Coupling spectral unmixing and trend analysis for monitoring of long-term vegetation dynamics in Mediterranean rangelands Patrick Hostert a, * , Achim Ro ¨der b , Joachim Hill b a Department of Geoinformatics/Cartography, Institute of Geography, Humboldt University Berlin, Chausseestr. 86, 10115 Berlin, Germany b Remote Sensing Department, Universita ¨t Trier, Behringstr. 15, 54286 Trier, Germany Received 19 November 2001; received in revised form 21 October 2002; accepted 3 June 2003 Abstract The development of vegetation cover is one of the primary indicators for land degradation, stability, or regeneration in regions threatened by overgrazing. This paper addresses the problem how spatially explicit information about degradation processes in European Mediterranean rangelands can be derived from long time series of satellite data. The selected test site in central Crete, Greece, is considered to be representative for the highly heterogeneous character of such landscapes. The monitoring approach comprises the time period between 1977 and 1996, covered by nine Landsat TM and four Landsat MSS images. Special emphasis has hence been put on the evaluation of potentials and drawbacks when coupling Landsat TM and MSS based results. The data sets were geometrically and radiometrically pre-processed in a rigorous fashion, followed by a linear spectral unmixing approach and a time series analysis of vegetation fraction images. Based on the resulting map, the spatio-temporal patterns of vegetation cover changes are explained. Even a test site such as central Crete, with its limited spatial extend, exhibits heterogeneous patterns of change, supporting the hypothesis that long time series of EOS data from Landsat-like sensors are mandatory to identify the relevant changes at landscape level. D 2003 Elsevier Inc. All rights reserved. Keywords: Landsat; Long-term monitoring; Spectral mixture analysis; Vegetation dynamics; Rangelands 1. Introduction The European Mediterranean has been under serious anthropogenic pressure since historical times. Among others, animal husbandry has always been one of the major influ- ences in rangeland ecosystems of Northern and Southern Mediterranean lands alike (Le Houe ´rou, 1981). In recent times, the enlargement of the European Union towards the Southern rim of the European continent significantly changed the boundary conditions for most ecosystems asso- ciated with Mediterranean grazing regimes. European subsidies for grazing stock on a per capita basis enforced a rapid growth of grazing animal populations in some European Mediterranean countries and stocking densities often exceed carrying capacities (Dubost, 1998; Papanastasis, 1998). Severe degradation of soil and vegeta- tion resources can be observed as a result of such pressure. However, while ecosystem degradation is evident in many cases, there is often only limited knowledge about the heterogeneous pattern of change in space and time. More- over, even in the case that ongoing changes are obvious, their severity is difficult to monitor and to assess in a continuous manner (Hill, Hostert, Tsiourlis, Kasapidis, & Diemer, 1998; Pickup, Bastin, & Chewings, 1998). Overgrazing may change vegetation communities in qualitative as well as quantitative ways. Excessive feeding on palatable plant species usually leads to an invasion of unpalatable varieties that may even culminate in an overall increase in vegetation density. However, Mediterranean grazing regimes are usually managed through intentional burning or mechanical clearing in a way that unpalatable species are suppressed. Excessive overgrazing during ex- tended time periods will therefore usually lead to a decrease in biomass, expressed through declining vegetation density. The development of vegetation cover may hence serve as a key indicator of degradation, stability, or regeneration. This paper consequently addresses the problem, how long time series of spatially explicit monitoring information can be drawn from satellite based remote sensing data. The first objective is to develop a suitable methodological 0034-4257/$ - see front matter D 2003 Elsevier Inc. All rights reserved. doi:10.1016/S0034-4257(03)00145-7 * Corresponding author. Tel.: +49-30-20936805; fax: +49-30- 20936835. E-mail address: patrick.hostert@geo.hu-berlin.de (P. Hostert). www.elsevier.com/locate/rse Remote Sensing of Environment 87 (2003) 183 – 197