numero30_7:Maquetaci\363n AET.qxd - Universidad Politécnica de ...

F. González,S. Kaiser, S. Merino, M. Huesca, A. Roldán, J.M. Cuevas y G. VenturaINTRODUCTIONIn times when many countries are concerned aboutthe reduction of CO 2 and other greenhouse gases tosatisfy the requirements of the Kyoto protocol, theavailability of updated cartography at different scalesbecomes more and more important. Internationalorganizations such as the European Space Agency(ESA) are involved in global projects that aim toprovide these products to decision and policymakersall over the world. One of the most determining projectsis the currently running European GLOBCO-VER project (DUP-ESA, 2006). The objective is todevelop a service which will produce a 300m globalland cover map for the year 2005 using mainly fullresolution data acquired by the MERIS (MediumResolution Imaging Spectometer) sensor on-boardENVISAT (Joint Research Centre – Terrestrial EcosystemMonitoring, 2006).As the medium spatial resolution of MERIS-FR(Full Resolution) products may restrict the processof obtaining high-quality end-products in somecases, other satellite-derived data with increased spatialresolution, such as the ones acquired by the AdvancedWide Field Sensor (AWiFS) of the IndianRESOURCESAT-1 (National Remote SensingAgency, 2006) should be tested for their aptitude toproduce high-quality land use - land cover (LULC)maps with a reasonable effort in terms of time andmoney. Benefits and disadvantages of using satelliteswith either increased spatial or spectral resolutionhave to be evaluated in the future to refine methodologiesfor LULC mapping.As a first step in this direction, the classificationdescribed in this paper aims to supply informationabout pros and contras of using satellite data withdifferent spatial and spectral resolution for the creationof forest or LULC maps at any scale for researchersand decision makers dealing with this issue.For that purpose, a classification based on an AWIFSimage of the Spanish province of Madrid was performedand the results were compared with previousclassification results for MERIS-FR data.STUDY AREAThe Spanish province of Madrid (figure 1) comprises8022 km² including the metropolitan area ofMadrid in the centre surrounded by evergreen broadleavedforests (Quercus ilex) and shrublands, cultivatedand sparsely vegetated areas in the South-Eastand the Sierra de Guadarrama with its evergreen anddeciduous forests (Pinus sylvestris, Quercus pyrenaicaetc.) in the North-West.Figure 1. The province of Madrid (grey) situated in the centre of Spain86Revista de Teledetección. ISSN: 1988-8740. 2008. 30: 85-91