Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The relationship of bacterial water quality and health of Lake Okanagan swimmers Naegele, Barbara Ellen 1974

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


UBC_1974_A6_7 N33.pdf [ 5.31MB ]
JSON: 1.0099886.json
JSON-LD: 1.0099886+ld.json
RDF/XML (Pretty): 1.0099886.xml
RDF/JSON: 1.0099886+rdf.json
Turtle: 1.0099886+rdf-turtle.txt
N-Triples: 1.0099886+rdf-ntriples.txt
Original Record: 1.0099886 +original-record.json
Full Text

Full Text

THE RELATIONSHIP OF BACTERIAL WATER QUALITY AND HEALTH OF LAKE OKANAGAN SWIMMERS by BARBARA ELLEN NAEGELE B.Sc, The University of British Columbia, 1 969 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in the Department of ZOOLOGY We accept this thesis as conforming to the req u i red s tanda rd THE UNIVERSITY OF BRITISH COLUMBIA 1974 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of ZOOLOGY The University of British Columbia Vancouver 8, Canada A B S T RACT This study was an investigation into the relation ship of bacterial water quality and the incidence of illness amongst lake and chlorinated pool swimmers. During the summer of 1972 three groups of swimmers were surveyed: 1. Lake Okanagan swimmers who swam at the Kelowna City Park beach 2. Ogopogo Aquatic Team who swam at the same beach 3. Chlorinated swimmers who swam at an unheated community pool The swimmers ranged in age from six to sixteen and were surveyed for upper respiratory illness, gastroenteritis, otitis externa, shistosomiasis and conjunctivitis. The lake water, proximal creeks, and chlorinated pool were sampled throughout the swimming season for the fecal contamination indicators; fecal coliforms and fecal streptococci. Bacterial water quality of the creek varied with the effluent input and quality of the lake water varied with inflow from the creeks and swimming density. The flake fecal i i coliform counts generally remained below 200 organisms per 100 milliliters. The lake fecal streptococci counts were slightly higher than the fecal coliforms and more responsive to swimming density. In the chlorinated pool samples fecal coliforms were not present while fecal streptococci counts ranged up to 168 organisms per 100 mis. Contamination at all sites was of both human and animal origin. Total illness incidence was highest amongst Hake Okanagan swimmers and lowest amongst the Ogopogo Aquatic team swimmers. The most distinctive illness difference observed between lake and chlorinated pool swimmers was the incidence of otitis externa. For Lake Okanagan swimmers the incidence was 17.82 per 1000 person hours water exposure, for the Ogopogo Aquatic Team the incidence was 16.48 and for chlorinated pool swimmers incidence was only 0.67. It was surmised that the causative organism for the auditory dermatitis was Pseudomonas aeruginosa and its presence was verified in the lake water and streams and negated in the chlorinated water. For lake swimmers the correlations between incidence of illness and lake fecal indicator counts were significantly positive for upper respiratory infections, gastroenteritis and otitis externa. The correlation coefficient exhibited i i i between otitis externa of lake swimmers and the fecal streptococci counts was .866. Throughout statistical analysis fecal streptococci counts displayed stronger positive correlations with illness incidence than the fecal coliform data. For chlorinated pool swimmers the illness patterns did not significantly correlate with fecal streptococci counts. It was recommended that fecal coliforms alone were an inadequate index of recreational water quality. It was also suggested that Ps. aeruginosa warranted further research as a potential water affiliated pathogen and water quality index. TABLE OF CONTENTS Page ABSTRACT ii LIST OF TABLES viiLIST OF FIGURES x ACKNOWLEDGEMENTS xi Chapter 1 INTRODUCTION 1 Outline of Investigation 2 Research Objectives • -5 2 METHODOLOGY ... 9 location of Investigation 9 Illness Survey 1 Bacteriological Sampling 13 Laboratory Analysis 17 3 WATER QUALITY RESULTS 22 Bacterial Water Quality 2Morphology and Identification of Fecal Streptococci 31 v Chapter Page Pseudomonas aeruginosa as a Water Quality Index 34 Additional Variables Measured 36 Temperature . .... . 37 Urine and Fecal Indicators 37 Water pH. 34 ILLNESS RESULTS 9 Inadequacy of Interviewing 3General Illness Patterns 40 Incidence of Specific Illnesses 43 5 INTERPRETATION OF WATER QUALITY DATA. ...... 49 Fecal Contamination of the Creeks 51 Bacterial Statistical Analysis 52 Source of Lake Water Fecal Contamination . . 53 Correlation of Creek and Lake Water Qual i ty 61 6 INTERPRETATION OF ILLNESS DATA 66 Incidence and ExposureOtitis Externa as a Swimming Related Problem. ..... 72 7 CORRELATION OF ILLNESS INCIDENCE AND WATER QUALITY 76 Statistical Analysis 7Discussion of Correlation Analysis 78 Rutland Correlations. . . . . . . . . .79 vi Chapter Page Lake Water Quality and Illness Correlations °u Value of Indicator Organisms. ...... 83 8 CONCLUSION 94 BIBLIOGRAPHY 98 APPENDICES: 1 Health Branch Recommend Water Quality Standards 10J 2 Limnological Information 105 3 Tables XVIII .-XXII 106 vi i LIST OF TABLES Tab!e Page I Fecal Coliform Counts: July, August, September 1 972 24 II Fecal Streptococci Counts: July, August, September 1972 26 III Fecal Streptococci Morphology 32 IV Fecal Streptococci Morphology: Preliminary Identification of Human and Livestock Strains. . . 33 V Ps. aeruginosa and Fecal Coliform Counts: July and August 1973 36 VI Fecal Indicator Counts in Urine Samples from Lake Swimmers 8 VII Cases of Illness Amongst Lake and Chlorinated Pool Swimmers, July and August 1 972 47 VIII Percentages of Illness Amongst Lake and Chlorinated Pool Swimmers, July and August 1 972 4a IX Ratios of Fecal Indicator Counts 52 Fecal Coliform: Fecal Streptococci X Statistical Analysis: Comparison of Fecal Indicator Counts for Lake and Creeks 56 v i i i Table Page XI Statistical Analysis: Correlation of Lake and Creek Fecal Indicator Counts 64 XII Illness Incidence Per Thousand Person Hours Swimming Exposure: July and August 1972 /0 XIII Statistical Analysis: Comparison of Weekly Illness Incidence for Chlorinated Pool and Lake Swimmers XIV Statistical Analysis: Correlation of Illness Amongst Chlorinated Pool Swimmers and Fecal Indicator Counts yb XV Statistical Analysis: Correlation of Illness Amongst Lake Swimmers and Fecal Indicator Counts 87 XVI Otitis Externa and Fecal Indicator Correlation: Ogopogo Pool Swimmers 93 XVII Lake Oganagan Air and Water Temperatures ln7 1 972 '. . ' IU/ XVIII Water pH of Sampling Sites 109 XIX Ogopogo Pool Fecal Indicator Counts: Surface vs. Sediment Water Samples ''° XX Fecal Indicator Counts: Ogopogo Pool vs . Hot Sands 111 XXI Weekly Illness Incidence Per 1000 Person Hours Swimming Exposure 112 XXII Ogopogo Fecal Indicator Counts Used for Product Moment Correlations 113 i x LIST OF FIGURES Figure Page 1 Map of Lake Okanagan water sampling sites 7 2 Water quality sampling sites: ' photographs 15 3 Laboratory analysis of fecal contamination indicators: photographs 20 4 Fecal streptococci: schematic outline of preliminary investigation . 21 5 Ogopogo pool fecal coliform and fecal „R streptococci counts 6 Mill Creek fecal coliform and fecal streptococci counts 7 Brandt's Creek fecal coliform and fecal streptococci counts 30 8 Ogopogo pool thrice daily counts 59 9 Lake fecal streptococci vs. air and fiQ water temperatures 10 Illness incidence per 1000 person hours exposure 69 11 Correlation weekly otitis externa incidence and fecal streptococci counts 92 ACKNOWLEDGEMENTS I am indebted to the staff of the Kelowna South Okanagan Health Unit who willingly offered their assistance and made room for my investigation in their already cramped quarters. I would also like to give special appreciation to Dr. David Clarke (Medical Health Officer) and Mr. Fred Alcock (Senior Health Inspector) for their constant encouragement. My interest in environmental health has been inspired by their enthusiastic and energetic approach to public health problems. xi Chapter 1 INTRODUCTION Lake Okanagan, located amidst semi-arid mountains and productive valleys, is situated in an attractive and rapidly expanding region of southeastern British Columbia. The lake is one in a chain of several lakes which together with their tributaries form 3,000 square miles of the Okanagan Basin (Figure l()i. The attributes of the Okanagan Basin include fertile land, government industrial incentives, Mediterranean summers, ski hiills, sport fishing and pleasant beaches. The region is understandably developing relatively quiiekly. For years it was assumed that Lake Okanagan and the surrounding Basin would remain an unchanging aesthetic attribute capable of providing for industrial, agri cHtural , recreational, and tourist needs of the region. If managed with foresight and discretion the lake has renewable potential but presently planning is in a state of flux with one "interest" group battling the priorities of another. 1 2 The observations regarding increased polychlorinated biphenyls in fish, weed growth along shorelines, Anabaena algae blooms, decreased water transparency and nutrient enrichment from sewage outfalls are not in balance with needs for increased irrigation, housing and highways. To contend with the problems of future planning in the Okanagan Basin, a two million dollar, four year, federal and provincial study was commenced in 1969. The object of this "Canada British Columbia - Okanagan Basin Agreement Study" was to prepare a framework for the management of water resources. Trh e t s t ii dy n e n d e a v o ured „ t o c p n oduc e a a d p ill a n - w h i c h •< i n c 1 u d e d; c o n -sideration for social, economic and environmental aspects. The controversies resulting from even the preliminary findings are numerous and the arguments will likely abound as to which is the preferable course to follow. OUTLINE OF INVESTIGATION Out of this mixed political environmental situation regarding Lake Okanagan, I extracted one area of research. It is an area whiich received attention in the early 1950's (Stevenson, 1953), has again been requestioned (Krishnaswami, 1971) but as yet has not been sufficiently documented. The question I examined was whether or not the bacteriological water quality of Lake Okanagan correlated with 3 the illness incidence observed in local swimmers. I undertook research on one Lake Okanagan beach to substantiate the hypothesis that illness patterns of swi.mmers corresponded with fluctuations in water quality. It was my assumption that the lake environment undermined the health of swimmers and I collected data during the summers of 1972 and 1973 to test this hypothesis. The project was located in Kelowna (population 47,000 1973) which is the largest of the several cities near Lake Okanagan. The participants were children who swam either at the main city park beach or at the chlorinated community pool. Thought was given to trie 1!Qding non-swimmers, however according to Stevenson (1953) and Ames (1969) swimmers demonstrate a greater illness rate than non-swimmers irrespec tive of location and water contamination. The explanation given for this difference is that water is an abnormal habitat for man and hence affects health regardless of its bacterial quality. Between the two research groups I expected illness incidence among lake swimmers to be greater than the incidence of similar illnesses among chlorinated pool swimmers. The swimming season, because of temperature changes, lasted a'short period beginning with the July 1st weekend and ending with Labour Day, September 3rd. Even by mid-August temperatures dropped sufficiently to lessen participation. 4 The days were often warm but the cold nights quickly affected temperatures in the chlorinated pool, which was unheated, and the lake. My investigations did not deal with serious water bourne illnesses such as Salmonella typhi or infectious hepatitis, but with the inconvenient nuisance illnesses such as colds and gastroenteritis. For determination of water quality I applied the contamination indicators fecal coliform and fecal streptococci. The concept of coliforms as an indicator of human sewage and hence potential pathogens has been used for almost seventy-five years with occasional reappraisal as to its reliability. Recently fecal coliform rather than the total coliform count has gained favour. Total coliforms included strains ubiquitous to the environ ment but not necessarily common in fecal material (Geldreich, 1965). Two criticisms against using solely the coliform indicator are firstly it is representative of only enteric pathogens and secondly in some cases enteric infections, such as salmonel1osis , Have been caused by ingesting water with no or few coliforms (Dutka, 1973). As a supplement to the fecal coliform analysis I included fecal streptococci counts. Fecal streptococci analysis has been recommended as being "more practical, efficient and accurate than coliform tests for evaluating the sanitary 5 quality of almost all types of water." Fecal streptococci standards are mandatory in parts of the United States, especially for chlorinated pools but are not included in the British Columbia recreational water quality standards (Appendi x 1). RESEARCH OBJECTIVES This project was the first study on an Okanagan beach combining intensive bacteriological testing and the surveying of illnesses amongst swimmers. The Okanagan Basin Study did not include inquiries into health of the swimming population. Some water sampling was done in which beaches were sampled twice a month (Lynch, 1972), but the actual bathing population was ignored and the conclusion arrived at by the Okanagan Basin Study regardingdrecreation water qualii'tyuwas tyi tisufif i'ci entadatat.a.." My objective was to delve further into this concern, questioning the influence of recreational water quality on the health of swimmers. The project was also prompted by information and records from Dr. D. Clarke, Medical Health Officer, South Okanagan Health Unit. During the past fifteen years he had C? Bartley, L. Slanetz, "Types and Significance of Fecal Strep," A.J.P.H., Vol. 50, 1960, p. 1552. 6 sensed an increase in rate of upper respiratory ailments but was lacking data. T.he total coliform counts which have been taken during these years were within the public health standards but the samples were too intermittent to be used as a baseline for epidemiological judgements. It was hoped that my study would help clarify the premise that Lake Okanagan water quality has a detrimental effect on the health of swimmers. My investigation involved three questions: 1) do lake swimmers exhibit a higher incidence of illness than chlorinated poo I sw i mmers? 