Extraction of soil properties from laboratory and imaging spectrometry data

The aim of this study is the assessment of soil parameters from remote sensing data sets which are important as input parameters for environmental information systems (e.g. modelling of hydrological cycles, evaluation of areas with erosional hazard). The soil parameters extracted from reflectance spectra are correlated with the soil properties and implemented in a soil data base. The existence of a spectral soil data base is of central importance for further applications of spectral evaluation routines - such as mixed signature analysis. The first step in creating a spectral data base is the extraction of soil parameters which can be assessed by the means of remote sensing. In a next step laboratory measurements of soils have to be evaluated in order to establish relationships between the reflectance spectra of soils and their physical or chemical properties. The following step uses these rules as a basis for the spectral soil data base, which links the information on the spectral properties with physical and chemical properties of the investigated soils. The main emphasis is focused on the investigation of laboratory spectra of soils and the extraction of soil parameters from these reflectance spectra. Methods are described how to parameterize soil properties such as the soil color, the content of organic material, mineral composition and soil texture from reflectance spectra. The study compares the information content of spectral data sets of different remote sensing data sources e.g. airborne imaging spectrometry with field and laboratory spectrometry.