Forestry and streamflow reductions in South Africa: a reference system for assessing extent and distribution

Abstract

Forestry is an important sector of industry in South Africa but the growing of timber places significant demands on the available water resources. Yet, a ready source of information on the extent, and probable hydrological impacts, of afforestation in specific locations in South Africa has not been available. This paper reports on the modelling exercise conducted to produce an easy-to-use, handy catalogue to fill this need, and presents some of the notable results of this exercise. Databases of quaternary catchment and magisterial district boundaries, rainfall, streamflow, forestry areas by tree genus, timber rotation lengths (years) and forestry growth potential were combined by means of a geographical information system to delineate uniform blocks of forestry. On these blocks a robust empirical model predicted total and low-flow reductions as a function of rotation, length, tree genus, water availability, growth potential and plantation age distribution. The results were summarised by quaternary, tertiary, secondary and primary catchment, and by magisterial district and province. The area of commercial timber plantations in South Africa is estimated at 1.5 million ha (57% pine, 35% eucalypts and 8% wattle), covering just 1.2% of South Africa. But the regions in which forestry is concentrated receive higher rainfall and yield a disproportionately large share of the streamflow, particularly low flow (dry-season flow). The commercial plantations are estimated to reduce mean annual streamflow by 3.2% (1417 x 10(6) m(3)) and low flows by 7.8% (101 x 10(6) m(3)/yr). Our estimate of an average reduction of 98.6 mm/yr per unit of planted area is 13% lower than the previous nation-wide estimate (113.6 mm/yr) of the net effect of forestry on total water resources (DWA, 1986). Mpumalanga Province with the highest concentration of forestry (7.2% of land area) experiences the largest reductions in flow - almost 10% of total flow and 18% of low flows. However, the largest relative impacts on low flow are seen in Northern Province where small areas of forestry are confined to humid upper catchments that are the principal source of dry-season flow in otherwise dry secondary catchments.