2) is there a correlation between observed illness and bacterial water quality? 3) how and why does the bacterial water qua I i ty va ry? Recent assessments were done by the Okanagan Basin Study on physical and chemical features of Lake Okanagan. Investigations were made into seasonal fluctuations of chlorides, phosphates, nitrogen and algae growth. The intention of a bacterial and illness survey was to add to the information available on Lake Okanagan. Questionnaire studies (Okanagan Bas,%9S£udy?tBul#etin #5) reported that 91% of tourists considered swimming important, 73% of residents swam annually and 80% of these did so on more than twenty occasions. The lake is obviously an important recreational resource with swimming a prime activity for tourists and residents. Figure 1A. Location of Lake Okanagan. SAMPLI NG SITES w Brandt's Creek LAKE OKANAGAN KELOWNA Hot Sands Beach Ogopogo Pool SCALE .45 miles Figure IB. Water Sampling Sites Chapter 2 METHODOLOGY LOCATION OF INVESTIGATION There are several public beaches scattered along the shoreline of Kelowna and the most central and popular one is the city park beach commonly called "Hot Sands." The Hot Sands beach includes a semi-enclosed boardwalk area called Ogopogo Pool (Figure 1). Ogopogo is referred to as a pool but in essence it is part of the beach demarcated with a floating walkway. Form this location I chose two types of swimmers all between the ages of six and sixteen. One group was the Ogopogo Aquatic Swim Team which swam for a minimum of one hour daily during the week in the area designated Ogopogo Pool. Participants in this group remained constant throughout the summer. The other lake group were swimmers from the Hot Sands swimming classes and these children swam one hour three times weekly for a total of two weeks. Participants in this group thus changed every two weeks. 9: 10 The comparison group for the study swam at the Rutland community pool which at that time was the only chlorinated pool open to the public. As with" the "Hot Sands" swimmers classes were held thrice weekly for two weeks. The pool was five miles east of the lake. I did not individually select the participants. All the children registered for swimming classes in the three locations were included in the illness survey. As a result, group size, age, and sex distribution varied with swimming location. The ratio of male and female participation in the illness survey was as follows: Location Total Participants Mai e Female Ratio Ogopogo Swim Team Hot Sands Swimmers Rutland Swimmers 52 148 248 26 64 115 26 84 133 1:1 1:1.32 1:1.16 Nor did I distinguish participants according to their socio economic backgrounds. Judging from the distribution of telephone numbers of participants it is my estimation that the Hot Sands arid Rutland swimmers represented similar residential cross sections of Kelowna while the Ogopogo Swim Team was comprised of children from the more affluent residential districts. 11 ILLNESS SURVEY I collected information on the children's health by interviewing both the children and the parents. Also I gave forms to the swimming instructors^ and requested them to keep records of specified health problems. Illnesses and infection surveyed were categorized into: respiratory infections, colds, otitis externa, gastroenteritis, conjunctivitis, shistosomiasis and miscellanea. For purposes of interviewing they were referred to as sore throats, colds, ear aches, stomach flu, eye infections, and skin irritations. These categQifi'es were based on research by Stevenson (1 953) and previous observations from Lake Okanagan (Clarke, 1959), in which eye, ear, nose, and throat infections and gastroenteritis were the infections associated with swimmers. Illnesses were categorized with respect to similarity of symptoms rather than similarity of cause. Upper respiratory ailments for example may be caused by several different viruses and some of these same viruses such as the adenoviruses may also cause gastrointestinal symptoms (Sartwell, 1965). Colds, which are caused by rhinoviruses, were in a separate classifi cation for purposes of interviewing and then were amalgamated with upper respiratory infections to simplify analysis. Two of the swimming instructors were related to the local medical health officer and may have been slightly biased to over estimate. 12 Otitis externa or dermatitis of the external auditory canal was specifically included because of its association with s swimming (Senturia, 1956; Wright, 1972). Shistosomiasis or swimmer's itch is a known hazard of lake swimming and caused by shistosome cercariae (MacKenthum, 1964) which, as a part of their life cycle, are released from snails two or three weeks after the onset of warm weather. Beaches in Kelowna are treated every four years with copper sulfate to eliminate the host snails. Conjunctivitis was included in anticipation of an infection sometimes experienced by chlorinated pool swimmers. No distinction was made between viral, bacterial or chemical conjunctivitis. I was often present at the swimming classes to check ailments with the children. At the completion of each two week swimming sessionlliriinterviewed parents by telephone. The parents of the Ogopogo participants were interviewed at the end of July and August. Incidence raifcher than prevalence was surveyed and so a condition was tabulated only once even when a similar infection re-occurred. If, for example, the same participant suffered a succession of ear inflammations the problem was listed as one case not several separate ones. I interviewed by telephone rather than written questionnaire in order to have an unbiased return rate. I opened the conversation by saying the survey was a South Okanagan Health Unit project on the health of children. 13 Rather than ask a set order of questions I threaded the questions into conversation and as the comprehension of parents varied, so did my manner. With each interview the following basic information was collected: age and sex of child types of i I Inesses or infections date of first symptoms previous incidence of similar problem servcces of physician utilized swimming locations swimming frequency Data was accepted from swimmers who swam either only at the Rutland chlorinated pool or in Lake Oganagan. Participants who swam in waters other than the designated pool or lake were excluded from the final data. BACTERIOLOGICAL SAMPLING Water samples were taken from Rutland Pool, Ogopogo Pool and the two creeks and were analyzed for fecal coliforms and fecal streptococci. Since Ogopogo pool and the Hot Sands are in juxtaposition, Ogopogo water samples were accepted as representing water quality fow both. To ascertain this I ran a preliminary comparative test series for the two swimming areas (Appendix 3). The creeks, Mill and Brandt's 14 (Figure 2), are located on either side of the Hot Sands and they were each sampled at the mouth for their potential contribution to the beach water quality. Mill Creek is 0.53 miles from Ogopogo Pool and is fed by ground water seepage from surrounding residential and cattle feedlots and from mountain runoff. Its peak flow is during the summer months and 1972 was a particularly heavy runoff year. Brandt's Creek, 0.84 miles from Ogopogo Pool, drains the industrial section of Kelowna and collects effluent from "Calona Wines" and the "Sun Rype Ltd." fruit processing plant. Brandt's Creek has a lower mean discharge and a lesser nitrate con centration than Mill but a higher Biological Oxygen Demand and total coliforma growth (Appendix 2). During July and August of 1972 the water sampling schedule for each site was as follows: Location Frequency T i me Rutland Pool Ogopogo Pool Mill Creek Brandt's Creek Daily Twice or thrice times dai ly Thrice weekly Thrice weekly 9 a.m. 8 a.m. and/or 12 p.m., 4 p.m. 8:00 a.m. 8:30 a.m. 15 Figure 2. Water quality sampling sites. 16 Water sampling of the lake and creeks continued through September although interviewing of swimmers was completed by the end of August. In the summer of 1973 additional sampling was done to investigate for the presence,of Pseudomonas aeruginosa. Each location was sampled three times a week in the morning and water was analyzed for fecal coliforms and Pseudomonas aeruginosa. Samples were collected in 6 oz. sterilized bottles from the water surface by facing the bottle' mouth against the flow direction. For the chlorinated pool sodium thiosulfate treated bottles were used. Surface sampling is the accepted mode in bathing water quality studies (Carpenter, 1972). Bacteria settle both in the sediment and at the surface but it is the top layer which contacts the nose, mouth, eyes and ears of swimmers. Resuspension of silt may interfere with Bacterial concentrations but generally it has been accepted that surface sampling is sufficient (Tennant, 1971). I cultured a series of sediment samples (Appendix 3) for a comparison with surface samples. Highest daily air temperatures were recorded and the lake and Rutland pool surface water temperatures were measured two to three times a week (Appendix 3). The pH of each sample site was determined weekly with bromthymol blue. In August 1972 urine samples from six lake swimmers were analyzed for fecal coliforms, fecal streptococci, and 17 Ps. aeruginosa. The tests were done to substantiate the suspicion that urine was a potential contributing source of these organisms. LABORATORY ANALYSIS I completed all bacteriological analysis with Millipore membrane filter apparatus following the techniques and culture medias outlined in "Standard Methods 13th Edition," and the "B.B.L. Manual," and "Biological Analysis of Water and Wastewater." The plastic "disposable" petri plates were sterilized in ethanol followed by a 24 hour immersion in 1% bleach and rinsing with distilled water. Mayonnaise jars sufficied in lieu of erlynmeyer flasks for autoclaving the media. The media was made very four or five days and stored in a 10°C fridge between use. Grids and filter paper were transferred to plates with flat tipped sterilized tweezers. Broth media were dispensed with 2 ml syringes. All plates were taped prior to incubation. Fecal coliforms were isolated using M-FC broth in the ampoule form and were incubated in a water bath at 44.5°C for 24 hours. At this elevated temperature only fecal coliforms grow (Kabler, 1961) and colonies are shiny ink blue. KF broth was used for fecal streptococci isolation since it is the 18 media most frequently recommended for accurate results (Kenner, 1961). To give colour differentiation to the streptococci colonies one ml of 1% tetrazolium solution was added to the media (Figure 3). Fecal streptococci colonies were incubated at 37°C in a dry heat incubator and colonies ranged in colour from pale pink to deep red. To further identify fecal streptococci colonies I transferred with a platinum loop a selection of colonies to-brain-heart infusion agar. Plates were incubated at 44.5°C two days and 10°C for five days. Growth at these temperatures distinguished strains according to the schematic outline in Figure 4 (Geldreich, 1 969); Ps. aeruginosa was isolated on pseudosel agar and incubated at 37° and checked after 18-24 hours and 48 hours. Colonies which diffused green pigmentation or f1uorescedtunder UV were quantified as Ps. aeruginosa (Dr. Weinstein, pers. comm.). Results were compiled as the number of organisms per 100 milliliters. For each indicator organism tested two sample plates and one control plate were incubated. The two sample counts were averaged and raised to the nearest integer. For samples which were based on less than 100 mis of water the counts were multiplied by the required multiple to again yield organisms per 100 mis. Samples which produced crowded growth were rejected. 19 The proximity of sampling sites allowed bacterial plating to be completed within an hour of sampling and hence die-off rate was minimal and accuracy of isolation maximal.^ The persistence of fecal coliforms and fecal streptococci varies with the species. Most species diminish 50% to 80% after a one to two day lag while others such as s. faecalis survive up to two weeks (Geldreich, 1969). Due to these differing survival rates I felt it was important to culture samples immediately. 'Public Health samples are ordinarily sent to the Provincial Laboratory in Vancouver which means a one to three day lag and it is conceivable that this time lag results in a significant loss of accuracy. It is contended by the S.O.H.U. that provincial lab results are invariably lower than local analysis. Fecal Streptococci KF Broth Fecal Coliforms M-FC Broth Ps. aeruginosa Pseudosel Agar Figure 3. Laboratory analysis of fecal contamination i ndi cators. 21 GROWTH KF Agar Brain Heart GROWTH at 45°C and 10°C Enterococci - gut warm blooded animals S. faeoal-is var. liquefaoiens-ubiqui, tous S. faeealis - (atypical) vegetation source Red-Pink Colonies Infusion Agar GROWTH at 45°C only S. bovis - livestock and poultry sources S. equinus - livestock and poultry sources Figure 4. Fecal streptococci: schematic outline of preliminary identification. Chapter 3 WATER QUALITY RESULTS BACTERIAL WATER QUALITY The details of the fecal indicator counts from each water sample are presented in Tables I to V and depicted in Figures 5 to 8. At the chlorinated pool fecal coliforms were unde tected while fecal streptococci were frequently isolated and ranged as high as 168 organisms/100 mis. Streptococci are slightly more chlorine resistant than coliforms and it is not uncommon for fecal streptococci to be present in chlorinated pools (Mailman , 1962). The counts yielded in the chlorinated pool samples were dependent on the density of swimmers at the time of sampling (Mailman, 1962).. The surface waters have pockets of contamination which vary with the distribution of swimmers and these pockets are destroyed by dissemination and chlorine. Additional sampling of the pool would have provided a more descriptive representation of the changing fecal streptococci 2 2 23 levels. I collected all the Rutland samples five to ten minutes after children entered the water and it would have been interesting to test levels later in the day when crowding was heavier. The lake water samples usually produced fecal coliforms and fecal streptococci counts which fluctuated in unison (Figure 5). However there were distinct differences between the daily vicissitudes of the two indicators. Fecal coliforms were more stable over time and sometimes remained zero for several days. The fecal streptococci counts tended to be higher and increased throughout the day. The 8 a.m. samples were taken before even the avid tourists were swimming but by noon, when the beaches were crowded, fecal streptococci counts had increased (Table II). Fecal coliforms did not display this daily trend nor did they reachsthe high levels of the streptococci. There weire six occasions when fecal streptococci were over 200/100 mis but only once was fecal coliforms this dense. In Mill Creek the fecal streptococci organisms usually predominated over fecal coliforms (Figure 6) while in Brandt's Creek the fecal col iformscounts were far higher than fecal streptococci (Figure 7). In Brandt's Creek there was one atypical week in which both fecal index organisms were destroyed by the chemicals released during "Sun Rype" processing. During July, the fecal streptococci counts in Table I Fecal Coliform Counts, July, August, September 1972 Date Rutland Pool (Chlorinated) Ogopogo Pool 8 a.m. Ogopogo Pool 12 p.m. Ogopogo Pool 4 p.m. Brandt's Creek Mill Creek July -11 0 8 0 38 584 .12 0 • 4 4 . 