Study About Scaling Effects In Topographic Correction And Validation Of Remotely Sensed Downward Shortwave Radiation

Ground measurements of surface solar radiation(SSR) are generally considered to be the true value and are extensively used for the monitor of surface energy balance of the earth. The remote sensed SSR products are capable of monitoring the dynamic variation of energy balance on a large scale of both the land and ocean, but such products lacks profound consideration of the redistribution of solar radiation caused by terrain, i.e., the shadowing effect of adjacent mountains. Thereafter, models or algorithms able to correct topographic effects are urgently to be developed to accurately estimate the real distribution of SSR. In practice, DEM data at various resolutions can be introduced to correct topographic effects for SSR products at certain scales. It is not the application of DEM, but the choice of DEM resolution that really matters because DEM at the chosen resolution should both effectively characterize the spatial heterogeneity of SSR and also minimize the data redundancy, which still remains a current difficulty. The validation reliability is another problem we need to take into consideration when the “point” value of insitu measurements is compared with the “pixel” value of remote sensed SSR products. Whether the point value can represent adjacent larger scales mean value needs further research. That is to say, the result of validation depends not only on the topographic correction models, but also the poor spatial representativeness of the observation sites. Based on the issues mentioned above, we focus the study on three main parts:(1) Evaluate the currently most used topographic correction models, and select the best one among them to quantitively analyze the effect of topography on SSR based on the high-resolution simulated horizontal SSR. Develop visual tools to calculate the distribution of SSR on the rugged terrain by only horizontal SSR and DEM data.(2) Research the scaling effects of variable resolution DEM when they are applied to correct topographic effects. For remote sensed SSR products at certain scales, select the best DEM resolution which can achieve the optimum correction results, and analyze the sensitivity of DEM resolution to the corrected results.(3) On the basis of previous two tasks, establish a new method to evaluate the spatial representativeness of meteorological radiation stations, then simulate the results of 97 meteorological radiation stations in China. At the same time, give an extended analysis of the influence factors of spatial representativeness, including the sun’s geometry position, atmosphere condition, temporal scales of data used, the installation of radiometer, the slope and aspect of stations, etc.