A GIS based analysis of meandering channel evolution focussing on channel curvature, meander cut-off events and bank erosion rate interactions over a 125-year time period



Regan, Siôn
(2020) A GIS based analysis of meandering channel evolution focussing on channel curvature, meander cut-off events and bank erosion rate interactions over a 125-year time period. PhD thesis, University of Liverpool.

[img] Text
200825451_Sep2019.pdf - Unspecified

Download (22MB) | Preview

Abstract

The development and evolution of meandering channels has long held a fascination for scientists of many disciplines. Researchers have used field, archival and modelling sources to investigate the dynamics and long-term behaviour of meandering channels to help understand and predict future changes. There is still a great deal of uncertainty about why different meander bends are active and why similar bends are stable. Questions also remain about how the impact of change on one part of the river is propagated upstream and downstream. These questions are important as dynamic rivers can be challenging for human settlements, but also provide an opportunity to increase biodiversity and habitat opportunity in the riparian zone. The aims of this thesis are to explore the variability of meander migration in space and time for two actively meandering rivers in the UK. The long-term evolutionary behaviour of individual bends is explored using historic Ordnance Survey maps. Finally, a data modelling approach is used to predict channel change within a large dataset, with rivers from a third catchment, the River Ribble included, with a view to predicting future channel change. It is shown a high degree of variability exists both along channel reaches and between different periods. The mean annual migration rate for a reach varied from a minimum of 0.04ma-1 and 0.15ma-1 in the Lugg and Till catchments respectively to a maximum 1.29ma-1 and 0.92ma-1 for the catchments. The River Migration Toolbox is used to measure the migration rate of individual bends for each reach and rates of up to 3ma-1 were measured, which represent some of the most active rivers in the UK. There appears to be a strong relationship between the rate of migration and length of period between two map dates as the shortest time between the map dates tended to have the highest migration rates. The implications for management of this phenomenon is discussed. The evolutionary behaviour of meander bends is then investigated and the relationship between channel radius of curvature and migration rate quantified. It is shown that the migration rate tends to increase as the radius of curvature decreased, although there was a high amount of scatter present in the results. The trajectories of individual bends in a bend curvature-migration rate phase space were analysed to help understand the behaviour of different types of bends. The changes of the curvature profile of individual bends was measured and showed bends would develop a long section with a low radius of curvature, before becoming compound or double-headed. The final section of the research uses a machine learning approach to understand the importance of the different factors on the migration rate of individual bends and to predict channel change. The approach showed human infrastructure has an important control on migration rate, along with channel curvature and riparian vegetation. The machine learning approach was able to predict the location of active meandering channels but had a worse performance when predicting the rate of erosion. The potential future applications of the approach are discussed.

Item Type: Thesis (PhD)
Divisions: Faculty of Science and Engineering > School of Environmental Sciences
Depositing User: Symplectic Admin
Date Deposited: 14 Aug 2020 08:10
Last Modified: 21 May 2021 07:11
DOI: 10.17638/03089856
Supervisors:
URI: https://livrepository.liverpool.ac.uk/id/eprint/3089856