2,000 720 13 0 20 40 14 68 75 _ 5,000 1 ,820 17 o 0 4 330 18 0 0 10 0 19 0 0 25,000 440 . 20 0 0 0 8 0 21 2 9 20,000 330 24 0 2 - 8 340 560 25 0 2 0 - 14,000 270 26 0 72 36 10 18,000 710 27 0 16 8 46 28,000 560 , 28 0 i 4 5 251 7,800 480 31 0 2 0 0 2,000 370 Aug. ' 1 0 2 2 40 80,000 316 2 10 8 -3 0 22 4 4 140,000 452 4 0 80 60 80 7 0 20 4 • 6 27,500 380 8 0 0 4 4 9 0 8 2 10 17,500 604 10 0 6 12 11 0 14 40 136 80,000 268 14 0 4 0 6 6,600 580 15 0 12 76 0 CONTINUED Table I (Continued) Date Rutland Pool (Chlorinated) Ogopogo Pool 8 a.m. Ogopog Pool 12 p.m. Ogopogo Pool 4 p.m. Brandt's Creek Mill Creek Aug. 16 0 0 22 6 * 0 664 17 0 6 0 18 0 0 2 0 0 280 21 0 2 16 0 196 22 0 0 6 0 23 0 100 0 0 2,000 84 24 0 0 0 4;800 25 0 2 108 1,700 676 28 0 18 18 0 240 30 10 2 700 31 0 0 0 1 ,400 500 Sept. 5 36 0 5,200 228 8 0 . 0 2,980 240 12 2 0 6,300 60 15 2 0 3,200 400 19 2 0 1,000 0 22 0 6 1 ,314 50 24 26 4 4 390 50 29 0 0 46,000 218 Zero readings due to NaoH from fruit processing. Table II Fecal Streptococci Counts, July, August, September 1972 Date Rutland Pool (Chlorinated) Ogopogo Pool 8 a.m. Ogopogo Pool 12 p.m. Ogopogo Pool 4 p.m. Brandt's Creek Mill Creek July •11 12 13 14 17 18 19 20 21 24 25 26 27 28 31 Au|. 2 3 4 7 8 9 10 11 14 15 0 0 3 3 1 14 16 0 4 82 15 8 2 0 3 1 168 1 2 22 10 39 7 2 4 1 2 32 28 18 9 75 6 0 52 15 38 15 10 2 1 113 88 75 80 113 . 28 96 350 775 134 • 226 226 0 . 63 26 27 22 110 9 92 6 113 '103 21 290 198 8 48 22 102 89 330 18 •1 87 85 12 27 87 206 34 32 9 7 0 10 340 40 44 4 92 2 312 560 2,500 2,600 3,000 950 290 2,500 4,500 4,500 2,000 4,000 232 144 1 ,000 48 102 474 68 700 1 ,600 3,000 3,200 2,800 1 ,500 466 492 228 604 204 224 PO ov CONTINUED Table II (Continued) Date Rutland Pool (Chlorinated) Ogopogo Pool 8a.m. Ogopogo Pool 12p.m. Ogopogo Pool 4. a.m. Brandt's Creek Mill Creek Aug. 16 0 6 49 2 8 0 17 0 0 1 18 30 35 22 0 8 0 21 12 0 41 88 900 22 10 12 1 7 24 1 1 56 25 41 37 84 3,000 8 28 2 8 58 1,750 41 30 0 0 57 12,000 400 31 0 1 0 14,000 390 Sept. 5 17 0 880 626 8 31 1 1 ,460 1,400 12 .- 1 1 15 0 0 1 ,200 60 19 0 6 1,000 0 22 0 0 8,000 38 24 26 o 0 - 0 29 4 o 60,000 0 ro OGOPOGO POOL, FECAL COLIFORM VS. FECAL STREPTOCOCCI < 720h % 640 o 560 2 480h ~ 400 g 320 _i | 240 I60h HIGHEST DAILY COUNTS FECAL COLI o-oFECAL STREP 20 JULY 30 9 19 AUGUST FIGURE 5 29 8 18 SEPTEMBER MILL CREEK, FECAL COL!FORM VS. FECAL STREPTOCOCCI k—A FECAL COLI 0--0 FECAL STREP 20 JULY o IK I \ IN3 9 19 AUGUST FIGURE 6 8 18 SEPTEMBER FECAL INDICATOR COUNTs/lOO mis LOGARITHMIC SCALE PO CD CO o O p o o o o O b o. o o o o o o o c o o o o o o o oe 31 Mill Creek were slightly higher than counts in Brandt's whereas during August, the fecal streptococci counts in Brandt's were noticeably higher than counts in Mill Creek. After September 15th 1972 counts dropped for both indicators at all locations except Brandt's Creek. Brandt's dropped temporarily and then peaked again towards the end of September. Morphology and Identification of Fecal Streptococci Fecal streptococci colonies were easily distinguished by colour and I have summari ze.ddi n the next table typical colours and colony sizes observed. I did not quantify colonies in reference to colour since there is not yet sufficient in formation to match morphology on KF broth with streptococcal species. Each location had its own anomalies. In Mill Creek a large purple colony appeared several times during the summer and at Rutland the streptococci colonies suspiciously resembled the colonies later isolated from urine. Although this colour differentiation is not of immediate value for this study, it may, in future, become a guideline for determining source of contami nation. The results from the preliminary fecal streptococci differentiation on brain heart infusion agar are listed in Table IV. Again this was a qualitative sampling of colonies 32 Table III Fecal Streptococci Morphology COLONY MORPHOLOGY ON KF AGAR Range in size of colonies: 0.1 mm - 1.0 mm Coloured colonies examined under microscope: were gram + ve some diploid only most in chains most colonies were shiny and moist COLOURS OBSERVED: Rutland Pool: Small bright pink colonies. Ogopogo Pool: Wide variety of colours. Brandt's Creek: Variety of pink colours. Mill Creek: Samples were often predominated by medium pale pink colonies. The only sample site containing purple colonies. Any colonies appearing as opaque white were counted as overgrowth. and not a measurement of the actual ratios of livestock and human contamination. Strains of human origin were isolated from each sample site. This was substantiated by growth at both 10°C and 45°C. According to the results at 10°C each site was also shown to harbour the livestock strains S. bovis and s. equinus. The livestock strains are particularly short Table IV Fecal Streptococci: Preliminary Identification of Human and Livestock Strains a. Growth of Fecal Streptococci on Brain Infusion Agar - 10° and 45° These colonies were either Enterococcl, atypical S. faeaalie or S. faeaalie vav liquefaoiene. Colony Source Size Colour on KF Broth Rutland Pool smal 1 pale pink Ogopogo Pool 1 arge dark red large orange ii medium pink " medium orange medi urn purple 1 arge pale pink medium dark red " 1 arge pink 1 arge orange medi urn pal e " 1 arge pale yel1ow-orange smal 1 red Brandt's Creek medium dark red Mill Creek medium dark red II medium dark pink II 1 arge pale pink II 1 arge pink b. Growth of Fecal Streptococci on Brain Infusion Agar - 10°C These colonies were either S. bovis or S. equinue of livestock origin. Rutland Pool medi urn pale yel1ow-orange II medi urn pink Ogopogo Pool large pink II smal 1 orange M smal 1 bright orange •1 large pink II medium pink it large pale pink. n medium orange ii j medium pale yellow-orange n smal 1 red it medium dark red II 1 arge orange Brandt's Creek 1 arge pale pink II smal 1 dark pink •I medium yel1ow-orange Mill Creek medi urn pink .1 mm small .5 mm medium 1.0 mm larae 34 short-lived and theref oreJ-i n order to have been isolated the contamination must have been fairly recent. The lake water samples displayed the most varied colour representives of both human and livestock strains. PSEUDOMONAS AERUGINOSA AS A WATER QUALITY INDEX 2 During my exploration of health and swimmers I inter viewed two doctors for their thoughts and observations. The ear, nose and throat specialist (Dr. P. Found, pers. comm.) mentioned that he had in previous years sent in swabs from patients complaining of otitis externa but since Ps. aeruginosa was always identified he now no longer bothered to verify the cause. He said he was not certain if the Pseudomonas was associated with the lake water and so I decided to pursue this aspect. Samples mailed to the Provincial Laboratory during August 1972 confirmed the presence of the potential pathogen at Brandt's Mill, Ogopogo and Hot Sands, but not in the Rutland pool . In the summer of 1973 I obtained the quantiative results listed in Table V. Mill Creek was positive for Ps. aeruginosa on 63% of the days sampled, Brandt's and Ogopogo on 30% of the days. In an environmental study of the Gatineau Lakes (Tannant, 1971), where fecal coliforms and fecal streptococci counts were comparative to those measured in 35 Lake Okanagan, Ps. aeruginosa was confirmed in only 8% to 10% of samples from lakes and tributaries. In comparison therefore I feel that the Kelowna beach warrants further investigation into the quantitative^ presence of Ps. aeruginosa and its potential role as pathogen and indicator organism. The pseudomonas data ddid not relate to the fecal coliforms fluctuations observed during the same summer. Ps. aeruginosa is present in 10% of healthy human fecas (Reitler, 1957; Sutter, 1961) and I had expected the organism to correspond with coliform counts. Ps. aeruginosa is capable of metabolizing over 100 organic compounds and hence would be able to survive well in the creeks but again its presence was not verified as consistently as I had expected. Not all strains of Ps. aeruginosa are pathogenic and even apparently identical strains may or may not be pathogenic (Dr. J. Campbell, pers. comm.). It is established as a pathogenic organism in skin necrosis, mixed intestinal infec tions and otitis externa. Without further study I cannot interrelate Ps. aeruginosa with any other variables. The results confirm and establish the presence of this organism in Mill Creek, Brandt's Creek and Ogopogo pool. ''The quantiative accuracy of pseudosel media in combination with membrane filter technique is unsettled (Dr. S. Weinstein, pers. comm.). I found that when pseudomonas did grow it was obvious (Figure 3) since each colony was enveloped in a diffusion of green pigment. Table V Ps. aeruginosa and Fecal Coliform Counts, July and August 1973 a. Pseudomonas aeruginosa counts Date Mill Creek Ogopogo Pool Brandt's Creek June 11 0 0 o 20 0 0 0 22 0 0 0 27 0 0 0 July 3 0 0 0 4 0 46 0 6 T.M.T.C. 0 0 9 0 0 0 11 13 0 0 13 1 2 0 17 6 0 0 18 6 2 0 20 0 3 0 23 0 0 .0 -25 6 0 0 27 1 3 0 30 16 7. 0 Aug. 1 6 2 0 3 1 0 0 6 10 0 4 8 17 0 12 10 4 1 6 13 1 0 4 15 9 39 1 12 17 4 2 2 20 1 0 16 22 3 1 0 24 T.M.T.C. 0 Fecal coliform counts Date Mill Creek Ogopogo Pool Brandt's Creek June 11 56 0 448 20 144 0 520 22 >172 0 50-27 436 0 100 July 3 320 138 280 4 400 0 320 6 650 96 T.M.T.C. 9 50 2 73,000 11 200 154 56,000 13 270 - 240,000 17 300 0 >100,000 18 230 0 129,000 20 - 6 600,000 23 210 5 130,000 25 490 0 150,000 27 1 00 0 2,600,000 30 130 32 -Aug. 1 152 12 580,000 3 332 98 22,500,000 6 160 2 80,000 8 1 ,100 100 >35,000 10 640 34 21 ,000 13 560 12 . >1,000 15 650 12 >1 ,000 17 475 66 >3,200 20 210 4 400 22 410 8 >1 ,300 24 1,250 4 14,600 37 ADDITIONAL VARIABLES MEASURED  Temperature Water temperature fluctuations clearly responded to air temperature changes (Table XVII). The Rutland pool temperatures were similar to those of Ogopogo pool and dependent on air temperatures. I did not measure swimming density but it was visibly obvious that hotter weather induced more children to swim more frequently. Urine and Fecal Indicators Toilet facilities were far enough from the beach to make urinating in the water easier. Urine in the bladder is sterile but on passing through the urinary tract it contacts with fecal coliform and fecal streptococci. Normal body flora is predominated by Gram positive staphylococci and streptococci (Carpenter, 1972) and therefore in urine fecal streptococci tend to outnumber fecal coliforms. I coaxed a few unwilling children to donate urine samples and the results obtained are listed below'. The female urine in these samples produced the higher counts. Water pH Chemical effluents, temperature changes, swimming load and season Biological Oxygen Demand increases were a few 38 Table VI Fecal Indicator Counts in Urine Samples from Lake Swimmers Sex Fecal Coliforms Fecal Streptococci Ps. aeruginosa M 0 TMTC*(> 200,000) 0 M 40 TMTC 0 M 4,000 45,000 0 F 80 TMTC 0 F 6,000 TMTC 0 F TMTC (> 200,000) TMTC 0 * TMTC = Too Many to Count of the variables which influenced pH changes (Appendix 3). The pH values fluctuated from 6.8 to 9 at Brandt's Creek, 7.5 to 8.6 for Mill and 7.2 to 8.4 for Ogopogo. At Rutland the pH readings were on occasion below the required standards, of British Columbia. Since colourimetry rather than a pH meter was used I question the accuracy of these results. During the summer, alkalinity increases in the lake are related to increased primary productivity, and C02 and bicarbonate depletion. During this period the average lake pH rises from 8.0 to 8.2. This subtle change was not measurable at the Ogopogo pool because shallow water combined with swimming density interfered with typical pH readings of the lake water. Chapter 4 ILLNESS RESULTS INADEQUACIES OF INTERVIEWING To uncover illness patterns, which were often asso ciated with subjective symptoms, was, at times, a confusing task. Children complained about imaginary problems, hoping for attention, mothers could not recall which child had been sick and what one parent considered serious another one over looked. There were occasions when parents suppressed infor mation already given by their children. In one case a young boy reported an obvious shistosomiasis rash but when the mother was contacted several days later she insisted her son was fine and denied the possibility of a skin irritation. When being interviewed certain parents immediately sensed the issue of a "pollution study" and volunteered their own interpretations. Some parents stated that they refused to allow their children to swim in the lake because of "pollution" or previous summers of ear infections. One parent announced that the lake must be improving because "the children weren't 39 40 as sick this summer." Mothers of the higher income Ogopogo swimmer's were more sensitive to the questions I asked, more distraught over unwell children but less opinionated than residents from other areas. The psychological awareness of an illness was sub jective as was my interpretation of what was said so the incidences recorded were undeniably biased. V. attempted to be consistent when interpreting data. For instance, if there were conflicting remarks between a parent and child I then believed whichever person claimed