Desertification is reported to be intensifying and spreading in Kenya dry lands, threatening millions of inhabitants and severely reducing productivity of the land. Concern over desertification acceleration status in the country has been raised and measures to address the problem called upon. Among these measures is assessment of desertification using available data and technological tools. Vegetation cover was used as a land degradation indicator to determine land degradation and rate of change using spectral change detection technique based on pixel-wise operation. In combination with ancillary data, vegetation degradation occurrence and areas at risk of desertification were assessed. The study area is located in Northwestern Kenya, one of the dry land areas. Multi-spectral and multi-temporal analysis was applied to NOAA/AVHRR 1km and Landsat TM/ETM 30 meter resolution for periods covering wet and dry season of 1986 to 2001. Normalized Difference Vegetation Index (NDVI) and Land Surface Temperature (LST) were used to detect change. The results show desertification is apparent and increased vegetation degradation. Arid areas were found to be increasingly degraded and at high risk of further degradation at a rate of 1.8% per year.