AN ARTIFICIAL NEURAL-NETWORK TO ESTIMATE SOIL-TEMPERATURE

This study was undertaken to develop an artificial neural network (ANN ) model for transient simulation of soil temperature at different dept hs in the profile. The capability of ANN models to simulate the variat ion of temperature in soils was investigated by considering readily av ailable meteorologic parameters. The ANN model was constructed by usin g five years of meteorologic data, measured at a weather station at th e Central Experimental Farm in Ottawa, Ontario, Canada. The model inpu ts consisted of daily rainfall, potential evapotranspiration, and the day of the year. The model outputs were daily soil temperatures at the depths of 100, 500 and 1500 mm. The estimated Values were found to be close to the measured values, as shown by a root-mean-square error ra nging from 0.59 to 1.82 degrees C, a standard deviation of errors from 0.61 to 1.81 degrees C, and a coefficient of determination from 0.937 to 0.987. Therefore, it is concluded that ANN models can be used to e stimate soil temperature by considering routinely measured meteorologi c parameters. In addition, the ANN model executes faster than a compar able conceptual simulation model by several orders of magnitude.