In Zambia, forests are essential resources for rural people. However, approximately 300,000 ha per year of forest has been turned into cultivated area. Thus, sustainable natural resource management is strongly needed in this area. Satellite images is a promising tool to assess the land-use changes. Also, it is important to evaluate the changes in the satellite images by land surveys to establish the site-specific parameters to assess the actual land-use changes. In this study, satellite images from Landsat for 35 years were used in Kabwe, one of the major cities in Zambia. Especially, we focused on NDVI (Normalized Difference Vegetation Index), which is the index of vegetation activity to identify the distributions of land-surface states with field surveys. We visited Kabwe twice in January (rainy season) and August (dry season). Consequently, characteristic NDVI shifts could be associated with each surface type: forest, cropland, grassland, and wetland. Especially croplands made by deforestation showed 0.4 NDVI drop in the shift and it could be seen in after 2004 mainly. I also calculated deforestation area mathematically and visualized the area in image by using the gap. Each result will be useful to study land type distributions and how these have been being changed.