Detecting changes in a wetland : using multi-spectral and temporal Landsat in the Upper Noun Valley Drainage Basin-Cameroon

Graduation date: 2009Since the early 1980s, satellite imagery has been commonly used to detect change\ud in wetlands. The overall objective of this study is to utilize remote sensing and\ud Geographic Information Systems (GIS) technology to determine the extent of change of\ud the wetland area and of other land use and land cover classes in the Upper Noun drainage\ud basin from 1973 to 2007. Specific objectives were 1) to map land cover and related land\ud use practices within the wetland and surrounding areas of the drainage basin using the\ud supervised maximum likelihood algorithm classification method; 2) to map land use and\ud land cover classes by seasons in order to determine human-induced pressure on the\ud wetland area and the drainage basin using available Landsat MSS, TM and ETM+\ud imagery; 3) to utilize GIS in confirming and finalizing the land use and land cover classes\ud and delimitation of the Upper Noun drainage basin by incorporating auxiliary data such\ud as settlements and hydrology. The Upper Noun drainage basin is an important wetland\ud that lies within the western high plateau of Cameroon where it supports a wide range of\ud wildlife. The study defines twelve different land cover and land use classes that are\ud grouped into four major categories. The first category is the humid floodplain class that\ud makes up the wetland area, including the floodplain lake (1973), the reservoir, the\ud permanent and seasonally flooded prairies, and irrigated farmlands. The second category\ud consists of the agropastoral classes which are made of upland grazing areas and mixed\ud farming areas. The third category is the montane and semi montane forest zone at\ud elevations above 1,500 - 3,000 m. The fourth category is comprised of major settlements,\ud and settlements with enclosures and openfields.\ud This study is an example of an application of spatio-temporal data in assessing the\ud close relationship existing between humans and their environment. This is a Sub Saharan\ud African community that has a long history of direct dependence on the available\ud resources for its daily livelihood. A detailed physical and historical background is\ud presented to provide a context for spatio-temporal variation discovered through change\ud detection using Landsat imagery without ground truth data. The data analysis reveals\ud considerable change within the Upper Noun drainage basin from 1978 to 2002. Within\ud the wetland area in the floodplain, the reservoir shows evidence of large fluctuations in\ud area since the construction of the Bamendjin dam in 1975. Within the reservoir area, an\ud acute siltation has been observed since 1988 and is increasing in area. A significant drop\ud in area of permanent and seasonally flooded prairies was observed. Irrigated farmland\ud areas also show downward trends from 1988 to 2002. Concerning the agropastoral\ud landscape, the upland grazing areas showed a general drop in area, while the mixed\ud farming area increased from 1978 to 2002. The montane forest also decreased in area;\ud however, it appeared to have recovered slightly in 2002 following the successful\ud implementation of the Kilum/Ijim community forest management project. Settlement also\ud showed some expansion in area as a result of growth of most enclosures and openfields to\ud larger villages and major settlements. This study thus provides base data for monitoring\ud human impacts on the Upper Noun drainage basin and its natural habitats, especially\ud within and around the wetland