Process studies for investigation of natural and anthropogenic factors contributing to arsenic release processes in Bangladesh

Background/Summary

In Bangladesh, 40 million people consume potentially toxic groundwater with dissolved arsenic (As) concentrations above the WHO recommendation of 10 µg l-1. Iron oxides and hydroxides (Fe(III)-(hydr)oxides are considered the most important host minerals of As forming stable surface complexes (Pienkowska et al., 2021). Typically, these processes vary strongly in space, with aquifer depth and time/season (Appelo and Postma, 2005).

Within the carbon cycle, electron acceptors such as nitrate (NO₃^-), sulfate (SO₄^2-) and Fe(III)-(hydr)oxides are reduced successively, usually with increasing depth. This iron reduction leads to the dissolution of As-rich iron minerals (Chakraborty, Mukherjee and Ahmed, 2015; Rowland et al., 2007) one the one side and the formation of methane (CH₄) and ammonium (NH₄) on the other side. In the last few years, CH₄ has been identified as a highly efficient electron donor of the carbon cycle (Buschmann and Berg, 2009; Ettwig et al., 2016; Glodowska et al., 2020b; Stopelli et al., 2021) but more recent research indicates that electron donors of the nitrogen cycle, such as ammonium (NH₄), might also be relevant for arsenic release (Xiu et al., 2024; Gao, Weng and Guo, 2021; Moore et al., 2023; Weng et al., 2017).

In May 2022, depth-specific groundwater and dissolved gas samples were taken at our two study sites in Korgaon (KB, Division Sylhet) and Bera (BB, Division Rajshahi) in Bangladesh. At both sites, three fully screened monitoring wells were installed in 2020 - one next to a pond, one next to a latrine and one on an agricultural site respectively. The results of geochemical composition of groundwater and extracted dissolved gases N₂-, CH₄- and CO₂-concentrations as well as stable isotope ratios (δ¹³C-CH₄, δ¹³C-CO₂, δ²H-CH₄) are presented in Ernst et al. (preprint).

At this pre-monsoon campaign, the dissolved As-concentration varied greatly within well depth, indicating different dynamics for the process of As-mobilization at the same time. Also, the concentrations differed significantly between the three wells, as the water from the two wells in settlement areas (latrine and pond) were contaminated by As above the WHO recommendation (10 µg l-1) at all depths, whereas in the field mostly no As was detected. The dissolved CH₄ was of biogenic origin in all the wells based on Bernard ratios and stable isotope ratios. Particularly in the village wells, dissolved As-, CH₄- and Fe(II)-concentrations correlated positively and the stable isotope ratios and showed a kinetic isotope fractionation characteristic for CH₄-oxidation. In conclusion, As-mobilization in pre-monsoon season is likely caused by CH₄ oxidation through Fe(III)-(hydr)oxide reduction.

In order to investigate seasonal effects, we repeated sampling at the same well infrastructure of the Korgaon site in November 2023 (post-monsoon). In contrast to the pre-monsoon sampling in May 2022, As was found in all wells in concentrations around 50 µg l-1. Simultaneously, CH₄ concentrations were higher and CO₂ concentrations lower in the post-monsoon sampling. This clearly indicates that seasonal effects should be considered when investigating arsenic release processes (Pathak et al., 2022). Moreover, it is evident that large scale hydrogeological processes, e.g. by agriculture, overlap local impacts by human settlements (e.g. latrines and ponds). It is scientifically accepted that the carbon and the nitrogen cycle are involved in arsenic release processes, by supplying electron donors (e.g. methane or ammonium) for the dissolution of Fe(III)minerals and therefore the release of adsorbed arsenic under anoxic conditions (Moore 2023).

Based on these results we think that intensive studies on the vertical influences of seasonal driven processes need to include isotope analyses, as concentrations of groundwater components alone are not sufficient to understand the complexity of ongoing seasonal processes