NORTH AMERICAN ZOOS By Christine E. Bohne A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Animal Science Master of Science 2020 ABSTRACT NORTH AMERICAN ZOOS By Christine E. Bohne Water is essential for life of all animals. However, drinking water might be of poor quality if it contains excess chemicals, nutrients or contaminants. Based on review of the scientific literature, the quality of drinking water in zoos has not been inve stigated. Therefore, objectives were to: 1) assess general quality of water in Association of Zoo and Aquarium zoos; 2) examine possible relationships among husbandry practices and water quality; and, 3) analyze iron concentrations in drinking water of zoo s with b lack r hino. Forty zoos without and 10 with b lack r hino agreed to participate when randomly invited from a pool of 174 eligible zoos (29% response rate). Water samples were analyzed for pH, hardness, total dissolved solids, calcium, phosphorus, ma gnesium, potassium, sodium, iron, manganese, zinc, copper, chloride, sulfate, nitrate, total coliform, and Escherichia coli . A water quality index was used to rank overall quality among zoos. A questionnaire about husbandry practices and drinking water a lso was completed by a subsample of zoos (n = 39). Over 85% of zoos used municipal water primarily . Three of 50 zoos (above the 90 th percentile) had poor quality water. M ajority (59%) of zoos responded that drinking water quality was consider ed in nutritional management . However, only 18% routinely analyzed drinking water . Zoos with Black Rhino were aware of the recommendation to formulate for low dietary iron to reduce Iron Overload Disorder. However, only 2 of 8 zoos with Black Rhino routin ely analyzed drinking water for iron. iii TABLE OF CONTENTS LIST OF TABLES ................................ ................................ ................................ .......................... v LIST OF FIGURES ................................ ................................ ................................ ..................... viii LIST OF ALGORITHMS ................................ ................................ ................................ ............ xiv KEY TO ABBREVIATIONS ................................ ................................ ................................ ....... xv CHAPTER 1 LITERATURE REVIEW ................................ ................................ ......................... 1 1.1.Introduction ................................ ................................ ................................ ........................... 2 1.2. Water Nutrition and Quality ................................ ................................ ................................ 3 1.2.1. Importance ................................ ................................ ................................ .................... 3 1.2.2. Water Quality ................................ ................................ ................................ ................ 3 1.2.3. Aesthetics ................................ ................................ ................................ ...................... 5 1.2.4. Hard Water ................................ ................................ ................................ .................... 8 1.2.5. pH ................................ ................................ ................................ ................................ .. 8 1.2.6. Temperature ................................ ................................ ................................ ................ 10 1.2.7. Biological Factors ................................ ................................ ................................ ....... 11 1.2.8. Nitrates and Nitrites ................................ ................................ ................................ .... 12 1.2.9. Metals ................................ ................................ ................................ .......................... 14 1.2.9.1. Copper and Molybdenum ................................ ................................ .................... 15 1.2.9.2. Lead ................................ ................................ ................................ ...................... 16 1.2.9.3. Zinc ................................ ................................ ................................ ...................... 17 1.2.9.4. Manganese ................................ ................................ ................................ ........... 17 1.2.9.5. Iron ................................ ................................ ................................ ....................... 19 1.2.10. Water Quality Index (WQI) ................................ ................................ ...................... 20 1.3 Iron in the body ................................ ................................ ................................ ................... 23 1.3.1 Importance ................................ ................................ ................................ ................... 23 1.3.2 Metabolism ................................ ................................ ................................ .................. 24 1.3.3 Transport ................................ ................................ ................................ ...................... 25 1.3.4. Storage ................................ ................................ ................................ ........................ 26 1.3.5. Regulation ................................ ................................ ................................ ................... 27 1.3.6. Iron - Manganese Interactions ................................ ................................ ...................... 29 1.3.7. Disorders ................................ ................................ ................................ ..................... 29 1.3.8. Dietary Recommendations for Black Rhino ................................ ............................... 31 1. 4. Black Rhino Biology ................................ ................................ ................................ .......... 32 1.4.1. Wild Biology ................................ ................................ ................................ ............... 32 1.4.2. Wild Diet and Eating Habits ................................ ................................ ....................... 33 1.4.3. Iron in the Wild Diet ................................ ................................ ................................ ... 34 1.4.4. Genetics of Iron Absorption ................................ ................................ ........................ 35 1.4.5. Gut Microbiome ................................ ................................ ................................ .......... 35 1.5. Black Rhino Husbandry ................................ ................................ ................................ ..... 36 1.5.1. Captive Diet and Eating Habits ................................ ................................ ................... 36 1.5.2. Iron in the Captive Diet ................................ ................................ ............................... 37 iv 1.5.3. Iron Control Methods ................................ ................................ ................................ .. 37 1.6. Conclusions ................................ ................................ ................................ ........................ 38 CHAPTER 2 ASSESSMENT OF DRINKING WATER QUALITY AND RELATED HUSBANDRY PRACTICES IN NORTH AMERICAN ZOOS ................................ ................. 39 2.1. Introduction ................................ ................................ ................................ ........................ 40 2.2. Materials and Methods ................................ ................................ ................................ ....... 41 2.2.1. Sample Groups ................................ ................................ ................................ ........... 41 2.2.2. Questionnaires ................................ ................................ ................................ ............ 44 2.2.3. Water Sampling Kits ................................ ................................ ................................ .. 44 2.2.3.1. Sample Collection from Non - Black Rhino Zoos ................................ ................. 45 2.2.3.2. Sample Collection from Black Rhino Zoos ................................ ......................... 45 2.2.4. Water Quality Index (WQI) Calculations ................................ ................................ .. 46 2.2.5. Exploratory Questions Developed for Statistical Analysis of the Information .......... 48 2.2.6. Statistical Analysis ................................ ................................ ................................ ..... 49 2.3. Results and Discussion ................................ ................................ ................................ ...... 50 2.3.1. Questionnaire Responses ................................ ................................ ............................ 50 2.3.1.1. General Questionnaire Responses (Questions 1 through 10) .............................. 50 2.3.1.2. Questionnaire Responses of Zoos with Black Rhino (Questions 11 through 28) 52 2.3.2. Results of Water Quality Index Analysis ................................ ................................ .... 55 2.3.2.1. WQI of Non - Black Rhino Zoos ................................ ................................ ........... 60 2.3.2.2. WQI of Black Rhino Zoos ................................ ................................ ................... 60 2.3.3. Examination of Statistical Questions Utilizing WQI and Questionnaire Responses .. 61 2.4. Conclusions. ................................ ................................ ................................ ....................... 70 APPENDICES ................................ ................................ ................................ .............................. 74 APPENDIX A : RANDOMIZED NUMBER GENERATOR OUTPUT ................................ .. 75 APPENDIX B : ZOO SAMPLING KIT DOCUMENTS ................................ .......................... 78 APPENDIX C : QUESTIONNAIRE ................................ ................................ ......................... 87 APPENDIX D : ANALYTE STANDARDS ................................ ................................ ............. 91 APPENDIX E : QUESTIONS USED FOR STATISTICAL ANALYSIS ................................ 93 APPENDIX F : FIGURES AND TABLES ................................ ................................ ............... 96 LITERATURE CITED ................................ ................................ ................................ ............... 150 v LIST OF TABLES Table F.1 . Analytes included in each of the four Water Quality Index (WQI) calculations performed for each participating zoo. The four WQI calculations were as follows 1) Low all analytes, 2) Low select analyte, 3) High all analytes, and 4) High select analytes. ...... 130 Table F.2. List of low and high standards used in the calculation of Water Quality Index (WQI) values for each analyte included in the WQI calculations. Phosphorus and magnesium do not have a standard value and were not inclu ded in any of the WQI calculations nor statistical analysis. ................................ ................................ ................................ ............... 131 Table F.3. Origin point Water Quality Index (WQI) values for all zoos (Non - Black Rhino and Black Rhi no) and all four analytes. Ranked in order of highest to lowest WQI value for the Low: All Analytes formula. (a) indicates a WQI value greater than or equal to 2.0, the highest 50 th percentile value across all analyte formulas. (b) indicates a WQI value g reater than or equal to 13.2, the highest 90 th percentile value across all analyte formulas. .......... 132 Table F.4. Exhibit Water Quality Index (WQI) values for all Black Rhino zoos and all four analytes. Ranked in order of highest to lowest WQI value for the Low: All Analytes formula. (a) indicates a WQI value greater than or equal to 2.0, the highest 50 th percentile value across all analyte formulas . (b) indicates a WQI value greater than or equal to 10.2, the highest 90 th percentile value across all analyte formulas. ................................ ............. 135 Table F.5. Difference in drinking wat er quality between the origin and Black Rhino exhibit sampling points, as shown by a change in calculated Water Quality Index (WQI) value between the two points across all four analyte formulas. The Difference was calculated by subtracting the Black Rhino exhibit WQI value from the origin WQI value (Origin - Black Rhino Exhibit=Difference). (a) Indicates a decrease in drinking water quality from the origin to the Black Rhino exhibit sample points. Negative zero ( - 0.0) being possible due to the rounding of m inor changes between water quality at the two sampling points (e.g., - 0.0067 rounding down to - 0.0). ................................ ................................ ........................... 137 Table F.6. Measures of Central Tendency for Non - Black Rhino zoo Water Quality Index (WQI) values at the origin sampling point calculated using the four different analyte formulas. . 139 Table F.7. Percentiles for Water Qualit y Index (WQI) values at the origin point for Non - Black Rhino zoos, calculated using the four different analyte formulas. ................................ ...... 140 Table F.8. Measures of Central Tendency for Black Rhino zoo Water Quality Index (WQI) values at the origin sampling point calculated using the four different analyte formulas. (a) Multiple modes exist for the data; smallest value shown. ................................ .................. 140 Table F.9. Measures of Central Tendency for Black Rhino zoo Water Quality Index (WQI) values within the Black Rhino exhibit calculated using the four different analyte formulas. (a) Multipl e modes exist for the data; smallest value shown. ................................ ............. 141 vi Table F.10. Percentiles for Black Rhino zoo Water Quality Index (WQI) values at the origin sampling point cal culated using the four different analyte formulas. ................................ . 141 Table F.11. Percentiles for Black Rhino zoo Water Quality Index (WQI) values within the Black Rhino exhibit c alculated using the four different analyte formulas. ........................ 142 Table F.12.: Sign Test summary table for the water quality difference between the origin and Black Rhino exhibit sampling points for Black Rhino zoos calculated using the four different analyte formulas. The Difference was calculated by subtracting the Black Rhino ex hibit Water Quality Index (WQI) value from the origin WQI value (Origin - Black Rhino Exhibit=Difference). ................................ ................................ ................................ ........... 142 Table F.13.: Sign Test summary table for the sign change of the drinking water quality between the origin and Black Rhino exhibit sampling points for Black Rhino zoos calculated using the four different analyte formulas. The change in sign indicates whether the water quality is worse at the origin point (+), worse within the Black Rhino exhibit ( - ), or whether there was no change in water quality between the two sampling points. ................................ .... 143 Table F.14.: Kruskal - Wallis H Test summary table for the effect zoo age has on the water quality difference between the origin point and Black Rhino exhibit sampling points for Black Rhino zoos calculated using the four different analyte formulas. T he change in drinking water quality between the origin point and Black Rhino exhibit decreases as the mean rank value increases; meaning the quality of the drinking water is lower as the mean rank increases. ................................ ................................ ................................ ..................... 144 Table F.15. Summary table showing the p - value, Mann - Whitney U statistic, z - value, effect size, and mean ranks for the difference in Water Quality Index (WQI) values sampled at the origin point for zoos that d id and did not use municipal water as their primary drinking water source. The WQI values for all four analyte formulas are shown in the table. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. ................... 