Modeling Pasture Dynamics in a Mediterranean Environment: Case Study in Sardinia, Italy

In this paper it was developed and tested a hydrologically based, spatially distributed crop yield model for the purpose of modeling pasture dynamics within the Mediterranean environment of Marghine Goceano, in Sardinia region, Italy. The model uses daily information of weather drivers given by temperature, precipitation, and solar radiation, and provides daily spatially distributed estimates of soil water content, runoff, and pasture growth, including leaf area index (LAI) and biomass. Pasture grass growth for 6years (2001-2006) was simulated, and yearly average biomass yield (tha-1) was compared against ground-based and satellite-based estimates from previous studies, finding acceptable model performance. Intraseasonal biomass dynamics was evaluated during year 2004 and compared against pointwise ground measures from former studies, with acceptable results. The spatially distributed leaf area index maps were compared against their satellite-derived counterparts (9.3-km pixel size, 2005-2006), available on a biweekly basis. The pasture model provides acceptable performance in depicting daily, spatially distributed dynamics of pasture grass, and may be useful when tackling a number of issues, including, e.g., (1)assessment of pasture (and fodder) productivity under current climate and management, (2)short-term to medium-term forecast of pasture yield for seasonal herding, and (3)modified pasture conditions under prospective climate change using climate forcing from general circulation models and other climate models. Future developments may regard inclusion of nutrient dynamics and study of (over)grazing effect upon pasture dynamics.