ABSTRACT:

This is the data used to create the published study of the same name published in the Journal of Hydrology in 2022 (10.1016/j.jconhyd.2022.104068).

Shallow (<30 m) reducing groundwater commonly contains abundant dissolved arsenic (As) in Bangladesh. We hypothesize that dissolved As in iron (Fe)-rich groundwater discharging to rivers is trapped onto Fe(III)-oxyhydroxides which precipitate in shallow riverbank sediments under the influence of tidal fluctuations. Therefore, the goal of this study is to compare the calculated mass of sediment-bound As that would be sequestered from dissolved groundwater As that discharges through riverbanks of the Meghna River to the observed mass of As trapped within riverbank sediments. To calculate groundwater discharge, a Boussinesq aquifer analytical groundwater flow model was developed and constrained by cyclical seasonal fluctuations in hydraulic heads and river stages observed at three sites along a 13 km reach in central Bangladesh. At all sites, groundwater discharges to the river year-round but most of it passes through an intertidal zone created by ocean
tides propagating upstream from the Bay of Bengal in the dry season. The annualized groundwater discharge per unit width at the three sites ranges from 173 to 891 m2/yr (average 540 m2/yr). Assuming that riverbanks have been stable since the Brahmaputra River avulsed far away from this area 200 years ago and dissolved As is completely trapped within riverbank sediments, the mass of accumulated sediment As can be calculated by multiplying groundwater discharge by ambient aquifer As concentrations measured in 1969 wells. Across all sites, the range of calculated sediment As concentrations in the riverbank is 78–849 mg/kg, which is higher than the observed concentrations (17–599 mg/kg). This discovery supports the hypothesis that the dissolved As in groundwater discharge to the river is sufficient to account for the observed buried deposits of As along riverbanks.

ABSTRACT:

This repository contains all the water chemistry data, sediment borehole lithology observations, and handheld XRF observations of elemental concentrations in sediments used in this study.

Study Abstract
Groundwater containing high concentrations of dissolved arsenic (As) and iron (Fe(II)) discharges to rivers across the Ganges-Brahmaputra-Meghna delta. Observed Fe(III)-oxyhydroxide (FeOOH)-As deposits lining the riverbanks of the Meghna River may have been created by bidirectional mixing in the hyporheic zone (HZ) from ocean tides. This process has been named the Natural Reactive Barrier (NRB). Sedimentary organic carbon (SOC) is deposited annually on floodplains. Floodwaters that infiltrate through this layer may chemically transform the groundwater prior to discharging through the HZ in ways that influence the capture and retention of As in the NRB. The goal of this study is to understand how the interaction of these two scales of river-groundwater mixing influence the fate of As trapped within an NRB. Monitoring wells were installed to 1-17 m depth, up to 100 m distance from the river’s edge during the dry season on the East (Site 1) and West (Site 2) sides of the river. They were sampled during the dry season (January) under gaining river conditions. The physical properties and elemental composition of the sediment was described by hand observation and hand-held X-Ray Fluorescence (XRF), respectively. Mixing with river water was quantified using the sum of charge of major cations (TC). Site 1 has a sloping bank that is only partially inundated during the wet season. The aquifer is composed of homogeneous sand. Site 2 is flat and therefore fully inundated in the wet season. The aquifer is composed of sand with thin (1-20 cm thick) clay layers. Both sites generate the dissolved products of FeOOH-reduction coupled to organic carbon oxidation, and silicate weathering beneath the floodplain. These products are dissolved Fe, As, silica, bicarbonate, calcium and phosphate. This chemistry is conducive to the formation of crystalline iron oxide minerals such as goethite which may co-precipitate with As, trapping it long-term.

ABSTRACT:

This data set contains anonymous information about 159 water production wells and shallow dug wells (Norias) across the Upper Rio Laja Basin in Guanajuato State, Mexico. The type of information contained includes locations of the wells, well type, depths, surface elevations, static and dynamic water levels, average electricity costs, well outer casing diameter and pump tubing diameter (inner casing), pumping rates which are set to be constant, and miscellaneous notes on the well including whether it is possible to obtain a manual water level and whether the well goes dry occasionally. The well type DWSW and IW means Drinking Water Supply Well and Irrigation Well, respectively. The well IDs (PW1, PW2 etc..) correspond to the same wells of previous datasets from this basin published on HYDROSHARE by Peter Knappett.

ABSTRACT:

This is the raw data used in the paper entitled "Rising Arsenic Concentrations from Dewatering a Geothermally Influenced Aquifer in Central Mexico" in the journal Water Research in 2020 by the authors listed here. Four files are included: 1) historical delta 18O concentrations ("1999 18O.xlsx") in 122 wells which were measured shortly before 1999 and assembled and reported in Jurgen Mahlknecht's PhD dissertation (Mahlknecht, J. (2003) Estimation of recharge in the Indpendence aquifer, central Mexico, by combining geochemical and groundwater flow models, University of Agricultural and Life Sciences (BOKU)); 2) historical water chemistry data ("1999_Wells_Chemistry.xlsx") in 246 wells which were measured shortly before 1999 and assembled and reported in Dr. Mahlknecht's dissertation; 3) water chemistry data which was collected between 2014 and 2018. Most of the wells were sampled between 2015 and 2017. Many of the wells were sampled 2-5 times during different times of the year between 2014 and 2019, however, the database included here only has the first sampled chemistry from each of the 137 wells. This was the main database used to perform the analyses in the paper; 4) water chemistry data from the 22 re-sampled wells between 1999 and 2016 ("Resampled Well Data_mM). These only include the parameters that were analyzed in 1999. The nominal detection limit for ICP-MS data measured in 2016 was 0.01 ppb. The nominal detection limit for the Ion Chromatograph data was 0.01 ppm. Missing values indicates the data was not available. Alkalinity values are reported as mg/L "HCO3" which is not the convention of reporting this parameter as mg/L "CaCO3", however, it was consistent with the format used for the 1999 data set.