Lake Sibayi variations in response to climate variability in northern KwaZulu-Natal, South Africa

Abstract

The optimal management of natural resources like lakes requires understanding the relationship with other environmental elements. Remote sensing techniques using multi-temporal and multi-sensor images for change detection purposes are important in this regard. This study used a combination of approaches to detect changes in the lake surface area of Lake Sibayi in relation to changes in past climates. Delineation of the study area is achieved using WR2012 endoreic maps and Landsat satellite images from 1992 to 2016. Using data from eight meteorological stations, past climate of the lake catchment was investigated. Thereafter, a multivariate correlation analysis is conducted to examine the relationship between the changes in the lake surface area and the changes in climatic (precipitation, temperature) variables and water levels. Results suggest that the lake surface area has decreased by 20% since 1992. There are significant changes in temperatures, while the annual rainfall totals declined significantly. The correlation between precipitation in the catchment and annual water level changes is 0.88. Statistically, significant increases in the water level and precipitation were experienced in 1993 and 2001. SPI analysis reveals that the study area is getting drier and the probability of recurrence of moderate dryness is 10%. The rate at which the lake is shrinking is not solely climatic related, as anthropogenic aspects are also responsible. To prevent further shrinkage of Lake Sibayi, it will be necessary to develop aggressive restoration policies and action plans aimed at maintaining inflows in the face of compounding climate change and water demand. Recommendations of the nature of further studies that can increase our understanding are included.