Efectos de diferentes tipos de vegetación mediterránea sobre la hidrología y la pérdida de suelo

Resum

Large areas of the Mediterranean basin are affected by land abandonment. Erosion risk is immediately increased and remains until a new vegetation cover develops. Hydrology is affected by this land use change process in the mid- and long-term. Vegetal characteristic and re-colonization dynamics are of paramount importance for a correct understanding of this process in order to assess and control their effects on soil and hydrology. Herbaceous vegetation is the first step towards colonizing soil, forming grasslands, appearing over time shrub specimens, being of great importance in the Mediterranean region due to their diversity and the vast areas they occupy. The effects of these abundant vegetation types should be analyzed from an integrated perspective as water is a scarce resource and soil erosion risk is high, yet these types of vegetation are poorly characterized. The objective of this Doctoral Thesis is to assess the effects of land agricultural abandonment over hydrologic and erosive processes in a cool-Mediterranean climate region in the center of the Iberian Peninsula, and to characterize the effects of different Mediterranean shrub vegetation over runoff, infiltration, interception fluxes and soil loss. Two experiments at different scales were carried out in order to characterize the effects of grasslands and those of four shrub species (Dorycnium pentaphyllum, Colutea arborescens, Medicago strasseri and Retama sphaerocarpa) over runoff generation and soil loss. A high intensity simulated rainfall experiment applied over a one-meter diameter plot has shown the effectiveness of grasslands being able to minimize sheet erosion just six months after land abandonment. Nevertheless, their effects on runoff regulation were less marked. At 18 months after land abandonment, the runoff rate continued to be moderately high (20.8 mm h-1). The shrub vegetation is very effective in controlling the runoff and soil loss rates, except R. sphaerocarpa with a moderate influence. This small-scale experiment has allowed to separate direct and indirect effects, the first ones associated to aerial cover presence, and the latter ones to the changes induced in soil properties. The grassland protection effects practically disappear when the aerial cover is lost, whereas the effects of shrubs persist for a long time due to the complex and deep changes induced on soil properties. A shrub revegetation experiment has been realized in USLE plots (20 meter length and 4 meters width) in order to assess the influence of three shrub species (D. pentaphyllum, C. arborescens and M. strasseri) on soil loss and runoff generation. These variables have been measured under natural rainfall in order to assess the relationships between these variables and the rainfall properties. The shrub vegetation takes nearly three years to minimize soil loss and regulate the runoff generation, being very effective after this period. Results point out that effectiveness is greater under high intensity erosive events. The scale of the experiment has little influence on results when these are expressed in relative terms as a proportion of what happens on bare soil, especially under high intensity rainfall. A second line of work has been developed in order to characterize the influence of the shrub vegetation on the rainfall interception processes. An experiment has been carried out to determine the water storage capacity, a key parameter for the rainfall interception models, and stemflow for nine shrub species using two different laboratory techniques, rainfall simulation (13 mm h-1) and immersion of plants on water. Results have revealed that shrubs store high rainfall volumes and that interspecific variability is high for water storage capacity (from 0.35 mm to 3.24 mm) and for stemflow percentages (from 3.8% to 26.4%). Some anatomic characteristics (the leaf pubescense, the branch rigidity and the insertion angles of leaves and branches) have a great influence on water storage capacity and stemflow. The immersion method underestimates water storage capacity in a different way for each species. A second experiment has been realized for measuring rainfall interception losses, stemflow and throughfall for D. pentaphyllum, C. arborescens, M. strasseri and R. sphaerocarpa under natural rainfall. An adaptation of the “interception flow collection box” designed by Belmonte Serrato and Romero Díaz has been used in field experiments to automatically measure rainfall interception fluxes. Results indicate that interception losses change from 10% for R. sphaerocarpa to 43% for M. strasseri and stemflow varies between 7% for M. strasseri and 20% for R. sphaerocarpa, showing a great variability in both fluxes with some higher values than those commonly observed for tree vegetation. The species and the rainfall volume have been identified respectively as the more significant biotic and abiotic factors that are more influential to rainfall interception fluxes. Interception losses and stemflow varies with rainfall volume. Both of which increase their absolute values as rainfall volume increases. However, when they are expressed in relative terms, a logarithmic increase has been identified for stemflow, and a logarithmic decrease has been observed for interception losses, reaching a threshold value for rainfall volumes greater than 10 mm. Threshold values depend on the shrub species. Interception losses and stemflow vary greatly in this region, being difficult to predict because the rainfall events smaller than 10 mm are the more frequent ones and there is a great variety of shrub species. Rainfall events greater than 10 mm are less frequent but they are key events for species having an adaptation mechanism to aridity based on stemflow promotion, as these species store high water volumes in deep soil layers during these events. The results of this Doctoral Thesis highlight that risk of sheet erosion is limited to the six first months after land abandonment. After this critical period, the erosion risk is due to the low efficiency of grasslands in runoff regulation and to the fact that the effects of this vegetation type are ephemeral and disappear when the canopy cover is lost. Shrubs take almost three years to minimize soil erosion, but once they do, important changes in soil properties are induced, increasing infiltration capacity that remains although canopy cover is lost. Nevertheless, shrub species can store high water volumes generating important interception losses when they are able to form dense communities, affecting the water availability in a region where water is a scarce resource. The variability associated to “kind of shrub” is so important that species could be considered at the management level. R. sphaerocarpa, the most characteristic shrub which is proliferating in Mediterranean grasslands in central Iberian Peninsula, shows low interception losses and a moderate soil protection capacity, being an optimal shrub species for complementing the protection effects of grassland without affecting the rainfall volume reaching the soil. The high erosion risk and the scarcity of water make necessary an integral characterization of shrub species. This kind of vegetation protects the soil when form dense communities, but it is necessary to know the effects over the rainfall interception fluxes in order to not endanger available water resources.