Field and Imaging Spectroscopy to Determine Soil Degradation Stages in Semi-arid Terrestrial Ecosystems

Soil degradation implies a decline in the capacity of the soil to perform environmental regulatory functions and yield agricultural goods. Soils are affected by regressive and/or degradative evolution processes as a result of natural causes as well as anthropogenic influences in semi-arid regions such as Central Spain. The main aim of this work is to evaluate the capacity of hyperspectral HyMap data to identify soil degradation stages applying field and imaging spectroscopy data from selected test plots with soils under semi-arid conditions applying a GIS support. This work has included data from hyperspectral airborne data, field spectrometry, soil analyses, cartographic and ancillary data. The methodology adopted in this work was to determine degradation stages by identifying selected soil types as endmembers from HyMap data. Spectral field data together with soil analyses was used to identify and verify the endmembers. The Spectral Angle Mapper procedure was used to determine the distribution of the selected soils within the hyperspectral data. A GIS analyses was implemented to derive terrain properties related to soil degradation stages. The spatial distributions of the individual soil surface conditions were verified according to field and cartographic surveys. Results have shown that the spectral variations of the soil surface features were satisfactorily related to differences in the spectral characteristics that were influenced by the mineralogical composition as well as physical and chemical soil properties. The identification and spatial distribution of semi-arid soils within the study area was successfully carried out with the HyMap data. The further incorporation of soil and landscape characteristics permitted to differentiate soil degradation stages. INTRODUCTION Agricultural pressure is the driving force responsible for an intense land use causing severe soil degradation within semi-arid Mediterranean environments. Soil degradation leads to the loss of actual or potential productivity and utility. This implies a decline in the capacity of the soil to perform environmental regulatory functions and yield agricultural goods (i). A timely identification of soil degradation forms an important basis for sustainable land use and land conservation. The effects of an irregular precipitation regime, the reduction of an extensive vegetation cover, as well as the changes in land use have caused major damage to the ecosystems of semi-arid Mediterranean environments, such as Central Spain in the Southeast of the Autonomous Community of Madrid. Remote sensing applications have proven to be an important approach to identify soil degradation features (ii, iii, iv, v). Soils affected by regressive or degradative evolution processes, mainly Regosols, Cambisols and Luvisols are very sensitive to soil degradation processes. As a result of truncation, these soils expose the underlying materials of soil horizons (cambic, argillic) or soil layers (gypsum, marls) at the surface and form a land cover mosaic of significant characteristics related to soil materials. In this work, these soil surface materials are related to the sensitivity of soil degradation processes (erosion) and are referred to as soil degradation stages. The soil degradation stages, as a result of regressive or degradative soil evolution processes, can often be recognised through typical soil composition and colour changes, which are due to anthropogenic or natural removal of the top soil by erosion processes (iv). The objective of this work is to evaluate the capacity of hyperspectral HyMap data to identify and to determine the spatial distribution of soil degradation stages applying © EARSeL and Warsaw University, Warsaw 2005. Proceedings of 4th EARSeL Workshop on Imaging Spectroscopy. New quality in environmental studies. Zagajewski B., Sobczak M., Wrzesień M., (eds) field and imaging spectroscopy data from selected test plots of soils under semi-arid conditions and with a GIS support. The study area (Figure 1) is located in the centre of Spain in the Southeast of the Autonomous Community of Madrid bordering with the Autonomous Community of Castilla La Mancha. This area is in a region, which has a Mediterranean semi-arid climate with a mean annual temperature of 13.8o, an annual precipitation 454 mm and a potential evaporation of 767 mm according to the meteorological station at Ocaña in the province of Toledo (vi). The topography is ondulating with agricultural influenced plateau areas, known as paramo, with a maximum height of 780 m above sea level (asl) and divided by the Tajo river basin which lies at 520 m asl with steep escarpments. The lithology is mainly composed of Quaternary and Tertiary sediments with marls, gypsum and clay. The main soil types found within the area are Regosol, Cambisol, Luvisol and Calcisol (vii) with a frequent surface exposure of underlying horizons (argic, cambic, calcic, petrocalcic, gypsic or salic) or rocks (limestone, marls or gypsum). These characteristics determine an increase of clay, iron oxides, carbonates, gypsum or soluble salt content; and a decrease of the organic matter content on the soil surface. There is a scarce natural vegetation cover consisting of calcicole and gypsophile flora. The land use is mainly vineyards, olive groves and pastureland. The area within the Tajo river basin is almost entirely taken up by irrigated cultivation and within the last few years the irrigation has expanded to the neighbouring more elevated areas. Figure 1. Study area represented by a digital terrain model and HyMap flight lines A and B (false colour composition RGB of bands at 2.082, 0.895 and 0.462 μm) and two corresponding topographic profiles taken along the flight lines.