Development and evaluation of new modelling solutions to simulate hazelnut (Corylus avellana L.) growth and development

The worldwide increase in the hazelnut demand and cultivated area are explained by the rising importance of this species for the food industry. However, no simulation models are available to support the analysis of the impact of environmental conditions and management practices on hazelnut production. This paper presents two modelling solutions to simulate hazelnut tree growth and development under potential and water-limiting conditions, partly based on generic approaches used in tree modelling and on new modules specifically developed to address hazelnut tree peculiarities. The two solutions differed in the simulation of the photosynthetic rate, in one case reproduced using a Farquhar-type model (gross photosynthesis), whereas a radiation use efficiency approach (net photosynthesis) was used in the other. The coherence of simulation outputs at leaf, organ and plant level and the models responsiveness to weather conditions was verified and discussed with literature data, and the performances of the modelling solutions were evaluated using experimental data collected between 2002 and 2013 growing seasons in Piedmont region (northern Italy). Results highlighted the reliability of both the solutions in reproducing phenological development (mean relative root mean squared error, RRMSE = 8.61%), as well as the time trend of specific leaf area (RRMSE = 26.32%) and leaf area index (RRMSE = 18.46%). Also, the simulation of the temporal dynamics of soil water content and temperature along the soil profile led to outputs very close to observations (RRMSE = 14.02% and 10.32%, respectively). The solution based on gross photosynthesis resulted slightly more accurate in reproducing the year-to-year fluctuation in yields (RRMSE = 25.03%) compared to the one based on net photosynthesis (RRMSE = 30.40%). These results proved the suitability of these modelling solutions to be used as simulation engines within a variety of applications, ranging from decision support systems for the management of the orchard to complex yield forecasting systems.