Russian River Pathogen Project (RRPP) Working Draft Report

Updated 01/28/2008

1. Summary

The goal of the project is to design the most efficient forensic and real-time monitoring and evaluation program for the Russian River watershed to identify: 1) potential primary sources of potentially pathogenic fecal bacteria (e.g., dairies, municipal waste/runoff, recreational sites), 2) potential secondary sources (e.g., benthic sediments, periphyton, side channels, stock ponds), and 3) contributing factors (e.g., agricultural fertilizer, managed flows, water temperature, available organic material). The proposed evaluation program will resolve spatio-temporal issues, problems with source identification, potential load allocations, and factors that contribute to in-stream exacerbation or remediation of the problem. The program will be designed to support TMDL planning and implementation.

Identification of pathogen sources, transport mechanisms and pathways, and storage sites in a complex watershed requires concurrent scientific assessment of many physical, chemical, and biological characteristics of the watershed. Currently in the Russian River watershed there is incomplete understanding of the actual sources of potentially-pathogenic fecal organisms and pathways for exposure of local populations. The scientific literature provides some coverage of the relevant issues in the Russian River watershed. Critical research issues in the technical and scientific literature include: 1) identifying major source areas and types of fecal matter sources during different seasons, 2) transport mechanisms and processes in surface and sub-surface waters, 3) extent of contribution and locations of bacterial hold-over and storage from winter storm runoff to summer loads, 4) identification of vertebrate host species for pathogenic organisms in various places and times, 5) relationship between environmental factors and measured bacterial concentrations, 6) use and identification of appropriate indicator fecal bacteria species and strains, and 7) whether a testable model (conceptual and physical) can be created that explains bacterial conditions.

There are 3 main areas of investigation that should be pursued in the pre-TMDL investigation: 1) spatio-temporal localization of sources and hold-over bacterial reservoirs throughout the watershed; 2) appropriate identification techniques for index organisms (including identification method and host-strain relationships); and 3) adequate monitoring program for monitoring TMDL implementation effectiveness (including sampling frequencies, locations, and index organisms). These investigations can be feasibly pursued for a normal water-year with a bacteria monitoring program that includes a subsequent data analysis and interpretation phase.

2. Background

2.1 Basin Conditions
Contact recreation (REC-1) is the primary waterway use that is affected by the presence of high concentrations of fecal organisms in the Russian River and its tributaries. The Basin Plan describes the numeric standard for fecal coliform bacteria as: “Median 30-day levels (based on a minimum of 5 samples/30 days) should not exceed 50 MPN/100 ml and that no more than 10% of those samples should exceed 400 MPN/100 ml.” The California Department of Public Health (formerly Department of Health Services) has the following guidelines for single samples taken at freshwater beaches: Fecal Coliform <400 MPN/100 ml; Enterococcus <61 MPN/100 ml; E. coli <235 MPN/100 ml. Drinking water quality in the Russian River watershed has not been identified as a beneficial use impaired by fecal bacterial contamination of surface water. However, ground-water used for drinking water can become contaminated by surface waters containing fecal material (Olsen et al., 2002)[1].

The numeric standards for recreational contact must be met be any means available to the Regional Board for waters within the Basin. The Russian River and several major tributaries are used for contact recreation and at certain times of the year, exceed these numeric standards. This resulted in 303(d) listing and initiation of TMDL development for the Russian River main-stem and Santa Rosa Creek.

2.2 Potential Russian River watershed sources

There are several major land-uses and point sources that could contribute fecal matter into tributaries of the Russian River. These include dairy and livestock operations, wastewater treatment facilities, sewer lines, septic systems, municipal area runoff, and manure applications on agricultural fields. Previously-published research has identified all of these sources as potential contributors of fecal material to waterways. These land-uses will serve as inputs at certain times of the year and/or when facilities and best management practices are failing.

Farming involving raising of cows and sheep, and/or application of manure on fields is likely to result in increased fecal bacteria into waterways due to overland flow and poor management of waste material. Many of the sub-watersheds in the southern end of the watershed have agriculture as the largest or one of the largest land-uses in the sub-watershed.

Rural residential development often is accompanied by on-site septic systems that vary in age and efficiency of capture and processing of input material. If and when these systems are over-whelmed or as they age, fecal matter may enter surface and ground waters.

Urban area stormwater runoff can contribute very high loads of fecal bacteria to waterways (Salmore et al., 2006), as can wastewater treatment plant effluent. In Maryland, urban areas are considered the primary contributors to waterway fecal contamination in mixed land-use (agriculture and urban) watersheds (Wickham et al., 2006). In one study in the Southern Appalachians, a stream contaminated with fecal bacteria while running through an urban area, became less contaminated once it ran through National Forest lands (Clinton and Vose, 2006).

2.3 Developing TMDLs

There is existing guidance for TMDL development from the USEPA (citation). This guidance and example TMDLs are instructional for the initial studies phase of TMDL development.

