What Lies Beneath? Predicting The Quality Of Groundwater For Well Users

Groundwater is the world’s most extracted raw material, with withdrawal rates currently in the estimated range of 982 km3 /year and providing almost half of all drinking water worldwide. Throughout the globe, groundwater is being extracted for domestic use through bored or dug water wells which tap into the water which lies beneath.

These groundwater supplies have some advantages over surface water-derived drinking water, as it is largely considered as having a higher water quality, naturally containing many minerals required for human health, and is generally better protected from pollution and evaporation. However, the quality of groundwater is extremely variable and is largely dependent on local environmental, agricultural, and human factors.

Fig 1: A typical bored well

In Europe, the quality of drinking water is governed by the EU Drinking Water Directive, and thus, all water for human consumption has to reach a certain standard, with a basic parameter being the absence of any fecal bacteria. However, there’s a loophole within the legislation: small supplies serving less than 50 people are not included in the directive, and, as a result, private groundwater wells are not covered, and the quality and remediation of these water supplies is the sole responsibility of the well owner.

In countries like Ireland, this means that over 750,000 people are drinking water from unlegislated supplies, and this can be a cause for public health concern. For example, Ireland has the highest incidence rate of Verotoxigenic E. coli, a pathogenic bacterium, which has been shown to be associated with private groundwater wells (see OhAiseadha et al, 2017).

Fig. 2: VTEC under the microscope

Groundwater microbial quality can be affected by numerous environmental and source-specific risk factors, including well design, location and maintenance, septic system (wastewater treatment) location, local hydrogeological setting, and significant climatic events (e.g. flooding, snowmelt, etc.). Moreover, groundwater pathogens may come from multiple human or animal fecal sources such as close-by septic tank systems, livestock grazing, manure spreading, and/or farmyards. Recent work has shown that up to 70% of well contamination occurs via “localized” pathways, as opposed to “generalized” contamination, with groundwater contamination risk typically increasing in areas characterized by “high” hazards (i.e. faecal source) with this relationship mediated by local hydrogeological characteristics e.g. hydraulic conductivity, aquifer productivity, etc.

In short, groundwater is complicated, and due to the sheer number of private wells (> 170,000 in Ireland alone), it is extremely hard to manage; for example, is it even possible to instigate a national or even local repeated sampling regime? Probably not, or at least not without an inordinate amount of time, resources and money. But — this is where science can help — by combining our knowledge of the natural environment and in particular, the subsurface environment, with targeted sampling regimes and statistical models, we can “predict” areas which are at a greater risk of contamination and, thus, can help in facilitating target remediation and groundwater management strategies.  And that’s exactly what we have done.

In a recent study, the spatial and temporal (time) distribution of groundwater E. coli (an indicator organism) presence in the mid-western region of Ireland has been examined via a mixed methods approach which comprised a 13- month field sampling program, a formalized source owner survey, and statistical analysis. We have identified risk factor variables based on three “categories,” namely: Intrinsic (environmental factors), Specific (local (sampling area) features), and Infrastructural (groundwater source and domestic wastewater treatment system characteristics). Using both the sampling results and the risk factors, we created a predictive model of private well contamination.

All in all, the model (which was externally validated) has a predictive capacity of around 90%. We discovered that factors like soil permeability (how fast something moves through the subsoil), the geomorphology (fancy word for the physical features of the bedrock), rainfall, and well depth and type (bored or dug) could help predict contamination. In terms of sources of pollution, both cattle and domestic wastewater treatment systems (septic tanks) were found to be predictive sources of microbial pollution.

The model developed (called the ISI-LR model) represents a potentially useful and internationally transferable tool for predicting bacterial contamination of private domestic water wells in geologically diverse regions. When viewed through the lens of the characteristic agricultural, infrastructural, and hydrogeological profile of rural Ireland and other similar regions, findings serve to highlight the magnitude of private groundwater source risks.

In conclusion, the developed model represents a comprehensive risk assessment and management tool that may be used by local authorities, water managers, and even well owners/users to develop effective water-quality management strategies to minimize pathogen exposure risks in Ireland and further afield. In fact, the researchers involved in this study are currently developing an app called GRAppLE, which will allow well users to perform “live” risk assessments to safeguard their drinking water supply — so watch this space.

These findings are described in the article entitled Development of a hierarchical model for predicting microbiological contamination of private groundwater supplies in a geologically heterogeneous region recently published in the journal Environmental Pollution. This work was conducted by Jean O’Dwyer from University College Cork and the Environmental Research Institute, Paul Hynds from the Environmental Sustainability and Health Institute and Kenneth Byrne, Michael Ryan and Catherine Adley of the University of Limerick.

*New DOI's are registered weekly and may not function until then.
You May Also Like:

Examining Hippocampal Volume In Older Adults With Depression

The overall prevalence of depression among the older population is around 7%...
Read More

Leave a Reply

Your email address will not be published. Required fields are marked *