Defining the Clogging Sites

The exact relationship of the clogging event to the well is of particular importance in establishing a suitable treatment and management strategy for the rehabilitation and ongoing operation of a clogging system. There are a multiple of observational strategies to determine the sites and magnitude of clogging events which may involve biological, physical, chemical and positional strategies. Primary concerns rest with the determination of the position and extent of the clogging particularly as it relates to the hydraulic conductivity through the afflicted zone. Each of the strategies rely to some extent on each other but the main focus in this ensuing section will be on each strategy in turn.

Biological Determination of Clogging

Clogging involves a major biological component in the formation, maturation and hardening of the clog. The microorganisms that may be involved in the clog will change during the maturation (clogging) process and much of the biological activity will be occurring within the clog itself. This is perhaps one of the first conundrums in trying to determine whether there is a clogging which has been generated through a biofouling of the system or process. Generally early in the clogging, there is a relatively free exchange of microorganisms between the clog formation and the water passing over the surface of the clog. Later on as the surface of clog becomes denser (with accumulated chemicals) and hardens (due to reduced water and organic content and the formation of amorphous and crystalline structures). When this hardening happens, there is gradually less exchange between the incumbent microbes within the clog and water passing over the clog formation. This means essentially that a young clog formation may be easier to recognize biologically because some of the microorganisms can be recovered from the water. As the clog matures, the population (and organic content) of the clog frequently declines and there is less exchange of microorganisms with the passaging water. To determine therefore if there is a clog forming, steps may have to be taken to try and induce the clog dwelling microorganisms to vacate the clog and become determinable as a part of the water microflora.

Traditional strategies for examining waters suspected of being biofouled by a clog formation is to determine the microbial content of waters sampled from a point downstream of the suspected clogging site. Techniques that have long been employed were not designed to determine critically the biofouling microorganisms but rather to obtain a general indication of the microbial loading (often expressed as the heterotrophic bacteria) of the water. Of these techniques it is the spreadplate technique which is the most widely accepted. This technique involves the streaking of a known volume of the water sample over an agar (semi-solid) surface. The microbes in the sample essentially have to "mine" water from the agar in order to grow. Selective nutrients are added to the agar culture medium in order to stimulate the growth of the microorganisms that are to be quantified. For example, if it desired to determine the number of iron related bacteria (IRB), ferric ammonium citrate may be included in the medium. Many of the IRB can utilize the citrate as a source of organic carbon while the ferric ion stimulates the bacteria to initiate the reduction oxidation (ferrous/ferric) cycle. The bacteria that can "mine" the water from the agar base and utilize the nutrients added to the agar culture medium are thus able to grow. Because the agar is gel-like, the microbes able to grow tend to form clusters which become visible. These are known as colonies and can be counted as being representative of the population size. So convenient is this that the attitude has been generated that all microorganisms can be grown on agar culture media and so this has become the standard technique for enumerating microorganisms. Unfortunately, the agar spreadplate technique has a tendency to underestimate the population since either the microorganisms will not all grow under the conditions presented, or they become competitive for "space" (ie, the agar surface or volume) and "food" (the nutrients supplied in the agar). Various researchers have noted underestimates to the population numbers of between one and four orders of magnitude. Furthermore when the first "crop" of colonies appear on the agar plate, there is a tendency to take these to be representative of the total enumerable microorganisms whereas, in reality, there may be further "crops" of colonies as other microorganisms adapt and grow on the agar medium.

Other techniques have appeared ranging from estimates of the total suspended solids (TSS) assuming that a fraction of this material will be biocolloidal in nature and support an active microbial flora. Laser particle sizing has probably become the more precise method for determining the TSS. A simpler technique is to determine the turbidity of the water as an indication of the likely presence of microorganisms. This may not however be a very suitable technique since a "clear" water may contain up to four hundred thousand microorganisms per millilitre which could represent a very significant biofouling event.

Direct microscopic techniques are also commonly used. These techniques vary in sophistication from a simple low magnification examination of a drop of the water sample to the utilization of staining as a means of highlighting the presence of microbial cells within the water. The former technique does allow the direct recognition of sheath (tube) and stalk forming bacteria because of the uniqueness of their form. For example, Gallionella produces a ribbon-like stalk which is very easy to recognize. The observation of the stalk does not however mean that the microbial cell of Gallionella is present. It could simply be a stalk that has sheared away from the cell and is simply detritus floating in the water. Of the more refined techniques, Acridine Orange Direct Count (AODC) is one of the most favored because this technique is able to highlight all of the microbial cells including those that may be closeted within colloidal (slime particulate) structures within the water.

Filtration of the water sample is another way to "concentrate" the cells. The usual method is to use a membrane filter (MF) which has a very specific pore size. For entrapping microbial cells, the most common pore size used is 0.45 microns although a higher efficiency can be achieved at 0.22 microns. Once the cells are entrapped on the surface of the MF, they can be cultured to form colonies by laying the filter on the surface of a selective culture medium. Again, the microorganisms have to "mine" the water through the filter. Where considerable debris is also in the water sample, this will also become entrapped to some extent on and in the MF. If this happens, this debris may change the nature of the nutrients available on the surface of the filter (due to the entrapped particulates). This may then bias the selectivity of the cultural conditions being used to enumerate specific types of microorganisms.

