The estimation of daily mean soil temperature at different depths in three Agroecological regions of Sri Lanka

Soil temperature is an important climatic variable in regulating ecosystem process, yet it is difficult and costly to measure daily. Several mathematical models are available to estimate soil temperature; however, most are with complicated independent variables. In this research, site specific parsimonious models were developed to estimate daily mean soil temperature in deeper depths (i.e.>10 cm) using air temperature, soil temperature at 5 cm depth and day of the year as independent variables. Different models were developed for four Coconut growing sites and results showed that 5 cm depth soil temperature along with day of the year estimate the soil temperature at deeper depths with high accuracy. Further, models with air temperature and day of the year estimate the soil temperature at 5 and 10 cm depths satisfactorily. These models can be used to estimate the daily mean soil temperature simply and cost effectively and it also allows the reconstruction of missing soil temperature values. COCOS Vol. 20: 9-25 (2013)