Multi-Scale detection, mapping, and modelling geomorphic change in gravel-bed rivers with UAV remote sensing

Scott, R

Abstract

Fluvial science is in particular need of surveying tools which can rapidly and accurately capture topographic data. The use of low-cost, consumer grade UAV (unmanned aerial vehicle) systems and Structure from Motion (SfM) processing methods has seen successful adoption by many other earth surface processes sub-fields however their use for monitoring within the field of fluvial geomorphology remains limited. This study tests the applicability of UAV photogrammetry to fluvial surveying, capturing centimetric resolution data across kilometric scales, providing an ideal perspective for geomorphic process interpretation. For a historically modified UK case study, four series of very high resolution DEMs (digital elevation model) and orthomosaic imagery are produced for a 2km reach of quasi-wandering gravel-bed river. Comparative analyses of DEMs between 2016 and 2018 reveals widening of the incised margin and significant geomorphic evolution characteristic of re-naturalization following the termination of gravel mining, channelization, and resultant aggressive incision. Whole reach volumetric analysis reveals a negative sediment budget approximating a net loss of 250m3/year. Budgetary segregation shows 22% of eroded material is sourced from the banks of the inset wandering margin and is a possible cause of a general fining (30% reduction in mean b axis) of bed material within active channels, detectable by grain scale analysis of high-resolution orthomosaic imagery. Vertical scour is seen to be prevented, even under extreme flows (~100 m3/s-1), by a bed armouring effect which is sustained by liberation of coarse clasts from the floodplain via lateral erosion and bank collapse. Woody debris dynamics, gravel bar creation and migration are intricately modelled throughout the site, their presence seen to be affecting flow-prioritization of sub-channels inside the incised margin. UAV surveying workflows and processing protocols are also developed for fluvial science: A means to neutralize and filter out surface error caused by vegetation occlusion in the SfM workflow, and a method to correct for geo-referencing error in large DEMs. Geomorphic findings at this UK case study hold valuable and transferable insights to river re-naturalization in the context of gravel extraction and channelization.

Citation

Scott, R. Multi-Scale detection, mapping, and modelling geomorphic change in gravel-bed rivers with UAV remote sensing. (Thesis). University of Salford

• Thesis Type: Thesis
• Deposit Date: Apr 8, 2020
• Publicly Available Date: Apr 8, 2020
• Award Date: Jan 1, 2019