A more recent technique which has been developed relies on the nurturing of the microorganisms in the water by minimising the applied stress (from the testing procedure) and providing a variety of environmental conditions. To do this the principles used by Winogradsky in the generation of a water column in which the bottom was reductive and the top, oxidative. In the patented modification a floating intercedent device (FID) is floated on the water to restrict oxygen entry while a solid nutrient deposit in the base of the column causes a nutrient front to diffuse up the column. The combination of an oxidative front diffusing downwards and a nutrient front diffusing upwards generates a range of environmental conditions within which the incumbent microorganisms may find a suitable niche for activity and growth. A range of these devices have now been developed as biological activity reaction tests (BART^TM)

. Given a wide variety of monitoring techniques each with a different level of sensitivity to the presence of microorganisms, it becomes important to determine what would be the specific trends of value in the evaluation of the likelihood of a clogging situation having developed. General trends in the water coming from an afflicted well would include the following phases:

1. Erratic microbial population, unstable chemical characteristics. Development Stage.
2. Consistently low microbial population, stable chemical characteristics. Initial Clogging Stage.
3. Rhythmic fluctuations in the microbial population, stable chemical characteristics, minimal capacity losses. Clog Formation Stage.
4. Asymmetric fluctuations in the microbial population, degenerating chemical characteristics, significant (>20%) capacity losses. Clog Maturation Stage.
5. Frequent low microbial populations with occasional pulses of very high populations, stratification of the microflora in the water column, severe (>40%) production capacity loss, very considerable drawdowns. Plugging Clog Stage.
6. Consistently high microbial populations, stratification in the water column, system essentially lost production capacity (may occur at between 10 and 60% of the original production capacity).

Given that there due considerable fluctuations in the microbial loading as clog growth moves through the stages from one to six, it becomes much more of a challenge to determine: (A) is there a clog formation occurring? and if so, (B) where and how fast is the clog formation moving through the stages listed above. Clearly an intensive evaluation of a water sample at one particular time is less likely to yield validatable information relating to the formation of a clog rather than a simpler sequence of tests performed on a more regular basis. Fortunately, the rate of clog formation is relatively slow but unfortunately the rate of clog formation has never been well documented on a large scale. From the experiential knowledge base it could be projected that the time taken for a clogging well to go through all of the six stages to total clogging may vary from a short a time as six months to as long as fifty years or more. In groundwater situations where clogging is a common occurrence, the experience may be that the average non-rehabilitated water well would last one to four years. It should be remembered that wells installed in hazardous waste sites as a part of the remediation strategy may be particularly susceptible to clogging because of the high nutrient loading that may be impacted upon the wells by the operation of the site. Clogging in the five to twenty year time period is a relatively common experience. The problem with these wells is that the clogging stages may generate so slowly that unless the operator is maintaining a watchful eye on the performance of the wells, the wells may enter stage six (totally clogged) before any rehabilitation is attempted. This is again particularly a problem in sporadic use wells such as the relief wells at dam sites which only function close to capacity at time of increased hydrostatic pressure as a result of rising water levels behind the dam. It can be projected that theoretically it is inevitable that all water wells would eventually clog in the fullness of time!

From the above comments, it becomes evident that a more frequent monitoring of the microbial flora in the water has advantages in that the microflora varies in size during the clogging process and infrequent monitoring is less likely to usefully monitor a microbially induced clogging event. There are two phases in the determination of a clogging event: (i) determine that a significant clogging event is taking place; and (ii) locate the clog sites so that a treatment strategy can be implemented.

It can be seen that phase (i) has to involve a frequent monitoring. The time period for effective monitoring of a well for clog formation which has been used routinely is monthly with simple site specific tests. These tests should be selected on the basis of an initial evaluation of the various test methods in order to select the most suitable test(s) for routine use. By way of an example, microbial activity could be monitored by the use of the IRB- and the SRB- BART^TM biodetectors. These two tests were performed in duplicate on an monthly basis. Duplicates were performed because during the clog formation there is so much variability in the water even from one fifteen ml sample to another! If the dd (days of delay) and the RX (reaction pattern) forming the reaction pattern signature (RPS) varies between the duplicate tests then this may be taken to indicate that the water being sampled is being impacted by a random sloughing from the clog formation. When these tests are performed monthly and the two replicates are in reasonable harmony, the major marker of a major clog formation occurring are a decline in the dd and sometimes a shift in the RX patterns recorded in the RPS. Erratic shifts in the dd would indicate that the clog formation is in stages 3 or 4 while the radical decline in the dd would indicate that stage 6 (total clogging)is a serious concern. Shifts in the RPS reaction patterns may be used to indicate that there is a change in the microbial community which may be associable with the formation of a clog.

Given that evidence is obtained that a clog formation is being generated (phase i), the second and more challenging phase (ii) can be initiated to locate the possible site of the clog formation. Since the clog usually concentrates at particular sites rather than in a diffusive manner throughout the formation, it becomes important to locate those sites so that a control management policy can be implemented. Clogs generally form at sites which are the most accommodating to the various microorganisms forming the clog. Affecting factors include the redox front position, nutrient loading, water flow patterns, and suitable surfaces. To determine these presences it becomes necessary to create a circumstance under which many of the incumbent microorganisms become traumatized to the point that they detach from the clog and move with the water flow. It should be remembered that a clog will form under conditions of a relatively stable environmental conditions (eg, periodic pumping on a regular basis). Deliberately changing the operational and/or environmental conditions around a well within a clogging zone can cause at least some of the microorganisms in the clog to detach. These can subsequently be determined through their presence in the water that has passed through the clogging zone.

The strategy to determine the location and intensity of a clog formation can be to use the strategy of a biofouling assessment management (BAM). To undertake a BAM evaluation it is important to appreciate the environmental conditions prevalent at the suspected clogging site. It should be remembered that these environmental conditions do not have to be constant but that the variations should be of a routine nature even if seasonal in occurrence. To stress the clog formation as a mechanism for causing releases of some of the incumbent microorganisms (as markers) into the groundwater has to be based upon the manipulation of the ambient environmental conditions. Such manipulations have to create such changes to (for example), hydraulic flow, redox conditions, pH, temperature, concentrations of specific nutrients or interactive chemicals. The latter chemical group may act in an inhibitory or stimulative manner upon the clogging microorganisms to cause trauma and detachment of some of the incumbent microbes. Clearly, the manipulation has to be selected taking into account the conditions within which the clog is forming.

By way of example, some reactive stressing scenarios are listed below for specific forms of operations (given in brackets):

(Routinely Pumped) - Do not pump for as long a time as may be conveniently achieved but not less than 4 hr or greater than 7 days.
(Routinely Flowing) - Seal off the well for a period of time that is reasonable achievable but for not less than 4 hr.
(Stagnant Non-Producing) - Pump the well for the equivalent of at least 100 well volumes; allow to sit for 24 hr and repeat.
(Deep Set Clog Formation) - For deeper set clogging, it may not be possible to use a flow on - flow off strategy to create stress and so a chemical input could be used. This may take the form of a disinfectant, dispersant, acid or alkaline chemical treatment or the application of a radical temperature shift upwards or downwards. In the former case, the temperature rise should exceed ambient by preferably 40^oC whereas in the latter case, the temperature should be lowered to close to, or within, the freezing range.

