Relative influences of different physical factors on salinity intrusion in coastal aquifer in Bangladesh

Abstract

Salinity intrusion in southwest coastal aquifers in Bangladesh is understood to be influenced by several physical processes, which operate over different spatial and temporal scales, with relative roles different from each other. These include lateral intrusion of seawater interface owing to sea level rise or pumping abstraction, saline river water interaction with shallow aquifer, storm surge flooding and tidal influences. This study was aimed at evaluating relative roles of these different processes on the dynamics of salinity intrusion and assessing how they may vary over different spatial and temporal scales considering future climate and sea level rise scenarios. The methodology included conceptualizing the hydrogeological system in the coastal aquifer in the greater southwest zone based on available hydrological and hydrogeological data, and setting up both regional and local (small) scale models for density-dependent flow and contaminant transport to simulate groundwater flow and salinity intrusion. In summary, the salt water interface reaches greater distances in the southwest zone and shorter distances in the southcentral zone. It encompasses most parts of Satkhira and Khulna districts, about halfway in Bagerhat district, most parts of Pirojpur and Barguna districts and some parts of Patuakhali and Bhola districts. Shallow groundwater salinities in most places are the result of interaction of saline river water with groundwater. Lateral movement of the seawater interface is an extremely slow process and impacts of sea level rise is limited to areas close to the shore. The changes in salinity are very small, in the range of 0.15 to 0.35 ppt, from the baseline to the end of the century. Sea level rise will have greater influence on groundwater salinity via increased river salinity which will allow increased salinization of shallow groundwater. In the northern parts of Satkhira and Khulna districts, rivers are of shallower depths and groundwater salinities at shallow depths may increase by approximately 70 to 30% of the change in river salinity. In areas further south, greater river depths allow increased salinization of groundwater at deeper depths via river-aquifer interaction. In the southcentral zone, even with moderate to high increases in river salinity, there are little changes in groundwater salinity as residence time of high river salinity levels are small and freshwater flushing largely restricts accumulation of salts in groundwater. Irrigation abstraction from groundwater may not substantially influence the ingress of the saltwater interface. But it can change the hydraulic gradient locally, causing salinity intrusion upstream, which is a probability in Satkhira and Khulna districts if large pumping takes place in Jessore district. Storm surge flooding might be a major salinization mechanism in the shallow aquifer by vertical infiltration of salt water from the saline inundated areas. Duration of surge inundation appears to be the most important factor. For a storm surge inundation event with seven days of saline flooding, salinization effect in groundwater may stay several months (e.g. 3 to 10 months found in numerical experiments), while several years for longer duration of flooding (e.g. 1 to 4 years for flooding duration of 1 month and 4 to 10 years for flooding duration of 6 months, for different inundation and top soil layer conductivity scenarios). For an Aila like event where inundation took place for more than 3 years in some areas, flushing of salts from the affected aquifers is thus expected to take considerable number of years. Aquifers thus may become more vulnerable to increased storm surge inundation events, often predicted in climate change scenarios. Tidal oscillation has an influence on the mixing of salt in the salt water transition zone and its movement, with propagation and retreat of the transition zone dependent on mainly tidal amplitude, with beach morphology and hydrogeologic parameters having roles to some degree. However, responses are limited to very short spatial dimension, extending a short distance from the shoreline through the neap and spring tide cycle.