Tomales Bay

In this TMDL, threats from fecal organisms to recreational contact (REC-1 and REC-2) and shellfish harvesting were the basis for the TMDL. Sources of problems included human and domestic animal waste entry into tributaries to the Bay. The detection methods for indicator bacteria were restricted to growth media-based approaches, which quantify the culturable fecal bacteria. Several potential pathogens were examined – Salmonella, coliphage, and E. coli H:O157. Some correlations were made in the background studies based on spatial or temporal proximity of problems (high concentrations) with particular tributaries or storm events. There were fecal bacterial concentration variations by tributary, predominant upstream land-use, and season. Human inputs to the system were thought to be primarily from failing septic systems, municipal runoff, and possibly wastewater treatment plants. Animal inputs to the system were thought to be primarily from domestic animals – grazing, dairies, and equestrian facilities. Hydro-dynamic modeling was conducted to target input tributaries for potential inputs to the Bay violating the SHEL standard. Finally, TMDL implementation was to include continuing E. coli/fecal coliform monitoring with medium spatial resolution (30 sites) and temporal resolution (weekly/monthly and event).

Napa River

In this case, fecal coliform and Enterococci bacteria presented threats to REC-1 and REC-2 uses and were the basis of the TMDL. Sources of problems were much the same as in the Tomales Bay watershed, including exceedance of standards from human and domestic animal sources. Detection was conducted using growth media.
The Kendall Tau statistic was used in correlation analysis to relate wet season E. coli concentrations to various urban area parameters and dry season E. coli to % agriculture. In addition, most of the wet season delivery of fecal bacteria was through surface water pathways and in the dry season through ground-water pathways. Human inputs to the system were determined to be primarily from failing or inadequate septic systems, sewer lines, municipal runoff (thought to be very important), and possibly wastewater treatment plants. Animal inputs were primarily from grazing lands and confined animal feeding operations. The TMDL was based on the REC-1 standard as the use most likely to be impaired.

Charles River, Massachusetts

This TMDL focused on fecal coliform and Enterococcus bacteria in a forested and residential watershed. Tributary watersheds/waterways were rated (e.g., Class A) according to goals for fecal bacterial concentrations. Inputs were thought to be primarily from various human waste disposal mechanisms, urban/storm-water runoff, domestic animals, and wildlife. Upstream areas are more contaminated than downstream and concentrations have generally been declining since 1989. The TMDL included river segment-specific potential causes in dry and wet seasons. Generally, there was a correlation between level of development (pristine to urban/agricultural) and E. coli concentrations. Single family residential areas tended to have higher concentrations than commercial and industrial areas. Finally, wet season concentrations tended to be higher than dry season. Human causes of inputs in the dry season were agriculture, failing septic and sewer systems, and illicit connection of sewer to storm drains. Human inputs in the wet season were domestic animal waste, storm-water runoff, and sewer overflows. Animal causes of inputs were minor inputs from wildlife and some input from domestic animals in the dry season. One statistical analysis in the baseline study was correlation analysis of fecal coliform concentrations and rainfall (using Pearson’s R and Spearman Rank Order Correlation). Current monitoring under this TMDL consists of “gap-filling” and measuring effectiveness of control measures and BMPs. Modeling using MIKE-21 is used to study actual and potential benefits from BMPs.

2.4 Bacterial sampling in waterways

The most commonly used indicator bacteria for fecal matter inputs into the environment are E. coli and Enterococcus sp. The most common sampling procedures, and the ones used in the Russian River watershed, are to take grab samples below the surface of a water-body, briefly store and transport the sample on ice, and grow culturable bacteria in challenging growth conditions in order to isolate the intestinal bacteria E. coli and/or Enterococcus sp. Some studies have sampled other media available for fecal organismal storage and growth, including river-bank soils, benthic sediments, landscape soils, and benthic macro-algae. E. coli and Enterococcus are able to survive and sometimes grow in environmental compartments (e.g., benthic sediments and algae), making their use as a fecal matter indicator more challenging (Anderson et al., 2005; Ishii et al., 2006; Power et al., 2005; Whitman et al., 2003; Whitman et al., 2006). E. coli concentrations can also vary in the same waterbody diurnally (citation) and by depth of sampling (Kleinheinz et al., 2006). Although there are short-comings to using E. coli as an indicator bacterium published in the scientific literature, indicator bacteria E. coli and Enterococcus sp. have been enumerated using the Colilert © and Enterolert © approaches, respectively.

An important study in the region of the Russian River to consider is one conducted by UC Davis and other scientists for the Regional Board (Atwill et al., 2007). In this study, the authors describe appropriate fermentation and chromogenic techniques for enumerating fecal bacteria. They also describe sampling regime considerations that they measured to impact fecal bacterial concentrations. These included a non-linear dependence on

antecedent rainfall, a correlative relationship of benthic fecal bacteria with fine sediments, and a linear relationship with depth of sampling.

2.4 Bacterial identification

Currently, fecal coliform bacteria, Escherichia coli, and Enterococcus sp. concentrations are all used as indicators of fecal matter input. This approach is widely used, but is also regularly criticized in the scientific literature (Anderson et al., 2005; Ishii et al., 2006; Power et al., 2005; Whitman et al., 2003; Whitman et al., 2006) for various reasons.

