Assessing the Impacts of Vegetative Cover Change over Mau Water Towers on the Discharge of River Njoro, Kenya

Abstract

Mau water catchment and its environs is a very fragile ecosystem whose dynamics exhibits oscillations in magnitude caused mainly by human impacts and other climatic factors. The most recent oscillation was accompanied by excision of the forested catchment by the communities living around, leading to additional decrease in vegetative cover. The main purpose of this study was to determine the relationship between discharge and normalized difference vegetative index over the catchment of interest. Monthly Normalized Difference Vegetative Index (NDVI), discharge and rainfall datasets for the period 1982 and 2000 were used in this study. Time series of the NDVI, discharge and rainfall were then obtained. In order to determine the relationship between NDVI and discharge correlation analysis was done between the two variables. The relationship between NDVI and rainfall was also determined through correlation analysis. From the results obtained it was evident that discharge has been relatively constant over time except for a marked increase between 1996 and 1999. NDVI and rainfall had a constant trend throughout the study period. From correlation analysis it is evident that there is no statistically significant relationship between discharge and rainfall. This is explained by the fact that the clearance of vegetation has been compensated by growth of other vegetation types. NDVI only reflects the vigor of vegetation but does not distinguish between vegetation types e.g. tea from forest. NDVI and rainfall only shows a slight relationship when lagged. This is explained by the fact that the NDVI at a region only affects the rainfall forming processes later in the season, though on a very slight scale. In conclusion, the variability in discharge is thought to be dependent on other catchment parameters e.g. vegetation type, soil type and slope .Rainfall is completely dependent on other synoptic scale parameters e.g. air masses and other mesoscale forcings e.g. Lake Victoria circulation. It should be noted that a statistically significant relationship could be attained only with the use of very high resolution NDVI.