Saturday, March 2, 2019
Brief Analytical Report III. Elevated Kahalu’u Bay Enterococci Associated with Large Tidal Flux
R. H. Bennett Ph.D., Applied Life Sciences Honaunau, Hawaii
When the local news recently (Jan 8, 2019) reported several beaches on Oahu were closed due to elevated indicator bacteria the Enterococci (ENT) and others on Hawaii island at about the same time, it raised technical questions. There had been no major rain events on the islands, and no sewer spills to cause health department warnings. Instead, the warnings arose for beaches widely separated geographically and simply because the levels of ENT elevated above the regulatory trigger of 130 CFU per 100 mL MPN. That trigger level legally requires agencies to post warnings, beach postings and mass media alerts. How did it come to pass that these beaches hit the warning level ENT at about the same time? What bacteriological events have this scale?
For over five decades, the EPA and the states struggled unsuccessfully to find a way to assess the microbial safety of recreational waters precisely. Funding for such an effort is not a priority for most governments, even though ocean recreation is a significant economic activity for all coastal states. Peng (2017) demonstrates that ocean users are concerned, and they are willing to pay about $42 per person per year for recreation water that does not exceed bacterial standards. This suggests that beach warnings and closures from bacterial exceedances have a cost from deterred recreation.
For simplicity, the EPA and the states seek a universal indicator level, that when present, it highly correlates with specific disease risk. To date, no such indicator exists and nor is one likely to be found. Microbial ecology is vastly complex and not amenable to simple presence measurements as a correlate of disease risk. To put this into an everyday context, if commercial kitchens had a similar bacterial warning indicator, very few kitchens would be deemed safe.
Sewage contamination of recreation waters occurs regularly and is often associated with significant rain events and sewer systems malfunction and overflows. Illness attributed to such incidents is rare. When they do occur, it manifests as gastrointestinal, upper respiratory and skin infections. The GI and respiratory illnesses are mainly due to virus and skin infections are due to various species of Staphylococcus. The assumption in almost all risk studies is that water contamination is solely a GI disease risk. In addition to GI illness, Fleisher (2010) monitored skin infections and respiratory illness among swimmers and non-swimmers in tropical waters. Swimmers were 5.91 times more likely to have skin lesions than non-swimmers. GI illness in swimmers was a lesser risk at 1.76 times greater, and there was no increase in GI illness with increasing ENT levels.
As the American Academy of Microbiology stated in 1995, bacterial indicators do not accurately forecast disease risk, and health agencies are advised to monitor the disease agents directly. Now in the age of molecular microbiology of the human gut microbiome, the unique human fingerprint of human gut bacteria is readily determinable. This technology has yet to penetrate the traditions of the regulatory world. We have formulated health policy on bacterial indicators that consistently fail to predict disease risk from ocean recreation.
Studies about the nature of the nearshore waters in Kahalu‘u Bay, Hawaii Island are ongoing. We show that large volumes of groundwater containing urban pollutants enter the bay every day. Researchers estimate the volume of groundwater volumes between 300 thousand to 3 million gallons per day every day flow into the bays along the Kona Coast (Peterson 2009). We also showed that the ebb and flow of the tides significantly alters the salinity and temperature of the nearshore waters. The daily tides, four per day, two high and two low, control the rate and volume of groundwater flows (Valle-Levinson 2011). Thus, it is likely that groundwater flows are influencing the movement of ENT as well?
The research of Lee (2017) and Boehm (2009) clearly shows that groundwater flows controlled by tidal forces, move indicator bacteria from the sand and rock strata into the water column, especially when the fluxes are the greatest. Large tidal fluxes occurred in Hawaii in the week of January 7, 2019. These large tidal fluxes appear to be the cause of elevated ENT levels and triggered beach warnings in the absence of known sewage release events.
To test the theory that Kahalu‘u Bay tides influence the ENT counts, two data sets were statistically queried. The first data set is that of the Hawai‘i Health Department, Clean Water Branch (HDOH CWB). They collect ENT data for most of the major beaches in the state. Sampling dates and time is a matter of a scheduled routine, and no special consideration is given to tides. The HDOH data is available online for those with patience and an aptitude for data management. This data set had observations for the years 2013 to 2017. In 2017 there was a sewer line leak, of unknown duration, near the restrooms for the Kahalu‘u Bay. Consequently, data from that from that year were excluded from this study. The dated observations were sorted by ENT count, those above 135 CFU per 100 mL; fourteen were found and constitute the HIGH ENT set. Fourteen sample dates, selected at random, where the ENT was ten or less were also used in the analysis and used for the LOW ENT set.