In each of these examples listed above, the incumbent microorganisms within the clog formation will become traumatised so that some of the organisms will be released from the clog to the free water flowing over the formation. When the pumping or other hydraulic activity is recommenced after the shock, these released microbes ("floaters") will move with the flow and be pumped or flushed out of the well. Given that detachment from the clog will occur throughout the traumatised part of the formation, the sequence in which these floaters appear out of the well will be a reflection of their position within the clogged zone. For example, floaters originating from the clog closest to or within the well water column would be flushed out of the well immediately there is a release of water from the well (eg, by pumping or opening the release valve). Floaters further back in the clog will appear sequentially as water continues to flow from the well. It is therefore possible to compute the approximate position of these various clog communities by the following formulation:

V_IC = P_T x P_R - (equation 1)

where V_IC is the volume of water flushed through to the initial recovery of the clog floater characteristic for that defined part of the clog; P_T is the time of pumping to that recovery; and P_R is the rate of pumping.

V_TC = P_T x P_R - (equation 2)

where V_TC is the volume of water flushed through to the final recovery of the clog floater characteristic for that defined part of the clog; P_T is the time of pumping to that recovery; and P_R is the rate of pumping.

V_CC = V_TC - V_IC - (equation 3)

where V_CC is the volume characterised by the present of specific floaters marking that particular part of the clog formation and is calculated as the difference in volume obtained between equations one and two. This volume would represent the water extracted from that part of the clogged formation where these particular microorganisms dominate.

This information gives some useful information on the size of the formation that would need to be treated in order to disrupt and disperse the clog formation and return the well to an acceptable production capacity. There are two methods in which this volume can be presented: (1) as a magnification factor of the well water volume; and (2) as a theoretical inside and outside radius projection based upon the axial midpoint of the well to the inner and outer edges of the particular part of the clog formation from which the characteristic floaters were determined.

Treatment Well Volume Projection. This may be calculated based upon the known well water volume (V_WW) when the well has not been subjected to any manipulable drawdown effects. The magnification factor for the zone to be treated would be from V_MIC to V_MTC using the formulae:

V_MIC = V_IC / V_WW - (equation 4), and
V_MTC = V_TC / V_WW - (equation 5)

where V_MIC represent the number of well water volumes that would have been flushed before the first floaters from that particular clog were observed (equation four); and V_MTC represents the number of well water volumes that would been flushed through the well before the last of the characteristic floaters was observed in the effluent water (equation five). The minimum volume of any treatment liquid that would need to be placed down hole to disrupt this particular part of the clog would be minimally need to be V_MTC. There is however a probability of tortuosity in the pathway that the treatment liquid is forced to move through and around the clogged formation target. To compensate for this the treatment liquid volume (V_TLV) should be:

V_TLV = V_MTC x 1.5 - (equation 6)

The 1.5 magnification factor helps to ensure that the targeted clog formation is treated together with any other clog formations occurring closer to the well than the targeted clog.

Treatment Well Radius Projection. While the practicality of well treatment is suited to the use of the volume of water in and around the well that requires treatment as the V_TLV, some well operators may also require to know the distance that the clog formation extends beyond the well itself. Such information may have value in deciding the location of new well installations within the same groundwater system. There may clearly be some benefits gained from installing new wells away from the zones of clogging extending out from existing wells. Another advantage of computing the radius is that a mental image of the size of the clog formation may be more easily comprehended.

To predict the size of the clog zone as a radius from the mid-point of the well, there are a number of characteristics that need to be considered. These may be summarized as follows:

V_WW = (R_W)² x Pi x L_W - (equation 7)

where the volume of the well water (V_WW) is calculated using the radius of the well (R_W) and the length of the water column in the well (L_W)while Pi is 3.142 ; and

V_DD = (R_W)² x Pi x L_DD - (equation 8)

where V_DD is the volume of water drawn down by the activation of pumping from the well and L_DD is the length of the drawdown into the well as a result of pumping; and

V_WA = V_WW - V_DD - (equation 9)

calculates the active water volume (V_WA) as the difference between the volumes of the water well column (equation seven) and the drawdown volume (equation eight).

The volume of water is considered equal to the void volume for simplicity in both the gravel pack (where present) and the porous aquifer medium around the well. The total volume occupancy of a clog will therefore occupy a greater dimensions depending upon the void volume (porosity) of these porous formations enveloping the well. The lower the porosity, the greater the distance away that the clog formation may occupy. These distances are calculated as mean radii from the axial midpoint of the well and are purely theoretical. This is particularly the case because the projection assumes an even flow of water into the well down the whole length of the screen and a perfectly cylindrical movement of water towards the well's epicenter. While this is almost always not the case, the computation of the data in this manner allows a theoretical comprehension of the planar form of the theoretical cylindrical clog formation in an averaged manner. To perform these calculations it is necessary to compute the total void volume of the gravel pack (V_GP):

V_GP = [(R_GP)² x Pi x L_S x P_GP] - [(R_W)² x Pi x L_S x P_GP] - (equation 12)

where R_GP is the radius of the gravel pack, R_W is the radius of the well itself, P_GP is the porosity of the gravel pack, and L_S is the length of the well screen.

Where the volume of water pumped (V_PEN) from the well to the event of note (eg, shift in the dominant flora; reduction in the flora measurable in orders of magnitude), it has been shown above that it is possible to compute the volume of occupancy in terms of the number of active water well volumes. To calculate V_PEN it is necessary to decide after what time of pumping (P_T) the water samples were observed to have a consistently different and lower microbial population. This population may be considered to be the "background" flora normally recovered from the water. Remember that it is rare for a groundwater sample to be absolutely devoid of active microbial cells (ie, sterile). The time at which the "background flora" is first recorded (T_PEN) is used to compute the extent that the clog formation has infested into the groundwater formation. The equation to be used depends on the V_PEN obtained using equation thirteen):

V_PEN = T_PEN + P_R - (equation 13).

