Relationships, occurrences, and trends
By Eric Duderstadt
Advances in science and technology have equipped the groundwater industry with more data than has ever been available at any point in recorded history.
Rapid screening techniques used to quantify and profile the overall bacterial populations, as well as identification of the more prominent bacterial species within water, now allow for a comprehensive assessment of the biological conditions within a well.
This approach was used within a laboratory setting to examine the biological communities in thousands of samples from potable wells over several years. Data from those analyses was compiled to determine relationships, occurrences, and trends that developed with regard to deposit formation, corrosion potential, unsafe conditions, and other fouling mechanisms associated with bacterial influences.
Testing Approach for the Well Environment
Bacteria are found universally—including some of the most extreme environments on the planet—and are part of our everyday lives. They play important roles within the food, pharmaceutical, and bioremediation industries as well as within our own bodies. They exist in soils, air, and water.
The potable wells we rely upon as a source for safe water are no exception. Groundwater wells are dynamic systems which vary considerably in construction and operation, as well as a multitude of external influences from the aquifers and environments they interact with. These variables may result in a variety of conditions downhole that can stimulate extremely diverse and expansive microbial communities.
The water well industry has traditionally relied heavily upon the total coliform test, commonly referred to as a “Bac-T test,” to determine the biological presence within a well. This is due to the notoriety these methods have gained because of regulatory actions put in place by the U.S. Environmental Protection Agency.
However, these indicator organisms do come with some qualifiers. Firstly, coliforms are a large group of bacteria, most of which occur naturally in the environment and pose little risk if consumed. In reality, only a small percentage of coliforms are pathogenic. Thus, the designation between total coliforms and fecal coliforms is important to consider.
Most importantly, the presence of coliforms only suggests the potential presence of more problematic organisms; they don’t confirm it. Furthermore, the absence of coliforms doesn’t confirm a water sample is free of harmful microorganisms.
As the influences bacteria have on wells are now becoming more and more well known—impacting production, mineral accumulations, water quality, and corrosion processes—the use of biological testing and its ability to provide additional insight into the fouling potential and operational aspects of a well requires a closer look beyond the singularity of the Bac-T test.
The extreme diversity and sheer number of bacteria capable of residing within wells makes testing for each culprit, if not impossible, at least impractical. Thus, a number of tactics have been developed to offer a more practical alternative.
Such tactics include quantifying the overall population, compartmentalizing bacteria into broader groups based on shared qualities, and identifying only a few of the most prominent species present. The implementation of these techniques offers a streamlined approach for assessing the biological makeup within a well and the threats it poses to both the well and water quality—and also provides valuable insight
to designing rehabilitation techniques and planning the long-term health of a well.
Nearly 2000 water samples were collected from hundreds of problematic potable groundwater wells across the United States and submitted for laboratory testing of biological activity with regard to maintenance, operation, and fouling potential. Sample collection was conducted over a seven-year period from 2010-2017.
In general, wells were analyzed as part of a proactive measure to investigate issues related to production or water quality changes. The size, design, location, and operation schedule of each well varied with no restrictions to any of these parameters.
Each sample was classified as either “Casing” or “Aquifer” to designate between static or active pumping conditions at the time of sampling. A series of tests was selected in accordance with the streamlined approach described above and conducted in stable sterile conditions. Data was then collected and analyzed to determine any noticeable trends or relationships which had developed.
Quantifying the overall bacterial count within a well is a first step in determining the potential for biofouling within the well. A number of methods are available for analyzing bacteria counts in water, ranging vastly in accuracy and cost.
One of the more traditional methods is the Heterotrophic Plate Count. HPCs estimate the number of heterotrophic bacteria (organisms that require organic compounds for nourishment) in a sample by recording the level of growth on non-selective nutrient-containing media. HPCs are useful because they don’t count dead organisms or inanimate particulates. However, one colony may develop from a single cell or
numerous cells, and cell clumping in colonies is random, making accurate bacterial counts difficult.
Perhaps the biggest disadvantage of this method is that only a small percentage of bacteria found in water systems are culturable using common laboratory media—with further limitations placed on the number of bacteria available due to the aerobic conditions in which the test is carried out. Therefore, culture-independent methods are essential to understanding the genetic diversity, population structure, and ecological roles of most microorganisms.
One such method which does not rely on the growth of the available bacteria is Adenosine Triphosphate, or ATP, analysis. ATP analysis is a chemical test that rapidly and accurately monitors microorganisms in a water sample through the measurement of a universal energy molecule within cells. The test exposes microorganisms to enzymes that catalyze a reaction, converting ATP into light energy. Each molecule of ATP consumed in the reaction produces a known quantity of light, the output of which is measured by luminometers and compared with a standard to estimate the number of cells present.
