Analysis of Change Detection of Birnin-Kudu Land Cover Using Image Classification And Vegetation Indices

Abstract

The study utilizes Landsat-7 ETM+ based Normalized Differences Vegetation Index (NDVI) and Normalized Differences Water Index (NDWI) from 1972 to 2012 at the study area situated in Birnin Kudu, Jigawa state, in North-western Nigeria. The classified satellite data based NDVI of 1972, 1986, 2003 and 2012, including NDWI of 1986, 2003 and 2012 were used to determine land-cover change; vegetation and water body that have occurred in the study areas. This study attempts to use a comparative change detection analysis to produce the best way to quantify changes that has occurred in the study area with a lag time of 40 years (1972-2012) for NDVI and 26 years (1986-2012) for NDWI. The results of the classifications of NDVI and NDWI were displayed on satellite imagery, of which the percentage differences of change detected from variations of land cover/vegetation using NDVI of 1972-1986 is 15%, 1986-2003 is 40% and 2003-2012 is 11.6%. In the same vein, the result of percentage differences of change detected from variations of water bodies using NDWI of 1986-2003 is 0.03% and 2003-2012 is 1.5%. In the final analysis the change detected using NDVI for the period of 40 years (1972-2012) is 43.4%, while using NDWI for the periods of 26 years (1986-2012) is 1.47%. The study recommends periodic examination of land-use changes for determining various ecological and developmental consequences over time. The study area is of great environmental and economic importance having land cover rich in agricultural production and livestock grazing.

Keywords: Analysis, change detection, land-cover, image classification algorithms, NDVI, NDWI