There are constraints upon which formula should be used to compute the radius from the midpoint of the well to the event of note (R_PEN) depending upon the volume of water pumped before this event is observed in the samples. The various options are described below.

Clogging is restricted to the borehole when the V_PEN is less than or equal to the V_WA. In this event the R_PEN would be taken to be equal the radius of the well (R_W). Where V_PEN <= V_WA ,

R_PEN = R_W - (equation 14)

If clogging extends into the gravel pack but not beyond that zone, then the V_PEN has to be greater than the active water well volume (V_WA) and less than volume of water occupying the saturated gravel pack (V_GP). For this to be applied, the following formula is used:

where V_PEN > V_WA and V_PEN < V_WA + V_GP - (equation 15)
then V_EGP = V_PEN - V_WA - (equation 16)

Equation 15 determines that the V_PEN is in the gravel pack zone while equation 16 determines the volume of water within the gravel pack phase (V_EGP) by excluding the active well water. To calculate the radius to the termination of the clog event (R_PEN) identified by the V_PEN can now be determined:

R_PEN = SQRT [V_EGP / (Pi x L_S)] - (equation 17)

If the clogging extends through the gravel pack into the groundwater formation then:

V_PEN > (V_WA + V_GP) - (equation 18)

In the event of equation eighteen being validated, the formulation for calculating the radius to the termination of the clog event (R_PEN) would be based upon the prospect of the clog event being in the groundwater formation beyond the porous packing (eg, gravel pack) around the well. The formulation is based the generation of the R_PEN which extends to a greater value than the R_GP:

V_BGP = V_PEN - (V_WA + V_GP) - (equation 19)
V_MFM = V_GP x (P_GW / P_GP) - (equation 20)
V_TFM = V_BGP + V_MFM - (equation 21)
R_PEN = SQRT V_TFM / (Pi x L_S) - (equation 22)

The volume of the groundwater (V_BGP) pumped from the formation beyond the gravel pack is calculated as the difference between the total volume of water pumped before the clog event was recognized (V_PEN) and the amount of water known to have been extracted from the well collum and the gravel pack (equation 19). In order to calculate the R_PEN where it extends into the formation, there has to be a correction for the volume of water (ie, void spaces) so that a volume consistent to the porosity of the groundwater formation (P_GW) can be assumed. This is achieved in equation 20 by modifying the volume of water saturating the gravel pack (V_GP) to be the equivalent volume had the gravel pack had the same porosity as the groundwater formation. The resultant volume (V_TFM) would be the theoretical formation groundwater volume (V_TFM) if the formation porous medium extended to the well casing (see equation 21). From this assumption it now becomes possible to calculate the approximate radius of the observed clogging event from the well (R_PEN) using equation 22 on the basis of screen length (L_S) and the V_TFM.

Using the above equations it therefore becomes possible to predict the radius from the well to a particular clogging event depending upon the volume of water which passed out of the well before the significant circumstances diagnosed as relating to a specific form of clog was observed. Fundamentally from the equations given above, it is possible to calculate the R_PEN based on the volume of water pumped (V_PEN). If the clog is within the water well column then V_PEN < V_WW and it may be assumed that the clogging event is restricted to the water column of the well itself. If the V_PEN is validated by equation 15 as being larger than the well water volume but smaller than the combined volume of the active water column and the gravel pack void volume, then the R_PEN may be calculated using equation 17. Where equation 18 is validated and the water volume pumped is greater than the combined volumes of water column and gravel pack voids, then the clog is present in the formation and can be calculated using equation 22.

Of the wells in which the BAM predictive calculations has been performed, it has been noticed that the clog may vary in its characteristics as components of the clog are pumped from the well during the BAM sampling. These (microbiologically) different parts of the clog are known as biozones and perform different functions within the clog. The act of flushing the microorganisms from a stressed clog does give a sequential release of the various components from the clog but due to the inevitably unequal degrees with which the sloughing occurs from the clog, there is a "fuzziness" to the data obtained from the samples. This fuzziness is created by two events:

(1) the gradual lateral sloughing and shearing away from the biofilms forming parts of the clog closer to the well so that the organisms are sampled within samples taken later in the BAM procedure;
and (2) the slow sloughing from less stressed parts of a clog again causing delayed releases of the organisms to the water and arrival in the samples.

In general, in spite of the "fuzziness" which is almost inevitably generated through sampling a natural system, there are a sequence of zones which are most commonly observed around a typical water well in which a redox front has formed in some manner. The sequence involves the development of a number of distinctive biozones within the clog as the environmental conditions shift from oxidative to reductive with the distance from the well itself. These biozones range from the first ones actually within the wells water column to those within the gravel pack and others forming within the groundwater formation itself. Sequences commonly follow a pattern which is described below in order from the most oxidative (within the water column) to the most reductive (in the deeper parts of the formation).

Highly oxidative biozone (Eh > +150 millivolts). Dominated by various iron related bacteria and aerobic bacteria. Iron may be bioaccumulated within slimes which may be most observable at air:water interfaces (eg, the drawdown zone on the casings), as sediments within the column due to the higher density of some of the biomass due to iron accumulation, as suspended particles within the water column which are often stratified at particular depths within the column, and at focus sites at radical water tortuosity which may associated with the water passing through the slots in the well screen and pump filters as well as on the impellers in the pump. Sheathed iron related bacteria may dominate when the water is examined microscopically including some stalks of Gallionella if there is a moderate to high total iron content in the water. Where there is a significant available organic carbon nutrient availability, the bacteria may be dominated by pseudomonads.

