Proceedings 5 th EARSeL Workshop on Imaging Spectroscopy. Bruges, Belgium, April 23-25 2007 1 THE USE OF HYPERSPECTRAL REMOTE SENSING DATA FOR THE ASSESSMENT OF CHEMICAL PROPERTIES OF DRYLAND SOILS IN SE-SPAIN Thomas Jarmer 1 , Joachim Hill 2 and Sebastian Mader 3 1. Freie Universität Berlin; Remote Sensing and Geoinformatics, Institute for Geographical Sciences D-12249 Berlin, Germany; jarmer@geog.fu-berlin.de 2. Remote Sensing Department, University of Trier, D-54286 Trier, Germany, hillj@uni-trier.de 3. Remote Sensing Department, University of Trier, D-54286 Trier, Germany, mader@uni- trier.de ABSTRACT Soil samples were collected in the catchment of the upper Rio Guadalentin basin in SE Spain (Prov. Murcia). Additional to soil chemical laboratory analysis the spectral reflectance of the soil samples was measured in the range of 0.35 and 2.5 μm. These reflectance measurements were convex-hull-normalised to derive individual absorption features. The continuous spectra were used to calculate colour parameters according to the Commission Internationale de l´Eclairage (CIE) colour scheme. Accordingly, a method for spectral detection of pedo-chemical properties in the investigated soil was developed based on statistical analysis which allows the prediction of the chemical concentrations in soils. Organic carbon contents were estimated from reflectance measurements based on C.I.E. colour coordinates with high accuracy (r² cv > 0.79). Inorganic carbon was predicted from laboratory reflectance measurements with an accuracy of r² cv > 0.71 considering the normalized absorption area of the carbonate absorption feature at 2.33 μm and the C.I.E. chromaticity value x and y. The transfer of the derived regression models to HyMap data allowed for the spatial prediction of organic and inorganic carbon contents in the Rio Guadalentin basin. Concentrations obtained are in accordance with the concentration range of the chemical analysis. The predicted chemical soil concentrations reflect the physio-geographic conditions of the investigated area. INTRODUCTION Three decades after the first United Nations Conference on Desertification 1977 in Nairobi reduction of soil and water resources and associated land degradation processes world-wide are considered as one of the most important environmental problems. It is agreed that land degrada- tion processes are climate sensitive and related to physio-geographical and anthropogenic conditions (i). Regarding the fact that the pressure on arid and semi-arid areas resulting from climatic variability, climate change, demands of increasing stocking rates and population develop- ment was probably never as high as nowadays (ii), arid and semi-arid environments are often considered as risk areas in the context of global climate change and desertification dynamics. Soils as a substantial part of terrestrial ecosystems are extremely important, since soils are. Soil chemical and physical properties are basic indicators for soil productivity which is strongly related to agricultural production. In semi-arid and arid environments the inorganic carbon content in soils developed on carbonatic bedrock material is a major soil development indicator. High inorganic carbon concentrations point to weakly developed or degraded soils while low contents often indicate more developed soils. Organic carbon is a second reliable indicator related to soil quality and depending on water availability. In drylands water is the limiting factor for vegetation growth and influences organic matter production. Increasing soil organic carbon concentrations improve soil conditions because of better aggregation, higher infiltration rates and water retention which are conductive to their resistance to erosion.