Mediterranean evergreen vegetation dynamics : detection and modelling of forest and shrub-land development in the Peyne catchment

Vegetation development in Mediterranean landscapes is often a slow process. The typical Mediterranean climate -with long dry periods in summer, mild winters and concentrated rainfall events in spring and autumn- is an important constraint on growth, enhanced by the often marginal and degraded soil conditions. Climate change scenarios predict that the extreme character of the Mediterranean climate will increase: higher temperatures especially in the summer, and decreasing summer precipitation. The low growth potential and slow succession rates make the Mediterranean landscape specifically vulnerable to disturbances and modern climate change. It is therefore important to create reliable predictions of future vegetation development and to have accurate methods to monitor the natural areas. Analysis of spatial patterns and temporal variability in key vegetation characteristics such as biomass, leaf area, water use and growth provides information on the ecosystem that can be used to create more accurate projections for future development of the landscape and vegetation. The main objective of this thesis is therefore: To increase our understanding of, and ability to monitor the growth and development of evergreen Mediterranean vegetation and to improve the projections of vegetation change as an effect of modern climate change. The evergreen forests and shrub lands in the Peyne area in southern France stand model for many areas in the European Mediterranean basin and other Mediterranean areas across the globe. This thesis combines remote sensing, plot-based field data on biophysical vegetation parameters, geophysical exploration, tree-ring analysis, and dynamical production modelling. Large improvements can be made in remote sensing detection of biophysical vegetation parameters by optimising the spatial and temporal support of the imagery. Rooting depths of evergreen Mediterranean species Quercus ilex and Arbutus unedo were found to reach far into the fractured bedrock. By using the geophysical exploration method of Electrical Resistivity Tomography (ERT) it is possible to detect vegetation water abstraction from the rock substrate during the summer down do six meters below the surface. Using a dynamic tree growth model combined with tree ring analysis it was found that the vegetation in the study area is especially vulnerable to a decrease in summer rainfall and a rise in summer temperature. A probable climate scenario for 2070 predicts a temperature increase of 2 °C in winter and 5 °C in summer plus a decrease of summer precipitation by 40%. Based on this scenario, the Forest-BGC vegetation productivity model predicts a decrease of leaf area by 25%, and potentially over 50% less production of woody biomass in the area. Such changes in vegetation growth may have large consequences for the landscape and land use possibilities in the region. This thesis contributes to the advance of our capability to monitor the growth and development of evergreen Mediterranean vegetation and to the improvement of projections of vegetation change in the next decennia as an effect of anticipated climate change.