An assessment of the effectiveness of atmospheric correction algorithms through the remote sensing of some reservoirs

Although satellite remote sensing techniques have been widely implemented for a variety of applications, using either single or time-series images, few studies have explicitly considered atmospheric effects, and how they can most effectively be minimized. Despite the considerable number of available atmospheric correction algorithms, there is little literature concerning their relative merits. Over water bodies, atmospheric effects account for the majority of the at-satellite measured radiance in the visible bands, and therefore targets of this type provide an opportunity for assessing the effectiveness of the different methods available. This paper reports a study of atmospheric effects and their correction, using multi-spectral satellite remote sensing data for an area to the west of London that includes eight large water reservoirs and a major international airport (Heathrow). Through comparisons of reflectance within a time series of 12 Landsat-5 Thematic Mapper (TM) images, the overall impact of atmospheric contributions is shown. The available atmospheric correction literature is then reviewed, and the methods applied to the satellite imagery of the reservoirs. The results are compared with reflectances acquired using a field spectro-radiometer. A critical appraisal of the results, and of problems encountered in applying the various methods, leads to an evaluation of their value in practice.