One critical element to using an indicator bacterium for fecal matter input is to be able to identify the host species for the input in order to manage the input. This is not possible using Colilert © or Enterolert © types of approaches because they are not strain specific. In many studies, this approach is still used and links are made to potential host-animal inputs based on the likely pathway for fecal matter (e,g,. agricultural area runoff vs. septic system).

Most fecal bacterial detection systems (e.g., Colilert ©) depend on the culturability of the bacteria; however, not all viable fecal bacterial cells in the environment are culturable and therefore can’t be detected by conventional methods (Awais et al., 2006).

2.5 Die-off rates and watershed loading

“Typically, conditions favorable to the survival of pathogens in water are lower amounts of light energy, lower salinity, elevated levels of nutrients and organic matter, and lower

temperatures.” (EPA guidance manual)

The rate at which bacteria and other potentially pathogenic micro-organisms die in various environmental conditions is important in understanding the fates and potential sources of these micro-organisms. Microbes living in feces may survive hours to months in the various receiving media (soils, water column, benthic sediments). The faster these microbes die, the less risk they pose in to human health. The slower they die and if they can grow outside host animals (Desmarais et al., 2002; Solo-Gabriele et al., 2000), the greater risk they pose to human health. Oocysts (dormant early life stage) of Cryptosporidium and Giardia lamblia can survive for 2 to 6 months in river water at cold and ambient temperatures (Medema et al., 1997; Adam, 1991; Bingham et al., 1979). Temperature is apparently the major limiting factor for virus and coliform bacteria survival in soils, with an estimated doubling of the die-off rate for each 10 oC rise (Gerba and Bitton, 1984; Reddy et al., 1981; Sampson et al., 2006). Temperature is also the dominant factor affecting virus survival in freshwater, with greater survival occurring at lower temperatures. Enteric viruses can survive from 2 to more than 188 days in freshwater (Novotny and Olem, 1994). In addition, different strains of fecal coliform bacteria may survive at different rates outside of the host organism and the distribution of bacterial strains initially present in fecal matter changes over time in the environment (Anderson et al., 2005).

Die-off, or decay, rates for Enterococcus sp., fecal coliform, and E. coli specifically are used in TMDL studies and planning and vary among TMDLs. The Charles River TMDL (see above) assumes a fecal coliform die-off rate of -0.6 day-1. Other rates range from -0.5 day-1 (Tomales Bay TMDL) to >>-0.5 day-1 (USEPA, 1985). Die-off rates in the literature vary depending on the culture/receiving medium, with fecal coliform bacteria in sediments tending to have lower rates (-0.02 day-1) than water (-0.24 day-1; Anderson et al., 2005). In comparison, Enterococci sp. have much higher die-off rates in sediment (-0.22 day-1) and water (-0.73 day-1; Anderson et al., 2005). Without knowing the actual rate of die-off of bacteria and viruses in a particular water-body, any modeling of fates and potential risks will be speculative. Loading of fecal bacteria and viruses in various parts of a watershed and delivery to receiving waters depends on the combination of die-off rates and environmental conditions conducive to survival and sometimes growth of pathogens. These environmental conditions can determine whether fecal matter entering the water-body will die and no longer pose a risk, or survive and grow.

2.6 Reservoirs for Bacteria and Secondary Growth

Once bacteria and other fecal organisms enter the environment they may survive and even grow in media that have the right physical and chemical conditions. These media include benthic sediments, benthic periphyton (e.g., attached filamentous green algae), and riverbank/floodplain soils where survival rates may be greater than in the water column (Sherer et al., 1992; Burton et al., 1987; Thomann and Mueller, 1987; Whitman et al., 2003; Hoyer et al., 2006; Sampson et al., 2006; Whitman et al., 2006). In waterways near and including the Russian River, investigators have found that fecal bacteria, including E. coli, appear to be deposited during wet season flows, along with fine sediments (atwill et al., 2007). Pathogenic organisms, including indicator bacteria like E. coli may have increased survival times if they are protected from sunlight and extreme temperatures. Enteric and pathogenic bacteria and viruses may survive for months in benthic sediments, increasing the chance of resuspension and health impacts (Burton et al., 1987; LaBelle and Gerba, 1980; Roper and Marshall, 1979; Burton et al., 1987; and Sherer et al., 1992). In some cases, resuspension of sediments can result in higher measured concentrations of fecal bacteria than municipal outfalls to recreational beach areas (Noble et al., 2006).

Survival of indicator fecal organisms in the environment, differential survival among strains, and even subsequent bacterial growth reduces their utility as indicators of primary fecal matter input. If E. coli can survive for months in media downstream from the initial input, then connecting the detection of E. coli months later and miles downstream to the primary input can be very challenging (see Anderson et al., 2005 for related discussion). This does not mean that E. coli have no role in fecal matter detection, but rather detection of primary feces inputs using E. coli must be proximal in time to the initial input. Detection of potential secondary storage and growth sites for E. coli and other fecal matter organisms is a critical element of determining extent and importance of inputs.