Slightly oxidative biozone (Eh for +50 to +150 millivolts). Generally the iron bacteria are located in lower numbers in this biozone and Gallionella may be the most easily identifiable (by microscopic examination). There is often a greater presence of pseudomonad bacteria together with other bacteria capable of growing aerobically. The type of bacteria (non-IRB) may give some indication of the type of biofouling and clog formation that may be occurring. Where the pseudomonad bacteria dominate, there may be a significant concentration of a limited range of organic compounds such as might be present in the pollution event involving specific (narrow spectrum range) organics. These organics could originate from the accidental or unexpected releases of (for example) solvents, gasoline, and jet fuel. Broader spectrum organics may shift the microbial flora to a broader band of microorganisms which would include the enteric and gRAM positive bacteria. Where this broader spectrum of organics originates from degrading organic wastes (e.g., manure, septic waste, sewage) the enteric and the denitrifying bacteria are likely to become dominant in this biozone. Where these enteric bacteria dominate there is a greater probability of a positive total coliform test result since many of the enterics are able to produce gas under the test conditions. In the event of the organics being in low concentrations with a diverse range of compounds, the microflora becomes much less aggressive and there is a growing body of evidence which suggests that the gRAM positive bacteria may now dominate. Often covert slime formers may be noticed under these conditions (reaction one on a SLYM-BART^TM) but they are often not aggressive (ie, do not extend to a secondary reaction type, RPS limited to dd, RX1).

Transitional oxidative-reductive biozone. This may be the biozone with the greatest biomass and void volume occupancy. It has been observed that there is the greatest microbial activity at the redox front where the Eh shifts from +50 to -50 millivolts. The aerobic microbial activity will tend to shunt the redox to a more reductive phase while the microorganism will also be searching for alternative electron acceptors to oxygen. The most common alternative is nitrate and so waters which contain nitrates may tend to support a larger aerobic flora which is comprised of organisms able to respire using the nitrate instead of oxygen. Complex consortia (communities) of various species of micro-organisms may develop at these sites into complex stratified layers within the biofilms forming the clog. SRB almost always may be found as a part of the deeper strata in these biofilms. While the SRB may not be normally recovered from the production water, the act of stressing the clog may cause the releases of these organisms from these zones. Generally the pseudomonad bacteria do not dominate in this zone. The bacteria that are found are often slime formers and may include enteric and gRAM positive bacteria.

Mildly reductive biozone (Eh between -200 and -50 millivolts). The microflora in this zone tends to be more dispersed and less likely to rapidly cause an occlusive clog formation. One of the problems with analyzing water samples from this zone is that there is often some degree of fuzziness occurring due to the presence of late-sloughing microorganisms from the more oxidative regions of the clog; and the transient population of microorganisms which may be moving relatively passively (e.g., suspended animates, ultramicrobacteria) through the water. Populations recovered from this zones may be relatively non-aggressive and may, or may not, include SRB and methanogenic bacteria. Denitrifying bacteria may be recovered in greater numbers from these samples where there is a significant inorganic nitrogen content in the groundwater as nitrate and nitrite.

Radically reductive biozone (Eh < -200 millivolts). This is a stressed environment for many aerobic bacteria but the facultative and strictly anaerobic bacteria can flourish in a dispersed clog formation. Since very reductive conditions are more likely to occur deeper in the groundwater formations, the relative void volume to microbial biomass potential is likely to be much higher and so there is likely to be less biofilm formation within a given void volume when compared to the redox front or oxidative conditions. Generally these conditions are likely to be present in the deeper (and slower moving older) groundwaters where the expected level of microbial activity (given the restricted nutrient substrates) is likely to be much less. Studies conducted on these deeper groundwaters indicate that the microorganisms present at those depths are different from the surface dwelling microbiota. One major difference is that the time it takes for a microbial cell to divide into two cells (i.e., reproduce) extends into months, years, decades and even centuries rather than the minutes, hours and days commonly associated with the surface (aerobic-oxidative) microflora. It may be expected that samples taken from this type of groundwater may also be contaminated with microorganisms present in recharge waters and passive cells.

It is important in the BAM scenario to attempt to delineate the various biozones so that the V_MIN (volume of the water in and around the well that needs to be treated) and the R_PEN (radius of the clogged zone around the well) can be calculated and used in the rehabilitation of the well. Of primary interest is the V_MIN since this allows the management to assess the volume of water within the well, the gravel pack and the formation) that would need to be treated in order to (1) shock; (2) disrupt; ad (3) disperse the clog formation of concern to the ongoing operation of the well. Of secondary interest is the R_PEN (width, radius of the clogged zone around the well) which allows management and users to understand the extent to which the environment surrounding the afflicted well has become infested with the clog formation.

Relationship of Microbial Type to Biozone Occupancy. It has become evident that different types of microorganisms are recovered from the water samples taken from particular biozones which can aid in defining the form of clogging occurring within the well system. Below is listed the experiential evidence to-date as to the interpretation of the shifts in microbial types from sample-to-sample and this value of this in indicating the form of clogging.

Iron Related Bacteria (IRB). The most easily identified forms of IRB are the stalk (Gallionella) and sheathed formers (eg, Leptothrix, Crenothrix and Sphearotilus). These are strictly aerobic organisms are commonly associated with the oxidative biozones. All may grow within or attached to a slime formation (biofilm). The stalks of Gallionella may shear away from the cell and are easily identified in the water sample by the characteristic ribbon-like shape. Sheathed IRB will not be identified so easily unless the slime growths are dispersed releasing the characteristic sheaths to the water where they may be microscopically observed directly or in filtered material. Individual cells are more difficult to identify since they are often relatively "typical" gRAM negative rods with no obvious features which separate them from say the pseudomonad bacteria.

The other IRB bear the common characteristic of being able to bioaccumulate iron in the ferric form within or upon the cell, or in the slime surrounding the cell. These microorganisms are heterotrophic by and large and will naturally generate these ferric (brown) deposits under oxidative conditions. The yellow, orange, red and brown slimes so commonly associated with clog formation represent biozones which are either transitional or oxidative. Under reductive conditions the IRB may be present but are generally not so aggressive (and therefore able to compete) with other microbial types which can flourish under these conditions.

