Assessing the influence of atmospheric and topographic correction and inclusion of SWIR bands in burned scars detection from high-resolution EO imagery: a case study using ASTER

摘要

In the present study, the effect of atmospheric and topographic correction to burned area delineation from Earth Observation (EO) imagery is explored. Furthermore, the potential added value of the inclusion of the shortwave infrared (SWIR) bands for improving retrievals of burned area cartography is investigated. In particular, the capability of ASTER imagery when combined with the maximum likelihood (ML) and the support vector machines (SVMs) classification techniques is examined herein. As a case study, a Mediterranean site on which a fire event occurred in Greece during 2007 and for which post-fire ASTER imagery was available is used. The combination of topographic correction (orthorectification) with the inclusion of the SWIR bands returned the most accurate results in burned area detection. SVMs showed the highest accuracy, showing the most promising potential in delineating the burned areas. The most accurate results for burned scar mapping were obtained from the combined use of SVMs with an orthorectified image and SWIR spectral bands, at least this was the case in our study site. Our results offer a very important contribution to the understanding of the capability of high-resolution imagery such as that from ASTER in burned area estimation. This study also corroborates the usefulness of topographic correction as an image processing step to be incorporated in modelling schemes for delineating burned areas from such data. Findings potentially provide very useful information towards the development of EO-based techniques that aim to operationally provide services related to the estimation of burned area. This is of considerable scientific and practical value to the wider scientific and users' community given the continuation of free access today to observations from space from high-resolution sensors globally.