Collaborative Research: Vertical control of groundwater arsenic concentrations in shallow Bangladesh aquifers

数千万人食用的恒河-雅鲁藏布江-梅赫纳三角洲(GBMD)地下水中砷含量升高，这可能是有记录以来最大的一起人类中毒事件。虽然砷的来源是自然的，但地下水中砷的空间分布是高度可变的。国际水文计划小组过去3年在孟加拉国Araihazar upazila进行的实地调查表明，就最浅的含水层而言，这种跨越101-102米的空间差异是由地表地质垂直驱动的：(A)无论沙质沉积物延伸到地表，地下水的浓度往往保持在低至约20米深；(B)因为在表层覆盖有细粉砂质或粘土的浅层含水层中，地下水的浓度通常较高。要证明这种关系在复杂的河流三角洲环境中的表现，需要使用地球物理和地球化学工具的新组合收集空间密集数据。在这项提案中，我们建议应用在阿拉哈扎尔开发的新方法，其中包括现场测量不稳定沉积物的性质，以调查孟加拉国其他几个地区浅层含水层中砷的空间控制。NSF标准1-科学：河流-三角洲沉积本质上是复杂的系统，因为它们的结构反映了河流通道的位置和顺序随时间的变化。巧合的是，这组PI发现了表面沉积性质与地下含水层氧化还原状态之间似乎在101-102米的空间尺度上非常一致的关系。如果这种关系在孟加拉国其他地方的这种尺度上成立，它也可以解释地下水在104-105米空间尺度上的平均浓度和变异性的区域差异，以及与已报告的地下水升高的类似环境，如越南和阿根廷。拟议的调查还将为同时研究地表地质对浅层含水层的微生物学和水文学的影响奠定基础。NSF标准2-影响：孟加拉国约1000万口管井中的绝大多数是浅层(20米)。随着更多这样的水井被测试，高砷水井用于饮用水和烹饪(而不是洗涤)的使用率有望下降。如果浅层含水层的氧化还原条件与当地地表地质条件有关的假设能够在GBMD的不同河流-三角洲环境中得到证实，那么这一知识就可以更有效地用于确定安全含水层。然而，政策制定者面临的一个关键问题是，随着主要从浅层含水层取水的灌溉继续进行，管井和浓度是否会随着时间的推移而变化。地表地质学假设的垂直控制的一个含义是，浅层含水层中地下水浓度的空间分布对大规模抽水灌溉引起的水文变化相对不敏感。了解这一点显然很重要，因为随着时间的推移，这将减少重复测试管井的需要，以及将灌溉从地下水转换为地表水的需要。最后，我们的观察可能会导致在野外范围内操纵地表沉积物，在几乎没有其他选择的地区生产低砷地下水。

0345688van GeenElevated levels of arsenic (As) in groundwater of the Ganges-Brahmaputra-Meghna Delta (GBMD) consumed by tens of millions of people has caused what may be the largest case of human poisoning in recorded history. Although the source of As is natural, the spatial distribution of As in groundwater is highly variable. Field investigations conducted by the team of PIs in Araihazar upazila, Bangladesh, over the past 3 years suggest that in the case of the shallowest aquifers, this spatial variability across 101-102 m is driven vertically by the surface geology: (a) groundwater As concentrations tend to remain low to ~20 m depth wherever sandy deposits extend to the surface, (b) As concentrations are typically elevated in shallow aquifers that are capped by a surface layer of fine silt or clay. Demonstrating the expression of this relationship in a complex fluvio-deltaic environment required the collection of spatially dense data using a novel combination of geophysical and geochemical tools. In this proposal, we propose to apply the new approach developed in Araihazar, which includes on-site measurement of labile sediment properties, to investigate the spatial controls of As in shallow aquifers of several other areas of Bangladesh. NSF Criterion 1 - Science: Fluvio-deltaic deposits are intrinsically complex systems because their architecture reflects changes in the position and order of stream channels over time. Somewhat serendipitously, this team of PIs has discovered what appears to be a very consistent relationship between the nature of surfacial deposit and the redox state of the underlying aquifers on spatial scales of 101-102 m. If this relationship holds on such scales elsewhere in Bangladesh, it could also explain regional differences in the mean and variability of groundwater As concentrations on spatial scales of 104-105 m throughout the GBM, as wells as similar environments where elevated As in groundwater As has been reported, such as Vietnam and Argentina. The proposed investigation will also set the stage for parallel studies of the impact of surface geology on the microbiology and hydrology of shallow aquifers. NSF Criterion 2 - Impact: The vast majority of the ~10 million tube wells in Bangladesh are shallow ( 20 m). As more of these wells are tested, the usage of high-As wells for drinking and cooking (as opposed to washing) will hopefully drop. If the hypothesis that redox conditions in shallow aquifers are anchored to the local surface geology can be confirmed in different fluvio-deltaic settings of the GBMD, this knowledge could be used for targeting safe aquifers more effectively. A critical question facing policy makers, however, is whether tube well As concentrations could change over time as irrigation, which draws water primarily from the shallow aquifers, continues. An implication of the hypothesized vertical control by surface geology would be that the spatial distribution of groundwater As concentrations in the shallow aquifers is relatively insensitive to changes in hydrology caused by large-scale groundwater pumping for irrigation. This is clearly important to know since it would reduce the need for repeated testing of tube wells over time and the need for switching irrigation from groundwater to surface water. Finally, our observation might lead to field-scale manipulation of surface deposits to produce low-As groundwater in areas with very few other options.