Persistent Waterhole 200m-Segment data of the Roper catchment NT generated by the Roper River Water Resource Assessment

Persistent waterholes provide important refuge habitats. However, identifying persistent waterholes in remote sensing data can be underestimated if based on the assumption that waterholes remain in the same location through time. This dataset allows for the fact that a waterhole can vary in shape and local location through time, by providing the percentage of years that at least one water pixel is identified within along-stream segments at the end of each dry season. Water pixels, derived from Landsat imagery, have been identified along the stream channels during the driest period of the dry seasons from 1987 to 2018. Waterholes were mapped at the end of each dry season (based on available cloud-free Landsat scenes) using a Normalised Difference Water Index derived from Landsat imagery and is described in Sims et al. (2016). Further information is provided as a supported attachment. The Roper River Water Resource Assessment provides a comprehensive overview and integrated evaluation of the feasibility of aquaculture and agriculture development in the Roper catchment NT as well as the ecological, social and cultural (indigenous water values, rights and aspirations) impacts of development.\nLineage: End-of-dry-season waterholes were derived from Landsat data extracted from Digital Earth Australia using the method described in Sims et al. (2016). The end-of-dry-season waterhole images from 1987 to 2018 were used along with an in-channel mask containing a 500m buffer from the watercourse which was divided into 200m segments along each streamline. This was done in ArcGIS where point locations at 200m intervals were extracted from the river lines (XTools Pro-> Feature Conversions->Convert Features to Points, fixed interval = 200m). These points were used to create Theissen polygons (ArcTools->Analysis Tools->Proximity-> Create Theissen Polygons) which in turn were clipped to a 500m buffer either side of the river line (ArcTools->Analysis Tools->Proximity->Buffer, distance = 500m). The percentage of end-of-dry-seasons containing at least one pixel (30m x 30m) of water within each 200m segment was calculated.