Estimation of hourly global photosynthetically active radiation using artificial neural network models

Photosynthetically active radiation (PAR) reaching the earth’s surface is a major parameter controlling many biological and physical processes related with the evolution of plant canopies, agricultural and environmental fields. Unfortunately, PAR is not often measured and therefore it must be estimated. The unavailability of measurements of global solar radiation at the place of interest and different factors affecting the linear relation between PAR and global solar radiation can preclude the estimation of PAR from global solar radiation. In this paper, a novel approach based on a simple multilayered feedforward perceptron has been used to analyse the non-linear relationships between PAR and different meteorological and radiometric variables in order to determine their relative relevance. An artificial neural network based model for the estimation of the hourly PAR involving hourly global irradiance as only measured variable has been successfully developed. The model was tested using data recorded at six radiometric stations covering a wide range of climates. The model performance has been compared with other existing empirical complex models showing important improvements. Next, a second artificial neural network based model involving only sunshine duration measurements has been developed and proved to be an acceptable alternative to calculate hourly PAR when radiometric information is not available.