145 Table F.16. Summary table showing the p - value, Mann - Whitney U statistic, z - value, effect size, and mean ranks for the difference in WQI values sampled at the origin point for zoos that did and d id not use well (bore) water as their primary drinking water source. The Water Quality Index (WQI) values for all four analyte formulas are shown in the table. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. ................... 146 Table F.17. Summary table showing the p - value, Mann - Whitney U statistic, z - value, effect size, and mean ranks for the difference in Water Quality Index (WQI) values sampled at the origin point for zoos that did and did not use river water as their primary drinking wat er source. The WQI values for all four analyte formulas are shown in the table. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. ............................. 147 Ta ble F.18. Summary table showing the P - value, Kruskal - Wallis H statistic reported as the X 2 , effect size, and mean ranks for the different zoo size categories reported in numbers of species. The p - value reported is asymptotic and not exact. The Water Qual ity Index (WQI) values for all four analyte formulas are shown in the table. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. ................................ ....................... 148 vii Table F.19. Summary table showing the p - value, Mann - Whitney U statistic, z - value, and effect size for the difference in Water Quality Index (WQI) values between the origin point and the Black Rhino exhibit for zoos that have replaced their drinking water pipes w ithin the last 5 years versus zoos that have not replace their drinking water pipes with the last 5 years. The WQI values for all four analyte formulas are shown in the table. Only Black Rhino zoos are included in this WQI grouping. ................................ ................................ ............ 148 Table F.20. Summary table showing the p - value, Mann - Whitney U statistic, z - value, and effect size for the difference in Water Quality Index (WQI) values sampled within the Black Rhin o exhibit for zoos that cleaned the drinking water receptacles provided to the black rhino daily or less than daily . The WQI values for all four analyte formulas are shown in the table. Only Black Rhino zoos are included in this WQI grouping. .............................. 149 Table F.21. Summary table showing the p - value, Mann - Whitney U statistic, z - value, and effect size for the difference in Water Quality Index (WQI) values sampled at the origin point for zoos that did and did not routinely analyze the drinking water provided to their an imal collections. The WQI values for all four analyte formulas are shown in the table. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. ................... 149 viii LIST OF FIGURES Figure A.1. Random number generator output to select the Black Rhino Zoo subsample invited to complete a questionnaire and submit drinking water samples. Twenty - five in the outline are included within the questionnaire subsample, and the remaining BR zoos only submitted a drinking water sample. ................................ ................................ ................................ ....... 76 Figure A.2. Random number generator output to select the Non - Black Rhi no zoo subsample invited to complete a questionnaire and submit drinking water samples. One hundred in the outline are included within the questionnaire subsample, and the remaining Non - Black Rhino zoos only submitted a drinking water sample. ................................ ........................... 77 Figure B.1. Cover page for all zoo water sample only sampling kits. ................................ ......... 79 Figure B.2. Cover page for all zoo water sample and questionnaire subsample sampling kits. .. 80 Figure B.3. Origin point water sample collection instruction sheet for Non - Black R hino zoo sampling kits. ................................ ................................ ................................ ........................ 81 Figure B.4. Origin point and Exhibit water sample collection instruction sheet for Black Rhino Figure B.5. Laboratory water sample submittal forms required for each water sample submitted by a zoo. ................................ ................................ ................................ ................................ 86 Figure C.1. Questions 1 through 10 provided to all zoos (Black Rhino and Non - Black Rhino) in the questionnaire subsample group. ................................ ................................ ...................... 88 Figure C .2. Questions 11 through 19 of 28 provided only to Black Rhino zoos in the questionnaire subsample group. ................................ ................................ ............................ 89 Figure D.1. Standard values and sources used for both th e low and high standard Water Quality Index (WQI) calculations. ................................ ................................ ................................ ..... 92 Figure F.1. Organizational chart showing definition and partitioning of candidate zoos in the stud y design for participation. ................................ ................................ ............................... 97 Figure F.2. Organizational flow chart of study invitation responses. ................................ .......... 98 Figure F.3. Organizational chart showing final disposition and fate of zoos initially agreeing to participate based on confirmation to one of the three invitations. Fifty total zoos participated in the study (10 in the Black Rhino [BR] group and 40 in the Non - Black Rhino [Non - BR] group). ................................ ................................ ................................ ................................ ... 99 Figure F.4. Pie chart showing the overall breakdown of primary drinking water sources used by all 39 zoos to complete questionnaires, including both Non - Black Rhino and Black Rhino facilities. ................................ ................................ ................................ .............................. 100 ix Figure F.5. Box plot chart showing the distributions for zoo age across Black Rhino zoo Water Quality Index (WQI) the distributions between zoo age group were not similar in shape. Due to the distributions no t being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ................................ ................................ 101 Figure F.6. Box plot chart showing the distributions for zoo age across Black Rhino zoo Water Quality Index (WQI) the distributions between zoo age group were not similar in shape. Due to the dist ributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ................................ ................................ 102 Figure F.7. Box plot chart showing the distributi ons for zoo age across Black Rhino zoo Water Quality Index (WQI) the distributions between zoo age group were not similar in shape. Due to the distributions not being similarly shaped, t he p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ................................ ................................ 103 Figure F.8. Box plot chart showing the distributions for zoo age across Black Rhino zoo W ater Quality Index (WQI) show the distributions between zoo age group were not similar in shape. Due to the distributions not being similarly shaped, the p - value presented was an asymptoti c value and not an exact p - value. ................................ ................................ ................................ ........... 104 Figure F.9. Pyramid chart showing the distribution of zoos that did and did not use municipal water as their primary drinking water source. All Water Quality Index (WQI) values are la. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ................................ ..... 105 Figure F.10. Pyramid chart showing the distribution of zoos that did and did not use municipal water as their primary drinking water source. All Water Quality Index (WQI) values are from the origin sampling formula. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not a n exact p - value. ................................ ................................ .......................... 106 Figure F.11. Pyramid chart showing the distribution of zoos that did and did not use municipal water as their primary drinking water source. All Water Quality Index (WQI) values are Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ................................ ..... 107 Figure F.12. Pyramid chart showing the distribution of zoos that did and did no t use municipal water as their primary drinking water source. All Water Quality Index (WQI) values are x formula. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ........................... 108 Figure F.13. Pyramid chart showing the distribution of zoos that did and did not use well(bore) water as their primary drinking water source. All Water Quality Index (WQI) values are Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ................................ ..... 109 Figure F.14. Pyramid chart showing the distribution of zoos that did and did not use well(bore) water as their primary drinking water source. All Water Quality Index (WQI) values are from the formula. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ........................... 110 Figure F.15. Pyramid chart showing the distribution of zoos that did and did not use well(bore) water as their primary drinking water source. All Water Quality Index (WQI) values are Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ................................ ..... 111 Figure F.16. Pyramid chart showing the distribution of zoos tha t did and did not use well(bore) water as their primary drinking water source. All Water Quality Index (WQI) values are formula. Both Black Rhino and Non - Black Rhino zoo s are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ........................... 112 Figure F.17. Pyramid chart showing the distribution of zoos that did and did not use river water as their primary drinking water source. All Water Quality Index (WQI) values are from the ori Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ................................ ........... 113 Figure F.18. Pyramid chart showing the distribution of zoos that did and did not use river water as their primary drinking water sour ce. All Water Quality Index (WQI) values are from the Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being sim ilarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ................................ ........... 114 xi Figure F.19. Pyramid chart showing the distribution of zoos that did and did not use river water as their primary drinking water source. All Water Quality Index (WQI) values are from the Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ................................ ........... 115 Figure F.20. Pyramid chart showing the distribution of zoos that did and did not use river water as their primary drinking water source. All Water Quality Index (WQI) values are from the la. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ................................ ........... 116 Figure F.21. Box plot showing the distribution of zoo size across categories in number of species. All Water Quality Index (WQI) values are from the origin sampling point and were - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. Due to the few exceptionally high WQI values making it difficult to see whether or not the box plots are similarly shaped, these values were removed in order to better assess the box plot distributions. All analysis was performed, and all results are presented with the exceptionally high values included in the data set. ................................ ............................. 117 Figure F.22. Box plot showing the distribution of zoo size across categories in number of species. All Water Quality Ind ex (WQI) values are from the origin sampling point and were - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. Due to the few exceptionally high WQI values making it difficult to see whether or not the box plots are similarly shaped, these values were removed in order to better assess the box plot distrib utions. All analysis was performed, and all results are presented with the exceptionally high values included in the data set. ................................ ............................. 118 Figure F.23. Box plot showing t he distribution of zoo size across categories in number of species. All Water Quality Index (WQI) values are from the origin sampling point and were - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. Due to the few exceptionally high WQI values making it difficult to see whether or not the box plots are si milarly shaped, these values were removed in order to better assess the box plot distributions. All analysis was performed, and all results are presented with the exceptionally high values included in the data set. ................................ ............................. 119 Figure F.24. Box plot showing the distribution of zoo size across categories in number of species. All Water Quality Index (WQI) values are from the origin sampling point and were - Black xii Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presen ted was an asymptotic value and not an exact p - value. Due to the few exceptionally high WQI values making it difficult to see whether or not the box plots are similarly shaped, these values were removed in order to better assess the box plot distributions. All analysis was performed, and all results are presented with the exceptionally high values included in the data set. ................................ ............................. 120 Figure F.25. Pyramid chart showing the distribution of zoos that had and had not replaced drinking water pipes within the last 5 years. All Water Quality Index (WQI) values are Both Black Rhino and Non - B lack Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ................................ ..... 121 Figure F.26. Pyramid chart showing the distribution of zoos that had and had not replaced drinking water pipes within the last 5 years. All Water Quality Index (WQI) values are from the origin sampling point and were calculated using the formula. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ........................... 122 Figure F.27. Pyramid chart showing the distribution of zoos that had and had not replaced drinking water pipes within the last 5 years. All Water Quality Index (WQI) values are f Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymp totic value and not an exact p - value. ................................ ................................ ................................ ..... 123 Figure F.28. Pyramid chart showing the distribution of zoos that had and had not replaced drinking water pipes within the last 5 years. All Water Quality Index (WQI) values are from the orig formula. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic val ue and not an exact p - value. ................................ ................................ ........................... 124 Figure F.29. Pyramid chart showing the distribution of zoos that cleaned the drinking water receptacles provided to the black rhino daily or less than daily. All Water Quality Index (WQI) values are from the Black Rhino exhibit sampling point and were calculated using grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ........ 125 Figure F.30. Pyramid chart showing the distribution of zoos that cleaned the drinking water receptacles provided to the black rhino daily or less than daily. All Water Quality Index (WQI) values are from the Black Rhino exhibit sampling point and were calculated using ck Rhino zoos are included in this WQI xiii grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ........ 126 Figure F.31. Pyramid chart showing the distribution of zoos that cleaned the drinking water receptacles provided to the black rhino daily or less than daily. All Water Quality Index (WQI) values are from the Black Rhino exhibit sampling point and were calculated using grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ........ 126 Figure F.32. Pyramid chart showing the distribution of zoos that cleaned the drinking water receptacles provided to the black rhino daily or less than daily. All Water Quality Index (WQI) values a re from the Black Rhino exhibit sampling point and were calculated using grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymp totic value and not an exact p - value. ................................ ................................ ........ 127 Figure F.33. Pyramid chart showing the distribution of zoos that did and did not routinely analyze the drinking water pro vided to their animal collections. All Water Quality Index (WQI) - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ........ 127 Figure F.34. Pyramid chart showing the distri bution of zoos that did and did not routinely analyze the drinking water provided to their animal collections. All Water Quality Index (WQI) k Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ............ 128 Figure F.35. Pyramid chart showing the distribution of zoos that did and did not routinely analyze the drinking water provided to their animal collections. All Water Quality Index (WQI) valu - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ................................ ........ 128 Figure F.36. Pyramid chart showing the distribution of zoos that did and did not routinely analyze the drinking w ater provided to their animal collections. All Water Quality Index (WQI) - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. ................................ ............ 129 xiv LIST OF ALGORITHMS Algorithm 1.1. Original Water Quality Index (WQI), w here C n is the rating scale, W n is the weighting factors, and M 1 is the coefficient for temperature, and M 2 is the coefficient for ................................ ................................ ....................... 21 Algorithm 1.2. Updated Water Quality Index (WQI), where w i is the weight of the parameter (a number between 0 to 1) and q i is the quality of the parameter (a number between 0 to100) (Brown et al., 1970) ................................ ................................ ................................ .............. 22 Algorithm 1.3. F ormula for calculating the quality rating scale, where C i is the measured concentration of the i th parameter and the S i is the standard value of the i th parameter (Akter et al., 2016) ................................ ................................ ................................ ................ 23 Algorithm 1.4. Formula for calculating the relative weight (Akter et al., 2016). ........................ 23 Algorithm 1.5. Weighted arithmetic Water Quality Index (WQI), where w i is the relative weight of the i th parameter and q i is the quality rating scale of the i th parameter (Akter et al., 2016) ................................ ................................ ................................ ................................ ............... 23 Algorithm 2.1. F ormula for calculating the quality rating scale. Where C i is the measured concentration of the i th parameter or analyte and the S i is the standard value of the i th parameter or analyte (Akter et al., 2016) ................................ ................................ .............. 46 Algorithm 2.2. Formula for calculating the relative weight (Akter et al., 2016). ........................ 47 Algorithm 2.3. Weighted arithmetic WQI. Where w i is the relative weight of the i th parameter or analyte and q i is the quality rating scale of the i th parameter or analyte (Akter et al., 2016) ................................ ................................ ................................ ................................ ............... 47 xv KEY TO ABBREVIATIONS BMP .................................................................................................bone morphogenetic protein BR ............................................................................................................................. ..bl ack rhino C i .......................................................................................observed concentration of i parameter C ..................................................................................................................Ho rating point Ca ............................................................................................................................. ........calcium CaCO 3 ...................................................................................... ........................calcium carbonate CT ......................................................................................................................condens ed tannin Cu ......................................................................... ..............................................................copper DMB ....................................................................................................................dry matter basis DMT - 1 ............................................ ..................................................divalent metal transporter 1 DNA ..........................................................................................................deoxyribonucleic acid EPA ......................................... .........................United States Environmental Protection Agency Fe ............................................................................................................................. ...............iron Fe 2+ ................................ .............................................................................................ferrous iron Fe 3+ ............................................................................................................................. ...ferric iron GI ................................................................................................................ ..........gastrointestinal H + ............................................................................................................................. ......hydrogen HAs .................................................................................. ..................................health advisories HCl ....................................................................................................................hydrochlo ric acid HFE .............................................................. ........................................hemochromatosis protein HJV ...........................................................................................................................he mojuvelin xvi IOD ................................................. ...........................................................iron overload disorder IREs .......................................................................................................iron responsive elements Ireg - 1 .................................. .........................................................................................ferroportin IRP - 1 .....................................................................................................iron responsive protein - 1 K .................... ...............................................................................................................potassium M s coefficie nt MCL ...............................................................................................maximum contaminant levels Mg ............................................................................................................................. ..magnesium Mn .................................................................................................................... ...........manganese mRNA ...............................................................................................messenger ribonucleic acid N .......................................................................................................... ............................nitrogen Na ............................................................................................................................. .........sodium NH 3 ....................................................................... ..........................................................ammonia NH 4 + ............................................................................................................................a mmonium NO 2 - ............................................................................................................................. ........nitrite NO 3 - .............................................................................................................. .......................nitrate NRC ...................................................................................................National Research Council NTU ............................................................................................ .....nephelometric turbidity unit ORVWSC ........................................................ P ............................................................................................................... ...................phosphorus q i ......................................................................................................................quality rating scale QI ................................................................................. Quality Index ROS .........................................................................................................reactive oxygen species xvii S i ..................................................................... ................................standard value of i parameter S e ............................................................................................................................. ........selenium SMCL ............................................ ...............................secondary maximum contaminant levels TC ............................................................................................................................t otal coliform TDS .......................................... ...................................................................total dissolved solids TfR - 1 ...........................................................................................................transferrin receptor - 1 TfR - 2 ...........................................................................................................transferrin receptor - 2 TON ..........................................................................................................threshold odor n umber w i ............................................................................................................................r elative weight W ......................................................................................................... weighting factor WHO .................................................................................................World Health Organization WQI .................................................................................................. ..............water quality index Zn ............................................................................................................................. ..............zinc 1 CHAPTER 1 LITERATURE REVIEW 2 1.1. Introduction Water is an essential nutrient for life (NRC, 2005; Ross et al., 2014). Clean drinking water is essential for maintaining optimal health. For centuries , humans have known that safe, clean, drinking water is essential for avoiding disease, a nd maintaining health. Across the country , municipal and private water treatment facilities provide clean drinking water for millions of people, and , by living in close proximity to humans, their pets. Livestock managers and nutritionists also are aware of the importance of water quality and the problems that poor - quality drinking water can pose to livestock health and production. Research shows that providing livestock adequate access to clean drinking water can increase feed intake and can potentially increase productivity because water and feed intakes are positively correlated (NRC, 2007; Pond et al., 2005). As a result, many animal agricultural facilities make water testing and treatment a routine part of livestock management. However, based on a r eview of the current literature, there appear s to be a lack of research into drinking water quality in zoos. Drinking water quality can negatively impact the health of animals when certain constituents are present in large enough amounts. Even essential nutrients can be present in drinking water in great enough concentrations to contribute to excess intake leading to adverse health effects in animals if not properly monitored and corrected (NRC, 1974). Due to the lack of peer - reviewed data and informatio n on nutritional water quality in zoos, this research project seeks to survey and obtain water samples from a random sample of Association of Zoos and Aquariums (AZA) accredited zoos across North America. The information and samples collected were analyze d and the data analyzed and summarized in order to create a resource for zoo veterinarians and nutritionists about drinking water sources, water husbandry practices, and 3 drinking water quality. Our goal is to assess the quality of drinking water provided to animal collections in AZA - accredited zoos; with iron concentrations in drinking water provided to black rhino being of particular interest in our study. Additionally, we seek to examine drinking water husbandry practices in zoos via a questionnaire foc used on water husbandry practices provided to a subset of participating zoos. 1. 2 . Water Nutrition and Quality 1. 2 .1. Importance Water is an essential nutrient required to sustain life. Water is crucial for metabolic processes, temperature control, fluid balance, gastrointestinal health, waste elimination, nutrient transport, and digestive function (Church, 1993; Dryden, 2008; Pond et al., 2005). Water makes up about two - thirds of the body mass in adults and more than three - quarters of the body mass of baby animals (NRC, 2005) . Water also is required in every biochemical process in the body (Pond et al., 2005). Due to the great many biological processes in which water is involved in the body, most mammals cannot survive for more than a few days without an adequate supply of clean drinking water. Because water is considered the universal solvent, it is vital for the absorption and transport of water - soluble nutrients. The ability of water to act as a solvent also makes it possible for d rinking water sources to contribute additional nutrients in excess of the diet to captive animals (NRC, 2005). Water also can carry pollutants, pathogens, and other constituents (Vigil, 2003). 1. 2 .2. Water Quality Water quality is defined as determinin g the suitability of a body of water for a specific purpose based upon the measurement of the characteristics of the water (Johnson et al., 1997; NRC, 2007). For the scope of this study, the specific purpose of the water is drinking by various 4 species of zoo animals. Water is rarely ever pure ; it has dissolved constituents either from nature or human activities (NRC, 1974). The water consumed by humans, livestock , and captive animals is sourced from either groundwater (aquifers) or surface water (rivers, lakes, and streams); because of this , the dissolved constituents in water can change over time (Vigil, 2003). An increase in contaminants in water can lead to decreased water intake, illness, and , in extreme cases, death of humans and animals alike (Chur ch, 1993; DHEC, 2013; Gray, 2008). The criteria that define the quality of water are categorized into chemical, physical, biological, and radiological characteristics (NRC, 1974; Vigil, 2003; Gray, 2008). Chemical characteristics consist of ess ential minerals ( and toxic elements and compounds (arsenic, cadmium, lead, nitrate, ) (Gray, 2008; Vigil, 2003). Physical characteristics consist of odor, taste, color, temperature, turbidity, and total dissolved solids (TDS) (NRC, 1974). Biological characteristics consist of small and microscopic living organisms; including, bacteria, protozoa, algae, and small invertebrates (Gray, 2008). Viruses are considered pathogens, so they are included as biological characteristics of water quality even though they are not truly living organisms (Alberts et al., 2013). Radiological constituents may consist of stron tium - 90, radium - 226, tritium, and other radioactive elements and substances that end up in the water supply either via natural sources, human activities (e.g., mining), and (or) environmental spills of radioactive waste (NRC, 1974). 