Figure 1. Kahalu‘u Bay monitoring sites 1-4 and HDOH monitoring site (yellow pin)
A NOAA online database (NOAA Tide Calculator) for daily tides for Kailua Kona Hawaii was used to extract the historical high and low tide information. The tides at the date and time the ENT sample were also compiled. This data was used to calculate the magnitude of the tidal flux. For the purposes of this study, the Flux is the highest tide level before sampling, minus the tide level at the time of the sample.
The sites used to monitor the bay are shown in Figure 1. Sites 1 -4 are used by the Kahalu‘u Bay Education Center (KBEC) for biweekly monitoring for non-microbial variables like temperature, salinity, and others. (Brief Report I.) Site 5 is not seen in the figure. It is to the north of the brown roof comfort station. The yellow pin is the location of the site used by the state HDOH for monitoring of ENT. From other investigations (Part I) sites 2, 3 and 4 have statistically similar salinities and temperature fluctuations with the tide. Site 1 shows a far more significant influence of groundwater. Site one sits in a channel lined by rock outcroppings, thus mixing with sea water is reduced. All sites are accessible by foot, are only 2 feet deep, depending on the tide, and have sandy bottoms.
HDOH results for ENT reveal that only 14 sample dates had exceedance for ENT over the years 2013-2016. The 14 sites over 135 CFU/100ML MPN had a geometric mean ENT count of 508. The overall ENT counts in the data set ranged from 2005 to less than 10 with the overall geometric mean of 15.7 CFU/100 mL MPN. Only 5.5% of the samples of 256 observations in four years were in exceedance of state and federal standards for recreation waters. The frequency of exceedances and the geometric mean suggest this site is not likely sewage contaminated as a matter of course.
The tidal and enterococci data and the computed terms are shown in Table 1. The ENT counts were converted to the log base 10 to normalize a geometric distribution for statistical analysis. CP in the table represents Clostridium perfringens; a HDOH secondary indicator. CP is common in sewage from treatment facilities and is routinely found in soil, compost and the natural environment, albeit typically lower numbers than wastewater treatment plants.
The tidal flux was calculated from the NOAA published high tide just preceding the HDOH sampling. That tide was determined from the NOAA online interactive tide data set. The tidal flux is calculated by subtracting the height of the tide at sampling from the level of the preceding high tide. The high to low (H/L) flux is shown on the table.
The HDOH testing procedure also includes water temperature and salinity. Between the two groups, the mean temperature for the HIGH ENT and the lOW ENT were 26.59 and 27.74 C respectively. While the means are only 1.15 degrees C different, they are statistically significantly different (p =0.04). The mean salinities were only 0.67 PPT different and not significant(p=0.3).
Neither the tide level at the time of sampling and the high tide preceding the sample time were significantly different. In contrast, the mean tidal flux between the high tide preceding the sampling was 1.28 ft for the HIGH ENT group and 0.76 ft for the LOW ENT group. The differences were significant (p=0.03).
On two occasions in the HIGH ENT, a very high tide preceded a modestly high Low Tide followed by a lower tide near sampling tide. These two observations were kept in the data set as the flux was in the downward trend. Whether the tide at sampling was ebbing or flooding did not appear to have a noticeable influence and the data set does not provide for calculating the magnitude of the flood events.
It is important to note that the times of sampling was very consistent around 10:00 HST for both sample sets had very similar means of 10:05 and 10:02. In contrast, the sampling schedule did not appear to have any cyclicity or regular schedule. Some months were sampled five times and others three. There is no suggestion in the data that sampling was scheduled to correspond with either the high tide or low tide.
Table 1. Tidal Data and Fluxes associated with Enterococci Counts, Kahalu‘u Bay, Hawai‘i
(Hawaii Dept of Health, Clean Water Branch, Beaches Act Monitoring Program data in outlined cells)
Figure 2. NOAA tide data for representing a high flux and zero flux at sampling.
In the high flux situation above a tidal level change of 1.81 feet occurred before sampling. On the lower chart sampling occurred at peak high tide and the flux is virtually nil.
Figure 3. Tidal Pattern for Jan. 9, 2019. A sample day that resulted in a beach closure.
The figure illustrates that even a small flux can be associated with elevated ENT levels. It is important to note that earlier in the day there was a large net flux of 2.14 feet.
Figure 4. Large tidal fluxes preceded the sample date Jan 9th.
Researchers that conduct near coastal studies are acutely aware of tidal influences and the variation it induces in the data. It was not until 2005 when Boehm documented how groundwater flows during outgoing tides could transfer and or resuspend indicator bacteria into the water above. In contrast ocean bacteriology studies seldom, account for tidal influences on bacterial levels and this science has not been applied to the rationale for ENT sampling by federal or state agencies.
Confounding this phenomenon is the observation the ENT can survive and reproduce in wetted beach sand. ENT has a doubling rate of 1.1 to 3.5 days in wet beach sand conditions (Yamahara 2005). Beach sand is now well established as an important reservoir of indicator bacteria such as ENT and others and the numbers in beach sand are significantly higher than the nearby water (Bonilla 2007, Goodwin 2012).