an illness had occurred. The major advantage of the conversation approach was that eventually 100% of children and parents were contacted. GENERAL ILLNESS PROBLEMS Health problems enumerated during the swimming season were compiled with reference to illnesses observed per group (TableeVII) and then for ease of visual and statistical com parisons converted into percentages (Table VIII). Unfortunately, sample size varied both between and within the three groups surveyed. Statistical transformations, however, were not deemed necessary for a generalized comparison of illness patterns. I have collated illnesses according to the weeks surveyed rather than individual days. Water samples were not 41 taken during weekends and so I could not statistically match illnesses with two of every seven days. Also when talking with children or parents they were often vague as to whether the onset of an illness was one day or the next. I attempted to redue-e the degrees of freedom and lost statistical power. I omitted the variables age and sex from illustra tions since I did not consider these variables meaningful for the elementary analysis desired. I also amalgamated the two categories "upper respiratory ailments" and "colds" into one. Although the respiratory illnesses represented several etiological agents and was more ambiguous than the "cold" category, combining the two categories was convenient and did not affect statistical relevance. With regard to total illnesses experienced during the summer of 1972, the highest percentage was from the Ogopogo aquatic team (88.5%) followed by Hot Sands swimmers (51.4%). The total illness rates of the lake swimmers included shistosomiasis and otitis externa, neither of which was prevalent amongst chlorinated pool participants. This difference accounted in part for the lower total illnesses observed with chlorinated pool swimmers. For comparative interest a 1959 Okanagan Lake illness survey based on identical categories and similar types of swimmers reported relatively lower total illness percentages (Clarke, 1959): 42 1956-59 Percentage of Illness Amongst Lake  Okanagan Swimmers Site 1956 1957 1958 1959 Ogopogo Aquatic 31% 40% 72.4% 50.9% Pool Hot Sands Beach 13% 21% 25% 20% At all three swimming sites upper respiratory infec tions represented a significant problem and for Rutland and Hot Sands swimmers colds were a main complaint. The one obvious sex related difference I observed in the raw data was with colds. The incidence of colds was more frequent among girls. Since the sex ratio at Rutland and the Hot Sands was in the favour of females this may explain the high percentages quoted for upper respiratory problems. Studies from other fresh water regions have measured "eye, ear, nose and throat" ailments as representing 60% to 70% of total illnesses (Stevenson, 1953; Dutka, 1973). Illness results from the present study yield an ever greater proportion of such ailments: 43 "Eye, Ear, Nose and Throat" Infections of Swimmers .'ecatlor, Eye, Ear, Nose and 0, 0/ ^ T 1 .. location Throat Infections /o % of Totil Infections Rutland Pool 43 87.8% Hot Sands Beach 55 72.4Ogopogo Pool 43 93.6% This does not categorically imply that the Okanagan has a sig nificantly higher eye, ear, nose and throat infection rate because it may also indicate a lesser problem with other ill nesses such as gastroenteritis. INCIDENCE OF SPECIFIC ILLNESSES Swimmer's itch or shistosomiasis was observed on the legs of younger lake swimmers and occasionally on instructors who taught standing in the water. Two cases were reported by Ogopogo swimmers but they may have acquired the rash wading through the shallower waters to reach the pool. Some indivi duals, owing perhaps to skin pH, are more sensitive than others to the cercariae embedding. One mother interviewed had four children who swam together but only one acquired the 44 rash. The highest incidence was reported during the second week in July, two weeks after the beginning of warm weather. Four cases of conjunctivitis were reported by pool swimmers but not until August. Red and sore or itchy eyes have in past been a condition observed in lake swimmers although there were no instances of irritated eyes during 1972. It has been suggested (Dr. D. Clarke, pers. comm.) that either algae or chemical effluents were responsible for previous lake associated eye redness. Gastroenteritis was reported by each group of swimmers but again not until August for Rutland swimmers. The last week in July four Rutland swimmers had gastroenteritis but because they had been swimming elsewhere the data was excluded. From Ogopogo swimmers three cases of gastroenteritis were reported, all during the fourth week in July, while from Hot Sands swimmers gastroenteritis was reported in July and August. Into the miscellaneous category, I put everything from pneumonia to low grade fevers. In early August several lake swimmers mentioned they had stomach aches combined with difficulty breathing. It seemed too non-specific to include with miscellanea but in future perhaps more attention should be taken with this category. The outstanding difference between lake and chlori nated pool swimmers was the incidence of external ear infections. 45 From 200 lake swimmers a total of 47 otitis externa cases were noted while at the chlorinated pool only one instance was reported out of 248 swimmers. Amongst the Ogopogo swimmers, otitis externa was prevalent during the entire swimming season but incidence peaked to 30.8% in July and then dropped to zero after the first weekuj nsAugustt.h WithsHqJu Sands participants the incidence of external ear infections was more evenly distributed throughout the summer, with a peak of 12.8% (Table VII). The highest incidence for both groups of lake swimmers was the final week in July. External ear infections have been a swimming associated annoyance for many years and it is my impression that the problem has increased within the past decade. The figures from a survey of the Ogopogo aquatic team in the late fifties (Alcock, 1960) were lower than the 1972 results. In this earlier study information was not given as to the age range of the group, swimming frequency or other variables and the comparison with present rates may be unjustified. Cases of Otitis Externa: Ogopogo Aquatic Swim Team 1957 1958 1959 1960j 1972 Number (in group 40 58 55 74 52 % ea>annfecti:ohsons20.0% 39.7% 18.2% 31.0% 57.7% 46 There were also several other brief surveys done on Lake Skaha in 1970-1971 in which the incidence of otitis externa did not fall below 15% for any given group (South Okanagan Health Unit, 1971). The ear infections varied in severity and after the initial one most children suffered a series of ear aches. Each person interviewed stated that he did not experience problems during other seasons although several said they had suffered ear infections in previous summers. Some children used isopropanol drops on occasion and three Ogopogo swimmers wore ear plugs. Such prophylactic measures were likely successful and may have lessened the incidence potential. With Hot Sands swimmers there was an age differential associated with otitis externa. The younger children were not as frequently afflicted and this was probably attributable to swimming ability and hence exposure of ears to water. Amongst 6 to 8 year olds the incidence was 14.8% and amongst 12-14 it was 43%. Although the participation age distribution for Ogopogo swimmers was similar these children did not exhibit the same otitis externa age distinction. Even the youngest wese accomplished swimmers and in competitive swimming undoubtedly were well exposed to water regardless of age. Table VII Cases of Illness Amongst Chlorinated Pool and Lake Swimmers, 1972 Location Date Group Si ze Upper Respiratory Infection Colds Otitis Externa Gastroenteri ti s Shistosomlasls (Swimmer's Itch) Conjunctivi tis Misc. Tota1s Rutland Pool July 1st Wk 2nd 3rd 4th 58 58 60 60 1 2 2 2 8 1 2 ---3 10' 3 2 Aug. 1st Wk 2nd 3rd 4th 80 80 50 50 1 4 3 1 7 3 1 1 3 1 1 1 1 1 ' 2 4 11 13 7 4 TOTAL 248 13 25 6 - 4 4 53 Hot, Sands July 1st Wk 2nd 3rd 4th 62 62 63 63 . 1 2 2 2 1 5 7 13 2 4 8 3 1 6 1 -3 18 14 27 Aug. 1st Wk 2nd 3rd 4th 23 23 16 6 -2 3 1 2 2 2 2 -5 9 TOTAL 148 7 31 17 11 10 - 76 Ogopogo Aquatic Pool July 1st Wk , 2nd' 3rd 4th 52 52 52 52 1 1 2 3 1 1 . 4 , 8 16 3 2 -2 2 1 5 12 22 Aug. 1st Wk 2nd 3rd 52 52 52 3 2 -1 - . - -4 2 TOTAL 52 9 4 30 3 2 - 4 46 Table V.I 11 Percentages of Illness Amongst Chlorinated Pool and Lake Swimmers, 1972 V Location Date Group Si ze Upper Respiratory (inc. colds) Otitis Externa Gastroenteritis Shistosomiasis Swimmer's Itch Conjunctlvi tis Misc. Total Per Cent 111ness Rutland Pool July 1st Wk 2nd 3rd 4th 58 58 60 60 5.17% 17.30 5.0 3.33 - 5.16% 17.3 5.0 3.33 Aug. 1st Wk 2nd 3rd 4th 80 80 50 50 2.5 13.75 12.0 2.0 1 .25% 3.75% 1125 2.0 2.0 1.25% 1.25 4.0 ' 5.0% 13.8 16.3 14.0 8.0 TOTAL 248 15. 32% jl .25% 2.42% 1.61% 5.0% 21 .37% Hot Sands July 1st Wk 2nd 3rd 4th 62 62 63 63 3.23 11.29 14.29 23.80 v 3.22% 6.35 12.7 4.85% 6.35 1 .63% 9.7 1 .59 4.85% 29.09 22.3 42.8 Aug. 1st Wk 2nd 3rd 4th ; 23 23 16 6 8.69 13.04 4.35 8.70 8.7 8.7 8.7 21 .7 39.2 TOTAL 148 25.67% 11.50% 7.43% 6.75% 51.35% Ogopogo Aquatic Pool July 1st Wk 2nd 3rd 4th 52 52 52 52 1.92 7.69 5.77 1 .93% 7.7 15.4 30.8 5.77% 3.85 3.85 3.85 1.92% 9.6 23.1 42.3 Aug. 1st Wk 2nd 3rd 52 52 52 5.77 3.85 1 .9 3 7.7 3.85 TOTAL 52 25.00% 57.70% 5.77% 3.85% 7.69s; 88.46% Chapter 5 INTERPRETATION OF WATER QUALITY DATA Fecal streptococci is beginning to become more popular in North America as an indicator organism to be measured in conjunction with fecal coliforms for the determination of recreational water quality. I purposefully used the word popular since the functional role of indicator organisms and the somewhat arbitary health standards for recreational water receives constant criticism (Ames, 1969; Dutka, 1973; Kevin, 1968; Krishnaswami, 1971). In my scanning of the literature I abstracted one author who presented epidemiological evidence correlating illness rates of swimmers with bacterial water quality and the level noted in his research (Stevenson, 1953) was 2,300 total coliforms. Whether it is salt, fresh or chlorinated water the arguments abound as to what is a "potential health hazard." It is claimed that 200 fecal coliform and TOOsfecal streptococci (Tennant, 1971) are the demarcation points but these conclusions are not founded on 49 50 sufficient empirical evidence. A comment by Mailman (1962) sums the criticisms for chlorinated pool standards. Swimming pool bacteriological standards are really standards of attainability and are not based onaepidemiological v evidence. 1 Recent Canadian studies on recreational water quality, which include parts of the Okanagan Basin Study and the Environment Canada report on the Gatineau Lakes, accepted as their initial premise the health standards summarized in Appendix 1. By testing assorted public beaches and finding them within total and fecal coliform limits both studies interpreted these results as conclusive evidence of acceptable water quality. It was not questioned further as to whether water quality could still have a negative impact on health. Bacterial results from the Kelowna beachtwere within the British Columbia water standards but I wish to go beyond this preset limit and interpret the "potential health hazard" to swimmers from association withithis one particular beach. Although the fecal indicator counts were within the health standards this does not deny the possibility that the observed water quality had a deleterious effect on health. My data lacks the depth and variability upon which to estimate Mailman, "Cocci Test for Determining Mouth and Nose Pollution of Swimming Pool Water,""AJPH, Vol. 47, p. 2001. 51 precise points or "magic number" at which water quality does or does not affect health. It does however have enough sub stance to describe certain relationships which exist between the streams, the lake and the swimmers. FECAL CONTAMINATION OF THE CREEKS Fecal coliform and fecal streptococci are both incubated in the feces of warm blooded animals, however, the relative ratios of the organisms varyy with the types of animal. Human feces contains roughly four times more fecal coliform than streptococci while domestic and wild animal feces grow greater numbers of fecal streptococci (Geldreich, 1969). This is a useful distinction for tracing sources of recent contamination. The basic relative ratios rule to be used with discretion is:-: - if the fecal coliform to fecal streptococci ratio is greater than 4.0 contamination is of human origin - if the ratio is less than 0.7 contamina tion is from animal origin (Geldreich, 1969). Due to discrepancies of die off rates this rule applies only to fresh fecal contamination. From the creek water quality data it was apparent that fecal streptococci predominated in Mill Creek and fecal 52 coliform in Brandt's. The ratios at Mill were commonly less than 0.7 and in Brandt's usually greater than 4.0 (Table IX). Since Mill Creek received more agricultural drainage and Brandt's the industrial effluents, the ratios comply with expectations. Ogopogo water quality from the morning samples was a mixture of ratios and will be discussed separately. Table IX Ratios of Fecal Indicator Counts Fecal Coliform: Fecal Streptococci n ? •:- a Date Ogopogo Pool 8 a.m. Samples Brandt's Creek Mill Creek July 26 4.0 6.92 0.355 27 1 .78 9.45 0.176 28 0.053 2.20 0.178 31 0.33 6.90 0.247 August 2 2.0 32.0 0.676 3 0.2 31 .2 0.92 4 13.3 — — BACTERIAL STATISTICAL ANALYSIS Did bacterial counts in Mill Creek fluctuate concomitantly with those in Ogopogo Pool? Were fecal streptococci counts in Ogopogo significantly higher than fecal 53 coliform and did the creeks contribute to these counts? To further interpret such question I made use of paired "t" tests and product moment correlation coefficients. The "t" tests were used to check whether the time and location differences were significant. The correlation coefficients were used to determine how closely aligned the index organisms of the lake were with those of the creek and to relate tempera ture and fecal indicators. Since sample sizes varied, the significance of the correlation coefficients was further tested with the "t" test. The statistical analysis for bacterial water quality is summarized in Table X. FECAL CONTAMINATION OF THE KELOWNA BEACH Prior to interpreting the water quality of the lake swimming area I wish to emphasize two sampling comparisons which