Sulfate Reducing Bacteria (SRB). These bacteria are strictly anaerobic and are indeed very sensitive to the presence of oxygen. They therefore tend to be very active in transitional and reductive biozones but also have the ability to grow in the deeper strata of biofilms (slimes) which have formed under oxidative conditions. These deeper strata are reductive since the oxygen has been totally utilised by the microorganisms in the upper strata of the slimes. A particular concern with SRB is the fact that these microbes generate hydrogen sulfide (H₂S) as a normal product of growth from sulfates or sulfur. Hydrogen sulfide is a well known initiator of corrosive processes in various metals. In addition H₂S also is the generator of the "rotten egg" odor which is a frequent problem with water supplies where reductive conditions prevail together with some level of organic carbon sources.

Clog biozones dominated by SRB will normally be black or grey. Where the growth is suspended in the water, the water may appear to be grey with black specks in it and a black deposit at the bottom of the water column. Corrosion may take the form of pitting with associated embrittlement of the metal structure. Over the pitting will be an active microbial slime formation which will be consortial form and include the SRB. Over this pit slime will often be some hardened bioaccumulates of various metals (often dominated by iron) in the form of a tuberculous shell. This tubercle may often have the vague appearance of a bubble which may be ellipsoidal in shape. Pit corrosion initiated by the SRB can often lead to a localised perforation of the metal structure which then compromises the integrity of the structure itself with significant after-event costs.

"Rotten egg" odors are one of the more common taste and odor complaints levelled at well waters. These events occur when the transitional and reductive biozones are close to and/or dominate the clogging activity. The problems are magnified when the water is passive for prolonged periods of time. This causes the conditions in the well to become more reductive and stagnant. Under these conditions the hydrogen sulfide produced by the SRB activities may tend to accumulate reaching higher (and detectable) concentrations while at the same time the system is saturated with sulfides of various forms often causing various forms of blackening.

Slime Forming Bacteria. A wide variety of bacteria are able to form slimes under various conditions. These slimes may range from covert (survival) forms which do not necessarily contribute to the clogging process itself, to consortial radical slime formers which often grow in association with IRB (oxidative conditions) or other heterotrophic bacteria. In general, slime formers are commonly associated with the IRB clogging in the oxidative regimes but they also found in transitional and reductive biozones. In the latter biozones, the slime formers may be covert (generating deep set slow growing slimes) or dispersive (generating fragile slime formations which disperse easily).

Gas Generating Bacteria. Many bacteria are able to generate significant volumes of gas during a biofouling process such as clogging. This gas may often becoming entrapped as bubbles constrained by biofilms (as the "walls" of the bubbles). If these gas bubbles become constrained within the void spaces by these slime formations, this becomes an occlusive (plugging) event which may magnify the impact of the clogging on the hydraulic transmissivity through the porous media. Radical clogging may be a reflection of gas plugging particularly where there is an equally sudden relief from the clogging. A number of different types of microorganisms can generate gas in this significant manner. These are described below and in general gas generation occurs most frequently under reductive and transitional redox conditions.

Methanogenic bacteria. These are anaerobic bacteria which generate methane as a part of their metabolism. They are very sensitive to the presence of oxygen which is toxic and are also antagonistic with the SRB. Generally the methanogens are recovered from more reductive conditions than the SRB. The gas generated will be a mixture of methane (50 - 95%), carbon dioxide (5 - 50%), hydrogen (0.1 - 1%) and nitrogen (0.1 - 5%). The exact composition will be dependent upon the temperature of the environment (the lower the temperature below 20^oC, the higher the methane content of the gas) and the physico-chemical conditions of the environment. Under oxidative conditions the methane is rapidly degraded by a group of bacteria which are able to degrade methane (methanotrophic bacteria). Methanogenic bacteria commonly generate the methane from either the shorter chained organic acids but can also generate methane from hydrogen and carbon dioxide. Where methane appears in a water column and is from methanogenic origin (rather than a venting of natural gases from deeper in the earth's crust) it may be extrapolated that the well is probably in a reductive state with very little or no oxidative biozones and a weak transitional zone.

Denitrifying bacteria. Under anaerobic conditions waters rich in inorganic nitrogen (in particular, nitrate and nitrite) are likely to be heavily infested with denitrifying bacteria. These bacteria are able to reduce the nitrate through nitrite (denitrification) to gaseous (di)nitrogen gas (complete denitrification). Denitrification involves heterotrophic activity and so organic carbon is an important component in the activity. Where the groundwater is contaminated with wastewaters containing organic nitrogenous materials originating from such sources as leachages from manure, septic tanks and compromised sewage pipes. These denitrifying bacteria tend to be most commonly found in the reductive and transitional biozones. Nitrogen, where it forms into bioentrapped bubbles within the void spaces, is relatively insoluble in water and is readily utilized by the biological activities within a clog formation. Consequently generated plugging caused by biogenerated nitrogen tends to be persistent.

Fermentative bacteria. A range of bacteria are able to ferment organic materials with the generation of gases. This mostly occurs under anaerobic (reductive) conditions with the dominant gas being carbon dioxide and hydrogen is also generated on some occasions. Carbon dioxide has a solubility which is pH dependent. Under more alkaline conditions (pH > 8.5) the gas may shift to solubilised forms as bicarbonates or crystallised forms as carbonates. In the latter event, the carbonates may become integrated into the clog as a structural component. Under acidic conditions (pH < 6.5) the carbon dioxide may remain insoluble and create a gas bubble occlusion of the void spaces. This form of plugging may be transient. In general, fermentative bacteria may become dominant in the transitional redox front biozone and are often recovered concurrently with the presence of SRB. In the IRB-BART^TM, a reaction five is taken to be an indication of the presence of gas generating bacteria and, in particular, the fermentative bacteria.

Pseudomonad bacteria. These are comprised of a large and diverse group of gRAM negative strictly aerobic bacteria (Section four in Bergey's Manual). These bacteria often dominate in oxidative biozones where there are a restricted range of organic carbon compounds present in concentrations of greater than 2 ppm TOC. This restricted range may be the result of the pollution of the groundwater with a specific organic pollutant such as jet fuel, gasoline, or solvents. In hazardous waste rehabilitation processes which are being managed under oxidative regimes, various pseudomonad bacteria often dominate as the principal biodegraders of the organic pollutants of concern. While many of the pseudomonad bacteria can only function under oxidative regimes in which oxygen is the primary source of respiration, some can function under reductive regimes using nitrate as an alternate electron acceptor (nitrate respirers). Pseudomonads are therefore very common members of oxidative biozones. Surprisingly perhaps, pseudomonads are often recovered from mildly reductive biozones where there are a narrow spectrum of organic carbon sources and alternate electron acceptors are present. There are three major types of slime formation which can be observed in the pseudomonad bacteria. These are:

(1) covert dense slime formers;
(2) thread-like slime formers;
and (3) consortial slime formers.

