Item

Abstract

In the frame of the EU-FP7 EUFAR (European Facility for Airborne Research) project, higher performing soil algorithms are being developed as demonstrators for end-to-end processing chains with harmonized quality measures. For this, a review of existing soil algorithms and methodologies based on soil spectroscopy currently successful for the identification and prediction of soil properties was performed. Based on this review, the HYSOMA (Hyperspectral SOil MApper) interface was developed at the GFZ German Research Centre for Geosciences in the Remote Sensing section. It is a software package written in the IDL language for fully automatic generation of semi-quantitative maps of surface soil moisture content, soil organic carbon (SOC) content, and soil mineralogical content (e.g. iron oxide, clay mineral , carbonate). The main motivation for HYSOMA development is to provide non-expert users with a suite of tools that can be used for soil applications. Also, HYSOMA tries to incorporate modern hyperspectral algorithms with an easy-to-use graphical interface based on simple menu-driven functions. It is achieved by providing a software with basic image file import based on ENVI format, soil mask option removing water and vegetation pixels based on simple water and vegetation indexes, and performing soil functions based on analytical and empirical algorithms where no user input data (e.g. spectral libraries, ground truth data) is needed. Additional tools that allow fully quantitative soil mapping are implemented for more experienced users. In this paper, we present the development strategy of the HYSOMA toolbox, current status, and examples of derived soil maps.