were done. The first was the comparison of Ogopogo Pool with the Hot Sands Beach and the second was the comparison of the surface with sediment samples (Table X). The "t" test accepted the null hypothesis for the first comparison and therefore all bacterial measurements taken can be related to Ogopogo Pool and Hot Sands swimmers. In the second com parison the null hypothesis was accepted for fecal streptococci 54 but rejected for fecal coliforms. This results implies that stirring of the sediments by wind or swimmers may have affected surface fecal coliform counts. Depth related variations have also been observed in other studies but they were con sidered of li'minor" importance (Tennant, 1971). When the Ogopogo 8 a.m. fecal coliform counts were compared statistically with the noon and 4 p.m. fecal coliform results the difference was insignificant (Table X). When the daily fecal coliform counts were compared with the daily fecal streptococci levels the differences were significant at 8 a.m., 12 p.m. and 4 p.m. As was initially apparent in the results, fecal streptococci counts were fractionally higher than fecal coliform. An additional significant difference was detected between 8 a.m. fecal streptococci counts and the 12 p.m. and 4 p.m. fecal streptococci counts. The morning fecal streptococci counts were statistically lower than the noon or afternoon counts. There was not, however, a significant difference between the 12 p.m. and 4 p.m. counts. The count differences observed, both between the two index organisms and during the daily fecal streptococci increases, likely relates to the presence and density of swimmers. Natural skin flora and urine, as mentioned pre viously, are predominated by Gram positive streptococci (Carpenter, 1972). When daily swimming desntiy increased 55 so did the amount of direct contact between water and skin. Fecal streptococci counts correspondingly increased with the desntiy of bathers and then dropped again each night as the water dispersed. As one of the mothers I spoke with aptly described, "There's nothing wrong with the lakes. People who swim in them are dirtier thannthey are." If hot weather inspires swimmers, who adds to the fecal streptococci levels it is logical that temperature should positively correlate with the indicator. The coefficient calculated (Table X) was .504 which, given 32 degrees of freedom, was significant. The correlation is also apparent in Figure 9 in which the minimum air temperatures of July 20thii sAugust 17th and August 30th coincide with the lowest inflexion points of the fecal streptococci counts. Since the coliforms were not equally susceptible to density changes the correlation with temperature was less with a coefficient of .216. The pH changes may also have been density dependent but I did not take sufficient samples during swimming classes to ascertain any trends. At the Rutland pool neither pH nor temperature correlated with the unpredictable fecal streptococci counts. Table X Statistical Analysis: Comparison of Fecal Contamination Indicator Counts Paired "t" Tests H„: d = 0 H,: d > 0 < d a: TWO TAIL *a - .05 for all one and two tailed tests d.f. = degrees of freedom Data from Table XX , 1. Ogopogo Pool vs. Hot Sands Beach a. June Fecal Col 1. b. August Fecal Coll. c. August Fecal Strep. n = 15 d.f. = 14 "t"a = 1.776 1 calc -J:>u . *. ACCEPT Ho Counts from Ogopogo Pool were accepted as equal and hence a substitute for counts from the Hot Sands n = 18 d.f. = 17 "t" =2.110 a * calc -J0U .-. ACCEPT Ho n = 17 d.f. = 16 "t" = 2.120 a "f'calc " -9809 .' . ACCEPT Ho 2. Ogopogo Pool: Surface vs. Sediment Samples Data from Table XIX H9: ..d" = 0 Hi: H" < 0 > d~ a: TWO TAIL a. Fecal Coli. b. Fecal Strep. n = 10 d.f. =9 "t" • = 2.262 -fcaU = 2.416 ! . ' . REJECT H o There was a significant difference between surface and sediment fecal coli. n = 12 d.f. = 10 "t" =2.201 a Mt"calc = "637 .'. ACCEPT Ho CONTINUED Table X (Continued) 3. Ogopogo: 8 a.m., 12 p.m., 4 p.m. Data from Tables I and II H0: d" < 0 H,: d > 0 a: ONE TAIL a. Fecal Coll., 8 a.m. vs. 12 p.m. b. Fecal Coli., 8 a.m. vs. 4 p.m. c. Fecal Strep., 8 a.m. vs. 12 p.m. n = 30 d.f. = 29 »t'^ = 1.699 "t'clc " -386 .". ACCEPT H0 n = 25 d.f. = 24 "t" = 1 .711 a "t" i =1.489 calc . * . ACCEPT Ho n = 30 d.f. = 29 "t" = 1 .699 a -t-calc = 1.832 .'. REJECT H0 d. Fecal Strep., 8 a.m. vs. 4 p.m. e. Fecal Strep., 12 p.m. vs. 4 p.m. n = 23 d.f. =22 "t" =1.717 •t-calc - 2.459 .-. REJECT H0 The fecal strep, counts at 12 p.m. were s ign i f1 cantly higher than 8 a.m. counts . n = 18 , d.f. = 17 "t"a = 1 .740 "'"calc = T'681 . * . ACCEPT H0 4. Ogopogo Pool: Fecal Coli. vs. Fecal Strep. Counts ' z Data from Tables I and II H0: d < 0 Hi: d~ > 0 a: ONE TAIL a. 8 a.m. Samples b. 12 p.m. Samples c. 4p.m. Samples n = 35 d.f. = 34 "t"a = 1.690 "t" , = 1.908 calc .*. REJECT H0 n = 27 d.f. =28 "t" = 1.701 a "*" " 2-871 .'. REJECT H0 n = 23 d.f. = 22 "t" = 1.717 •fcalc = 2.623 . - . REJECT H0 There was a significant difference between each set of fecal coli. and fecal strep, counts Table X (Continued) Temperature Correlation with Fecal Indicators 'Data from Tables I, II and XVII < n Hj H0: r < 0 poiitive correlation not significant r < 0 significant positive correlation a = ONE TAIL a. Temperature and Ogopogo Fecal Coll, n = 37 d.f. 35 1.69 "*"calc = K31° ACCEPT Ho r = .216 b. Temperature and Ogopogo Fecal Strep. 35 d.f. = 33 "t" = 1 .693 a Mt"calc = 3-352 .'. REJECT H0 r = .504 Significant correlation between fecal strep, counts and air temperatures c. Temperature and Rutland Fecal Strep, 33 d.f, 31 "t" = 1 .896 a "fdc - -698 .-. ACCEPT Ho r = .1.24 Pseudomonas aeruginosa (1973) Data from Table V H0: r < 0 positive correlation not significant Hi: r > 0 a: ONE TAIL - significant positive correlation a. PB. aeruginosa and Temperature b. Pe. aeruginosa and Ogopogo Fecal Coli. n = 24 d.f. =22 "t" = 1.717 i,tMcalc = "-496 . •. ACCEPT Ho I r = -.042 n = 22 d.f. = 20 "t" = 2.086 a -t-calc -.857 ACCEPT Ho r = -.180 <J1 CO OGOPOGO, THRICE DAILY COUNTS OF FECAL STREPTOCOCCI O O < O 400 c/) ^ 300 Z> o o Q_ Ld £ 200 r— 00 00 0 1 I I 27 31 JULY • 8:00 A.M. 112:00 P.M. 4:00P.M 1 1 In i_Q 1 P I I 4 7 AUGUST i I I I FIGURE 8 HIGHEST DAILY FECAL STREP COUNTS/lOOmis LOGARITHMIC SCALE — _ ro oj •£» (T> co TEMPERATURE UF 61 CORRELATION OF THE CREEK AND LAKE WATER QUALITY Crowding was not the sole source of the colonies which grew from the lake waters. Into Lake Okanagan through numerous villages and cities lead 2,340 square miles of tributaries. With regard to fecal indicators, which survive at most several weeks, the nearby outfalls provided another source of fecal organisms. For Ogopogo pool and the Hot Sands beach, Mill and Brandt's Creek were closest with wind direc tion favouring Mill Creek. Another five miles to the south of the Hot Sands beach was the outlet from the Kelowna primary waste treatment plant and of all inputs into the lake, the sewage plant releases the highest bacteria and nutrient loading (Health Branch, 1973). I did not investigate the possible impact of this effluent on the Hot Sands beach because I assumed that the outfall was too distant to cause a substantial effect. To determine the influence of Mill and Brandt's Creeks on the water quality of Ogopogo pool or the Hot Sands beach I correlated the bacterial data from the separate locations using first a four hour lag and then a 24 hour lag (Table XI). Computations from another series with an eight hour lag were not listed since results were analogous to the first set. For the four hour lag the morning creek counts were compared with the noon lake counts and for the one day lag the morning creek counts were paired with the 62 8.a.mi: 1 ake-counts?.of the next day. As the mouths of both creeks were within one mile of the Ogopogo pool, it seemed reasonable that potential bacterial correlations would occur within a one day time span. The data were analyzed separately for July and August because the fecal indicators in Brandt's Creek were more irratic in the latter month. From Mill Creek both fecal coliform and fecal streptococci counts correlated positively, during July, with Ogopogo fecal indicators. The coefficients imply that Mill Creek contributed bacterial loading to the Hot Sands Beach. Although the fecal streptococci counts in Mill Creek were higher than the fecal coliform counts it was the fecal coliform count correlation which produced the stronger coefficient (r = .887). This irregularity was due probably to the sensitivity of fecal streptococci to the influence of alternate variables such as swimming density. Brandt's Creek and Ogopogo Pool provided one, sig nificant but likely accidentia!, correlation during July when the coefficient for fecal streptococci was r = .578. The coefficients during August for both fecal coliforms and fecal streptococci were insignificant. As a verification of the creeks and lake coefficients I then cross related the counts from the two creeks. The July fecal streptococci were coincidentally in phase (r = .812). It was this correlation which probably accounted for the 63 chance fecal streptococci correspondence between Brandt's Creek and Ogopogo Pool . Not one of the creek and lake correlation coefficients during August was significant. Does this indicate that after July neither of the creeks continued to influence beach water quality? It may, but it may also indicate that wind and water current changes created too much interference for a significant correlation without more elaborate analysis. Since there appears to be a potential fecal contamination correlation between the creeks and the lake, spring would be a preferable season toqquestion, without swimming load inter ference, the association of lake and creek water quality. Table XI Statistical Analysis:Correlatlon of Fecal Indicator Counts H »: r < 0 H,: r > 0 Positive Correlation Not Significant Significant Positive Correlation 1 ct : .05, ONE TAIL Data from Tables I and II 1n Product Moment Correlations. 1. Mill Creek vs. Ogopogo Pool Samples a. July fecal coli.: 4 hour lag (8 a.m. and 12p.m.) b. August fecal coll.: 4 hour lag c. July fecal strep.: 4 hour lag n = 13 d.f. = 11 "t" = 1.796 a "'"calc " 6-365 .*. REJECT H0 r = .887 July fecal coll levels In Ogopogo correlated strongly with Mill C reek fecal co1 I n = 10 d.f. = 8 "t" = 1.860 a '^'calc ° -207 . '. ACCEPT Ho r - .073 n = 12 d.f. » 10 "t" = 1.812 a "fcalc = 2.265 . ' . REJECT H o r =• .582 July fecal strep also correlated with Ogopogo counts d. August fecal strep.: 4 hour lag e. July and August fecal coll.: 1 day lag (8 a.m. & 8 a.m.) f. July and August fecal strep.: 1 day lag n = 9 d.f. = 7 "t"a = 1.895 Mt"calc - --450 .•. ACCEPT Ho r • -.450 n = 19 d.f. = 17 "t" = 1.729 a "tMcalc " -8526 ACCEPT Ho r = .203 n = 18 d.f. = 16 "t"a = 1.746 "fcale - 1.712 . '. ACCEPT Ho r - .393 2. Brandt's Creek vs. Ogopogo ' a. July fecal coli.: 4 hour lag b. August fecal coll.: 4 hour lag c. July fecal strep.: 4 hour lag n = 12 c.f. = 10 "t"a = 1.812 -t-c-lc - .102 .*. ACCEPT H0 r = .032 n = 10 d.f. = 8 ."t"a = 1 .860 "*"c«lc = -231 ACCEPT H0 r = .231 ' n = 14 d.f. = 12 "t"a = 1.782 -t"ealc - 2.455 .'. REJECT H o r = . 578 _ • CONTINUED t: Table XI (Continued) 2. Brandt's Creek vs. Ogopogo (Continued) d. August fecal strep:: 4 hour lag e. July and August fecal coli;: 1 day lag f. July and August fecal strep;: 1 day lag n = 8 d.f. = 6 "t"0 = 1.943 "f'calc = ^ . * . ACCEPT H0 r = .408 n = 22 d.f. - 20 "t"0 = 1.725 nt"calc = -642 .•. ACCEPT H0 r = .142 n = 17 d.f. =15 "t"a = 1.753 "t"calc :- 1.165 . *. ACCEPT H0 r = .288 3. Mill Creek vs. Brandt's Creek a. July fecal coli. b. August fecal col 1. c. July fecal strep. n = 13 d.f. = 11 "t"a 1.796 "*"calc = "-071 ACCEPT Ho r = -.022 n = 9 d.f. = 7 "t"a = 1.895 «t"calc = -1.145 .' . ACCEPT Ho r = -.397 n = 12 d.f. = 10 "t" = 1.812 "t'e.le " 4-397 . - . REJECT H0 r = .812 Fecal strep, levels corre lated in Brandt's and Mill Creeks during July . d. August fecal strep. en n = 12 d.f. =10 "t"a = 1.812 Mt"calc = -215 !.'. ACCEPT H0 r = .068 Chapter 6 INTERPRETATION OF ILLNESS DATA INCIDENCE AND EXPOSURE Throughout the study I deleted the variables which I did not consider necessary for exposing the initially hypothesized interactions. I extracted only those facets which were required for a linear description of the relation ship between water quality and the health of swimmers. One of the neglected variables was economic status. I assumed a minimal social and economic difference between Rutland and Hot Sands swimmers but there were likely certain distinguish ing reasons as to why some parents and children preferred one location to another. Whatever the reason, I have con tinued to assume that such differences did not affect illness patterns described earlier. Another important variable which I only partially accounted for was swimming exposure. I did not attempt to assess total swimming time for participants either during the weeks they were being surveilled or prior to their par ticipation. I treated each two-week swimming session as a 66 67 separate group even though some of the Rutland and Hot Sands swimmers either had been enrolled in an earlier session or continued into the next. Total swimming exposure was relevant to the opportunity for contact with other swimmers and water quality and hence was relevant to the probability of con tracting an illness. Although the maximum was not measured the number of minimum swimming hours was known and for the Rutland and Hot Sands swimmers it was equivalent. Therefore all illness incidences listed in Table XII were based on identical minimum water exposure. Ogopogo swimmers however swam not only regularly for the entire summer but an additional two hours per week more than other participants. To clarify the illness comparisons between Ogopogo pool and other sites the inci dences per 1000 person hours exposure were calculated as shown in Tables XII and XXI. Since children tend to romp in and out of the water all day the figures quoted may be severely underestimated for the Rutland and Hot Sands swimmers. The Ogopogo swimmers, especially the older ones, were con cerned with competition and not frolicing and so in their case the discrepancy between maximum and minimum exposure depended on the frequency of swim meets. One of the introductory hypotheses was that incidence of illness was higher amongst lake swimmers. The most concise 68 result supporting the hypothesis was the incidence of otitis externa and although percentages from Table VIII indicate the incidence at Ogopogo to be far greater than at Hot Sands the calculation in Table XII described the two location as having a surprisingly close incidence in terms of water exposure. I had assumed that the two lake groups would exhibit concurring illness patterns for each category's not just otitis externa. Again from Table XII this was evidently not the case for upper respiratory infections or gastroenteritis. The Ogopogo incidences per 1000 person hours exposure were even lower than incidences for the same ailments at Rutland. This contradiction of expectations may be explained by two possibilities. The chlorinated pool water, which is recycled for several days before renewal, may harbour viruses not prevalent in the continually moving lake water. Also the Ogopogo swimmers may develop an immunity to some of the in fectious strains of gastroenteritis and upper respiratory viruses. An increased resistance would explain the lower incidence.ofOf those Ogopogo swimmers who did suffer infec tions, attack was not confined to one age bracket or new members. There were six new members in 1972 but even the youngest had learned to swim in the lake in years previous and perhaps they too had developed immunity. When compared with the Hot Sands swimmers, Rutland swimmers exhibited lower incidences per 1000 person hours UJ cr 40.0 ID 00 o Q_ X LU GO rr 30.0 Z> O X o CO fr LU 20.0 Q_ O O o \ LU O 10.0 z: LU Q O z: ILLNESS INCIDENCE PER 1000 PERSON HOURS EXPOSURE JULY and AUGUST 1972 • RUTLAND M HOT SANDS §1OGOPOGO JZ1 Upper Respiratory Otitis Externa Infections ILLNESS CATEGORY FIGURE 10 Gastroenteritis Shistosomiasis Conjunctivitis en Table XII Illness Incidence per Thousand Person Hours Swimming Exposure _. Ju.