1. Mapping groundwater dependent ecosystem potential in a semi-arid environment using a remote sensing-based multiple-lines-of-evidence approach
- Author
-
Stephen G. Fildes, Tanya M. Doody, David Bruce, Ian F. Clark, Okke Batelaan, Fildes, Stephen G, Doody, Tanya M, Bruce, David, Clark, Ian F, and Batelaan, Okke
- Subjects
coefficient of variation fractional cover ,coefficient of variation NDVI ,wetness percentiles ,General Earth and Planetary Sciences ,barest Earth ,groundwater potential ,actual evapotranspiration ,Software ,analytic hierarchy process ,Computer Science Applications - Abstract
Refereed/Peer-reviewed Groundwater dependent ecosystems (GDEs) are vulnerable to groundwater regime changes. However, their protection is often hampered by challenges in their identification. Within is presented a remote sensing-based GDE potential mapping approach based on the persistency of relevant vegetation parameters during prolonged dry periods as an indicator of potential ‘consistency’ of water supply (e.g. groundwater). The study uses a novel approach to characterising persistency for selected vegetation parameters based on a normalised difference measure and an adaptation of the coefficient of variation statistic. Aggregation of parameters was facilitated through the analytic hierarchy process providing a structured weighting approach to minimise parameter bias. The approach is demonstrated in the semi-arid Flinders Ranges of South Australia where new groundwater resources are being sought to support local domestic and industry needs. Variations in GDE potential were mapped to better target areas where exploration of groundwater should be avoided. Mapping results indicated a high-level of agreement of 77% with an independent springs dataset, along with an 87% agreement with areas coinciding with known phreatophyte species and depths to groundwater. The index-based mapping approach has potential applicability across landscapes, as it normalises for variations in vegetation cover, minimises technical overheads, and employs continental-wide remote sensing data-products.
- Published
- 2023
- Full Text
- View/download PDF