2.7 Nitrate and oxygen isotopes as indicators of fecal matter input

Bacteria are not the only indicator of fecal matter inputs to streams.


3. Sampling and identification regime

Pathogen species and strains need to be identified to a resolution that permits 1) identification of sources/potential sources and 2) adequate protection of public health.

3.1 Current sampling regime

The Regional Board has historically focused on swimming beaches along the middle and lower reaches of the Russian River. Other programs, conducted by the CCWI, SCWA, and Russian River-keeper monitor bacteria in tributary waterways and watersheds. The Regional Board has used this sampling program to monitor compliance with the Basin Plan and not to directly measure recreational exposure or forensically identify sources of contamination.

In virtually all cases, the sampled medium was the water column. In exceptional cases, including one special study for this report, other media have been studied.

Sample collection has been through grab sampling using sterilized 100 ml bacteria bottles. Samples are stored on ice for up to 4 hours, until delivery to the laboratory for analysis. Prior to 2002, concentrations of total and fecal coliform bacteria were determined using the multiple tube fermentation method; from 2002 onward, total coliform and E. coli concentrations were determined using the Colilert © Quanti-Tray method. In 2006, Enterococcus concentrations were also determined using a similar method – Enterolert ©.

3.2 Sampling Intensity

Sampling and identification of fecal coliform bacteria (including E. coli) have occurred irregularly over the last decade (Figure 5) and unevenly over the watershed (Figure 4). Temporally, sampling has been throughout the year, with the most focus on summer sampling along the Russian River and wet season sampling in tributaries near urban areas.

4. Bacteria Identification Methods

4.1 Identification using permissive growth media and labeling approaches

Contemporary, commonly-used methods for measuring concentrations of the fecal indicator bacteria E. coli and Enterococcus sp. rely on culturing bacteria obtained in grab samples to allow their detection. These approaches are valid for measuring the concentrations of live, culturable E. coli or Enterococcus, but don’t permit strain identification or measuring the concentrations of live, but non-culturable bacteria.

Bonadonna et al 2006

4.2 Identification using genetic approaches

There are several DNA-based approaches to detecting the presence of fecal organisms, as well as methods using antibodies to detect bacteria-specific proteins.

The DNA approaches can be lumped into the following categories: polymerase chain reaction (PCR) amplification of DNA using specific DNA primer sequences (e.g., Al-Ajmi et al., 2006); ribosomal RNA sequence detection (“Ribotyping”, e.g., Anderson et al 2005); DNA microarray (Hamelin et al., 2006); fluorescence in situ hybridization (“FISH”) for environmental bacterial cell counting (Tank et al., 2005); and antibody labeling and microscopic detection (e.g., Li and Su, 2006).

These approaches are used to identify particular strains of fecal bacterial species in order to determine the host organism sources of fecal material (Yan et al., 2007). Any of E. coli, Enterococcus sp., or Bacteroides (a common fecal bacterium) can be identified using one or more of these approaches. In a recent study of fecal contamination in tributaries to the San Joaquin River, Johnson et al., (2007) used polymerase chain reaction with DNA primers specific to human, bovine, and chicken fecal Bacteroides to track potential sources of fecal contamination. They used Bacteroides because it is an indicator of recent fecal contamination, due to the fact that this genus is an anaerobic obligate and dies quickly in surface waters. They were able to use this approach to identify human fecal contamination of surface waters during the summer-months study. Bovine contamination from lots and pastures was less prevalent, though the authors noted that there was little overland flow to mobilize dispersed cow manure into waterways. There was not good agreement between Bacteroides (detected by PCR) and E. coli (detected using conventional methods) results, probably due to the longer survival times of E. coli in the environment.

DNA primers exist to identify Bacteroides, E. coli, and Enterococcus strains and thus isolate host organisms for the original feces inputs to waterways. DNA microarrays can be used to test for many strains simultaneously in a water sample. These DNA-based tests are a relatively inexpensive way to forensically point to potential fecal matter inputs, though they may be too expensive for a regular monitoring program across a large area.

5. Findings to date

5.1 Temporal and spatial distribution of sampling

Sampling and measurement of E. coli and Enterococcus sp. was infrequent throughout the watershed before 2001 and since then has been consistently in the summer with less frequent first-flush and winter/wet-season sampling (Figures 5 & 6). Sampling stations for the 4 primary monitoring programs are fairly spread out throughout the watershed, with most sites being in the lower and Southern parts of the watershed (Figure 1). Enterococcus sampling was much more restricted temporally and spatially than E. coli. The highest Enterococcus concentrations were in Santa Rosa Creek/Prince Memorial Greenway. Interestingly, after the high concentrations were found in 2001, Enterococcus concentrations were not measured again in Santa Rosa Creek, even though E. coli concentrations were measured at the same sites and thus grab samples could have been available.

