A multi-resolution method to map and identify locations of future gully and channel incision

Abstract

While channel erosion is recognised as a major, often-dominant, source of river sediment, channel geometry and its change remain impractical to measure for anything but small experimental watersheds. Designing remediation strategies in landscapes affected by channel erosion requires information on the extent and location of current incised channel features, as well as a method to determine locations where incision may occur in the future. We present a multi-resolution algorithm that uses topographic information to concurrently map both existing incised landform elements and areas at risk of future incision. The former uses elevation, slope and profile curvature to identify topographic signatures of incised landform elements, and the latter uses landscape position, topographic wetness index and stream power index to isolate areas likely susceptible to future incision.

We aimed to develop a computationally efficient method capable of operating across a broad range of landscapes. The algorithm was tested in three contrasting environments in eastern Australia with promising results. Sensitivity analysis indicates the method performs reasonably consistently across landscapes, but that outputs become more sensitive as the average slope of the landscape increases. A comparison between cleared and uncleared hillsides suggested that areas indicated at risk of future incision are plausible, and that cleared areas were more susceptible to channel incision.

The only required input is a digital elevation model, and outputs can provide a rapid visual assessment of landscapes affected by incisional erosion. This technique enables the identification of gully erosion and the planning of remediation works across landscapes of thousands of square kilometres. It may assist in prioritisation of works and further insights into the processes associated with channel incision.