Assessment of Deforestation in the Arsi Region of Ethiopia: A Remote Sensing and GIS ApproachEPA Grant Number: U916252
Title: Assessment of Deforestation in the Arsi Region of Ethiopia: A Remote Sensing and GIS Approach
Investigators: Carrell, Aziz K.
Institution: North Carolina Central University
EPA Project Officer: Michaud, Jayne
Project Period: January 1, 2003 through January 1, 2005
Project Amount: $123,152
RFA: Minority Academic Institutions (MAI) Fellowships for Graduate Environmental Study (2003) RFA Text | Recipients Lists
Research Category: Academic Fellowships , Fellowship - Natural and Life Sciences , Biology/Life Sciences
The objective of this research project is to analyze two forms of satellite imagery during a 15-year period to attain an accurate spatiotemporal analysis of deforestation and an inventory of remaining natural forest in Ethiopia, East Africa. It is hypothesized that no significant reduction in the use of forest resources is evident.
Data acquisition for this project will begin with Advanced Very High Resolution Radiometer (AVHRR) satellite data obtained from the National Oceanic and Atmospheric Administration for the period of 1985 to 2000. Supplemental Landsat Thematic Mapper (TM) satellite data needed for this analysis will be obtained from the United States Geological Survey Earth Resources Observation Systems (EROS) Data Center for the corresponding study period. Ground truthing of land cover types then will be performed for the purpose of spatial analysis, and a meeting will be held with local Ethiopian officials to attain primary information and materials supporting the research objective. A temporal land-use analysis of the region using remotely sensed data will be performed using a supervised classification scheme. Change-detection algorithms will be performed on the Landsat™ satellite data to ascertain the percent change of forest cover over a 15-year period. An accuracy determination of the classification scheme will be performed to ensure the viability of the classification results. The assessment will be performed to evaluate the level of forest recovery generated by local initiatives for the study region. The analysis of both Landsat (TM) and AVHRR data will give a combined level of precision spatially and temporally, thereby laying the groundwork for accurate data and better assessment of the deforestation problem. This research project seeks to generate a land cover change map for the period between 1985 and 2000. Change detection identified during the period will provide much needed insight into the dynamics of vegetable growth for the research area. This information then will serve to support the ongoing efforts to curtail the deforestation process by providing local officials with accurate and detailed maps needed to make informed decisions. The maps generated from this study will include: (1) Deforestation Versus Transportation Network, (2) Deforestation Versus Population, (3) Deforestation Versus Elevation, (4) Land Cover Change Map, and (5) Forest Inventory Map.