Spatio-temporal variations of aquatic weeds abundance and coverage in Lake Chivero, Zimbabwe

Information on the spatial distribution of aquatic weeds is required for understanding the evolution of weed invasion and propagation rates. Such information is also vital for identifying affected areas and relating weed abundance to probable changes in environmental conditions and human actions including management practices within the take and its catchment. Information on aquatic weed distribution also assists in evaluating the effectiveness of control measures and management actions. In Zimbabwe, Lake Chivero has been characterised by aquatic weed proliferation since the 1970s. Field surveys done between December 2005 and March 2006 showed concentrations of 1.2 mg/l and 0.3 mg/l up from 0.3 mg/l and 0.03 mg/l in 2001 for phosphates and nitrates respectively. Proliferation of aquatic weeds will continue unless nutrient loadings to this lake are reduced. The aim of this paper was to assess the feasibility of mapping the spatial extent and abundance of aquatic weeds in Lake Chivero, Zimbabwe using Landsat images. Landsat images of 1976, 1989 and 2000 were used to calculate the normalised difference vegetation index (NDVI) which was used for estimating the spatial extent of aquatic weeds and weed biomass. Field data and actual biomass measurements were obtained between December 2005 and March 2006 by harvesting weeds from the lake. This was subsequently related to NDVI and used to estimate the abundance of the different weed species. The results indicate that the weed coverage in Lake Chivero declined from 42% in 1976, 36% in 1989 to 22% in 2000. The research also demonstrated that Typha capensis has more biomass, 11 l kg per square metre, than any other weed type and hence higher abundance in all the years. It was concluded that remote sensing is an invaluable asset for detection of invasions, assessment of infestation levels, monitoring rate of spread, and determining the efficacy of weed mitigation measures. (c) 2008 Elsevier Ltd. All rights reserved.