Riverbank Filtration Effectiveness in an Arid EnvironmentEPA Grant Number: R829009
Title: Riverbank Filtration Effectiveness in an Arid Environment
Investigators: Langford, Richard P. , Pillai, Surresh , Schulze-Makuch, Dirk
Institution: The University of Texas at El Paso , Texas A & M University
EPA Project Officer: Page, Angela
Project Period: September 1, 2001 through August 31, 2004 (Extended to August 28, 2005)
Project Amount: $437,418
RFA: Drinking Water (2000) RFA Text | Recipients Lists
Research Category: Drinking Water , Water
Description:This experiment is a field test of bank filtration at a site where water level and salinity vary on an annual basis, as they do in many arid and semi-arid streams. No other studies of bank filtration have been performed in this kind of setting. Along the border with Mexico, shallow wells in the Rio Grande alluvium provide untreated water to the residents of many communities. Knowledge about the effectiveness of bank filtration can help prevent disease and provide affordable water supplies for residents of both countries. Additionally, the large cities along the Rio Grande are rapidly drawing down their aquifers and are planning on using Rio Grande water as their primary supplies. Although most cities are planning to use treatment facilities to supply municipal water, bank filtration may provide a cost-effective pretreatment method.
Bank filtration is the cleaning of contaminated water by pumping it from a stream through the banks to a well. The objective of this study is to determine whether bank filtration is effective at removing particulates and microbial pathogens in the Rio Grande, an arid stream that exhibits significant annual fluctuations in water quantity and chemistry. This experiment will test bank filtration at a site where water level and salinity vary on an annual basis, as they do in many arid and semi-arid streams. No other studies of bank filtration have been performed in this kind of setting.
Waters of different chemistries mix during bank filtration, resulting in complex interactions between soil, bacteria, pollutants and dissolved solids. During bank filtration, organic solids in the river water are strained out into alluvial sediments (Brand et al., 1989; Schwarzenbach and Westall, 1981). Microbial reduction of these organics often creates an anaerobic zone, a few meters wide, in which heavy metals are mobilized. Beyond the anaerobic zone, the subsurface alluvial environment becomes aerobic. In the aerobic zone, trace elements and heavy metals are immobilized and organic pollutants are reduced to harmless compounds.
In humid climates, bank filtration has been shown to remove the majority of dissolved organic constituents and heavy metals [e.g. (Sontheimer, 1980)]. Although bank filtration has been successful in humid climates (Hoetzl et al., 1989; Laszlo and Szekely, 1989), there is little data from arid climates. Bank storage/filtration technology is more difficult to apply to the Rio Grande because of (1) the arid climate, (2) heterogeneous sand-dominated river deposits, and (3) possible long-term salt or arsenic accumulation in the sediments.
The overall objective of this research proposal is to determine whether bank filtration is effective at removing particulates and microbial pathogens. The experiments are designed to address the following questions:
1. Is the stream the source of the water? Do stratigraphic heterogeneities reduce the effectiveness of bank filtration?
2. Does variation in salinity and water depth change the effectiveness of bank filtration?
3. Does pumping rate change the effectiveness of bank filtration?
Approach:The experiment will be a field experiment conducted at a research site established by the University of Texas at El Paso. The bank-filtration research site offers important advantages as the researchers can manipulate well spacing and pumping rates to best fit the design of the experiment. For this experiment, a very short riverbank to pumping well distance is used to minimize dilution of pathogen species.
Wells will be drilled in a cross pattern at the experiment site. Wells along the axis of the cross will document decreases in pathogen, while wells along the cross-bar will document dispersion of the flow between the river and the pumping well. We will characterize the experiment site by detailed analysis of cores collected from the wells. Slugs of Bromide from three injection wells will be used to document the flow paths between the stream and the pumping wells. Multi-level wells will be used to document vertical heterogeneities in flow.
Water samples will be collected twice each month during each of the three years of the project. Samples will be collected as soon as flooding begins, and then at two-week intervals during the November to February flooding of the channel. Water samples will be collected twice a month and analyzed for water chemistry. Samples will be analyzed monthly for Giardia, Cryptosporidium, and E. coli. A three dimensional model of water flow between the stream and the well will be created and the effectiveness of bank filtration will be measured.
Repeated sampling will document the effects of changes in water chemistry and each year the site will be sampled using a different pumping rate for the experiment well.
Expected Results:This proposal addresses riverbank filtration in an arid setting and will allow independent evaluation of the influences of seasonal variation in pumping rate and water level. The study will also help determine the utility of bank filtration in a region that desperately is suffering from a limited and polluted water supply.
This experiment will test bank filtration at a site where water level and salinity vary on an annual basis, as they do in many arid and semi-arid streams. No other studies of bank filtration have been performed in this kind of setting. This field study will also clearly identify the paths taken by water from stream channel to pumping well, allowing us to isolate the effect of bank filtration from mixing and dilution from existing groundwater.