REGIONAL-SCALE HYDROSTRATIGRAPHY AND GROUNDWATER CHEMISTRY IN THE WESTERN BENGAL BASIN, INDIA

Within the densely populated Bengal basin of India and Bangladesh, the alluvial aquifer system has suffered extensive non-point source pollution by arsenic. We are investigating regional groundwater flow and trends in solute compositions (including As) in the districts of Murshidabad, Nadia, North 24 Parganas, and South 24 Parganas of the state of West Bengal, India. The study area is bounded by the River Bhagirathi to the west, by the India-Bangladesh border to the east, by the bifurcation point of the rivers Bhagirathi and Padma from the Ganges to the north, and by the Bay of Bengal to the south.

A preliminary, conceptual, hydrostratigraphic model, based on 70 lithologic logs from depths ≤ 320 m in the study area, indicates the presence of a continuous, semi-confined sand aquifer underlain by a thick clay aquitard. The aquifer deepens from a maximum of ~ 50-80 m below land surface in the north to ~ 180-200 m in the south. In the southern half of the study area, discontinuous clay layers locally divide the near-surface aquifer from several deeper, confined aquifers. In general, the thickness and frequency of the clay layers increase toward the east. Discontinuous aquifers containing brackish water have been identified at greater depths (~ 200-250 m) in some localities.

During summers 2003 and 2004, we sampled 37 deep public-supply wells in a band ~ 250 km from north to south and ~10-60 km from east to west. In unfiltered samples from 2003, As concentrations were ≥ 10 ppb for 14 of 24 wells. Poor correlations between As and Fe (r²=0.22) and between As and Mn (r²=0.05) suggest that As concentrations in deeper groundwater do not reflect sorption onto Fe-Mn minerals. Values of ²H and ¹⁸O plot along the global meteoric water line and appear to vary with distance from the Bay of Bengal. Analyses of filtered samples from 2004 indicate anaerobic conditions: field test kits identified Fe(II), but no nitrate or nitrite, in 24 of 25 wells (nine of which had also been sampled in 2003). Sulfate was undetectable for 11 wells; the five highest CH₄ concentrations coincided with a lack of sulfate. However, O₂ was detected in all samples, albeit at low levels (0.49 to 2.3 ppm), which suggests mixing of more and less oxic waters over screened intervals of as much as 30 m. Analyses of metals and stable isotopes for 2004 samples are pending.