A Modelling approach to estimate carbon dioxide exchange at regional scale

At the moment, it is hard to find studies dealing with carbon flux estimates at regional scale in Mediterranean climate areas. In this study, it is presented a modeling framework to obtain 3 Km resolution carbon exchange maps for the Mediterranean island of Sardinia.The multilayer SVAT model Advanced-Canopy-Atmosphere-Soil-Algorithm (ACASA) was coupled “on-line” to the mesoscale meteorological model Weather-Research-and-Forecasting (WRF). The WRF-ACASA model was initialized with the ERA-Interim meteorological reanalysis data at a resolution of 0.7° (77 Km), while the European Environment Agency Corine land use map was exploited to acquire the land use data.Results were validated in for vineyard ecosystems using Eddy Covariance (EC) observations. A linear regression of modelversusobserved data with an R2= 0.91 indicates that the model was able to explain more than 90% of the total observed variation in carbon dioxide flux. A further analysis was performed comparing model results with other studies involving the three most diffused vegetation categories of the island covering about 60% of the island:Mediterranean forest,Mediterranean maquis, andCropland and pastures. This analysis confirmed the robustness of the modeling framework for the above land use type. The methodology tested in this work proved as the WRF-ACASA model represents a flexible instrument to provide a cogent simulation of the spatial and temporal evolution variability of CO2flux at regional scale.