Yield modelling in a Mediterranean species utilizing cause–effect relationships between temperature forcing and biological processes

Abstract

Crop yield forecasting has progressed clearly in the past years, mainly due to the growing demand for modelling. In the present research, considering the Mediterranean vocation for the olive cultivation, olive yield forecasting modelling was analysed, to understand the principal relationships between olive tree production and climate features in a large southern Italy area. The simulation of olive yield has practical application in different price formation activities: in management of cropping systems, formation of stocks, commercialization, making of agricultural policies and zoning. Both meteorological and biological variables were considered, pollen indexes were constructed from daily pollen concentrations evaluated by use of aerobiological methodology. The monitored pollen quantities should be considered as a Proxy variable of the olive orchards located around the volumetric pollen traps, even if the final harvest could be highly dependent on local summer meteorological conditions. The model carried out by use of Panel data analysis has demonstrated a great interpretative capacity, considering contemporary meteorological and biological variables (pollen monitored with volumetric traps). The aggregation of historical data (1999–2008) with cross-section data (10 different pollen monitoring stations) permitted to investigate the variability information related to temporal and geographical dimensions. The entire volume of pollen grains emitted during flowering was redundant to investigate the future fruit formation while the specific and concentrated mass of pollen grains during the days before full flowering was related positively with the final production. The statistical technique utilized permitted to obtain interpretative results common to different geographical areas and not limited by local meteorological conditions.