinez and Psuty, 2004). However, in dunes from wet temperate climates where salt can be washed from plants and through soils during heavy rain events, disturbance by sand deposition has been shown to be the crucial factor driving plant community zonation before salinity (Moreno-Casasola, 1986; Forey et al., 2008; Maun, 1998). Maun (Maun, 1994) also showed that partial sand burial, for some species, stimulated growth of ramets, stolon, roots and leaves leading to an increase of total leaf area, number of tillers and total dry biomass. Moreover, sand deposition also increases soil volume, soil resources and mycorrhizae activity promoting ecosystem functioning (Forey et al., 2008; Maun, 1998). However, in these studies, the effects of increasing disturbance on ecosystem functioning might have been confounded by a decrease in stress also occurring towards the ocean, as observed in salt marches by Proença et al. (Proença et al., 2019). Indeed, Forey et al. (Forey et al., 2008; Forey et al., 2010) have documented a beneficial effect of the ocean spray for plant growth and survival in the incipient foredune, with lower vapor pressure deficit values and maximum temperatures than in the inland transition dunes protected by the incipient foredune and the established foredune. In order to assess the potential role of disturbance for ecosystem functioning in stressed environments, we have to control for variation in stress.

Recently, scientists and coastal dune stakeholders have been attempting to reintroduce a certain dynamic in stabilized coastal dunes characterized by high plant cover and biomass (Arens et al., 2020; Creer et al., 2020; Pye and Boltt, 2020). This is particularly the case in Northern Europe where some remobilization projects are attempting to restore a dynamic of sediment transport from the beach to the back of the dune by removing vegetation or foredune notch excavation (Arens et al., 2013; Eamer et al., 2013; Konlechner et al., 2015; Kuipers, 2014). In these low-stress coastal dunes, these methods have been shown to be successful in increasing sand transport and thus disturbance in the dunes (Ruessink et al., 2018), leading to an increase in diversity, by reducing competition, and a rejuvenation of the soils (Brunbjerg et al., 2015; Nordstrom et al., 2007), confirming the CSR model (Grime, 1974). In contrast, sand remobilization in dunes from southern latitudes with lower summer rainfall and, thus, higher stress may reduce ecosystem functioning and diversity, consistent to Huston (Huston, 1979). To the best of our knowledge, there has been no experimental study assessing the impact of sand remobilization on ecosystem functioning in more stressed coastal dunes. These impacts are crucial to assess in the context of changing management paradigms.

We designed an original experiment in the coastal sand dunes of South West France, where we applied experimental disturbances simulating blowout effects from marine and aeolian origins at two positions along the complex stress/disturbance gradient. Indeed, in northern Europe, some studies have attempted to reintroduce sedimentary dynamics within the coastal dunes by blowout restoration or notches excavation (Pye and Blott, 2016, Ruessink et al., 2018, Van Boxel et al., 1997). However, these studies have only focused on morphological changes without assessing the impact on plant communities and ecosystems. As interactions between physical and biological processes play a major role in coastal dunes evolution while evolving across many spatial and temporal scales, it is important to set up experiments able to understand these interactions across different dune habitats. We choose to quantify litter decay as good proxy of nutrient cycling and availability along transects delineated from the beach to the back dunes, along the experimental blowouts and in controls. We also measured in all treatments environmental variables (i.e., wind abrasion, elevation variation, and grain size, (Forey et al., 2008)) and plant species composition as indicator of community response to environmental disturbances and changes in ecosystem functioning. Additionally, in order to take into account the context-dependency of the effect of disturbance (Grime, 1974), we also assessed if community and ecosystem responses were affected by the beach dune system geomorphology (Davidson-Arnott, 2010). We aim to answer the following questions: (1) do contrasting disturbance types (e.g., marine and aeolian) have different effects on ecosystem functioning and (2) does dune geomorphology (e.g., foredune elevation, oceanward slope and width, transition dune elevation) alter the effect of disturbance on ecosystem functioning.