-ly and lu'g'§_1 972_ Location Upper Respiratory Otitis Externa Gastroenteritis Shi s Conjunctivtis Rutland Pool 25.538 .672 4.032 2.688 Hot Sands Beach 39.832 17.820 11.530 10.482 --Ogopogo Aquati c Pool 7.142 16.484 1 .648 1 .099 --J Observed Illness x 1000  I (#in Group)(Min Swim Hrs Per Week)(Weeks Exposed) o Ta-ble XIII Statistical Analysis: Comparison of Weekly Illness Incidence for Chlorinated Pool and Lake Swimmers Paired "t" tests ' ! Illness Comparison of Chlorinated Pool and Lake Swimmers Data from Tables VIII and XXI H0: d < 0 Hi: d > 0 •• a: ONE TAIL 1. Rut!and Illnesses vs. Hot Sands a. Upper respiratory infections b. Gastroenteritis c. Otitis Externa n = 8 d.f. = 7 "t" = 1 .895 a "tncalc " -416 .•. ACCEPT H0 n = 8 d .f. = 7 "t" = 1.895 a "tMcalc = 1'798 .*. ACCEPT H0 2. Rutland Illnesses vs. Ogopogo n = 7 d.f. = 6 n = 7 d.f. = 6 "t" = 1.943 a "t" = 1.943 a "'"calc = 2-425 Mt"calc = -5992 .'. REJECT H o . * . ACCEPT Ho Significant difference in incidence of upper respira tory infections between Rutland and Ogopogo sw i mme r s. 3. Hot Sands Illnesses vs. Ogopogo n = 7 d.f. =6 n = 7 d.f. = 6 n = 7 d.f. = 6 "t" = 1 .943 a "t" = 1.943 a "t" = 1.943 a Mt"calc = 3-355 "*"calc " 2-413 Mt"calc = -055 . • . ACCEPT Ho . • . ACCEPT H0 . - . ACCEPT Ho Significant difference be tween lake swimmers Significant difference be-tween lake swimmers 72 exposure in all categories except conjunctivitis. However when the upper respiratory and gastroenteritis incidences of the two sites were compared on a weekly basis using paired one tailed "t" tests the differences measured were not sig nificant (Table XIII). In other words, for upper respiratory ailments and gastroenteritis the weekly percentage differences were small enough to be accepted by a "t" test but the total differences for the summer were significant in terms of person hours exposure. Generally the results indicate that Hot Sands swimmers suffered the highest incidence per 1000 person hours exposure for each category except conjunctivitis while Ogopogo swimmers experienced the lowest incidence of the gastroenteric and upper respiratory illnesses. OTISIS EXTERNA AS A SWIMMING RELATED PROBLEM Ps. aeruginosa is the causative organisms most often associated with otitis externa (Hardy, 1954; Wright, 1974) and as was recorded earlier this potential pathogen was present in the lake and creeks but not in the chlorinated pool. I do not have enough evidence to verify a correlation between the quantitative presence of pseudomonas and the incidence of otitis externa but the literature lends support to the probability that Ps. aeruginosa is, at least, a secondary 73 etiological! agent of the ear problems which for years have afflicted Okanagan swimmers. If one were to be suddenly immersed in water densely contaminated with Ps. aeruginosa an ear infection would not develop. A series of changes must occur to the ear canal prior to infection and these changes are related to the fre quency of water exposure. The resident bacterial flora of a healthy ear canal isSGram positive (Wright, 1974) and does not vary with geographic location, sex or season (Lindsay, 1958). The canal has an oily acidic coating released from sebaceous glands and this coating acts as a mechanical barrier and bacteriocidal agent against penetration by micro-organisms (Carr, 1961). When the canal is exposed frequently to water the epithelial lining is disrupted, ducts become plugged, lipids are removed (Senturia, 1954) and the pH becomes alkaline. This process is hastened by warm temperatures and high humidity and it would also, I presume, be hastened by alkaline water. At this stage the ear is predisposed to otitis externa and if Ps. aeruginosa is introduced into the ear the inflammatry dermatitis of otitis externa is produced. This process has been observed to occur naturally in ^swimmers (Wright, 1 974) and divers (Cobet, 1970) and has been artificially induced in cats (Senturia, 1954) and guinea pigs (Wright, 1972). Recently Wright (1974) hypothesized that the crucial period 74 is when the flora in the ear canal changes from Gram positive organisms to Gram negative. Gram negative organisms other than pseudomonas such as E. coli have also been confirmed as causative agents but not as frequently. Swimmers in Lake Okanagan were in contact with fecal coliform and Ps. aeruginosa, both Gram negative organisms. According to the literature, it is the water which damages the meatal surface presetting conditions for infection and according to my data it is the lake water which harbours the Gram negative pathogens responsible for otitis externa. The water plays the sequential role of leaching the lipid coating and then introducing the pathogens. It would have been interesting to ascertain how the incidence of otitis externa corresponded with the presence of Ps. aeruginosa. In Table XI (page 64) pseudomonas did not positively correlate with either air temperature or the 1973 fecal coliform counts and samples were too infrequent to test for a lag correlation. Presumably if Ps. aeruginosa did not correlate with fecal coliforms it would also not correlate with fecal streptococci. I hesitate to draw infer ences from the intermittent data §f> one summer. One of the physicians interviewed remarked that the swimmer's ear problems occurred generally two weeks after the beginning of summer weather. This concurs with the data collected in which the highest incidence of inflammation was 75 the latter half of July. Otitis externa does not spread to other areas of the ear and so is usually treated with pain relievers and auralgen or glycerol . Swimmers in the chlorinated pool, except for one case, did not develop otitis externa.1 Had Ps. aeruginosa been present in the chlorinated water then conditions would also have favoured the development of external ear dermatitis amongst Rutland swimmers. In a survey of 205 Colorado chlorinated pools, 32 were deemed "unsafe" because of the qualitative presence of Ps. aeruginosa and of the pools registered "unsafe" only 54.3% were positive for total coliforms (Palmquist, 1973). The conclusion from this survey was that Ps. aeruginosa* because of its "ability to cause otitis externa infections in swimmers," should be included as a bacterial indicator in the examination of chlorinated pools. I agree with this suggestion and think it should be extended to include fresh water bathing areas if and when useful quantitative Ps. aeruginosa levels can be agreed upon. It is the responsibility of public health departments to settle on suitable and useful preventative standards. During 1973 the chlorinated water of a Kelowna motel was checked because of complaints of external ear infections. Counts average 14/100 mis Ps. aeruginosa. 76 Chapter 7 CORRELATION OF ILLNESS INCIDENCE AND WATER QUALITY STATISTICAL ANALYSIS Product-moment correlation analysis was used to describe the concomitant fluctuations between illness incidence and water quality (Tables XIV and XV). In the first series illnesses were paired with bacterial counts from the same week. Two sets of bacterial counts were used to counter balance the fact that within each, swimming group different children swam at separate times throughout the morning and afternoon, and so were exposed to differing levels of con tamination. In one set weekly illness incidence was paired with the highest fecal index counts from the same week. In the alternate set* illness incidence was matched with the corresponding weekly mean count. Then, because of differences in illness incubation periods, two additional correlation series were calculated. Into one series a two day lag was introduced which was intended particularly for the gastroenteritis incubation period and 77 into the other a seven day lag was introduced to account for upper respiratory problems. Since correlations were on a weekly rather than a day to day basis the implications of introducing a lag would depend on when during the week illness events occurred relative to maximum fecal contamination levels. For instance if illnesses were maximum toward the end of a given week and bacterial counts towards the beginning then the resultant correlation coefficients would be similar both with or without a two day lag. The correlation coefficients were transformed into "t" values to test whether the correlations were positively significant. The length of swimming season varied for locations and so coefficients were based on different degrees of freedom. There were two inherent weaknesses in the correla tion analysis. One as mentioned earlier was loss of degrees of freedom due to amalgamation of illness data and the other weakness concerned the relevance of fecal contamination data. Since water quality, especially fecal streptococci counts change significantly within hours or even within minutes it must be assumed that the measured water samples were typical of the bacterial density to which individual swimmers were exposed. In actuality the bacterial level encountered by a swimmer was dependent on swimming density, time of day and location with the pool or beach and even within one sub-group 78 of swimmers the range of water quality varied far more than the tabulated figures indicate. DISCUSSION OF CORRELATION ANALYSIS Does the measured disparity in illness incidence vary dependently or independently with the measured fluctuations in water quality? Are the periods of high illness incidence paralleled by higher levels of fecal contamination? The main objective of the study was to answer this question of correla tion between recreational water quality and the health of swimmers. If significant positive correlations were defined it signified support of the basic hypothesis. Conversely if positive correlations were not statistically revealed then either the hypothesis was unsupported or the wrong indicator organisms were used. Positive statistical correlations give credibility to the hypothesis but the lack of correlations does not necessarily invalidate the hypothesis. I stressed the logic of correlations because the intended function of water quality indicators is often overlooked and water is judged to be clean or degraded on pre-set unquestioned standards. The coefficients calculated in Table XV offer con vincing support to the hypothesis of water influenced illnesses. Fecal streptococci and illness data consistently provided 79 stronger and more frequently significant correlations than fecal coliforms. There were eight instances in which illness correlations with fecal streptococci were significant while the corresponding correlations with fecal coliform were not significant. Rutland Correlations As the control group, Rutland illness incidence ideally would not have required correlation analysis. How ever, not only were fecal streptococci present but on occasion chlorinated pool counts were higher than morning lake counts. It was therefore necessary to determine the coefficient in Table XIV. Not one of the coefficients was significant but the gastroententis coefficient (r = .599) was more suggestive than the other illnesses of a positive correspondence and was certainly stronger than the zero value expected. The .599 correlation was in fact identical to the fecal streptococci coefficient for gastroententis amongst Hot Sands swimmers. Fecal indicators do not confirm the presence of pathogens and so the correlation of gastroententis and fecal streptococci does not implicate the water as a disease transmitter but merely reinforces the usefulness of fecal streptococci as an indicator measurement for chlorinated water. 