Mahmoud, Ahmed Mutasim Abdalla
(2022)
Monitoring Sand Dune Movement using Remote Sensing.
PhD thesis, University of Nottingham.
Abstract
Deserts are arid and semi-arid regions with low rainfall and sparse vegetation, which makes them suitable hosts for the formation of sand dunes. Deserts cover an area of more than one-fifth of the Earth’s total land cover. Despite the fact that some of these desert dunes have been monitored for many years, the size and volume of sand dunes as a whole are not adequately monitored and updated, due to both the difficulty of using traditional methods of measurement and the continuous movement of the sand dunes. Many attempts have been made to quantify desert areas around the world, but quantifying desert dunes requires a thorough investigation that considers the different sand dunes’ behaviour, as well as the impact of influencing factors such as wind, vegetation, topography, and sand supply on the dune movement. This must also involve analysing the dune activity both horizontally and vertically.
The movement of sand dunes is considered one of the major environmental issues in arid and semi-arid regions, that threatens livelihoods and rural communities by causing them to be submerged in sand. It also contributes to the degradation of land, causing poverty and food insecurity. Sand movement can be experienced on different scales: individual dune movement, dune field changes or in the form of dust storms. The key to mitigating these risks is to understand the movement of the dunes.
Sand dunes can also be found in coastal areas, where sediments are carried into the shores by sea tides and winds. Coastal dunes play an important role in coastal erosion risk management, where they act as a dynamic natural sea defence. In addition, they provide habitats that enrich coastal biodiversity and add resilience to the ecosystem. The world’s sandy beaches are undergoing significant changes, with 24% eroding and 28% accumulating material, while the remaining 48% are stable.
Therefore, a comprehensive understanding of dune activity is urgently needed. This requires more accurate measurement techniques that match the frequency of observation to the rapid dynamic movement of the sand dunes.
There are two main sets of techniques for monitoring sand dunes: conventional techniques, such as sand traps, Global Navigation Satellite System (GNSS) and Terrestrial Laser scanner (TLS); and remote sensing techniques such as optical, Synthetic Aperture Radar (SAR) and Airborne Light Detection and Ranging (LiDAR). The conventional techniques provide vital information about sand dunes, such as the sand particle size, which can be detected by sand traps, in addition to highly accurate ground truth data collected by GNSS, total station and levels that can be used for validating the monitoring results detected by the remote sensing techniques.
However, these conventional techniques have significant limitations related to time consuming data collection processes and the complexity of monitoring large, inaccessible sand areas. Additionally, they only provide an approximation of the dune movement, as the data are collected for discrete dune locations rather than the dune field as a whole. Moreover, with few repeated surveys, conventional techniques are limited to detecting the movement that occurs within days or hours due to the rapid movement of the sand dunes. This can be overcome by applying time series analysis using remote sensing techniques, providing continuous observations of the dunes over periods of years that are contiguous over very large areas.
Therefore, the aim of this research is to investigate the capabilities of various novel remote sensing techniques (i.e. optical multi-spectral satellite sensors, SAR techniques, airborne LiDAR) for detecting and monitoring sand dune movement and its impact on urban areas, crop fields, forests, water bodies and archaeological sites, to determine the most vulnerable areas to sand dune movement. As part of this study, the impact of the influencing factors that control the movement of the dunes, such as wind speed/direction, vegetation, topography and sand supply, is also considered. This aim is fulfilled by four objectives: (1) develop an automated framework that uses Google Earth Engine and machine learning classifiers applied on multi-temporal satellite images to detect the areal changes in sand dunes, in addition to computing the displacement and direction of movement for individual sand dunes; (2) investigate the capabilities of the SAR Offset Tracking technique for detecting horizontal sand dune movement; (3) investigate the use of multi-temporal Airborne LiDAR DTMs for monitoring the dynamic activity of the coastal sand dunes; (4) investigating the capabilities of Differential Interferometric Synthetic Aperture Radar (DInSAR) for detecting the vertical deformation of sand dunes. This was carried out in two study areas: a desert sand dune area in Northern Sudan, and a coastal sandy beach near Formby in the Northwest of England.
This research highlighted the capabilities of the novel remote sensing techniques in addition to defining the limitations of using the more traditional land surveying techniques for monitoring sand dunes. Moreover, it has been found that detecting the deformation of individual dunes could be provided from moderate spatial resolution images, but the higher resolution of the images, the better the footprints of the individual dunes. Additionally, using digital terrain models time series data demonstrated high capability in monitoring sand dunes both horizontally and vertically, providing rates of horizontal and vertical dune motion in addition to the volumetric changes of the dunes.
Based on that analysis, a strategy has been developed for monitoring sand dune movement, that consists of three main implementation stages: (1) the detection of sand dunes using different surveying techniques, measuring the changes and movement of the dunes; (2) a monitoring stage, where time series analysis is applied to distinguish patterns in the dune movement, in addition to identifying the impact of the sand movement influencing factors (i.e. wind, vegetation, topography, …etc.) and its relationship to the sand movement behaviour; and (3) a prediction stage of sand movement, based on previously detected dune behaviour and the influencing factors from the monitoring results. This strategy could have wide applicability and could also be modified to study other environmental challenges, such as glaciers.
| Item Type: |
Thesis (University of Nottingham only)
(PhD)
|
| Supervisors: |
Marsh, Stuart
Psimoulis, Panos
Novellino, Alessandro
Hussain, Ekbal |
| Keywords: |
Sand Dune Movement, Sand Monitoring, Remote Sensing, Sand Monitoring Techniques, SAR Pixel Offset, Airborne LiDAR, Optical Imagery |
| Subjects: |
G Geography. Anthropology. Recreation > G Geography (General)
G Geography. Anthropology. Recreation > GB Physical geography
T Technology > TA Engineering (General). Civil engineering (General) > TA 501 Surveying |
| Faculties/Schools: |
UK Campuses > Faculty of Engineering > Department of Civil Engineering |
| Item ID: |
71310 |
| Depositing User: |
Mahmoud, Ahmed
|
| Date Deposited: |
06 Jun 2024 11:15 |
| Last Modified: |
06 Jun 2024 11:15 |
| URI: |
https://eprints.nottingham.ac.uk/id/eprint/71310 |
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