Although ATP and HPC are both quantification techniques, they vary in the scientific methods in which they derive their results. Given these differences, ATP and HPC counts from the collected samples were cross-referenced to observe any correlations between the two methods.
Samples were categorized based on their HPC results and organized into a “Low,” “Medium,” and “High” category based on the range of the test method and the EPA’s “levels of concern.” ATP levels from each sample from within the three classifications were then averaged together. The results showed a steady growth in ATP levels consistent with the three levels of HPC growth, suggesting the two methods are
analogous to some extent.
However, a closer look at HPC values showed a strikingly disproportionate distribution among the samples. This was unexpected given the fact various degrees of fouling were to be encountered and the extent of fouling was considered random among the samples.
Further examination of the data found samples containing just one colony-forming unit (cfu) from the heterotrophic plate count recorded an average ATP value of 783,733 cells per milliliter (cpm). Furthermore, samples that showed no plate growth recorded an average ATP level of 141,211 cpm.
Although the majority of the samples were known to be from problematic wells, each of these ATP levels are considered
high especially considering the low amount of plate growth. As a point of reference, active potable water wells typically exhibit ATP values between 10,000 and 70,000 cpm for an active well sample, with values more than 100,000 cpm generally indicating biofouling.
The high level of ATP in samples with very low or no heterotrophic plate growth appears to explain the uneven distribution of plate counts among the samples while also illustrating the two primary disadvantages of the HPC method:
- Colony growth and cell clumping is random.
- A large portion of bacteria found in water systems are unable to be cultured within a laboratory setting.
Most Common Bacteria Types
As noted earlier, rapid screening techniques have made the identification of bacteria within water samples easier and more affordable in recent years. Over the course of the study period, the Biolog system was used for identifying bacteria present within the samples.
This system is referred to as a nutrient characterization assay in that it measures a bacterium’s response to a series of carbon compounds (amino acids, peptides, sugars, etc.).
The method uses a series of small wells containing the carbon compounds which can be inoculated with the isolated bacterium. A color change is then produced within the corresponding wells based on the bacterium’s response to the various compounds and a pattern is formed, producing a metabolic fingerprint specific to each bacteria tested. The pattern is then mapped in a computer database to select the bacterium present.
At least one bacterial identification was made in almost 85% of the samples with a total of 3216 identifications made over the study period. Identification of the most prominent bacteria within the observed samples identified three genus that accounted for nearly half (48.0%) of all the identifications made. A genus is a class or group marked by common characteristics—specifically here a category of biological classification ranking between the family and the species, comprising structurally related species or an isolated species exhibiting unusual differentiation.
Gallionella, the most commonly identified genera of bacteria, accounting for 24.3% of all the bacteria identified in the samples, is a group of iron-oxidizing bacteria naturally occurring and found in a variety of aquatic environments, including aquifers.
These bacteria utilize iron and manganese as energy sources and will accumulate oxidized iron and manganese in the stalks they produce as waste. The secreted stalks are often shed during change in flow (operation) and are effective fouling mechanisms within flow pathways and pumps. Surges of the shed stalks during start-up of a well can result in the occurrence of red water and spikes in total iron readings.
In its attachment to iron-bearing surfaces, Gallionella will actually pit metal surfaces in an effort to secure the iron necessary for energy, making it one of the chief forms of microbially influenced corrosion (MIC). All iron-bearing structures, including stainless steel, are susceptible to this form of pitting.
Pseudomonas, responsible for 13.7% of all the bacterial identifications made, are a large genus of gram-negative bacteria. Pseudomonads exhibit a high degree of metabolic diversity and consequently are able to colonize a wide range of ecological niches. Most members of this group are aerobic (require oxygen for metabolism) and are easily cultured, making them good candidates for scientific research studies.
Studies have shown a significant number of Pseudomonas members can produce large amounts of exopolysaccharides, a key component of biofilm.
Biofilm is a naturally occurring expression of bacteria resulting from the extrusion of slimy polysaccharide polymers which they exude to attach themselves to a smooth surface for propagation, nutrient capture, and growth.
Present throughout nature, biofilms are effective at plugging flow pathways and decreasing production in wells. They are also an excellent source for the development of mineral scale by providing excellent surface for adhesion of mineral-forming ions, fine-grained sediments, and other debris mobilized toward a well during operation—further increasing the fouling potential.
Bacillus, the third most common genus identified, is a group of gram-positive bacteria that, like Pseudomonas, are known to exude large secretions of biomass as a means of attachment and nutrient capture. They can be obligate aerobes or facultative anaerobes, meaning they can reside in areas with ample oxygen or little to no oxygen present.
This is especially noteworthy, as a phenomenon commonly observed in biofilms as they mature is a stratification effect in which the primary layering differentiates an exposed aerobic stratum overlaying and protecting lower anaerobic layers. Thus the presence of these bacteria can increase the anaerobic zones within a well and indicate more mature biofilm growth.
