ESTIMATION OF SOLAR RADIATION FROM CLIMATIC PARAMETERS Citation: Ali, M.H., A.K.M. Adham and M.S.U. Talukder (2005). Estimation of solar radiation from climatic parameters. Bangladesh J. Agril. Sci.32(1): 99 -104 ABSTRACT: Solar radiation is often required for many simulation models of evapotranspiration, crop growth and productivity, and climatic impact studies. For locations where measured values are not available along temporal and/or spatial scales, it can be estimated using empirical models. Techniques of multiple linear regression were used to develop model for site-specific forecasting of solar radiation viz. Mymensingh, Bangladesh from climatic variables. Inputs for the regression model include maximum temperature, minimum temperature, bright sunshine hour and day-length. The model was tested on an independent data set. Graphical comparison and different statistical indicators (root mean square error = 1.65 MJ/m 2 /d, relative error = 20 %, index of agreement = 0.81) indicate that the regression model provided reasonably accurate estimates of irradiance and hence, can be used for non-instrumented periods and at sites surrounding the calibrated site. Keywords: Global radiation, photosynthesis, climate, Bangladesh INTRODUCTION Agriculture is an exploitation of thermal and radiative energy. Solar radiation affects many physiological processes, particularly photosynthesis. Most crop growth and water use models require total incident solar radiation data because photosynthesis, the production of biomass, requires light energy (McCown et al., 1996; Heiniger et al., 1997). This is also needed for inputs in simulation models used in ecological systems (Pickering et al., 1994). Photosynthetically active radiation is measured infrequently, but it can be derived from total direct plus sky radiation (herein called solar radiation).