Remote sensing-aid assessment of wetlands in central Malawi

Abstract

<p>Many areas in Malawi undergo extreme seasonality: floods in the wet season and drought in the dry season. Each year, this extreme seasonality poses formidable challenges for local farmers to sustain their crops. Often in the dry season, farmers use the water in the surrounding seasonal wetlands (dambos) for small-scale irrigation to supplement their rainy season harvest. In Malawi, the agricultural use of wetland is growing year by year and these areas play significant roles in regulating food price shocks and price. Such intensive use of wetlands can negatively affect the sustainability of wetland eco-system and their crop production, with communities even affected by the drying up of wells. Farmers, especially small-scale farmers, will face even more challenges for sustaining wetland production, as climate changes cause more frequent occurrence of droughts as Malawi has experienced in recent years. With the increasingly intensive use of these seasonal wetlands for agricultural purpose and the expansion of wetland degradation generally across the country, more attention is required toward effective management of these wetlands through identification, mapping, monitoring and data analysis. To achieve the sustainable use of these seasonal wetlands, it is essential to establish systematic monitoring and assessment procedures. Widely used assessment protocols (i.e., WET-Health) which evaluate the wetlands based on physical indicators such as land cover, hydrology, geomorphology, soil organic matter and natural vegetation have been successfully implemented in South Africa. However, obtaining those indicators across the full length of an individual wetland, let alone all wetlands in one district in Malawi, is labour intensive and time-consuming and difficult to complete. In this research, we utilise both unmanned aerial vehicle (UAV) and satellite imageries. These data sources are being tested in nine different seasonal wetlands in central Malawi to provide an accurate derivation of key indicators such as gully formation, sedimentation, water extent, changes in land use and natural vegetation. Additionally, using satellite imageries and GIS, the condition of each individual wetland has been quantified, with land cover and the extent of inundation determined through multi-temporal data analysis. Our results can be applied across a larger area, i.e. several districts to help identify where more detailed ground assessment is needed and technical support required to improve wetland management, feeding into both policy and technical guidance which can help sustain the range of ecosystems services of these important areas.</p>