Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Doctorat → Allemagne → 2010 → Geospatial analysis to study environmental change Climate variability and vegetation cover dynamics in Ethiopia and the Horn of Africa

Freie Universität Berlin (2010)

Geospatial analysis to study environmental change Climate variability and vegetation cover dynamics in Ethiopia and the Horn of Africa

Beyene, Ephrem Gebremariam

Titre : Geospatial analysis to study environmental change Climate variability and vegetation cover dynamics in Ethiopia and the Horn of Africa

Räumliche Analysen zur Untersuchung umweltbedingter Veränderungen. Klimatische Schwankungen und Dynamiken in der Vegetationsbedeckung in Äthiopien und dem Horn von Afrika

Auteur : Beyene, Ephrem Gebremariam

Université de soutenance : Freie Universität Berlin

Grade : Doctoral thesis 2010

Rainfall variability and changes in long term amounts and distributions is a condition currently affecting many countries over recent decades. This situation controls the vegetation pattern significantly. In order to understand the existing variability and trends of the seasonal rainfall in east Africa at large and Ethiopia in specific, an in depth spatio temporal analyses is required. Spatial data on rainfall and vegetation cover can be used make quick assessments. Normalized Difference Vegetation Index (NDVI) maps have a good allow comparisons of the spatial and temporal variability in the amount and condition of vegetation. These can be used to assess agricultural production and trends in desertification in areas that rely on rainfed agriculture. Rainfall Estimates (RFE) can be used to asses trends in rainfall. To meet the Millennium Development Goals (MDGs) of eradicating extreme poverty and hunger, such tools can be used to enable proper planning. The main research questions are : Can the RFE data be used reliably to analyse seasonal rainfall variability ? How well NOAA satellite rainfall estimates (RFE) and NMA rainfall records are correlated ? How does the seasonal rainfall and vegetation cover trends and variability look like ? The study analysed monthly satellite rainfall estimates (RFE) from NOAA (National Atmospheric and Oceanic Administration) and monthly rainfall records (January 1996–December 2006) collected from weather stations by NMA (National Meteorological Agency of Ethiopia). The study further analysed seasonal trend of vegetation cover using Normalized Difference Vegetation Index (NDVI) from Advanced Very High Resolution Radiometer (AVHRR) between 1982 and 2008 (27 years). After doing spatio-temporal analyses of the satellite RFE and monthly weather stations rainfall records, it is found that the two datasets correlated well during the important rainy seasons ; summer and spring. Many weather stations have correlation values above 0.5 and 0.75. As a result RFE images are reliable enough to be used for timely spatio-temporal analyses of disasters in times characterized by late inception, dry spells and early cessation of rainfall. Therefore, the RFE images can be recommended for early warning systems in the country and to inform decision makers on the consequences of changes in the magnitude, timing, duration, and frequency of rainfall both temporally and spatially For vegetation cover trend the study period that is from 1982 to 2008, it was assessed in two time slices. The first part is between 1982 and 1995 and the 2nd part which is from 1995 to 2008. During the first half, the overall vegetation cover was in constant decline in most parts of east Africa with the exception of Sudan where relative stability was observed. During the second half, the overall vegetation trend was better than the first half. Though relatively stable, there are areas with fast decline particularly in western escarpment of the Ethiopian rift valley, southern Kenya and western Uganda. The trend showed higher decline of Vegetation during summer and spring seasons. There is an increasing deterioration of vegetation cover near and around Victoria region, most part of central and northern Tanzania, significant portion of southern Ethiopia and northern part of the rift-valley escarpment. The changes in the current rainfall with a rainfall amount forecasted using trend analyses were compared and it showed a decrease in the rainfall amount in most parts of east Africa particularly during winter and spring. For example, in more than 90% of the total area of Kenya, the rainfall amount in spring after 10 years of time is expected to decrease by 7 % or more. On the other hand, in southern Sudan autumn and spring rainfall amounts show an increasing trend that will increase by up to 10% particularly around the central southern Sudan

Mots clés : Vegetation cover Trend ; Climatic Variability ; Satellite and weather stations’ Rainfall Data ; , Monitoring food security


Version intégrale (7,7 Mb)

Page publiée le 17 octobre 2014, mise à jour le 28 décembre 2018