Enterococci is so prevalent in Southern California beach sand that the Orange County Health Department has cast credible doubt on the validity of ENT for recreation water safety determination. “ High levels of Enterococcus in intertidal sediments indicate retention and possible regrowth in this environment. The Resuspension of enterococci that are persistent in sediments may cause beach water quality failures and calls into question the specificity of this indicator for determining recent fecal contamination” (Ferguson 2005).
Lee (2017) identifies three mechanisms for bacteria transport from groundwater to the near shore. Groundwater levels within the tidal zone are subject to the ebb and flow of the tides. Tidal currents and certainly wave action can resuspend bacteria resident or transient in the sand and sediments.
It is consistent with flow hydrology that the greater the tidal flux, the greater the volume of water that will flow from the land to the sea. Since this flux is occurring in six hours, larger volumes of water may well have greater flow velocities and turbulence. The volume of water is this flux cannot be estimated from these data. Natural markers like Radon isotopes are used and will be helpful in more precise tidal flux volume and velocity studies (Sadat-Noori 2015).
The US Geological Service provides a conceptual model that aids in the visualization of submarine groundwater flows. As seen in the figure sea water fresh water mixing occurs at the interface and under the land. This helps to explain the mixing of warmer and more saline sea waters with the groundwater both inshore and the intertidal zone. The work also suggests how differences in the geological porosity control the flow rates.
The release of ENT from beach sand may well be controlled by factors other than simple hydraulic flows during tidal events. Research on marine bacteria biofilms suggests that the biofilms are dependent on the seawater osmolarity from the salts including calcium ions. The data indicate that less saline groundwater alters the physical chemistry allowing greater numbers of bacteria to be released from the biofilms (Kierek 2003). It is likely that repeated large ebb tide fluxes as often occurs in high Spring Tides, produce several days of low saltwater flows through the sand, magnify the degradation of biofilms, allowing the subsequent release of bacteria. The concentration of ENT in beach sand is comparable to raw sewage, and a significant release by tidal events could mimic concentrations of ENT seen in ocean sewage spills (Phillips 2011). Therefore, it may be likely that several days of high tidal flux before ENT testing could produce ENT counts of the magnitude seen during sewage spills (Fujioka 1999).
Figure 7. Monthly lunar and tidal cycles of high flux occur in a biweekly cycle of Spring Tide
As shown in Figure 7, the early Spring Tides are associated with the highest tidal fluxes, and flow events that liberate bacteria into the water column may well be more than a single large flux, but rather the accumulative physical events and the influence on bacterial suspension. Boehm (2005) shows this accumulative effect with increasing ENT numbers as two weeks past the full moon is approached. This suggests that sampling for recreation water monitoring be conducted around day 7 and 21 past the full moon, to minimize the presence of environmental source ENT that is conveyed by a series of sizeable tidal flux events.
There remains a distinct possibility that elevated ENT counts are an artifact of inherent time of sampling bias and as such official warnings may not be appropriate. However, given that the region near Kahalu‘u Bay is not served by sewer and most homes are utilized cesspits for over 3 to five decades sewage components may be conveyed to the ocean by groundwater.
However, given the relative paucity of bacterial pathogens in the feces of the average population ( Huttenhower 2012, Human Microbiome Consortium) regulatory policy is best informed by a continued scientific inquiry into the prevalence of human feces specific bacterial markers and viral pathogens in human wastewaters and less by assumptions about bacteria that live in, and influenced by the natural environment.
1. To avoid the intrinsic false positive errors due to tidal fluxes, the sampling program schedule can be organized to avoid high tidal fluxes by sampling on or about the time of a higher tide of the day time during Neap Tides and avoid sampling during low Spring Tides
2. Adopt DNA/RNA technologies for human bacteria markers including pathogenic Staphylococcus and virus marker surveillance of recreation waters.
3. Adopt Chemical markers of human sewage like those validated for Sucralose and others.
4. Develop comprehensive ocean recreationist disease risk assessments that precisely estimate risk from specific pathogens in marine waters and communicate the science- based risk assessments.
5. As Enterococci continues to be used, the warning statements need to be modified to include the high degree of uncertainty regarding disease risk to ocean users. This is particularly important when there is no evidence of a sewage release.
Boehm, Alexandria B., and Stephen B. Weisberg. "Tidal forcing of enterococci at marine recreational beaches at fortnightly and semidiurnal frequencies." Environmental science & technology 39, no. 15 (2005): 5575-5583.
Bonilla, Tonya D., Kara Nowosielski, Marie Cuvelier, Aaron Hartz, Melissa Green, Nwadiuto Esiobu, Donald S. McCorquodale, Jay M. Fleisher, and Andrew Rogerson. "Prevalence and distribution of fecal indicator organisms in South Florida beach sand and preliminary assessment of health effects associated with beach sand exposure." Marine pollution bulletin 54, no. 9 (2007): 1472-1482.
