A framework for coupling land use and hydrological modelling for management of ecosystem services

It is well known that land-use changes influence the hydrological cycle and that those changes in the hydrological cycle influence land use. The sophisticated spatial dynamic planning tools that have been developed in the last decades to support policy makers in the decision making process do not take into account the mutual feedbacks between land use and hydrology. In this study a framework for an integrated spatial decision support system is presented where the feedbacks between land use and hydrology are taken into account by coupling the SITE (Simulation of Terrestrial Environments) land-use model to the SWIM hydrological model. This framework enables policy makers to assess the impact of their planning scenarios on ecosystem services using a web-based tool that interactively presents trends in space and time of spatial indicators derived from both models. This approach is tested for the uThukela area, which is located along the northern areas of the Drakensberg Mountains which form the border between Lesotho and South Africa. The region is extremely important for its catchment-services as water derived from it is pumped into the Vaal River supplying water to the city of Johannesburg. Because of poor management of ecosystem services, less water is produced by the catchment more erratically, siltation levels are increasing and less carbon is retained in the soil. Biodiversity is threatened by grazing livestock, alien plants and other poor land management practices. In addition, overstocking, frequent burning and lack of soil protection measures have caused rill and gully erosion in areas of communal ownership where an overall management policy is lacking. The presented framework for a spatial integrated decision support system is currently being implemented and will be used by policy makers to assess policies developed for an Environmental Management Framework (EMF). Scenarios will be defined during stakeholder workshops. A prototype of the decision support system has been developed, but not all data necessary for modelling and calibration is yet available. From the analysis of land-use maps of 2005 and 2008 it was observed that forest and bush decreased, while settlements, subsistence farming, commercial farming and grassland increased.