80 Rutland illnesses were also correlated with Ogopogo fecal streptococci (Table XIV). Coefficients were reassur ingly and insignificantly negative. Lake Water Quality and Illness Correlations Generally the correlations without lags yielded the strongest coefficients. Even when two day lags were signifi cant, coefficients were still weaker than results without lags. For Ogopogo swimmers the use of fecal streptococci highest counts rather than of mean counts yielded the stronger corre lation while with Hot Sands swimmers the reverse occurred and use of fecal streptococci mean counts rejected three hypothesis where highest count correlations had rejected two. Since Hot Sands swimmers were dispersed over a more expansive area than the Ogopogo swimmers, mean counts rather than highest were likely a more realistic representation of water quality. The use of fecal coliform mean or fecal coliform highest counts did not affect the significance of any tests due probably to the fact that daily fecal coliform counts were far more stable than fecal streptococci levels. With Ogopogo upper respiratory infections neither fecal coliform nor fecal streptococci provided a significant positive correlation whereas with Hot Sands incidence fecal streptococci counts yielded a coefficient of .836 without a 81 lag and .639 with the two day log. The weak Ogopogo correla tion was compatible with the low incidence per 1000 person hours exposure (Table XXI) and the rejected "t" test (Table XIII) and again a possible reflection of the resistance factor mentioned previously. The substantial fecal streptococci correlation with Hot Sands respiratory data was a decisive indication that the 39.83 incidence per 1000 person hours was directly associated with swimming water quality. In Rutland the incidence for equal exposure was 25.53 with a correlation coefficient of -.317, a distinct contrast to the beach correlation. In response to upper respiratory correlations for all locations the fecal coliform were per ceptibly less sensitive. For Ogopogo swimmers gastroenteritis was signifi cantly positive with both fecal streptococci and fecal coliform. The three recorded cases occurred not only during the same week but on the same day which happened to be July 26, 1972, the day of the highest fecal streptococci counts isolated throughout the summer (775/100 mis). Two days earlier the count had been 350/100 mis which was the second highest sample recorded. It is apparent why the Ogopogo gastroenteritis correlated with water quality in spite of its low incidence per swimming hours exposure. With respect to otitis externa all fecal coliform correlations were insignriificant. For Hot Sands swimmers the 82 fecal streptococci and otitis externa correlation were sig nificant without a lag (r = .866) and with a two day lag (r = .802) while for Ogopogo participants the fecal streptococci coefficients were strongly positive (r = .629) but not significantly so. The correlation for Ogopogo would likely have been significant had prevalence rather than incidence been measured. The correlation for otitis externa and lake swimmers becomes even clearer when as with gastroenteritis it is viewed relative to days rather than weeks. TableaXVI (page 93) lists water quality and the respective days on which Ogopogo bathers reported ear aches. Again July 24th and 26th were Unfavourable days for fecal streptococci counts and otitis externa. The fecal streptococci and otitis externa correla tion expressed certainly warrants further investigation and it is to my knowledge the first time such a correlation has been demonstrated. If as indicated Ps. aeruginosa is the causative organism it would be interesting to clarify what relation, if any, pseudomonas has in time and density with fecal streptococci. The 1973 Ps. aeruginosa counts did not correlate with fecal coliforms but then otitis externa also did not correlate to a significant degree (r = .498) with fecal coliform. Does the presence of high fecal streptococci density imply the presence of Ps. aeruginosa and if so why? Since pseudomonas is not common to skin flora why would 83 Ps. aeruginosa increase conjointly with fecal streptococci? There is really negligible information on the correlation of Ps. aeruginosa or fecal streptococci with otitis externa, on the behaviour of Ps. aeruginosa in natural waters and on therrelationship of fecal streptococci with pseudomonas. Because of this dirth of information or investigation, I cannot fully explain why the 1972 Lake Okanagan fecal streptococci correlated with otitis externa. There is perhaps a propor tionality between fecal streptococci and pseudomonas similar to Geldreich's ratio in which case fecal streptococci counts would implicate potential pseudomonas levels. For future investigations on the behaviour of Ps. aeruginosa and its relation to other fecal contaminants I would recommend that water samples greater than 100 mis be filtered in order to set a baseline of fluctuations and avoid the likelihood of zero readings. Value of Indicator Organisms The fecal contamination indicators, fecal streptococci in particular, corresponded more closely with the illness patterns of lake swimmers than with the illness incidence of chlorinated pool swimmers. The strongest correlations were displayed with upper respiratory and otitis externa infections and then to a lesser extent with gastroenteritis. These 84 positive correlations confirm the introductory hypothesis that the incidence of illness amongst lake swimmers was associated wi th water quali ty. Fecal streptococci was throughout the study the more flexible and responsive indicator and it was evident that fecal streptococci correlated more positively than fecal coliform with all lake illnesses and with gastroenteritis from the Rutland swimmers. However, neither fecal coliform nor fecal streptococci correlations can clarify causes, and water quality, although it correlated with illness was not at any time actually proven as the direct carrier of pathogens or the direct cause of observed illness incidence. Contact with crowds at the beach, the fruit season, tourist influx plus other factors also contributed to the routes and rates of illness incidence. If recreational water quality does influence the incrideneee of some illnesses then where is the demarcation point for fecal indicators at which correlations either become positive or convincingly more positive? The data collected from one beach for one season was not sufficient in scope to set a basic limit. The highest coinciding levels of bacteria and illness were during the fourth week in July when fecal streptococci reached 775/100 mis and fecal coliform 251/100 mis, and for Hot Sands swimmers the incidence of upper respiratory infections was 24% and for otitis externa 85 13%. However, standards cannot be extrapolated from the evidence of one week of unreplicated data. Extracting the limits of water quality acceptability is not only a statistical task but a subjective one arbitrarily dependent on the potential health risks one is willing to accept. H,: 1. Table XIV Correlation of Fecal Indicator Counts and Illness Amongst Chlorinated Pool Swimmers r 5 0 r > 0 Positive Correlation Not Significant Significant Positive Correlation ° " .05, ONE TAIL Rutland Fecal Streptococci Counts and Illness Incidence Data from Tables VIII and XXII n • 7 d. f. - 5 "t" - 2.015 a UPPER RESPIRATORY GASTROENTERITIS CONJUNCTIVITIS "f'calc " °"748 . • . ACCEPT Ho "f'calc * 1-306 ACCEPT Ho "'"calc " -814 .-. ACCEPT H, r - -.317 r • .504 r = .342 Fecal Streptococci Uslnq Mean Counts, No Lao, n = 7 d. f. = 5 nt"a - 2.015 •fcalc - .0499 ACCEPT,H0 "tncalc " '-673 ACCEPT Ho "t" , = .659 calc ACCEPT H0 r = .022 r = .599 4 - .283 Fecal Streptococci Using Highest Counts, 2 Day Laq n = 8 d. f. ' 6 "t"a - 1.943 "'"calc • -0713 . ". ACCEPT H0 "t" " 1-332! .- . ACCEPT H0 . "'"calc • -828 .- . ACCEPT H0 r * .0319 r = .512 r = .347 2. Rutland Illness Incidence Correlated with Ogopogo Fecal Streptococci, No Laq n • 7 d.f. - 5 "t"a - 2.015 Rutland fecal streptococci levels did not significantly correlate with Illness Incidence In previous tests. Following tests done to check possibility of coincidental positive correlation between Rutland Illness Incidence and lake fecal streptococci counts. UPPER RESPIRATORY GASTROENTERITIS : . ' -"t" • - 779 1 calc -719 "^calc = -294 ACCEPT H0 ACCEPT Ho CO r - -.329 r - -.131 YcT'abd e;;«II Stati sti cal A n a 1 y s i [sg > ^ C 0 rf.eilflit i 3 nd of a €eg ad u cl n dii eait OKC! C oil nit>s;\a..n.d.. 'Id:l n.esst Amongst Lake Swimmers r < 0 Positive Corr ' o •; S ' •• If leant r > 0 Significant Positive Correlation a : .05, ONE TAIL Ogopogo Fecal Indicator Counts and Illness Incidence Data from Tables V111 aa rid/XX11 ^ — - • •• • ii--' - II'-. I- a_ /-» _ . . ._ J M _ I ~ „ a n = 6 d. f. = 4 "t" =2.132 a UPPER RESPIRATORY GASTROENTERITIS OTITIS EXTERNA Fecal coli. Fecal strep. Fecal coli . Fecal strep. Fecal coli . Fecal strep. "t" n = .0243 calc .'. ACCEPT H0 r = .0124 "t" n = .783 cal c .*. ACCEPT H0 r = .364 "f'caU = 2-418 . ". REJECT H0 r = .734 "f'calc - 3-598 . * . REJECT H0 r = .874 "t" , = 1.149 calc . ' . ACCEPT Ho r = .498 "t-calc " ]'617 .'. ACCEPT H0 r = .629 CONTINUED CO V a b iRa bxljei iXiV (Continued) b. Fecal Indicators Using Mean Counts: No Lag n = = 6 d.f. = 4 "t" = 2.132 a UPPER RESPIRATORY GASTROENTERITIS OTITIS. EXTERNA Fecal coli. Fecal strep. Fecal coli. Fecal strep. Fecal coli. Fecal strep. "t" = - 271 z calc " t" = 751 z calc -/t51 '•t"calc = .854 "f'^T, = 3.054 c a '1: c "t,Jcalc " -409 ,,tNcalc = 1-325 .•. ACCEPT Ho . ' . ACCEPT H0 . '. ACCEPT H0 . * . REJECT H o . ' . ACCEPT Ho . ' . ACCEPT H0 r = -.134 r = .351 r = .393 r = .837 r = .200 r = .552 c. Fecal Indicators Using Highest Counts: Two Day Lag n = = 6 d.f. = 4 "t" = 2.132 UPPER RESPIRATORY GASTROENTERITIS OTITIS EXTERNA Fecal coli. Fecal strep. Fecal coli. Fecal strep. Fecal coli. Fecal strep. "t»calc-= 1.689 Mt"calc = -489 "fcalc.- .0975 "f'calc • 2'441 "tMcalc = --0716 "t"calc = 1.0132 .'. ACCEPT Ho . ' . ACCEPT H0 . ' . ACCEPT H0 .'. REJECT H o .'. ACCEPT Ho . * . ACCEPT Ho r = .645 r = .238 r = .0487 r = .774 r = -. 358 r = .452 CONTINUED CO CO TabTablelXV (Continued) 2. Hot Sand.Fecal Indicator Counts Data from Tables XII and XXVIII a. Fecal Indicators Using Highest Counts: No Lag n = 7 d.f. = 5 "t" = 2.015 a UPPER RESPIRATORY GASTROENTERITIS OTITIS EXTERNA Fecal coli . Fecal strep. Fecal coli . Fecal strep. Fecal coli. Fecal strep. "t" = 432 z calc . ' . ACCEPT Ho "t" , =2.352 cal c . ' . REJECT H o Mt"calc " «836 .*. ACCEPT Ho "t" , = 1 .674'-cal c . * . ACCEPT Ho "t" , = .880 cal c . ' . ACCEPT Ho "t" . = 3.421 ca 1 c .'. REJECT H0 r = .174 r = .725 r =-.350 r = .599 r = .366 r = .837 CONTINUED CO Tab'iTiaWL'enXiV (Continued) b. Fecal Indicators Using Mean Counts: No Lag n = 7 d.f. = 5 "t"„, = 2.015 UPPER RESPIRATORY GASTROENTERITIS OTITIS EXTERNA Fecal coli. Fecal strep. Fecal coli. Fecal strep. Fecal coli. Fecal strep. 111" = 117 1 calc ' M/ .'. ACCEPT Ho r = .068 "t" al„ = 3.400 calc . *. REJECT H o r = .836 Mt"calc = K326 . •. ACCEPT H0 r = .608 "'"calc•- 2-083 .*. REJECT H0 r = .682 "f'calc = -583 .*. ACCEPT Ho r = .253 "'"calc - 3'865 . * . REJECT H o r = .866 c. Fecal Inc iieator Using Highest Counts: Two Day Lag n = 8 d.f. = 6 "r = 1 .943 UPPER RESPIRATORY GASTROENTERITIS OTITIS EXTERNA Fecal coli. Fecal strep. Fecal coli. Fecal strep. Fecal coli. Fecal strep. "t" n = -.631 cal c . *. ACCEPT Ho r = -2.71 "t" n = 1.855 cal c .*. ACCEPT H0 r = .639 "t" , = 6630 cal c .>. ACCEPT H0 r = .271 "t" i = 1.881 calc . * . ACCEPT Ho r = .644 "t" n = -.220 calc . ' . ACCEPT H0, r = -.098 "'"calc - 2-998 . * . REJECT H0 r = .802 CONTINUED o Table XV (Continued) 3. Fecal Indicators Using Highest Counts: One Week Lag n = 5 d.f. = 3 "t" = 2.353 Ogopogo Swimmers UPPER RESPIRATORY Fecal coli . Fecal strep. "tMcalc = '°656 "fcalc-= -1.587 .'. ACCEPT HQ .*. ACCEPT H0 r = .0378 r = -.676 Hot Sands Swimmers UPPER RESPIRATORY Fecal col i . Fecal strep. "f'calc " -1'341 . ' . ACCEPT Ho r = -6612 "tncalc = -1'059 . • . ACCEPT Ho r = -.522 OTITIS EXTERNA INCIDENCE/IOOO PERSON HOURS and WEEKLY MEAN FECAL STREP ! FIGURE II 93 Table XVI Otitis Externa and Fecal Indicator Correlation Ogopogo Pool Swimmers Date Cases of Otitis Externa Fecal Highest Coli form Daily Count Fecal Streptococci Highest Daily Count July 7 10 1 2 - : 11 2 8 5 16 3 75 (July 14th)* 39 (July 14)* 17 2 4 15 18 1 0 18 20 2 0 1 23 2 9 (July 21st) 113 (July 21 ) 24 5 8 350 25 1 2 75 26 4 72 775 31 1 0 226 August 6 1 80 (August 4th) 113 (Aug. 4) If counts not available for that particular day closest alternate sample listed. Chapter 8 CONCLUSION Both pertinent and tenuous relationships were observed from the data documented. In brief, lake swimmers experienced higher illness rates than chlorinated pool swimmers and these incidences correlated positively in varying degrees with fecal bacteria counts. These counts in turn were contributed by several sources including the creeks and swimmers. The evidence gathered, which pertained to one specific age bracket and one beach, affirmed the hypothesis that illness patterns varied concomitantly with bacterial recreational water quality. The lake water quality, which remained within the bounds of today's standards, was traced to contamination from the creeks and swimmers. Fecal coliform and fecal streptococci data provided the basis for this conclusion. The relative ratio of fecal coliform to fecal streptococci also assisted in the differentiation of agricultural and human contamination. The contamination in Mill Creek was mainly from livestock origin while the lake water contamination was a combination predominated by fecal streptococci of human origin. 