5.2 Daily and monthly concentrations

Concentrations of fecal bacteria can change dramatically over short time periods (e.g., during storms) necessitating frequent sampling during those periods. Meays et al. (2006) found that measured E. coli concentrations could vary over

24 hour periods by as much as 30-fold, depending on when samples were taken. Previous sampling in the Russian River watershed shows that summer concentrations don’t vary as much among days in a given week or month, but wet season concentrations can vary widely (Figure 5). There is also considerable variation between the mean monthly E. coli concentrations in spring/summer and fall/winter (Figure 7). This is probably due to storm runoff and high in-stream flows causing the most input and suspension of fecal matter and bacteria. There were insufficient Enterococcus data to justify calculation of monthly mean concentrations.

5.3 Event loading and sampling

Storm events may contribute to water quality issues by overwhelming control facilities (e.g., ponds) or by suspending settled material in the channel benthos and floodplain. Non-storm events (e.g., exceptionally low flows) may have different time-spans than storms, but may provide exceptional conditions that contribute to overall water quality problems and require special sampling approaches.

5.4 Seasonal loading and sampling

Seasonal fluctuations in flow, water temperature, land-use practices, recreational uses, and in-stream biotic conditions and processes may all affect the storage and potential propagation of live pathogens. Winter conditions will tend to not contribute to in situ growth of pathogens, but may provide disturbing and flushing flows. Summer conditions will tend to contribute to both in situ growth and sources of pathogenic bacteria. In the Russian River watershed, fecal bacteria concentrations are highest in the late fall and early winter, when the first large storms occur. The measured concentrations are well above federal limits, but there is also not much contact recreation in waterways at this time. The high concentrations after storms may be due to inputs of fecal material from overland flows, or disturbance of existing sources in-stream, or in newly inundated areas.

5.5 Long-term change and sampling

In order to track deteriorating conditions or conversely to measure performance of management programs, specific measures should be consistently taken during a long-term monitoring program. The number and frequency of samples and variation in measurements will determine the accuracy of reporting of effective management of pathogenic bacteria. Currently, the sampling program in the watershed includes consistent sampling on the main-stem Russian River during all times of year and inconsistent sampling in sub-watersheds. This will make performance of control programs difficult to ascertain due to the fact that sub-watersheds are the sources of fecal matter and that wet season loading from sub-watersheds may be the cause of summer/recreational season problems.

5.6 Spatial distributions of mean concentrations

Because of the varied potential point and non-point sources of pathogenic bacteria in the watershed, the sampling site selection should efficiently capture the sources as separately as possible. This means identifying the lowest number of sites that also provide information about sub-watershed sources and particular watershed uses. Existing sites provide the highest resolution in the lower, more developed areas of the watershed (Figures 1 and 8). Mean concentrations at tributary sites in the lower watershed (e.g., Laguna de Santa Rosa) are regularly above federal limits, whereas mean concentrations on the mainstem Russian River are generally below the limits.

5.7 Sub-watershed loading

Spatial distribution of sampled sites should capture inputs from

individual sub-watersheds, with resolution at least at the Santa Rosa Creek scale. Figure 9 shows one possible grouping scenario, with relatively gross resolution. When means of all E. coli concentrations are calculated for each group, there appear to be differences among them (Figure 10). However, because the ranges of values used include dry-season baseflows (low concentrations of E. coli) and wet season storm flows (high concentrations), there is overlap among the groups. In general, tributary concentrations are higher.

Land-use association

5.8 Association with benthic algae

E. coli and Enterococcus have been found to be associated with Cladophora sp. in surface waters and beach sands (Whitman et al., 2003). These bacteria can survive 6 months at 4oC in dried algae, suggesting that these algae are an important secondary source of fecal bacteria.

In the Russian River, E. coli were found associated with live benthic algae and benthic sediments at recreational beaches during the summer (Figure 9). Although this association was not characterized fully, it is consistent with associations reported in the scientific literature.

6 Gap Analysis

The existing Russian River watershed bacterial sampling

programs tell part of the story about where and when fecal bacteria are entering the system, but there are gaps in the data and in our knowledge. As a proposed pilot study, we have developed a proposed minimum set of sites to deal with spatial and temporal gaps, which also cover several main land-cover types. Event and bi-weekly sampling at these sites during the early wet season, which is when most fecal matter and bacteria are mobilized into the system, would tell us a lot about watershed process effects on observed concentrations.

Analysis Process

We considered the following information in determining the classes and distributions of 13 proposed sites: 1) previous sampling intensity (# of sampling events), 2) previously-high E. coli concentrations, 3) previously-monitored location, and 4) representation of residential, agricultural, and wild-land sub-watershed land-covers.

1) Sampling counts among the existing tributary sample sites we analyzed closely, ranged from 1 (low intensity) to 49 (high intensity). We selected sites that had this wide range in order to cover places that were consistent wet season problems as well as places that may be problems, but sampling intensity was very low (Figure 10).

2) Previously-measured mean concentrations for the wet season varied among existing tributary sampling sites from 495 MPN/100 ml to >20,000 MPN/100 ml. We selected sites that were well above the WQO and that met other criteria.

