Exploring the need for identifying suitable pseudo-invariant targets for applying atmospheric correction in multi-temporal studies using satellite remotely sensed imagery

Atmospheric correction is a complex process, which requires substantial modelling and computation, and a major difficulty is to obtain appropriate input parameters for the models. Numerous investigators have dealt with the development of simple or sophisticated approaches for the atmospheric correction of satellite images. However there is uncertainty about the effectiveness of such techniques especially when dealing with historical datasets in which input parameters for atmospheric models prove difficult to be obtained. The use pseudo-invariant targets in conjunction with radiative transfer calculations is an alternative atmospheric correction technique which offers a relatively simple mean of removing atmospheric effects in multi-temporal series of image data; providing that suitable pseudo-invariant targets can be easily identified on the satellite images and records on the their spectral characteristics are available. The spectral data of the proposed pseudo-invariant targets can be easily found in the literature from other studies. Indeed, this paper explores the need for identifying suitable pseudo-invariant targets, which are large in size, distinctive in shape and common in many geographical areas. This paper presents an application of use pseudo-invariant targets for removing atmospheric effects from Landsat TM and ETM+ satellite imagery acquired over different geographical areas such as in UK, Cyprus, Kazakhstan and Greece for environmental applications