Modelling and analyzing the water and carbon dynamics of Mediterranean macchia by the use of ground and remote sensing data

Mediterranean ecosystems are particularly vulnerable to the environmental changes which have occurred in the last decades. Evaluating the ecosystem response to these changes is therefore a top priority, particularly concerning water and carbon dynamics. The Pianosa Island is a well known test site where numerous environmental surveys have been performed to fully characterize the most typical vegetation type, Mediterranean macchia. A first measurement campaign concerning both vegetation and soil properties was performed in 2001, and was repeated in 2010 only for soil properties. Vegetation cover was characterized again by means of an aircraft high resolution LiDAR dataset taken in 2009. Additional medium and low spatial resolution satellite images (Landsat OLI and MODIS) and aircraft photos were recovered from various sources. The availability of these datasets offers a unique opportunity to develop and test a methodology capable of modelling and analyzing the water and carbon dynamics of Mediterranean macchia during the 2001-2010 decade. To this aim, simulation procedures integrating remotely sensed and ancillary data were first tuned towards the available observations of vegetation biomass and soil carbon content. These procedures were then applied to produce daily estimates of macchia actual evapotranspiration (AET), gross primary production (GPP) and net ecosystem exchange (NEE), whose accuracy was assessed against corresponding eddy covariance flux tower observations taken in two years (2007-2008). The results obtained are satisfactory and support the capability of the modelling approach to reproduce both the water and carbon dynamics of macchia during the study decade. From an ecological viewpoint, both fluxes are increasing in this time period, mainly depending on similarly increasing spring rainfall. The macchia ecosystem behaves as a net sink of carbon, which is stored primarily in soil (congruent to 90%) and secondarily in vegetation. (C) 2017 Elsevier B.V. All rights reserved.