3) We tried to select as many previously-monitored sites as possible, but in some cases had to choose new locations that appeared likely to have reasonable stream access (Figure 11).

4) We chose 1 to 3 sites per major land-cover types (Figure 11). We segregated sites by predominant adjacent and upstream/up-watershed land-cover in order to get a measure of the potential relative contributions from these different land-cover types. The land-cover categories were: [Res 1] High density/urban residential and commercial development, [Res 2] Moderate to low density residential development, [Res 3] Very low density rural-residential development; [Ag 1] Dairy/confined animals, [Ag 2] Vineyard; [Wild 1] West-side wildland of Russian River watershed, and [Wild 2] East-side wildland.

Sampling frequency

We suggest a combination of regular and event sampling at these sites to capture baseline (pre-first-flush) concentrations in early October (n=1 per site), event-related concentrations during and after the first 3 major storms (n=6), and bi-weekly sampling starting early October until the end of December (n=6). This will cover the period that has been found in our analysis to be the peak of fecal bacteria mobilization in the Russian River tributaries and in the River itself. The total number of samples would be 169 (13 sites X 13 samples/site).

7. Proposed sampling and identification approach

The following strategy for regular and event sampling of fecal indicator organisms would permit identification of possible source areas, transport processes, and host/source organisms for the fecal material inputs.

7.1 Sampling regime

The sampling regime is the combination of site-level sampling decisions, seasonal timing of sampling, and intensity of sampling.


Sampling depth in the water column can determine the relative concentrations measured at a site (Kleinheinz et al., 2006). Because fecal bacteria concentrations in the water column above undisturbed sediments tend to be highest near the surface, grab samples should be taken consistently within the top few inches of water in order to capture the highest occurring concentrations at a site.

For beach and shore sediments and potentially for different points across still water, fecal bacteria concentrations can vary depending on where the sample is taken (Kleinheinz et al., 2006). To account for this, sites where beach/shore sediments are to be sampled, or where water is relatively still, samples should be taken in a transect across the site.

Sample number (Meays et al., 2006)

7.2 Proposed sampling sites

Spatial distribution of sampled sites should capture inputs from individual sub-watersheds, with resolution at least at the Santa Rosa Creek scale. The list of sub-watersheds that would be at this resolution, including sub-reaches of the Russian River main-stem is shown in the map

7.3 Proposed sampling schedule

Three main types of sampling times should be considered: 1) Wet season base-flow, 2) Wet season event-related sampling, and 3) dry season base-flow.

Wet season tributary and mainstem base-flow may have lower fecal bacterial concentrations than during and after storms, but may still represent a significant proportion of wet-season transport of fecal bacteria. In addition, as storm flows subside, fecal-bacteria-containing sediments may be deposited in-stream or on banks and floodplains and function as reservoirs later in the year. In other words, differences in storm and base flows may represent deposition (as well as death and disintegration) of fecal bacteria within the system. Wet season base-flow sampling should be regular (weekly) and distributed in a way to capture transport from potential source areas to potential deposition areas.

Previous research has found that peaks in E. coli concentrations can depend on the type of waterway (position in the watershed) and antecedent rainfall (Gentry et al., 2006). Storm event-related sampling should include investigations at headwater reaches on tributaries, tributary mainstems, and river mainstem sites. These investigations would include sampling before, during, and after storms for several days to determine when peak concentrations and loads of fecal bacteria occur.

Dry-season base-flows occur when contact recreation is most likely to occur and is also when fecal indicator bacterial concentrations in the river water column have been lowest. However, concentrations have historically been found greater than the basin standard for contact recreation in waters above undisturbed sediments. Sampling should continue at regular intervals during dry seasons base-flows, especially during periods of contact recreation (weekends), because of evidence that sediment and algae disturbance can result in higher water column concentrations of fecal indicator bacteria. In addition, special sampling of benthic sediments and algae should take place in focus locations to determine the possible contributions of these media to surface water concentrations.

7.4 Bacteria identification

Historically, E. coli and Enterococcus sp. concentrations have been used to indicate fecal matter input to waterways, including the Russian River, and therefore are intended to represent concentrations of potentially pathogenic organisms. E. coli may survive for months after introduction into the environment, so does not make an ideal indicator of recent fecal material input. Other bacteria, such as Bacteroides sp. may make better indicators of recent fecal material input. Both E. coli and Bacteroides strains specific to particular vertebrate hosts may be identified based on their DNA. Anderson (2005) found that bacteria from different host organisms may survive in the environment at different rates, complicating the interpretation of linking strains found in the environmental samples with specific vertebrate hosts. This problem can be addressed by also sampling and identifying strains of bacteria with very short survival rates in the environment (e.g., Bacteroides sp.). A combination of E. coli, Enterococcus sp., and Bacteroides sp. sampling and identification may provide answers to multiple questions about long-term fecal matter loading, host organisms, and recent fecal matter inputs. Ideally, DNA finger-printing would be used for strain identification and Colilert/Enterolert types of tests for measuring concentrations of E. coli and Enterococcus sp., respectively.