94 95 Fecal streptococci was a more versatile indicator than fecal coliform. It represented chlorine resistent microbial growth in the Rutland pool and was sensitive to increases in swimming density. Fecal streptococci was also the pre ferable organism in terms of positive correlation with lake illnesses. I concur with the commonly expressed opinion that coliform organisms alone are not a sufficient criteria for estimating recreational water quality. The overall illness patterns observed were distinctly different for each of the three groups surveyed. The Hot Sands swimmers suffered a significantly greater incidence than chlorinated pool swimmers of upper respiratory and otitis externa infections. The hardier Ogopogo swimmers were afflicted continuously with otitis externa but suffered in frequently from respiratory ppoblems. The fecal streptococci convincingly correlated with gastroenteritis upper respiratory and otitis externa infections of lake swimmers while fecal coliform correlated only with gastroenteritis. I did not determine whether the measured concordance between water quality and illness was causally related to water harboured pathogens or to other variables such as contact with beach density. The incidence, however, of otitis externa amongst lake swimmers was quite striikiiing riavn"dt I don't think it would be premature to implicate water quality as the variable responsible. The presence of 96 Ps. aeruginosa in the lake and surrounding creeks was verified but whether this organism was the provoking pathogen was not proven. My research evolved from the longstanding supposi tion that the health of swimmers suffered because of exposure to degraded or unsafe water quality and in essence the data validated the hypothesis. However it is equally important to realize that the water quality correlation with illness incidence was far less damaging than it potentially could have been. From the viewpoint of most tourists and residents it is the intangible aspects of relaxation, fresh air or sunshine which are memorable and not the inconveniences related in this study. One cannot chart the enjoyment gained from splashing about in a blue lake as concisely as one can count bacteria grown from the same water and the comfort of ignorance helps the lake to feel as refreshing as it looks. For some families the problems which I outlined in this study are just now beginning to become serious enough to interfere with leisure enjoyment. The influence of liake Okanagan's bathing water quality is starting to spread beyond the immediate relationshp with health and into the realm of social implications. In consideration of both the research findings pre sented and the future potential impact of water quality on recreational enjoyment I have two conclusions or recommenda tions to make. One suggestion is that additional public 97 beaches should be made available in Kelowna to relieve the density of the park beach. The beach itself has only a narrow periphery of sand where sometimes crowding is so ext- me there is insufficient space on the sand and people must occupy the bleachers. The Hot Sands or city park beach is the only lake bathing area in the vicinity of Kelowna which provides life guarding. Extended services to other beaches would also help reduce density. My second suggestion is that the British Columbia recreational water standards for both chlorinated and fresh water areas should be re-examined for their purpose and pro tective value. If checking water quality intermittently for total or fecal coliform is not relevant to existent problems such as the otitis externa incidence described, then the standards should be changed or augmented to suitably represent recreational water quality. •B I B L I 0 6 R A P H Y Alock, F.R. 1960. "Survey of Swimming Associated Infections at the Kelowna Aquatic Pool and Rutland Pool." Unpublished paper, South Okanagan Health Unit. American Public Health Association. Standard Methods for Examination of Water and Waste Water." 13th Edition, New York. Ames, Wendel. 1969. "Double Standards for Swimmers." New England Journal of Medicine* Vol. 281, p. 52. Bartley, C, L. Slanetz. 1960. "Types and Sanitary Signifi cance of Eecal Streptococci." American Journal of Public Health* Vol. 50, pp. 1545-1552. Campbell, J.J. October 10, 1972. Personal communication, Dept. of Microbiology, The University of British Columbi a. Canada, British Columbia Okanagan Basin Agreement. June 1972. Preliminary Study Data - Bulletin No. 2. Water Quality of the Okanagan Basin. Finlay Printing, Kelowna, B.C., 4 p. Canada ,BBri:tishC66iliumb\i.a Okanagan Basin Agreement. June 1972 . Preliminary Study Data - Bulletin No. 4. Land Use and Water Demand. Finlay Printing, Kelowna, B.C. , 4 p. Canada, British Columbia Okanaga Basin Agreement. June 1972. Preliminary Study Data - Bulletin No. 5. Recreation and Aesthetic Resources3 Finlay Printing, Kelowna, B.C. , 4 p. Carpenter, Philipp. 1972. Microbiology. 2nd Edition, Philadelphia, Saunders, 476 p. 98 99 Carr, CP., B.A. Bawmann, B.H., B.H. Senturia. 1961. "Studies of Factors Considered Responsible for Diseases of the External Auditory Canal," The Annals of Otology* Rhinology and Laryngology* Vol. L'XX, No. 4, pp. 1055-1060. Clarke, D.A., il*972-1973. South Okanagan Health Unit. Pers. comm. Clarke, D.A., F.R. Alcock. 1960. "An Epidemiological Survey of Kelowna Beaches and the Kelowna Aquatic Pool, including a Four Month Water Sampling Program and a Study of Infection Incidence in Various Groups of Swimmers." Unpublished paper, South Okanagan Health Unit. Cobet, A.B., D.M. Wright, P.I. Warren. 1970. "Tektite I Program Bacteriological Aspects." Aerospace Medicine* Vol. 41, pp. 611-616. Dutka, B.J. T97/3ifQ";6O1 <i.forms anedanulnadequ'a.teoIndex^of Water Quality." Journal of Environmental Eealth, Vol 36, No. 1, pp. 39-46. Found, Paul. August 1972. Personal Communication, 543 Lawrence Avenue, Kelowna, B.C. Geldreich, Edwin E. 1965. "Detection and Significance of Fecal Coliform Bacteria in Stream Pollution." J.W.P.C.F.* Vol. 37, No. 12, pp. 1723-1726. Geldreich, Edwin E., Bernard A. Kenner. 1969. "Concepts of Fecal Streptococci in Stream Pollution." J.W.P.C.F.* Vol. 41, No. 8, Part 2, pp. 336-352. Geldreich, Edwin E. 1970. "Applying Bacteriological Parameters to Recreational Water Quality." J.A.W.W.A.* Vol. 62, pp. 113-120. Health Branch, Victoria, B.C., Dept. of Health Services and Hospital Insurance. 1968. "Regulations Governing Public Swimming Pools," Division of Public Health Engineering, p. 6. 1 00 Health Branch, Victoria, B.C. "Recommend Water Quality Standards," Division of Public Health Engineering 1973. "Okanagan Lakes Water Quality Survey, 1965-1967," Division of Environmental Engineering, 37 p. Kabler, P.W., H.F. Clark, N.B. Clark. duni 1971. "Patho genic Micro-organisms and Water Bourne Disease." Randolps Research Conference, Public Hazards of Microbial Pollution of Water, pp. 9-56, Rutgers The State University, June 1961. Keirn, M.A., H.P. Putnam. 1 9681 stilResistancepbf 1 Staphyloeocci to Halogen as Related to Swimming Pool Environment, Health Lab Soienee, Vol. 5, No. 3, pp. 180-193. Kenner, B., H.F. Clark, R. Kabler. 1961. "Fecal Streptococci." Applied Microbiology3 Vofli. 9, pp. 15-20. Krishnaswami, S.K. 1971. "Health Aspects of Water Quality." A.J.P.E., Vol. 61, No. 11, pp. 2259-2264. Lindsay, John (ed.). 1954-1955. "The Ear and the Nose and Throat." Yearbook of Eye} Ear3 Nose & Throat3 Vol. 3, p. 336. Chicago Yearbook Publishers. Lynch, A.J., E. Michnowsky, J. Campbell, J. Plummer. Dec. 1972. Canada Okanagan Basin Agreement, Task 135, Pre liminary Study No. 36. "Bacteriological," Pollution Control Branch,VVictoria, B.C., 23 p. Mackenthum, K.M., W.M. Ingram. 1967. "Biological Associated Problems in Freshwater Environments." Washington, U.S. Dept. of the Interior Federal Water Pollution Control Administration, 287 p. Mailman, L. 1962. "Cocci Test for Determining Mouth and Nose Pollution of Swimming Pool Water." A.J.P.E.3 Vol. 47, pp. 2001-2008. Millipore Company. 1972. "Biological Analysis of Water and Waste Water," 23 p. 101 Palmquist, A., D. Jankow. 1973. "Evaluation of Pseudomonas and Staph Aureus as Indicators of Bacterial Water Quality of Swimming Pools." JvuofaE.Hi, Vol. 36, No. 3, pp. 230-233. Reitler, R., R. Seligmann. 1957. "Ps. aeruginosa in Drinking Water." J. App. Baot.3 Vol. 20, pp. 145-150. Rohde, Paul (ed.). 1973. "B.B.L. Mannual of Products and Laboratory Procedures." 5th Edition, Division of Bacton, Dickinson and Co., Mississauga, Ontario, 211 p. Sartwell, Philip (ed.). 1965. Chapter 5, "Diseases Transmitted from Person to Person, "Preventive Medicine and Public Health*" 9th Edition, Meridith Publishing Co., New York, 1070 p. Senturia, B.H., C. Carruthers. 1954. "Prophilaxis of External Otitis." The Annals of Otology, Rhinology and Laryngology, Vol. LXIII, No. 1, pp. 97-100. Senturia, B.H., F.M. Liebman. 1956. "Evaluation of Factors Which may be of Importance in the Production of External Ear Infections." J. of Invest. Derm., Vol . 27, pp. 291-317. South Okanagan Health Unit, Unpublished Paper. 1971. "South Okanagan Fecal' Coliform Project, Penticton." Opportunity for Youth Programme, 51 p. Stevenson, A. 1953. "Studies of Bathing Water Quality and Health." A.JJp.H., Vol. 43, pp. 529-538. Sutter, V.L., V. Hurst, C. Lane. 1 967 . "Quantification of Ps. aeruginosa in Feces of Healthy Human Adults." Health Lab Science, Vol. 4, No. 4, pp. 245-249. Tennant, A.D., J . A. P. Bastien. 1971. "Bacteriological Water Quality Study, Gatineau Park Lakes," Bacteriological Haboratories Water Pollution Control Directorate, Environment Canada, Report No.: OR 71-3, 169 p. 1 02 Weinstein, S. Oct. 29, 1972. Personal Communication, Bioquest Cockeysvi11e , Maryland. Wright, Donald, Majorie, Dineen. 1972. "A Model for the Study of Infectious Otitis Externa." Arch Otolaryng., Vol. 95, pp. 243-248. Wright, Donald, J. Alexander. 1974. "Effect of Water on Bacterial Flora of Swimmers 1 Ear •." Arch otolaryng., Vol . 99, pp. 15-18. APPENDIX 1 HEALTH BRANCH PROVINCE OF B.C. RECOMMENDED.WATER • QUALITY STANDARDS Public Swimming Pools (1968) 6.04 Water sample from pools for bacteriological examination shall be taken at a time when the pool is in use at a point near the outlet of the swimming-pool and at such additional sampling points as may be selected to indicate the quality of the water is being maintained throughout the swimming-pool. Not more than 15 per cent of any series of samples, nor more than two consecutive samples in any series of samples, taken at least one week apart, shall show the presence of organisms of the coliform group in any of the five 10-millilitre portions examined. 6.04 All bacteriological tests shall be made in accordance with the procedure recommended in the latest edition of "Standard Methods for the Examination of Water and Waste water," published by the American Public Health Association. 6.06 The pH (hydrogen ion concentration) test shall be made on the pool water at least once every day the pool is in operation. A pH value of not less than 7.4 shall be maintained. Recreation Waters (1969) The standards stipulate that bathing, swimming and recreational areas in clear water lakes and upper areas of streams should have median less than 240 organisms per 100 milliliters for total coliforms and that this value should not be exceeded in more than 20% of the samples. 1 03 104 Recently, the Health Branch has adopted, for primary contact recreational waters, a fecal coliform standard of a log mean of 200 organisms per 100 mis; with not more than 10% of the total samples during any 30 day period to exceed 400 organisms per 100 mis. This standard is based on a minimum of not less than 5 samples for any 30 day period of the recreational season. APPENDIX 2 LIMNQLOGICAL INFORMATION, 19721 Lake Okanagan Mean Annual Temperature 46°F Average Annual Precipitation 12.2" Average Net Inflow 355,000 acre feet pH Range 7.4 - 8.6 pH Mean 8.0 Streams pH Mean Fl ow C.F.S. Max. Fl ow C.F.S. Nitrate N mg/1 B.O.D. Average Total Coli forms Medi an MPN Fecal Coli forms Log Mean Mill , Creek J8.C0 • 40 178 3.5 1 .6 » 5,420 <1 ,122 Brandt's Creek 8.0 2.9 19.8 .5 287 >169,000 < 980 Mill Creek: Highest flow is during July and August, 1972 Unusually high run off Brandt's Creek: Highest flow is during March. Lynch, 1972, Okanagan Basin Study Bulletin 2, Okanagan Basin Study Bullentin 4. 1 05 APPENDIX 3 v Tables XVII - XXII 106 Table XVII Lake Okanagan Air and Water Temperatures, 1972 Air: Highest Daily Temperature °F Water.: Surface Temperature °F Date Highest Air Temperature Water Temperature July 1 72 2 79 3 82 59° 4 87 60° 5 91 6 88 7 80 8 65 9 64 10 71 11 72 12 63 60° 13 76 62° 14 76 15 81 16 82 17 88 68° 18 70 60° 19 80 20 80 63° 21 81 63° 22 85 23 83 24 87 25 83 62° 26 86 67° 27 88 67° 28 91 67° 29 87 30 89 31 91 69° Aug. 1 86 70° 2 85 3 88 70° 4 91 5 91 1 C 95 7 97 72° 8 95' 72° 9 92 74° 10 91 72° 11 83 72° 12 81 • 13 81 14 73 70° 15 73 69° Date Highest Air Temperature Water Temperature Aug. 16 76 70° 17 66 18 72 19 79 20 87 21 75 22 76 68° 23 80 68° 24 85 25 86 26 87 27 89 1 28 93 72° 29 89 30 72 31 75 Sept. 1 76 2 79 3 82 4 80 5 75 6 65 7 70 8 63 64° 9 63 10 65 11 , 67 12 73 58° 13 74 '14 80 15 80 16 71 17 63 18 61 19 60 20 62 21 62 22 53 60° 23 56 24 48 25 54 26 56 56° 27 49 .28 47 29 60 58° 30 64 o 108 Table XVIII Water pH of Sampling Sites Brandt's Creek Mouth Mill Creek Date pH Date pH July 4, 1972 9.1 July 4, 1972 8.5 11 I !J. 8.0 11, 7.5 18, "• 8.7 18, " 8.6 27, 8.6 27, 8.5 Aug. 4, 7.8 Aug. 4, 8.1 11. " 6.8 11, 8.4 25, 8.1 25, 8.4 Sept. 3, 8.4 Sept. 3, " 8.2 8, " 8.0 8, " 7.9 20, 8.4 20 8.3 Rutland Pool Ogopogo Pool Date pH Date pH July 22, 1972 8.4 July 4, 1972 8.4 27, 8.1 11, " 7.2 Aug. 6, " 7.4 18, " 8.0 11, " 6.9 27, " 8.3 23, 11 6.8 Aug. 4, 7.9 Sept. 3, " 8.4 11, " 8.3 8, " 7.1 25, " 7.6 17, " 8.6 Sept.3, " 8.2 8, (8.4) 20, " 8.4 24, 8.1 1 09 TABLE XIX Ogopogo Pool: Fecal indicator counts: surface vs. sediment water samples. Date August 1972 Surface Fecal Coliform Sediment Fecal Coliform Surface Fecal Streptococci Sediment Fecal Streptococci 8 0 40 8 88 9 8 36 1 02 64 10 4 44 18 35 11 14 52 12 72 14 4 14 206 6 15 12 0 9 1 16 0 5 6 0 17 6 0 0 2 18 0 0 35 68 21 0 8 12 1 24 22 12 150 23 0 0 - Depth: 6 feet to 9 feet. Table XX Fecal Indicator Counts: Ogopogo Pool vs. Hot Sands Date Ogopogo Hot Sands Ogopogo Hot Sands Fecal Coliform Fecal Coliform Fecal Streptococci Fecal Streptococci June 1972 4 4 8 5 0 2 7 0 0 9 14 0 12 12 2 13 40 10 15 2 32 17 100 126 20 16 2 21 7 10 22 7 16 23 30 24 26 2 18 27 0 2 28 1 4 August 1 2 14 0 50 2 10 27 3 22 10 110 16 4 80 6 7 20 18 21 135 8 0 34 8 62 9 8 1 102 94 10 4 16 18 35 11 14 22 12 82 14 4 0 206 4 15 12 0 9 16 0 4 6 57 17 6 0; - 18 0 0 35 9 21 0 20 12 338 22 0 2 12 0 23 100 .0 0 2 24 • 0 1 25 2 8 37 45 28 18 12 8 62 30 10 4 0 21 31 0 0 1 0 Table XXI Weekly Illness Incidence per 1000 Person Hours Swimming Expo July and August 1972 Rutland Swimmers Date Upper Respiratory Otitis Externa Gastroenteritis Conjunctivitis Shistosomiasis July 1st wk 2nd wk 3rd wk 4th wk Aug. 1 2 3 4 17.23 57.66 .16.66 11.10 8.33 45.83 40.00 6.66 4.16 12.5 4.16 6.66 6.66 4.16 4.16 13.3 -Hot Sands Swimmers July 1 2 3 4 Aug. 1 2 3 4 10.76 34.30 47.63 79.33 28.96 43.46 10.73 21 .16 42.3 14.5 29.0 16.16 21.16 29.0 29.0 -5.43 32.3 5.30 29.0 Ogopogo Swim Team July 1 2 3 4 Aug. 1 2 3 3.84 15.38 11 .54 11 .54 7.77 38.60 15.38 30.78 61 .54 3.86 11 .54 -7.69 Table XXII Ogopogo Fecal Indicator Counts Used for Product Moment Correlations Weekly Highest Total : Total of highest daily counts for five days in a given week. Weekly Mean: Arithmetic mean of all counts in a given week including zero counts. DATE WEEKLY HIGHEST TOTAL WEEKLY MEAN Ogopogo Fecal Coliform Ogopogo Fecal Streptococci Rutland Fecal Streptococci Ogopogo Fecal Coliform Ogopogo Fecal Streptococci Rutland Fecal Streptococci July • 2nd wk 127 111 3 31.286 24.428 1.0 3rd wk 13 142 0 1 .875 14.700 0 4th wk 339 1 560 38 27.625 133.334 6.33 August i 1st wk 172 603 107 24.428 55.714 26.75 2nd wk 168 758 173 16.667 77.134 34.60 3rd wk 122 135 32 8.714 9.650 8.00 4th wk 242 267 54 16.500 —L 23.50 9.00 I 


Citation Scheme:


Usage Statistics

Country Views Downloads
United States 21 0
Germany 14 26
China 4 0
France 2 0
Japan 2 0
Malaysia 2 0
Russia 1 0
Poland 1 0
City Views Downloads
Unknown 23 26
Ashburn 7 0
Beijing 4 0
Wilkes Barre 3 0
Burbank 3 0
Cambridge 2 0
Tokyo 2 0
Redmond 1 0
Sunnyvale 1 0
Clarks Summit 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items