It must be remembered that not all pseudomonads will form a slime but those involved in clog formation commonly do.

Covert dense slime forming pseudomonads grow tightly to surfaces and do not necessarily slough easily when stressed. Thread-like slime formation occurs when the pseudomonad bacteria form web-like, string-like or floating floc-like growths within the void spaces of the porous media. In well matured clog formations, thickening strands may be observed connecting the different biofilms attached to the various surfaces within the medium. These strands can materially reduce the hydraulic conductivity in the clogged formation. Floating flocs may also be seen floating on the air:water interface or as suspended plate-like growths within the water column of a well. These flocs are formed by the pseudomonad bacteria anchoring the cells together with a fine series of polymeric strands of slime. Once anchored together, the cells become integrated into a floc which may then float on the surface of the water or at specific depths in the water column. These flocs are most likely to be observed in the oxidative biozones.

Pseudomonad bacteria are very commonly found to consortial partners alone with other microorganisms in the various stratified biofilms which form into the clog biomass. They particularly occur in the more oxidative (upper) strata and are often associated with deeper strata (reductive) where the SRB are present.

Enteric bacteria. These bacteria are also gRAM negative but are facultative anaerobes. In other words these bacteria are able to grow in the presence or absence of oxygen and so are more able to grow under reductive conditions. The enteric bacteria are most commonly associated with their occurrence in sewerage, septic waste, manures and fecal material. Features that differentiate the enteric bacteria is that they are almost all denitrifying bacteria, are fermentative and able to generate gas from glucose and often from lactose. Of particular concern is the tribe one enterics which include some of the species which can cause either typhoid, bacterial food infections, dysentery or gastroenteritis. The indicator microorganism for the presence of the tribe one enterics is Escherichia coli, the fecal coliform. The presence of this species in a water is taken to indicate a hygiene risk to the potential users. There is some concern that the tribe two enterics (including the genera Enterobacter, Klebsiella, and Serratia) can be found as components in clog formations on some occasions and form a part of the total coliform group which is used as a broader spectrum indicator of hygiene risk. Generally these enteric bacteria become significant components in a clog formation where there has been a significant passage of organic carbonaceous and nitrogenous materials through the clog formation.

Fungi and Mycelial bacteria. There is a wide range of microorganisms which grow over the surfaces as radiating threads. These growths are known as mycelial and can extend for considerable distances through voids when attached to the surfaces. The frequent constraint on these fungi and mycelial bacteria is that they require (ideally) a radically oxidative semi-saturated porous medium to grow optimally. Such conditions do not exist in active well but may exist above the well where water is recharging into the groundwater formation. While these mycelial growths can clog the recharge zone (particularly when there is a high nutrient load in that recharge water), a major concern is that these mycelium forming fungi and bacteria can generate spores (small cells capable of surviving for long periods in hostile environments). These spores can move through the groundwater formation and be recoverable when cultured using the correct techniques. If the groundwater enters a radically oxidative semi-saturated environment, the fungi and mycelial bacteria may become dominant. One impact of this is that a "woolly" type of growth may be generated earthy-musty (geosmin) odors.

Microbiology summary. There a wide variety of microorganisms which can infest the formations associated with the operation of a water well of any type. Each microorganism type has its own environmental constraints which renders a clog event complex involving a multiplicity of different niches each with its own unique consortial components. It therefore becomes difficult to analyze a water to the full extent of science and continue to obtain useful information with regards to the site, composition and control of a clogged formation. This is an instance of "no one size fits all" and a generalist approach recognizing the global vulnerability of the microorganisms fouling up the formation may prove to be a more practical approach.

Physical Determination of Clogging
One universal symptom of clogging is the loss of production capacity by the afflicted well. This may or may not be first easily observable feature depending upon the manner in which the well is being managed. There are a number of symptoms which may be recognized as being relatable to clogging and these are discussed below. It should be remembered that the impact of clogging on production capacity is due to the reduction in the useable void volume for the conductance of water at some point between the (source) groundwater flow and the wells water demand (created by either pumping or hydraulic pressure release requirement). This reduction in void volume could be due to the action of "silting" in which relatively inanimate particulates collect at a particular site relative to the well that the production capacity is diminished. It has been traditionally thought that such silting was purely a physical phenomenon and did not involve any microbiological activity. Research in the last two decades has shown that these processes may have a significant if not dominant microbiological component.

Drawdown. As a water well experiences significant clogging, there is an increase in the drawing down of the water level as the pumping cycle is initiated. Once a balance has been established between the pressures created by groundwater flowing into the well and the water column head, the drawdown stabilises at a particular level.

Over time, the net effect of clogging on this type of well is to disrupt that pressure balance and destabilize the draw-down. As clogging proceeds to reduce the potential flow rates of groundwater into the well, so there is a compensatory increase in draw-down. In other words in a clogging system, a well being pumped to a constant drawdown will gradually pump less and less water as the production capacity goes down. If the well is being pumped to a constant flow then the drawdown will become continuously magnified as the clogging proceeds. To determine whether a well is indeed clogging, the former scenario of pumping to a constant drawdown allows the formation of the clog to be monitored. The disadvantage of this technique is that the production capacity of the well becomes directly dependent on the state of clogging. However the daily production rate from the well becomes a clear signal as to the state of clogging in that well. It used to be thought that clogging was a relatively linear (first order) event but there is now a growing body of evidence which suggests that clogging is a complex interactive series of events involving harmonic (cyclic) components associated with biological activities as well as linearized chemical activities. Frequently the daily production from clogging wells which are drawing to maximum capacity from a stable drawdown reflect these events. Cyclic events may be reflected in minor shifts (5 - 15%) in the production capacity as the clogs moves through expansion (decrease in production), sloughing (increase in production) and stabilisation (constant production).

