Siltation Mitigation In A Data Poor Environment

Abstract

The typically mild slopes of the muddy coastline in the south of Myanmar, enable the tide to propagate deep into the mainland, carrying muddy and saline water. The Nga Moe Yeik Creek (a tributary of the Yangon River) is closed off at the upstream end during the dry season and is subject to high siltation rates. Extensive dredging is necessary to ensure navigability. However, dredging activities stopped in 2015 because dredging appeared to be insufficient.

Flushing the creek, by opening the Nga Moe Yeik Creek Sluice Gates more often could contribute to a solution to minimise the siltation rates, leading to lowered dredging volumes and an improved navigability of the creek. However, it may decrease the availability of irrigation and drinking water behind the sluice in the dry season. The responsible authorities are looking for an optimal solution. The current insight in the siltation problem is limited due to a lack of measured data collected in the Nga Moe Yeik Creek. The general consensus is that siltation rates can be decreased by an improved operating scheme of the Nga Moe Yeik Sluice Gates.

Since data resources are scarce or not available at all, data on bathymetry, water levels and suspended sediment concentration is collected from the field between May and July 2017. The observed water levels show a flood-dominant tidal wave with a tidal range in the order of 4 - 6 meters. Depths are measured in a zigzag pattern in the creek stretch between mouth and sluice. Suspended sediment concentrations are measured between 0.3 and 1 gram/L. Tidally and depth averaged concentrations in the order of 0.5 gram/L and 0.4 gram/L are observed at respectively 2 and 17 km from the mouth of the creek.

Part of this data is used to set up a Delft3D-flow model of the system that simulates the main sediment transport processes in the Nga Moe Yeik Creek. The initial hydrodynamic model is calibrated against observed water levels. In the next phase, the sediment transport model is used in a qualitative analysis to determine best suitable sediment parameters. Using a combination of model and collected data it is aimed to increase the insight in the siltation problem.

Both model and data showed similar hydrodynamic and sediment transport processes. The water level observations showed that the tidal wave deforms asymmetrically and becomes more flood-dominant as it propagates into the creek. A similar phenomenon is observed in the model results. Forced with a constant 2 gram/L the model generated tidally and depth averaged sediment concentrations in the same order of magnitude as the observations. From field observations and personal communication with local authorities, it is known that the siltation is most severe in the upstream end of the creek. This is also confirmed by the model results. In the scope of this work it is assumed that the model gives reliable results on the assessment of water levels and siltation.

The final model was used to assess several flushing scenarios (continuously or pulsatile flushing). Although all scenarios reduced thalweg siltation in the upper domain, pulsatile flushing reduced siltation slightly better. The scenario with 3 pulses gives best results followed by the scenario with 6 pulses. Less effective is the current flushing scheme with a continuous discharge. However to be able to give an quantitative answer on the extend of mitigation,more specific values of river discharge and critical bed shear stress are necessary.