ASSESSMENT OF DROUGHT RISK OVER THE SUDAN -SAHEL REGION OF NIGERIA

Abstract

This study assesses the drought risk over the Sudano-Sahelian region of Nigeria. Remote Sensing and Geographic Information Systems techniques were engaged to examine the relationship between rainfall and Normalized Difference Vegetation Index (NDVI). It was also used to establish the value of NDVI as a tool for assessing drought conditions. The relationship between rainfall and NDVI (surface reflectance from Pathfinder Advanced Very High Resolution Radiometer (AVHRR) satellite during 1981 to 2001 in the region was examined using spatial analysis methods. A strong positive correlation was found. However, the vegetation response, as reflected in NDVI is not exhibited until the following year. It was observed that the correlation is strongest during years of heavy rainfall, an indication that the relationship between rainfall and NDVI is not simple linear. A drought risk map that will be useful for decision makers that are involved in drought risk assessment was produced. It was revealed that the dynamics of vegetation is not static as it varies from year to year in accordance to changes in climate over the entire region, though it is dependent on the spatial location of the region. There is much decrease in vegetation around northern part of the study area especially in Borno. Drought risk maps were classified relative to the level of risk experienced in the region. Based on NDVI anomaly, 1987 and 1998 were wet years. The region experienced no risk except for some areas like Borno, Yobe, Sokoto and Kebbi which had slight to moderate drought risk. During dry years, the region experienced severe to extreme drought risk conditions. In 2002, Yobe, Sokoto and Borno experienced severe drought. There was also a transition from dry to wet year between 2002 and 2003. A strong correlation exists between Vegetation Condition Index (VCI) and NDVI (0.96) which also correlates with the annual rainfall. The threshold used for VCI is 54%. The threshold helped to indicate the dry and wet period on the Vegetation Condition Index maps. VCI drought maps depicts that 2002 was dry year with extreme drought condition. Wet period were observed in 2003 to 2005, in which no drought was experienced in the region. There is a strong linear relationship between NDVI anomaly and Standard Precipitation Index (SPI) values, especially in Kebbi and Yobe. SPI analysis indicated that, there was a long period of dryness in the 1980s when compared to the 1990s. This accounts for the high level of drought risk experienced in the 80s and low drought risk in 90s. The positive NDVI and SPI values derived across all the stations between 1992 to 1994 suggests that there were no drought risk in the region in those years. Analysis and interpretation of these maps which give different drought scenarios revealed that remotely-sensed drought indices can accurately detect and map the local and regional drought spatial occurrence. This study provided information that fine spatial resolution satellite data are useful aids to assist decision makers in monitoring and detecting drought. This will also guide the allocation of money in drought relief funds and drought impact mitigation schemes.