5 The typical analysis u sed to assess drinking water quality for livestock is called the suitability test, which evaluates pH, hardness, TDS, Ca, P, Mg, K, Na, Fe, Mn, Zn, Cu, chlorides, sulfates, and nitrates (CVAS, 2019). It also is possible to have water t esting done to analyze and report a more limited list of water constituents. Due to the potential constant change in composition of a water source via erosion, transpiration, evaporation, precipitation, oxidation, reduction, cation and anion exchange, aci d - base interactions, and microbial transformation , at least one water test per year is recommended to assess the safety of drinking water provided to animals in human care (Clauss et al., 2012; NRC, 1974). The MCL is the maximum level allowed of a contaminant delivered to any user of a public water system (EPA, 2018). 1. 2 .3. Aesthetics Color, odor, taste, and turbidity (cloudiness) are the main factors contributing to aesthetic quality of water (Gray, 2008). Whereas these factors often do not relate to the safety of the drinking water, they can influence consumption by animals and people, as well as perceived water safety by the latter (Gen ther and Beede, 2013; Vigil, 2003). Color, odor, taste and turbidity characteristics can be due to a variety of different constituents including combinations of more than one. For instance, color can be affected by presence of Fe, Mn, Zn, Cu, Pb, 6 hydroge n sulfide, microorganisms, and other factors (DHEC, 2013; Gray, 2008). Although a change in water color does not necessarily pose an immediate health risk to consumers, it can decrease the palatability and consumption of water. The EPA guideline for colo r was set at 15 color units (EPA, 2018) . Another visual water quality indicator is turbidity. Turbidity measures the clarity of water and is determined by shining a light through a water sample in order to determine the amount of light that is scattered by particles or materials suspended in the water sample (USGS, 2016). Turbidity is affected by particles in the water that are visible to the human eye, such as, clay, silt, soluble colored organic particles, microorganisms, and other factors. Although t he turbidity of water itself likely does not pose a threat to human or animal health, it can provide an environment in which pathogens can grow or regrow after treatment (USGS, 2016). Both the color and turbidity of water are easily recognizable indicator s of water quality or change to water quality. The standard for turbidity, as set by the EPA, specifies that at no time can turbidity exceed 5 Nephelometric Turbidity Units (NTU) (EPA, 2018) . Odor of water is most often related to the presence of chlori ne or hydrogen sulfide (DHEC, 2013; Gray, 2008). Chlorine is used in small amounts, residual readings of 0.5mg/L or less, as a disinfecting agent in the final stage of drinking water treatment (CDC, 2014; Vigil, 2003). The presence of a can decrease water consumption of some people, but there is scant information on the effect of odor on water intake of livestock or other animals (Gray, 2008). Hydrogen sulfide when present in water imparts a smell that may b e offensive and decrease water consumption by humans and may contribute to decreased water intake in dairy cows, but this has not been verified with research (Beede, 2009; DHEC, 2013). Although there were studies to evaluate water preferences of livestock , primarily with cattle, it is 7 very difficult to attribute exactly which characteristic(s) (odor, taste, metabolic impact, or a combination of many factors) impacts animal preference the most (Lardner et al., 2013). The EPA set the S MCL for odor at 3 Thre shold Odor Number (TON) (EPA, 2018) . Another aesthetic component of water, that is closely related to odor, is taste. Taste, or distaste can be affected either directly or indirectly by almost every constituent that can be present in a water sample. Met allic, bitter taste can be imparted by Fe, Mn, Zn, Cu, change in pH, high TDS, or an increase in the corrosive qualities of the water due to corrosion and release of metallic substances from pipes into the water during transport to the consumer. A salty t aste can be imparted by sodium and sodium - containing compounds, chloride and chloride - containing compounds, and elevated TDS. While excess Na can alter the taste of water, it also can cause dehydration and increased thirst of animals and people. In 2018, a group of endangered black rhinoceros ( Diceros bicornis; black rhino) were relocated within Kenya from a Lake Nakuru National Park to Tsavo East National Park in a conservation effort to boost a breeding population. Of the 11 black rhino originally relo cated, 8 died due to the higher salinity in the drinking water source at their new location (Van Sant, 2018). The increased Na content of the new drinking water source caused dehydration, which increased consumption of the saline water and led to salt poi soning. A or taste can be imparted on water in the presence of hydrogen sulfide, bacteria, and algae (DHEC, 2013; Gray, 2008; Vigil, 2003). Whereas taste and smell are independent senses, studies in humans show a link between taste and olfaction that can enhance the experience a person has with food or drink ( Taste and 2012). Whether or not this same interaction is present in livestock and other animals is unknown, but it may be possible for 8 the odor and taste of water to play a combined role in the intake of water by animals; more research is needed t o confirm such a connection. 1. 2 .4. Hard Water Hard water contains increased concentrations of alkaline earth metals, such as Mg and Ca in the form of salts (e.g., calcium carbonate) (USGS, 2016a). Because these salts are alkaline earth metals, as shown o n the Periodic Table, hard water is typically alkaline (basic on the pH scale). Because hard water is associated with an increased mineral content it very often has an increased concentration of TDS (DHEC, 2013). Hard water is not considered a health con cern, but it can promote mineral build - up in pipes and plumbing and decreased effectiveness of soaps and cleaning agents (Vigil, 2003). The USGS defines the categories of water hardness as follows: 0 - 60 mg/L of calcium carbonate as considered 61 - 1 20 mg/L as 121 - 180 mg/L as hard and greater than 180 mg/L calcium carbonate as water (USGS, 2016a). Although it is not considered a health concern, a range of between 100 and 300 mg/L is considered the taste threshold for Ca ions in water for humans; with the taste threshold for Mg ions believed to be even lower (WHO, 2017). Treatment for hard water typically involves an ion exchange system that replaces the Ca and Mg in the salts with Na (DHEC, 2013). The introduction o f Na as a water softening agent does increase the overall Na content of the water and may pose problems for people and animals that need a lower Na intake diet. 1. 2 .5. pH The term pH refers to the acidic or basic (alkaline) nature of a solution, in this in stance water, based upon the concentration of hydrogen ions present in the solution (Vigil, 2003). Acidic solutions have a range between 0.0 and 6.9 whereas alkaline solutions have a range 9 between 7.1 and 14.0 on a pH scale with 7.0 being neutral (neither acidic nor basic) (Alberts et al., 2013). Pure water, containing just molecular H 2 O, is neutral. Any factor that affects the hydrogen ion concentration in water will change the pH of water; both beneficial and non - beneficial constituents contribute to t he change in pH of water. Carbon dioxide, carbonic acid, and the rare occasion of exposure to chemicals (e.g., hydrochloric acid) will decrease the pH of water making it more acidic. Acidic water is corrosive to exposed metal in pipes and other materials used to transport and store water (DHEC, 2013). The longer acidic water is in contact with the exposed metal the more the solution will corrode the metal, releasing metal particles or ions into the water. Acidic water can be a source of Fe, Pb, Cu, and Zn in a drinking water due to the corrosion of metal pipes and joint surfaces (DHEC, 2013). Alkaline water typically has an increased presence of Ca and Mg compounds (e.g., calcium carbonate). Water pH is considered a SMCL and is generally not an issue u nless the pH approaches