Vegetation traits modulate resilience to fire in Mediterranean woodlands

Resumen

The response of an ecosystem to disturbance like fire reflects its stability, which has two main facets: resistance and resilience. Resistance is commonly defined as the ability of a system to withstand a disturbance; it is measured as the magnitude to which the target attributes or variables change due to the disturbance. Resilience has been defined from two perspectives: engineering and ecological resilience. Engineering resilience is defined as the rate at which a system returns to a single steady or cyclic state following a disturbance. Ecological resilience considers how much disturbance is required to move the system from one stable state to another alternative stable state, to another stability domain. Resistance and resilience to fire of a given ecosystem results from a combination of biotic and abiotic factors that are related to (1) fire regime (fire severity and frequency), (2) physiographic and climatic conditions, and (3) community structure and plant traits. These factors interact and act at different scales. This dissertation is particularly focused in the mediation role of plant traits in the response to fire of plant community and soil functions. Many plant traits, either regeneration or structural features, have been showed to correlate with the response of plants and communities to fire. Vegetation traits play a decisive role in determining flammability, fire severity and post-fire recovery, which in turn affect the composition of the post-fire community and lead to feedback processes between vegetation and fire. In Mediterranean regions, many species can recruit abundant seedlings (seeder) or resprout from vegetative organs (resprouter) after fire; other species are capable of using both regeneration mechanisms (facultative). These species traits allow pre-fire species or individuals to persist after fire, making Mediterranean communities highly resilient to fire. However, obligate seeders and resprouters differ in their response to fire regime, and therefore fire frequency and/or severity have the potential to modify post-fire seeder/resprouter ratio in burned areas. Several studies have assessed the effect of fire frequency on this ratio, while the role of fire severity is much less known and no previous studies have assessed it in Mediterranean plant communities. Plant and soil interactions and feedbacks, which drive important ecosystem processes such as biomass production, water availability, and nutrient cycling, are mediated by multiple plant traits. For example, traits such as litter production and chemistry, plant biomass, quality of root exudates and labile C inputs, and the rate and pattern of root growth modulate the structure and activity of the soil microbial community, leading to small-scale variations in soil organic carbon and nutrient content, pH, and soil microbial diversity. Groups of species (functional groups) share particular combinations of traits that influence ecosystems function in specific ways. For example, litter decomposition rates and pattern vary between resprouter and seeder species, which is attributed to differences in leaf chemistry between these two functional groups. Understanding how plant functional traits modulate the response of soil functions to fire disturbance is critical to predict the overall impact of increased fire frequency in fire-prone woodlands and shrublands and would help to identify plant-soil interactions and feedbacks behind the ecosystem response to fire. Plant-soil interactions and feedbacks may shape the magnitude and ecological impact of disturbances. However, little is known about the role played by plant traits and functional groups in the stability (resistance and resilience) of soil functions against disturbances. The general objective of this dissertation was to assess the role of vegetation traits in the resilience to fire of Mediterranean woodlands, with a particular focus on (a) vegetation fire-persistence (regeneration) traits, and (b) plant community and soil functioning responses to fire. The underlying general assumptions of this work are: both plant community composition (and associated functional traits) and soil functions are involved in ecosystem resistance and resilience. Soil functions depend on plant composition (functional groups) through a variety of feedback mechanisms. Plant functional groups modulate fire severity and resistance of both vegetation and soil functions by influencing fuel properties. They also modulate ecosystem resilience through differences in their capacity for re-growth, colonization, and the re-establishment of plant-soil feedbacks. The dissertation is structured into six chapters: General introduction (Chapter 1); four chapters that correspond to research articles published or submitted to scientific journals (Chapters 2 to 5) and a final chapter on Discussion and Conclusions (Chapter 6). Chapter 2 discusses how successional stages following land abandonment (dry grassland, dense shrubland, and pine stands) and their inherent characteristics of community composition and structure influence fire severity and resilience of Mediterranean ecosystems. Successional stages differed in total biomass, litter quantity, vertical distribution and horizontal continuity of plant cover, and overall flammability. The lowest fire severity was recorded in dry grassland, while dense shrubland and pine stands showed no differences between them. One year after fire, plant cover was inversely associated with fire severity, yet this negative relationship faded with time, becoming unappreciable seven years after fire. Pine stands and dense shrubland resulted in similar shrubland communities, dominated by highly flammable obligate seeder species such as Ulex parviflorus and Cistus species, which contributed to homogenize the landscape. Pine stands showed the largest changes in community structure after fire and the lowest short-term post-fire plant recovery, so pine stands can be considered the most vulnerable of the three successional stages studied, particularly if wildfires are followed by relatively dry years. Chapter 3 addresses the question of how fire severity influences the post-fire assembly of plant regeneration traits. It reports on post-fire dynamics of obligate seeders and resprouters and the variation in post-fire seeder/resprouter abundance ratio as a function of fire severity in three Mediterranean communities (dry grassland, gorse shrubland and pine stand). The seeder/resprouter abundance ratio increased non-linearly with increasing fire severity. This response was consistent for the three plant communities studied. The positive relationship between seeder/resprouter ratio and fire severity appears to be related to the mortality of individuals of resprouter species, which creates gaps of bare soil that are favourable microsites for seed germination (either from the existing soil seedbank or from adjacent unburned seed sources) and growth of seeder species. Chapter 4 focuses on how fire-related plant functional traits modulate the response of soil functions to fire disturbance in Mediterranean shrublands. The chapter analyzes the role played by plant regeneration traits (resprouter versus seeders), life-forms (grasses versus shrubs) and flammability-related traits in the amount of change (resistance) and the recovery (resilience) after fire of soil stability, infiltration, and nutrient cycling functions in repeatedly burned shrublands. The study focused on vegetated patches, each dominated by a particular species (Brachypodium retusum, Quercus coccifera, Ulex parviflorus, Cistus albidus, Rosmarinus officinalis) and inter-patch areas (bare soil). The selected species have contrasting fire-related plant traits that might affect fire behaviour and the response of vegetation to fire. Three experimental burnings were applied in three different shrubland communities. Soil functions were estimated by the Landscape Functional Analysis method (LFA; CSIRO). The response of these functions to fire was assessed by indices of resistance and resilience. The results showed that the variation in resistance and resilience of soil functions largely depends on the variation in structural and functional fire-related traits of the species. Resistance was more related to plant structural traits (e.g., the accumulation of dead fuel); while resilience was related to functional traits, such as the resprouting capacity of the species studied. Chapter 5 evaluates the resistance and resilience of two soil enzyme activities involved in phosphorus and carbon cycling (acid phosphatase and ß-glucosidase, respectively) as a function of the dominant mechanism (resprouting, seeding, or both) involved in the post-fire recovery of vegetation. Using an experimental fire, the recovery dynamics of enzyme activities, soil organic carbon, and vegetation cover were monitored in four types of microsites: resprouter-shrub patches; seeder-shrubs patches; patches co-dominated by seeder and resprouter shrubs; and inter-patch areas (bare soil). The functional groups related to post-fire regeneration traits (resprouter, seeder) modulated the activity of both enzymes (acid phosphatase and ß-glucosidase), with high activity in patches dominated or co-dominated by resprouter species. Fire similarly reduced the activity of both acid phosphatase and ß-glucosidase in all microsite types, but subsequent post-fire dynamics of these two enzymes largely varied. While ß-glucosidase was very resilient fully recovered three years after fire, acid phosphatase showed no signs of recovery. Overall, the results point to a positive influence of resprouter species in the recovery of enzyme activities.