Resumen de Evolución temporal de la vegetación en el Sahel mediante el NDVI y su relación con la precipitación en el periodo 1983-2012
The development and population prosperity in the transition zone between the Sahara desert in the north and the tropical humid savannah in the south, known as Sahel, depends to a great extent on the rainfall regime. The climate is defined by a long and dry season and a short season with rains between the months of June and September. Rainfalls are determined by a strong north to south gradient with small and variable precipitation in the north, which translates into continuous change of vegetation species and life forms that range from sparse thorny shrubs in between grass in the north to a greater amount of vegetal cover and ligneous species in the south. The vegetation growth shows a considerable inter-annual fluctuation that generates a generalized concern in the predominantly rural population, since their way of life is mainly based on dry land agriculture and pasturing. The severe droughts that happened during the 1970¿s, which were extended until the middle of the following decade, have been widely studied because of their serious consequences on the population, and because it is considered as one of the first signs of global climatic changes. Sahel is the region that has shown greater precipitation and vegetation variations in the last three to four decades. Technological advances in remote sensing have allowed constant monitoring of ecosystems, mainly because of the availability of improved spatial, time and spectral data. The present thesis investigates the time and spatial vegetation patterns and the precipitation variability in Sahel through interrelations based on the analysis of temporary series from the normalized vegetation difference index (NDVI) for the period 1983-2012 and rainfall satellite estimations. Whereas the precipitation emerges as a dominant cause from the increase of the vegetation, there is evidence of other anthropogenic factors overlapping the climatologic trend. Measuring rainfall in the semi-arid regions continues to be a pivoting point because of great spatial and weather variability. On top of this, note that the number of available weather stations in Sahel is quite limited. Satellite data, along with meteorological stations measurements, are currently the best source of information with regards to spatial-temporary precipitation in Sahel. Three temporary series of precipitation have been analyzed to cover the period between 1983 and 2012: GPCC, ARC2 and TAMSAT. These series have been compared with more recent ones to provide better quality estimations: RFE2, TRMM 3B42 and GPCP-1dd. The result shows a good correlation between series GPCC, ARC2, RFE2 and TRMM 3B42. The analysis of 30 years of precipitation data shows a recovering of the annual precipitation, although the trends in the annual temporary distribution have changed: nowadays it rains more but during less time. There are NDVI data available coming from the AVHRR sensor since 1981, as well as from its newer GIMMS3g version. Nevertheless, there are aspects in the design and processing of the sensor data that result in noise. There are data sets of the most recent NDVI coming from the sensors VEGETATION and MODIS installed in the SPOT and Terra satellites respectively, which are considered to be an improvement on the AVHRR GIMMS3g data. These sensors provide the possibility of evaluating the accuracy of the tendency analysis of the series of time of the AVHRR GIMMS3g in the period of time in which data overlap, from 2001 to 2012. This evaluation shows that the AVHRR GIMMS3g data set is correct in temporary studies of the vegetation in areas that do not record rainfalls greater than 1,000 mm, whereas the temporary tendencies in more humid zones should be interpreted with reservations.
