Spring frost climatology in Trentino: Past and future scenarios | La climatologia delle gelate primaverili in Trentino: Il passato e gli scenari futuri

There is a strong evidence of a generalized tendency to an anticipation of flowering from many tree species, particularly for fruit trees. Phenological modelling may be helpful in estimating bloom dates in periods with missing surveys, both in the past and for the future, relying on climatic scenarios; that allows to estimate frost risk by crossing simulated bloom dates with simulated (or recorded) minimum temperatures in the same period. Such an analysis allows to detect possible trends in frost risk, comparing both the past with the present and the present with the future. Six phenological series were surveyed in Trentino on apple cv. Golden d.; with these data, a phenological model was developed starting from the "Utah" approach, based on the fulfilment of chill requirement and on the subsequent heat accumulation with a variable efficiency. We found a very good agreement between heat requirement and topo-climatic features of each site: plants growing in more "favourable" conditions need more heat to budburst. If this principle is extended, a "climate- dynamical" heat requirement might be assumed, instead of static one. The phenological model was applied to past data (from 1960) and to future simulations, after downscaling the climatic model HadCM3 at two significant sites in Trentino and according to two different IPCC's atmospheric evolution scenarios. For every season, simulated dates have been compared with temperatures predicted for the same years, and frost days have been assessed. The results point out that, under present conditions, frost risk is reduced in comparison to past decades, while it is stationary, or even in light decrease, for uture decades. This trend is more significant under the hypothesis of dynamic heat requirement, increasing with the general temperature increase.