Abstract

Knowledge of the prevailing weather conditions regarding air temperature and humidity is critical for the definition of the Fire Weather and Fire Danger indices. In this study, two fitting algorithms were developed for the simulation of the diurnal air temperature and relative humidity cycles, using 2-hourly measurements at the forested site of Mount Parnes in Attica, Greece. Both algorithms use as input values the observed daily minimum and maximum values of air temperature and humidity to simulate the temperature and relative humidity daily cycles, respectively. In general, the algorithms provide satisfactory results especially for the three summer months (June, July and August). A 2-day prediction of the diurnal temperature and humidity cycles was also attempted using as input the temperature or humidity data of the previous days in the area of interest. The prediction may be considered satisfactory for the three summer months and less satisfactory for May and September due to weather instability. Finally a dynamic model was also used for supporting the performance of the developed algorithms, especially in the event of sudden weather changes.