Confronto tra l'uso del tempo termico e idrotermico nei modelli di previsione delle emergenze delle infestanti

Comparison between the use of thermal and hydrothermal time in weed emergence predictive models Numerous studies have been conducted to improve knowledge on weed emergence in the field. Many of these are addressed to develop prediction models in order to provide information to famers about opportunities, methods and timings of weed control. The first-generation models for predicting weed emergence, still widely-used, were based on the thermal time (growing degree days) concept. These models consider temperature as the only factor influencing the germination process. The second generation of model integrated the temperature factor with soil water potential in the concept of hydrothermal time. To calculate thermal and hydrothermal time it is necessary to know base temperature and base water potential. In some cases to calculate the thermal time the same base temperature is used for all the considered species, in other cases it is calculated for each species empirically (by selecting the base temperature that gives the best simulation) or estimated with laboratory experiments. The base water potential is calculated using the same method. In 2007-2008-maize was sown on two different dates (early sowing in March and traditional sowing in April) and in two sites, Padova and Pisa. Weeds were counted and eliminated twice a week. The emergence dynamics of Amaranthus retroflexus L. and Chenopodium album L., two species found in aIl fields, was modelled using a Gompertz function based on thermal time calculated using the same base temperature for both species, on thermal time with the specific base temperature estimated for each species and on hydrothermal time calculated with the specific estimated base temperatures and water potentials. The results show that the thermal time model can efficiently simulate emergences in years without dry periods. On the contrary, in years with low rainfalls and dry periods, it is necessary to use hydrothermal time to explain the emergence dynamics. In conclusion, if the objective is to built a model usable in different climates and situations (e.g. different sowing dates), only the use of hydrothermal time can allow a more reliable estimation of emergence dynamics to be obtained.