7.5 Nitrogen and oxygen isotopes

7.5 Optimizing consideration of sampling for distributed land-uses

In order to combine both spatial and temporal sampling rules, knowledge is needed of timing of conditions/activities/effects. This could take place a priori based on existing information about potential sources and times of year. For example, if recreational uses in the River are an important source of bacteria, then sampling would be spaced to capture recreational sites and intensity (very popular vs. infrequently-visited beaches) and timed to capture frequency and intensity (early season week-day vs. Memorial Day weekend).

Urban and agricultural land-uses are likely to contribute fecal material to waterways in the Russian River basin. In almost all studies, these general land-uses are found to be source areas for fecal contamination to surface waters and sometimes to ground waters. Certain of the existing monitoring stations used by the Regional Board or others (e.g., CCWI) in the watershed are downstream of one predominant land-use (e.g., agricultural). These existing stations and a select set of new stations can be used to determine the relative contributions of different predominant land-uses to fecal contamination in sub-watersheds draining to the Russian River.

In the gap analysis above (Section 6), stations CCMWC004, UCDRRPP001, and CCML001 are downstream of predominantly rural/wildland sub-watersheds; stations UCDRRPP004, etc.check

In order to provide a complete picture of land-use contributions of fecal contaminants, this set of stations would be augmented to include a larger representative set of land-use types. For this larger set of stations, E. coli and Enterococcus sp. concentrations would be measured using conventional techniques for one water year. In addition, for one or two focus watersheds with varying land-uses, Bacteroides strains would be identified using strain-specific DNA amplification (PCR). The stations for this approach would be representative of predominant land-uses, such as urban, agrictulture, and wildland and would sampling would be over one water year. Finally, for the same focus watershed, nitrogen and oxygen isotope measurements would be conducted to quantify fecal waste loads to waterways over one water year.

8. References

Adam, R. 1991. The biology of Giardia sp. Microbiol. Rev. 55(4): 706-732.

Al-Ajmi, D., J. Padmanabha, S.E. Denman, R.A. Gilbert, R.A.M. Al Jassim, and C.S. McSweeney. 2006. Evaluation of a PCR detection method for Escherichia coli O157:H7/H- bovine faecal samples. Lett. Appl. Microbiol., 42(4): 386-391.

Anderson, K.L., J.E. Whitlock, and V.J. Harwood. 2005. Persistence and differential survival of fecal indicator bacteria in subtropical waters and sediments. Appl. Environ. Microbiol. 71:3041-3048.

Atwill, E.R., D.J. Lewis, R.F. Bond, M. das Gracas C. Pereira, M. Huerta, and S.B. Ogata. 2007. Protocol consideration for monitoring fecal coliform and E. coli in Northern California Estuaries. University of California, Davis School of Veterinary Medicine and University of California Cooperative Extension, Sonoma County, Santa Rosa, California. 49 pps.

Awais, R., H. Fukudomi, K. Miyanaga, H. Unno, and Y. Tanji. 2006. A recombinant bacteriophage-based assay for the discriminative detection of culturable and viable but nonculturable Escherichia coli O157:H7. Biotechnol. Prog., 22(3): 853-859.

Bingham, A.K., E.L. Jarroll, Jr., and E.A. Meyer. 1979. Giardia sp.: Physical factor of excystation in vitro, and excystation vs. eosin exclusion as determinants of viability. Exp. Parasitol. 47:284-291.

Burton, G.A., D. Gunnison, and G.R. Lanza. 1987. Survival of pathogenic bacteria in various freshwater sediments. Applied and Environmental Microbiology. 53(4):633-638.

Clinton, B.D. and J.M. Vose. Variation in stream water quality in an urban headwater stream in the Southern Appalachians. Water, Air, Soil Pollut., 169(1-4): 331-353.

Desmarais, T.R., H.M. Solo-Gabriele, and C.J. Palmer. 2002. Influence of soil on fecal indicator organisms in a tidally-influenced subtropical environment. Appl. Environ. Microbiol., 68:1165-1172.

Gentry, R.W., J. McCarthy, A. Layton, L.D. McKay, D. Williams, S.R. Koirala, and G.S. Sayler. 2006. Escherichia coli loading at or near base flow in a mixed-use watershed. J. Environ. Qual. 35(6): 2244-2249.

Gerba, C.P., and G. Bitton. 1984. Microbial Pollutants: Their Survival and Transport Pattern to Groundwater. In Groundwater Pollution Microbiology, ed. G. Bitton and C.P. Gerba, pp 65-88. John Wiley and Sons, New York.

Hoyer, M.V., J.L. Donze, E.J. Schulz, D.J. Willis, D.E. Canfield. Total coliform and Escherichia coli counts in 99 Florida lakes with relations to some common limnological factors. Lake Reserv. Manage., 22(2): 141-150.

Ishii, S., W.B. Ksoll, R.E. Hicks, and M.J. Sadowsky. 2006. Presence and growth of naturalized Escherichia coli in temperate soils from Lake Superior watersheds. Appl. & Environ. Microbiol., 72(1): 612-621.

