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

Accueil du site → Master → Pays Bas → Environmental reconstruction of Mara River in Kenya and Tanzania : using remote sensing images and GIS

UNESCO-IHE Institute for Water Education, Delft (2013)

Environmental reconstruction of Mara River in Kenya and Tanzania : using remote sensing images and GIS

Gbadebo A. O.

Titre : Environmental reconstruction of Mara River in Kenya and Tanzania : using remote sensing images and GIS

Auteur : Gbadebo A. O.

Etablissement de soutenance : UNESCO-IHE Institute for Water Education, Delft

Grade : Master of Science (MS) 2013

Résumé
The Mara River is a trans-boundary River that cuts across Kenya and Tanzania and ends in Tanzania at Lake Victoria. It is an important River because ecosystem in Massai Mara and Serengeti and inhabitants of Mara depend on it for survival. Mara River is used for large scale and small scale irrigation farming in Kenya and Tanzania. It covers an area of 13,504km2and is 935km long. For the past 40 years, land cover in Mara watershed has gone through changes as a result of natural and anthropogenic activities in the area. The principle of GIS and remote sensing was adopted to study change in Land cover that occurred in the Mara watershed from 1973 to 2012. Multispectral image analysis and GIS were used to analyze satellite images. Image pre-processing (image enhancement), image classification and Image post -processing was done to identify change in land cover in Mara watershed. First, unsupervised classification was done to determine number of classes used for the land cover thematic classes, followed by supervised classification using Maximum Likelihood Classifier. Post classification by change detection was done for 10 images used to identify the thematic land cover classes that have changed from 1973 to 2012 and confusion matrix was done only for image of L7 ETM+ Sensor acquired on January 27, 2000 with Africover standard image to determine the accuracy of the classified image. The accuracy of the classified image used in this study was 94%. The hydrograph of Amala, Nyangore and Lalgorian Bridge located in the upper part of Mara River watershed was plotted against year (1955 - 2007). The effect of land cover change was seen on the average increase in hydrograph that occurred from 1983 to 2007. The results of change detection indicate that the classes of thematic land cover used in this analysis have changed and this was reflected on the Amala, Nyangore and Lalgorian bridge hydrograph (1955 - 2007). The land cover classes of dense forest, close shrubland, open shrubland (grassland) and water have reduced by 47%, 50%, 40%, and 60% respectively. The classes of sparse forest, agriculture, wetland, bare soil/urban area and burnt area have increased by 49%, 100%, 100%, 18% and 15% respectively. The degree of sinuosity ranges from 3.5 to 4.5 in the watershed.

Sujets  : watersheds ; remote sensing ; GIS ; Kenya ; Tanzania

Présentation

Page publiée le 11 janvier 2017, mise à jour le 13 octobre 2018