Grassroots September 2016, Vol. 16, No. 3 | Page 23

Feature The role of remotely piloted aircrafts (RPA’s) in rangeland monitoring Christiaan Harmse and Hannes Gerber Northern Cape Department of Agriculture, Land Reform and Rural Development (Research Unit) [email protected] / [email protected] One of the primary objectives of rangeland monitoring is to determine if grazing management strategies meet the goals for sustainable resource utilization and to prevent rangeland degradation from taking place (James et al., 2003). However, monitoring is an ongoing process nneeded to determine whether progress is being made towards the achievement of objectives on given variables such as prolonged droughts, veld fires, grazing pressures, land-use use and climate change. Ground surveys can be time consuming, labour intensive, logistically tically challenging in remote areas and, therefore, very expensive (Gardner et al., 2008). Depending on the scale, the objectivity and representativeness of ground based assessments can at times be questioned since surveys cannot be conducted at the frequency ncy required for proper analysis and monitoring of trends due to high cost implications and inaccessibility. Therefore, a significant weakness in ecological research and rangeland monitoring in southern Africa’s rangelands remains its inability to easily ttrack changes in vegetation structure and composition resulting from management strategies, land land-use and climate change (Koh&Wich, 2012). This can further be attributed to the lack of access to high resolution geospatial data sets (Mishra et al., 2015). Therefore, refore, rangeland monitoring is often neglected and the question must be asked, whereto with rangeland monitoring in the 21st centur century? Grassroots Figure 1: High quality geo-referenced geo image captured by a RPA (a) with a clear distinction possible between individual grass tufts compared to (b) a lower quality satellite image with only larger woody plants barely visible as darker patches. Ground resolutions and sensor capabilities of satellites have improved immensely within recent years (Whitehead et al., 2014). While Whil remote sensing technologies are evolving at a rapid rate, most researchers currently rely on satellite based remote sensing for mapping and monitoring (Broich et al., 2011). Even though certain low resolution satellite images are freely available (e.g. Landsat andsat and MODIS), other sub-meter sub resolution images (such as IKONOS, Worldview and Quickbird) can be extremely costly and may not always offer timely high resolution images to fulfil the objectives required (Koh&Wich, 2012). Remote sensing images from satellites sat are also often obscured by cloud cover, preventing any realreal time monitoring (Hansen et al., 2008; Koh&Wich, 2012). Therefore, only a few studies utilizing high resolution remote sensing data, such as Wessels et al. (2007) and Wessels et al. (2012) have been undertaken in southern Africa rangelands. September 2016 Vol 16 No. 3