Relief wells function differently and a measurement of drawdown may not be appropriate since the hydrostatic pressure being created in the well is indirectly controlled by the water level in the confined water body (eg, reservoir, lake, canal). The net effect of this may be that the passive flow from the relief well should be directly relatable to the head pressure created in the confined water body. In other words the greater the height of the water level in that body of water, the greater the flow should be out of the well. Clogging may be expected to be more complex around a relief well in part because a series of redox fronts may become established alone the conduit pathway from the confined water to the relief well. The net effect is however that flow patterns from the relief wells shift from simply reflecting the water level in the confined body of water to reflecting the hydraulic conductivity of water through the various clog zones which may have formed. There are two potential serious net effects of this clog induced reduced conductivity. First, the flow of water from the relief wells becomes reduced and no longer can be clearly correlated with the confined body water level. Second, the relief wells now become "isolated" structures separated by the clogging. This loss in connectivity between the relief wells causes radical variations to occur sometimes between the different relief wells functioning in a common environment. One of the most salient features of a successful rehabilitation of clogged relief wells is the recovery of a hydraulic connectivity between the wells so that the flows are generated in unison with each other and the water level in the confined body of water.

Pump Rate (Impaired Production Capacity). As clogging occurs, the flow of water to the well becomes reduced. One symptom commonly recognised as being indicative of clogging is the dramatic loss in pumping capacity. In other words, the pump suddenly begins to pump "air" and has to be throttled back in order to continue to produce water. One of the unfortunate components in using this phenomenon to assess clogging is that this may only be a satisfactory observational tool when the pump rate (P_R) is equivalent to the production capacity (P_C):

P_R = P_C

When this is the case, any loss in the production capacity due to clogging is immediately registerable through an inability in the well to provide water at the original pump rate. Generally the pump rate is set as a fraction (F_PC) of the production capacity and herein lies the problem.

P_RS = P_C x F_PC

where P_RS is the pump rate selected, P_C is the production capacity of the well, and F_PC is that fraction of the production capacity selected for the production activity. F_PC has to be less than one (F_PC < 1.0) and the smaller the fraction, the less the demand being made on the production capacity of the well. While this would appear to be a sensible option for creating less stress on the well, there is the unfortunate side-effect that the earlier stages of clogging may pass by unrecognized. For example let us take a well (A) in which a reduced production activity was designed with F_PC being set at 0.5. The well is then operating at 50% of its production capacity and for the impact of clogging to be observed on production, the clogging would have to reduce the flow of groundwater into the well by 50%. Now let a second well (B) be designed to operate with an F_PC of 0.1. This would mean that the well is operating on only 10% of the production capacity maximum for that well and so clogging would have to reduce the water flow by 90% before evidence of clogging would be observable on the water production rate. In the case of well A, the observation of clogging would be quite sudden since wells which have become subjected to a 50% loss in production capacity are in a very degenerative state and occlusive clogging ("sealing" of the surrounding formation around the well) is a significant possibility but remedial action is still very possible. For well B, the well would have to have been subjected to a 90% loss in production capacity before the pumping activity becomes affected. A well which is 90% clogged is essentially occluded and rehabilitation is that much more challenging since the majority of the void spaces in the clogged zones are either filled with clog material or isolated from the conductive water pathways by the clogs. As an observational mechanism it is therefore more convenient to operate a well close to its production capacity (eg, F_PC within the range of 0.65 to 0.85). The larger the F_PC value used, the sooner the clogging phenomenon becomes observable and therefore relatively speaking easier to treat.

Reduction-Oxidation Potential (Redox). In the last decade it has become increasingly common to find microbial activities focussing over narrow ranges of redox values. Maximal biomass is sometimes created when the redox potential is between -50 and +150 millivolts. Clogging can have an impact on the redox potential for groundwaters recovered from biofouled wells. In general, the net effect of biologically induced clogging is to stabilize the redox potential so that the front (shift point between oxidative and reductive conditions). A deep clog formation will therefore in general have production waters from the well which are moderately oxidative (eg, redox values of +20 to +200 millivolts). If, however, the clog formations are close to the well casing and there is a considerable level of stratified and attached slime formation in the well itself, the redox values may be lower in the range from -100 to +10 millivolts. Where a clogging is forming but has not yet stabilised, the redox potentials from the production water may vary considerably from sample to sample and from time to time.

After a well has been subjected a successful rehabilitation back to the original production capacities, the redox values will again stabilize to reflect the reductive state of the groundwater formation and the degree of oxidation being imposed by the intrusion of the well into the formation. This intrusion acts in part as a conduit for oxygen into the formation via the wells water column and through the fracturing of the formation around the well; and shifts the hydraulic transmissivity through the impacted groundwater formation. Such effects combine to focus the initial microbial fouling of the structure. Providing there are no radical changes in the formation, this initial fouling will grow into a clogging over time. The redox potential in the product water will reflect these stages and so may vary somewhat. As a general rule, a redox potential of less than +50 millivolts in the product water would suggest that there is likely to be either clogging close to, or inside the wells water column and there is a potential for downstream clogging from the reduced chemicals still being carried in the water. If the redox potential is over +100 millivolts, it is probable that any focussed clogging would be within the porous packings and outside in the groundwater formations around the well. A reductive redox potential (eg, less than +10 millivolts) does not preclude the possibility of extensive (anaerobic) clogging. Many drainage systems under sanitary landfill operations can become severely clogged (compromised) under extremely reductive conditions but these conditions are not likely to be so pronounced in a producing water or relief well.

pH. The relative acidity or alkalinity of the groundwater can have an impact on the degree of clogging. Generally, clogging is not badly impaired when the water pH is between 7.5 and 8.5 and indeed the biofilms within the clogging zone do have a capacity to buffer the pH into this mildly alkaline range.

Return to BIOFOULING/CLOGGING MENU
Return to Water Wells Menu
Return to MAIN MENU