Kleinheinz, G.T., C.M. McDermott, M-C. Leewis, E. Englebert. 2006. Influence of sampling depth on Escherichia coli concentrations in beach monitoring. Water Res., 40(20): 3831-3837.

LaBelle, R.L. and C.P. Gerba. 1980. Influence of estuarine sediment on virus survival under field conditions. Applied Environmental Microbiology 39(3):588-596.

Li, Y. and X-L. Su. 2006. Microfluidics-based optical biosensing method for rapid detection of Escherichia coli O157:H7. J. Rapid Methods Automat. Micro. 14(1): 96-109.

Meays, C.L., K. Broersma, R. Nordin, A. Mazumder, and M. Samadpour. 2006. Diurnal variability in concentrations and sources of Escherichia coli in three streams. Can. J. Microbiol. 52: 1130-1135.

Medema, G.J., M. Bahar and F.M. Schets. 1997. Survival of Cryptosporidium parvum, Escherichia coli, faecal enterococci and Clostridium perfringens in river water: influence of temperature and authochthonous microorganisms. Water Science Technology 35(11-

Noble, M.A., J.P. Xu, G.L. Robertson, and L.K. Rosenfeld. 2006. Distribution and sources of surfzone bacteria at Huntington Beach before and after disinfection on an ocean outfall – A frequency-domain analysis. Mar. Environ. Res., 61(5): 494-510.

Novotny, V. and H. Olem. 1994. Water Quality: Prevention, Identification, and Management of Diffuse Pollution. Van Nostrand Reinhold, New York.

Olsen, S.J., G. Miller, T. Breuer, M. Kennedy, C. Higgins, J. Walford, G. McKee, K. Fox, and P. Mead. 2002. A waterborne outbreal of Escherichia coli O157:H7 infections and hemolytic uremic syndrome: Implications for rural water systems. Emerging Infectious Diseases, 8(4): 370-375.

Power, M.L., J. Littlefield-Wyer, D.M. Gordon, D.A. Veal, and M.B. Slade. 2005. Phenotypic and genotypic characterization of encapsulated Escherichia coli isolated from blooms in two Australian lakes. Environ. Microbiol., 7(5): 631-640.

Reddy, K.R., R. Khaleel, and M.R. Overcash. 1981. Behavior and transport of microbial pathogens and indicator organisms in soils teated with organic wastes. Journal of Environmental Quality 10(3):255-265.

Roper, M.M. and K.C. Marshall. 1979. Effects of salinity on sedimentation and of particulates on survival of bacteria in estuarine habitats. Geomicrobiology Journal. 1(2):103-116.

Salmore, A.K., E.J. Hollis, and S.L. McLellan. 2006. Delineation of a chemical and biological signature for stormwater pollution in an urban river. J. Water Health. 4(2): 247-262.

Sampson, R.W., S.A. Swiatnicki, V.L. Osinga, J.L. Supita, C.M. McDermott, G.T. Kleinheinz. 2006. Effects of temperature and sand on E. coli survival in a northern lake water microcosm. J. Water Health. 4(3): 389-394.

Sherer, B.M., R. Miner, J.A. Moore, and J.C. Buckhouse. 1992. Indicator bacteria survival in stream sediments. Journal of Environmental Quality 21:591-595.

Solo-Gabriele, H.M., M.A. Wolfert, T.R. Desmarais, and C.J. Palmer. 2000. Sources of Escherichia coli in a coastal subtropical environment. Appl. Environ. Microbiol. 66:230-237.

Tang, Y.Z., K.Y.H Gin, T.H. Lim. 2005. High-temperature fluorescent in situ hybridization for detecting Escherichia coli in seawater samples, using rRNA-targeted oligonucleotide probes and flow cytometry. Appl. Environ. Microbiol., 71(12): 8157-8164.

Thomann, R.V., and J.A. Mueller. 1987. Principles of Surface Water Quality Modeling and Control. Harper & Row, New York.

U.S. EPA. 1985. Rates, Constants, and Kinetics in Surface Water Quality Modeling. Athens, Georgia: Environmental Research Laboratory, Office of Research and Development. EPA/600/3-85/040.

Whitman, R.L., D.A. Shively, H. Pawlik, M.B. Nevers, and M.N. Byappanahalli. 2003. Occurrence of Escherichia coli and Enterococci in Cladophora (Chlorophyta) in nearshore water and beach sand of Lake Michigan. Appl. Environ. Microbiol., 69(8): 4714-4719.

Wickham, J.D., M.S. Nash, T.G. Wade, L. Currey. 2006. Statewide empirical modeling of bacterial contamination of surface waters. J. Am. Water Resour. Assoc., 42(3): 583-591.

Yan, T., M.J. Hamilton, and M. Sadowsky. 2007. High-throughput and quantitative procedure for determining sources of Escherichia coli in waterways by using host-specific DNA marker genes. Appl. Environ. Microbiol., 73(3): 890-896.