Improving the spatiotemporal availability of lower Mekong River water levels with multi-mission satellite altimetry | |
| Author | Llera, Maria Teresa Gerardo |
| Call Number | AIT Thesis no.WM-23-12 |
| Subject(s) | Satellite geodesy--Mekong River Watershed--Measurement Earth sciences--Mekong River Watershed Geographical information systems--Mekong River Watershed |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Water Engineering and Management |
| Publisher | Asian Institute of Technology |
| Abstract | River water level plays a vital role in numerous applications such as water resources management, flood forecasting, hydrologic modeling, and policymaking. It is therefore crucial to monitor water levels, particularly in transboundary rivers where multiple countries rely on this shared resource. The most common method of water level monitoring is the use of physical gauging stations, which poses challenges such as limited spatial coverage and high costs of installation, maintenance, and operation. Satellite altimetry presents a viable solution for enhancing the availability of water level data. It allows measurement of water levels at ungauged locations and addresses limitations associated with physical gauging stations. However, a significant challenge of satellite altimetry lies in its sparse spatiotemporal resolution of individual missions. Combining multiple satellite missions and performing spatiotemporal interpolation is therefore necessary to obtain water levels at any given day and location. To produce water levels from altimetry water levels in the Lower Mekong River, Long Short Term Memory (LSTM) and Ordinary Kriging models were developed to interpolate data temporally and spatially. Altimetry data was extracted from 27 virtual stations in the DAHITI database, and gaps in the time series were filled by fitting a harmonic model. For LSTM, two sets of regression variables were used – the first set consists of soil moisture and rainfall data, while the second consists of an additional variable which is the water level at the upstream virtual station. The results were validated by comparing predictions at gauging stations. The results show that the interpolated water levels generally exhibit good agreement with the actual water levels, although the peaks were missed. Higher deviation was found in the wet season than in the dry season. In the case of one gauging station, errors are higher due to the absence of nearby virtual stations. The comparison of results between the two sets, lower error values were found in the second set, which could be due to the better agreement of LSTM predicted water levels with the original gap-filled time series. |
| Year | 2023 |
| Type | Thesis |
| School | School of Engineering and Technology |
| Department | Department of Civil and Infrastucture Engineering (DCIE) |
| Academic Program/FoS | Water Engineering and Management (WM) |
| Chairperson(s) | Shanmugam, Mohana Sundaram |
| Examination Committee(s) | Shrestha, Sangam;Ho Huu Loc;Virdis, Salvatore G.P. |
| Scholarship Donor(s) | Asian Development Bank-Japan Scholarship Program (ADB-JSP) |
| Degree | Thesis (M. Sc.) - Asian Institute of Technology, 2023 |