Estimates of the impacts of invasive alien plants on water flows in South Africa

Estimates of the impacts of invasive alien plants on water flows in South Africa
David Le Maitre, Greg Forsyth, Sebinasi Dzikiti and Mark Gush
Natural Resources and the Environment, CSIR, P O Box 320, Stellenbosch

This paper summarises the results of a new assessment of the impacts of invasive alien plants on the water flows in South Africa based on the mapping done by Kotzé et al. (2010). The approach took into account basic hydrological principles and factors which limit plant water-use and the accumulated information on water-use by these species in both plantation settings and those typically invaded since the publication of the previous national assessment in 1998. The unit area reduction of 97 mm/yr found in this assessment is about half the 190 mm/yr reported by Versfeld et al. (1998). The total reduction in mean annual runoff is about 1 444 million m3/yr or 2.9% of the naturalised mean annual runoff (see Table A below) which is much lower than the 3 300 million m3/yr estimated in 1998. The main reason for this is the lower flow reduction, but the decrease in the condensed are from 1.76 million ha in 1998 to 1.50 million ha in this study also reduced the impacts. The greatest reductions were recorded in primary catchment T (former Ciskei & Transkei,

322 million m3/yr, 4.5%), followed by U (southern KwaZulu-Natal, 154 million m3/yr, 5.0%) and W

(Northern KZN, 149 million m3/yr, 2.3%). The greatest proportional reductions were in K (Mossel

Bay-Tsitsikamma, 8.4%),  M  (Port  Elizabeth  coast-Coega, 6.5%),  H  (Breede,  6.1%)  and  G  (Berg- Agulhas, 6.0%). The taxon with the greatest impacts was the wattles (Acacia mearnsii, A. dealbata, A. decurrens) which account for 33.5% of the total reductions, followed by Pinus species (18.9%) and Eucalyptus species (15.1%). The unit area flow reductions due to pines were 212.1 mm/yr, followed by Hakea species (199.5 mm/yr) largely because they occur mainly in high yielding montane fynbos catchments.

This estimate is considered conservative for the following reasons. The approach we used probably underestimates the extent of riparian invasions within the mapped areas. The mapping was limited to catchments which cover only about 54% of the country (but do yield more than 80% of the MAR). There are also extensive riparian invasions by eucalypts, wattles and other species along perennial rivers in semi-arid and arid environments (e.g. middle and lower Orange R, lower Vaal River) whose impacts have not been included in this assessment.  In addition, Prosopis invasions in the Northern Cape result in a further reduction of about 8.94 million m3/yr. There are extensive invasions of Prosopis species in the North West, Free State and Western Cape provinces which have not been adequately mapped so their impacts cannot be quantified at the moment.