Ferguson, D. M., D. F. Moore, M. A. Getrich, and M. H. Zhowandai. "Enumeration and speciation of enterococci found in marine and intertidal sediments and coastal water in southern California." Journal of Applied Microbiology 99, no. 3 (2005): 598-608.
Fleisher, Jay M., Lora E. Fleming, Helena M. Solo-Gabriele, Jonathan K. Kish, Christopher D. Sinigalliano, Lisa Plano, Samir M. Elmir et al. "The BEACHES Study: health effects and exposures from non-point source microbial contaminants in subtropical recreational marine waters." International journal of epidemiology 39, no. 5 (2010): 1291-1298.
Fujioka, R. S., A. J. Bonilla, and G. K. Rijal. "The microbial quality of a wetland reclamation facility used to produce an effluent for unrestricted non-potable reuse." Water science and technology 40, no. 4-5 (1999): 369-374.
Goodwin, Kelly D., Melody McNay, Yiping Cao, Darcy Ebentier, Melissa Madison, and John F. Griffith. "A multi-beach study of Staphylococcus aureus, MRSA, and enterococci in seawater and beach sand." Water research 46, no. 13 (2012): 4195-4207.
Huttenhower, Curtis, Dirk Gevers, Rob Knight, Sahar Abubucker, Jonathan H. Badger, Asif T. Chinwalla, Heather H. Creasy, et al. "Structure, function and diversity of the healthy human microbiome." Nature 486, no. 7402 (2012): 207.
Kierek, Katharine, and Paula I. Watnick. "The Vibrio cholerae O139 O-antigen polysaccharide is essential for Ca2+-dependent biofilm development in sea water." Proceedings of the National Academy of Sciences 100, no. 24 (2003): 14357-14362.
Lee, Eunhee, Doyun Shin, Sung Pil Hyun, Kyung‐Seok Ko, Hee Sun Moon, Dong‐Chan Koh, Kyoochul Ha, and Byung‐Yong Kim. "Periodic change in coastal microbial community structure associated with submarine groundwater discharge and tidal fluctuation." Limnology and Oceanography 62, no. 2 (2017): 437-451.
NOAA Interactive Tide Calendar, https://tidesandcurrents.noaa.gov/noaatidepredictions.html?id=1617846
Peng, Marcus, and Kirsten LL Oleson. "Beach recreationalists' willingness to pay and economic implications of coastal water quality problems in Hawaii." Ecological Economics 136 (2017): 41-52.
Peterson, Richard N., William C. Burnett, Craig R. Glenn, and Adam G. Johnson. "Quantification of point‐source groundwater discharges to the ocean from the shoreline of the Big Island, Hawaii." Limnology and Oceanography 54, no. 3 (2009): 890-904.
Phillips, M. C., Solo-Gabriele, H. M., Reniers, A. J., Wang, J. D., Kiger, R. T., & Abdel-Mottaleb, N. (2011). Pore water transport of enterococci out of beach sediments. Marine pollution bulletin, 62(11), 2293-8.
Sadat-Noori, Mahmood, Isaac R. Santos, Christian J. Sanders, Luciana M. Sanders, and Damien T. Maher. "Groundwater discharge into an estuary using spatially distributed radon time series and radium isotopes." Journal of Hydrology 528 (2015): 703-719.
Santos, Isaac R., Bradley D. Eyre, and Markus Huettel. "The driving forces of porewater and groundwater flow in permeable coastal sediments: A review." Estuarine, Coastal and Shelf Science 98 (2012): 1-15.
Yamahara, Kevan M., Sarah P. Walters, and Alexandria B. Boehm. "Growth of enterococci in unaltered, unseeded beach sands subjected to tidal wetting." Applied and environmental microbiology 75, no. 6 (2009): 1517-1524.
Special appreciation goes out to Cindi Punihaole Kennedy, the founding director of the Kahalu‘u Bay Education Center (KBEC) and educator coordinator Kathleen Clark. Together with many volunteers, they have collected high-quality data for over seven years. The quality control for the data acquisition enables this analysis and those to follow, to be scientifically sound.
Appreciation is expressed to the State of Hawaii, Department of Health, Clean Water Branch for publishing the raw sampling data so that the public and scientists can independently analyze the data and make those findings public.
R.H. Bennett declares no conflict of interest and this work is provided pro bono for the benefit of KBEC, our community and the people of Hawai‘i.
Brief Analytical Report III. Elevated Kahalu’u Bay Enterococci Associated with Large Tidal Flux, by RH Bennett PhD. Creative Commons, CC BY 4.0. Free to use, distribute and copy with citation above and Healthy Hawaiian Oceans