Using Geographic Information Systems and Remote Sensing for Sustainable Forest Resources Management in the Mai-Ndombe Region (Democratic Republic of the Congo)
Tarek Rashed, Ph.D.
Karen K. Kemp, Ph.D.
Mark Kumler, Ph.D.
This paper presents the methodology used to build a forest zoning map and to produce a vegetation map for the Mai-Ndombe District using Landsat Thematic Mapper (TM) and/or Enhanced Thematic Mapper (ETM+) images. Nine images with different cloud coverage covering a period from 1986 to 2001 were used. All the TM images used were radiometrically corrected and orthorectified, while ETM+ images required geometric corrections. Moreover, one ETM+ image required radiometric rectification to reduce variability of environmental conditions during the image acquisition. Some scenes had intense haze and thus could not be used for comparative studies. Principal component and scatterplot analyses were carried out for the nine images. The analyses revealed data redundancy between bands one and two; thus all the first bands were excluded from the analysis. Image classifications were performed on all the images separately using Isodata clustering and image segmentation techniques. The result of the classification shows that the study area is mostly covered by swampy forests (dense swampy forest, riparian palm forest, open swampy forest and swampy forest) with 29% of the total area, followed by semi-deciduous forests (single dominant and dense semi-deciduous) with a total 27.7%. The dense humid forest and grassland cover each 10% of the Mai-Ndombe region. Sensitive vegetation classes were extracted to produce various models for forest resources management and a zoning plan using spatial analytical techniques. This zoning map proposes priority allocation for various activities in the region, including local community agriculture land allocation, forest harvesting, and forest and natural resources conservation areas. The use of such models can help prevent conflicts in land utilization in the region.
Bwangoy-Bankanza, J. B. (2004). Using Geographic Information Systems and Remote Sensing for Sustainable Forest Resources Management in the Mai-Ndombe Region (Democratic Republic of the Congo) (Master's thesis, University of Redlands). Retrieved from https://inspire.redlands.edu/gis_gradproj/23