The hydrogeomorphic distribution of the wetlands in Swaziland and their prediction

Abstract

Swaziland became a signatory to the Ramsar Convention on the 15th of June, 2013. To date, the country does not have a national wetland inventory. This study aimed to apply the newly developed wetland probability mapping technique developed in South Africa to Swaziland, in order to provide potential baseline information on the distribution of wetlands across the country. Prior to this study commencing, there was little understanding of this wetland probability mapping technique. Results of this study show that when applying the mapping technique as it was applied in South Africa, other watercourses (rivers, drainage lines, and riparian zones) are more frequently mapped than true wetlands. Given that Swaziland currently uses the broad Ramsar definition of a wetland, the wetland probability map is well suited to identify wetlands falling under such definition. However, it does not suffice as a wetland map in countries such as South Africa that use a more specific definition for wetlands. In order to improve the initial wetland probability map, this study further made use of attribute data, obtained from 2000 randomly distributed points across the initial wetland probability map, to improve the latter through refining it to distinguish wetlands from other types of watercourses. It also classified areas of the map with the highest probability of being true wetlands into hydrogeomorphic units. This resulted in three types of wetland probability maps being produced. The initial wetland probability maps developed here can be used used to identify watercourses across Swaziland, which includes wetlands, drainage lines, riparian zones and rivers. The refined wetland probability map, which partially distinguishes wetlands from other types of watercourses, acknowledges the dynamic nature of wetlands and that the distinction between wetlands and other watercourses is not always exact. The refined wetland probability map also allows the government of Swaziland to locate watercourses with a highest probability of being wetlands. Furthermore, the classified wetland probability map provides the government of Swaziland with the baseline information needed to understand the relationship between organisms and the environment, as well as how the wetland is connected to the drainage network and how water moves through the landscape. The three wetland probability maps produced in this study also indicate that the location and distribution of the larger wetland systems across Swaziland are controlled by topography, soils, as well as the contact zones between different geologies.