An improved model for estimating degree days

Although using hourly data offers the greatest accuracy for estimating growing degree days, daily maximum and minimum temperature data are often used to estimate degree days by approximating the diurnal temperature trends. In this paper, an empirical model (TM model), recently developed for estimating hourly mean temperature, is used to calculate degree-day values. The TM model describes the diurnal variation using a sine function from minimum temperature at sunrise until reaching maximum temperature, another sine function from maximum temperature until sunset, and a square root function from then until sunrise the next morning. Degree day estimates from the TM model were compared with values obtained from single triangle (ST) method. The results showed that the TM model was superior to the ST method. The differences between observed phenological stage occurrence and predicted date from the TM model were always less than 13 days for bud break and flowering, and 10 days for harvest.