While noticeable differences exist between these three most common groups of bacteria, they do share significant similarities, especially when viewed in the context of their potential to impact fouling within wells.
Firstly, it can be said each of these groups of bacteria are widely dispersed in nature, commonly found in soils and water, which is no doubt one of the reasons they enjoy such a good attendance record within the analyzed samples.
Further, many of them exhibit a high degree of metabolic diversity, not only allowing them to exist in a variety of different conditions but also making them more adaptable to changes in the well environment and efforts to stress or remove them.
While most are not classified as true pathogens, many members of these groups are considered opportunistic pathogens, meaning they could become disease-causing in immuno-compromised individuals (infants, elderly, etc.). This is even truer as they increase in density.
Perhaps the most noticeable similarity is they each play a key role in biofilm development or the formation of other deposit accumulations. This trait undoubtedly makes them each effective contributors to fouling mechanisms within a well which act to reduce flow spaces within and limit production.
Given the notoriety coliforms have gained within the water industry as means to determine biological presence within a well and the regulatory emphasis which has been placed on them, the rate of coliform occurrence over the study period was analyzed against other biological parameters.
As discussed previously, as biofilms mature they often begin to stratify and form layers. As these accumulations develop, the lower layers are sealed off or restricted from the flow of oxygen, creating anaerobic layers.
In addition to increasing the relative density and fouling potential of the biofilm, this stratification can result in the harboring of coliform bacteria, which are anaerobic organisms themselves, by providing nutrients and acting as a shield from environmental stresses.
Measurement of anaerobic growth among the samples was cross-referenced with coliform presence to observe any distinguishable trends. Samples were categorized based on the observed level of anaerobic growth, reported as a function of the total bacterial population, and organized into a “Low,” “Medium,” and “High” category.
Total coliform presence from samples within the three anaerobic growth classifications was then evaluated. The results showed increases to the anaerobic population and anaerobic conditions within the well setting correlated with an increased occurrence of coliform bacteria.
In samples where the anaerobic population was 10% or less of the total bacterial count, total coliform occurrence was limited to just 1.9%. However, as anaerobic growth rose to 15% or greater of the total population, total coliform bacteria were present in 25.6% of the samples with total coliform presence jumping to nearly 50% when samples displayed a level of anaerobic growth of 30% or greater.
Given the nature of biofilm development outlined earlier, the consistent rise in coliform presence associated with increased anaerobic growth clearly supports the notion that more mature biofilm growth can indicate an increased likelihood for coliform presence.
However, when coliform presence was further refined to just E. coli-specific organisms, or those bacteria which are more indicative of true contamination events, E. coli bacteria were identified in just 0.4% of all the samples over the course of the study.
This low level of occurrence was somewhat unexpected given the problematic nature of the wells being analyzed. However, it clearly supports the fundamental shortcomings of only using total coliform testing as the singular means of evaluating a microbial community.
Additionally, it supports the theory that the presence of coliforms only suggests the potential presence of more problematic organisms, without confirming it, and the absence of coliforms doesn’t confirm a water sample is free of harmful microorganisms.
One final look at the data summarized the relationship between coliforms and other bacteria markers in the context of overall well health. Of all the samples analyzed over the course of the study, less than 10% were positive for total coliform occurrence with only 0.4% positive for E. coli bacteria. Yet more than 60% of the samples displayed at least one or more of the following conditions associated with well fouling: iron related bacteria presence, a high level of anaerobic bacteria, or an elevated bacterial population (as defined by ATP analysis).
The high rates of occurrence of these additional markers for biological fouling clearly illustrate the gap between coliform testing and using other bacterial parameters to monitor the well-being of groundwater wells.
Laboratory testing is often viewed as an unnecessary and expensive use of resources within our industry. Yet the data gained from this simplified series of tests has yielded valuable insight into the ways in which wells are experiencing biological fouling through a new viewpoint beyond the lenses of coliform testing.
While data may not be the most exciting deliverable to our clients, it has historically driven change. Data has continually changed our views of water from the ancient Greeks, to the industrial revolution, to the advent of coliform rules, and to the rise of lead and copper issues.
The question has shifted now from “What is the data telling us?” to “What will we do with the data to make changes for the better?”
Eric Duderstadt is an environmental biologist with Water Systems Engineering Inc. of Ottawa, Kansas, where he works as a consultant. He earned his bachelor’s degree in biology at Ottawa University in 2007 and has since become certified as a corrosion technician within the National Association of Corrosion Engineers. He also works within the firm’s research department and investigative laboratory centering on microbiology and chemistry. Duderstadt can be reached at eduderstadt @h2osystems.com.