UCL Department of Geography


Mohammad Shamsudduha (PhD, 2007-2011)

Groundwater dynamics and arsenic mobilisation in Bangladesh: a national-scale characterisation


Project Partner

Department of Geology, University of Dhaka (Bangladesh)

Thesis publications
Project rationale

Elevated concentrations of aqueous arsenic (As) in groundwater are a major public health concern in Bangladesh, West Bengal (India) and several other similar deltaic and low-lying countries in south and south-east Asia. The mobilisation of As in groundwater-fed water supplies in Bangladesh is recognised as the largest mass poisoning in history affecting more than 35 million people. Research over the last two decades has broadly answered geochemical questions relating to the origin and release of As from mineral sources, but the critical control exerted by groundwater flow and storage changes on its patchy and unpredictable distribution in drinking-water supplies remains unclear. This shortcoming is of fundamental importance as it limits our ability to predict in time and space As concentrations in groundwater. Moreover, intensive groundwater abstraction for irrigation to sustain dry-season Boro rice cultivation in Bangladesh has, over the last 40 years, altered natural pathways of recharge and discharge transporting As from release hotspots to sites of abstraction. There is, at present, a range of conflicting hypotheses regarding the role of groundwater abstraction in mobilising As. Each hypothesis has been developed at the village-scale and none has been tested at the national-scale. Furthermore, it is also unclear how climate change (rainfall intensity and shift in south Asian monsoon) will affect groundwater storage dynamics by changing recharge and thus future As pathways in groundwater supplies. My research investigates how changes in spatio-temporal storage of shallow groundwater in Bangladesh due to abstraction for irrigation and climate change influence groundwater dynamics and As mobilisation in drinking water supplies.

Methodology and data analysis

I have compiled a national groundwater level database which consists of a total of 1.8 million weekly data points from a network of 1,267 stations across Bangladesh. I have then applied novel statistical techniques to determine long-term (1985-2005) trends and seasonal variations in groundwater levels and identified areas of rapidly declining and rising trends. The statistical analyses also provide valuable insight into groundwater dynamics and seasonal variations due to abstraction, rainfall and flooding. These exploratory analyses provide, for the first time, a national-scale understanding of spatio-temporal variations in shallow groundwater flow and storage.

I have estimated actual recharge to shallow groundwater at the national scale using newly compiled groundwater level database which highlights the fact that net groundwater recharge has increased in many parts of Bangladesh due to intensive abstraction for irrigation. The spatio-temporal dynamics of recharge in Bangladesh expose the fundamental flaw in definitions of “safe yield” based on recharge estimated under static (non-pumping) conditions and also reveal areas where (1) further groundwater abstraction may increase actual recharge to the shallow aquifer, and (2) current groundwater abstraction for irrigation (and urban water-supplies) is unsustainable.

Using a statistical framework, I am currently examining how changes in the groundwater dynamics (components such as recharge and trends in groundwater levels) relate to distributions of As in shallow aquifers.

Research progress

An examination of long-term trends and seasonality in shallow groundwater levels at the national-scale (Thesis chapter 3) has already been published in Hydrology & Earth System Sciences. I am currently writing up another two chapters on groundwater storage changes and recharge estimates for Bangladesh and plan to submit these to Geophysical Research Letters and Hydrogeology Journal respectively. I will shortly begin to work on a vital chapter of my thesis testing As mobilisation hypotheses at the national-scale using generated data sets on groundwater dynamics and As concentration data. I expect two further publications from my thesis chapters on (i) the validity of current As mobilisation hypotheses at the national-scale, and (ii) how variations in rainfall intensities affect groundwater storage and As mobilisation.

Research impacts

I intend to communicate the key messages from my doctoral research to (1) the scientific community through the submission of papers to leading scientific journals; and (2) policy makers (Bangladesh Centre for Advanced Studies), public health officials (International Centre for Diarrhoeal Disease Research Bangladesh) and water managers (Bangladesh Water Development Board) in Bangladesh through workshops and online publication of short (1-2 page), well-illustrated policy-relevant summaries of major findings of my research. Policy-relevant outputs already generated through my research include: (1) identification, for the first time, of the unsustainability of groundwater-fed irrigation in certain areas of Bangladesh, and (2) rising trends in shallow groundwater levels in coastal Bangladesh associated with sea-level rise.