the extreme ends of the pH scale (too acidic or too basic). The EPA set the recommended SMCL for water pH to fall within the range of 6.5 to 8.5 (EPA, 2018) . As a reference, natural water bodies (e.g., lakes, rivers, and streams) typically have a pH between 6.0 and 8.5, whereas well (bore) water sources usually range in pH from 5.0 to 9.0 (DHEC, 2013; Vigil, 2003). Adjusting the pH of water can be a tool to control or influence water quality. When a water source has increased con centrations of Fe and a pH that is not near neutral, bringing the pH as close to neutral as possible is required to utilize removal processes (DHEC, 2013). Adjusting the pH of water also can correct the undesired effects of water that is either too acidic (e.g., dissolved metals) or too basic (e.g., bitter taste) . If water is acidic (corrosive) and Zn or other dissolved metals are being released from pipes, the addition of basic solutions (e.g., sodium 10 hydroxide) will raise the water pH to neutraliz e the corrosive properties of the water (Vigil, 2003). Once the corrosive properties are neutralized, running the water for adequate time will flush the metal - laden water from the lines. Acidic solutions also can be added to alkaline water to bring the p H closer to neutral. This is a common water treatment method used in industrial applications to neutralize wastewater before release back into the environment at the end of production processes (Vigil, 2003). 1. 2 .6. Temperature The temperature of drinking water can affect water quality. Algae and bacteria grow more easily in warm water versus cold water (Vigil, 2003). Growth of algae and bacteria in drinking water decreases water quality as it introduces the possibility of pathogens, as well as, affectin g color, taste, odor, and corrosive properties (WHO, 2017). Research with livestock, including horses, ponies, and dairy cows showed that the temperature of the water in relation to environmental temperature can influence water intake with the general pre ference being for ambient or warm water versus cold water (Huuskonen et al., 2011; NRC, 2007; Wilks et al., 1990). Research showed water intake by dairy cows and horses increases when environmental temperatures are warmer, above 81°F (NRC, 2007; Ragsdale et al., 1949 as cited by Beede, 2005). In horses, low environmental temperatures, between - to - 17 decrease water intake by 6 to 14% (NRC, 2007). Neither the EPA nor the WHO have specific temperature value or range recommendations for water quality, but the WHO does state, water is generally more palatable than warm in terms of human wat er temperature preferences (WHO, 2017). Whereas colder water temperatures may be the more palatable choice for people, it is not necessarily the case for livestock, as stated above. 11 1. 2 .7. Biological Factors Biological factors influencing water quality in clude bacteria, viruses, protozoa, and algae, which can lead to a wide array of diseases depending upon the specific pathogen. Bacterial agents in drinking water can lead to diseases, such as, typhoid fever, diarrhea, shigellosis, and gastroenteritis; vir al agents can cause hepatitis and gastroenteritis; and protozoan agents can cause giardiasis, amoebiasis, and cryptosporidiosis (Gray, 2008). Algal agents include cyanobacteria, formerly known as blue - green algae that can produce toxic blooms in drinkin g water sources. The cyanotoxins of algae can damage nerve and liver tissue in humans, and cause disease in birds and other animals (WSDH, 2019). According to the EPA, the most common types of cyanotoxins found in U.S. waters are microcystins, cylindrosp ermopsin, anatoxins, and saxitoxins (EPA, 2019). Cyanotoxins can be lethal to aquatic species, and to livestock consuming contaminated water sources (WHO, 2003). A relatively recent case of cyanobacteria bloom in a moat around an exhibit in an unnamed No rth American zoo led to the death of several yellow - bellied sliders ( Trachemys scripta scripta ) and increased awareness and concern for cyanotoxin presence in zoo waters and the potential threat they pose for zoo animals (Doster et al., 2014). Although th e EPA does not have a specific MCL for cyanobacteria, the agency published 10 - day Health Advisories (HAs) for cyanotoxins in drinking water; but, these contaminants are not subject to any national primary drinking water regulation (EPA, 2019a). The 10 - day HAs for cyanobacteria vary by state, cyanotoxin type, and the age of the person consuming the water. Generally speaking, 10 - day HAs range from 0.3 - 3.0 µg/L depending upon the stipulations listed previously (EPA, 2015). 12 1. 2 .8. Nitrates and Nitrites An increased concentration of nitrogen (N) in water is cause for concern as it can be an indication of increased nitrate and (or) nitrite concentrations. Nitrates (NO 3 - ) are more stable than nitrites (NO 2 - ) and are more common in soils (DHEC, 201 3). Soil nitrates are readily dissolved in water and can travel rapidly through the environment. Both NO 3 - and NO 2 - are naturally occurring ions present in the N - cycle. During the N - cycle, ammonia (NH 3 ) and ammonium (NH 4 + ) can be oxidized in aerobic con ditions via ammonia - oxidizing bacteria to NO 2 - , which can be further oxidized to NO 3 - (Stein and Klotz, 2016). Increased concentrations of NO 3 - are most commonly a result of inorganic fertilizer application during crop production and (or) the presence of concentrated human or animal waste. Nitrite is less common in water sources and is most often used as a preservative in cured meats. High NO 3 - concentrations can encourage the growth of bacteria and other pathogens in water. Algal and bacterial blooms cause health concerns for humans and animals consuming affected water, as discussed in the previous section. Nitrate is readily converted to NO 2 - via reduction in the body and then readily converted to N - nitrosamines (Gray, 2008). N - nitrosamines are carcinogenic compounds that increase the risk of gastric cancer in experiments using mice offered both dietary and water N - nitrosamine sources (Trick er and Preussmann, 1991). High NO 2 - intake is associated with methemoglobinemia, baby in human infants, and other mammals (DHEC, 2013; NRC, 2007). Methemoglobinemia occurs when high concentrations of NO 2 - displace the oxygen bound to Fe in hemoglobin, creating methemoglobin, decreasing the oxygen - carrying capacity of red blood cells (Gray, 2008). The decreased oxygen carrying capacity of the red blood cells leads to decreased delivery of oxygen to tissues, and in extreme cases can lead t o the death of the infant or animal due to lack of oxygen. 13 In ruminant animals, NO 3 - is rapidly reduced to NO 2 - by microbes in the rumen (Church, 1993). This can become a problem for cattle and other ruminant species when high concentrations of NO 3 - build up in plants used as forage. Outside factors typically contribute to the accumulation of NO 3 - in forage plant s, such as drought, frost damage, shading, herbicide application, and high concentrations of nitrogenous compounds in the surface and (or) groundwater where the plants are grown (Church, 1993; Costagliola et al., 2014; NRC, 2005). The increased presence o f NO 3 - in the forage can cause nitr i te toxicity when gut microbes reduce the NO 3 - to NO 2 - in the rumen. The effects are identical to methemoglobinemia in humans, and if not treated can lead to the death of the animal (Church, 1993; NRC, 2005). A recent c ase of nitrate toxicity in dairy cattle occurred in the Campania region of Italy. A small herd of 50 dairy cows was fed waste scraps of fennel meant for human consumption that did not sell at a local market; within 48 hours of eating the fennel 15 of the cows had died. Tissue and blood samples taken during necropsies showed high levels of NO 3 - present in the animals, and it was determined they had died of acute nitrate toxicity (Costagliola et al., 2014). The fennel had been grown in a contaminated regio n of Campania, and samples from both the green and white portions of the fennel showed high concentrations of NO 3 - present in the plants (Costagliola et al., 2014). While fennel and other novel produce are not common feedstuff for livestock in the United States, many zoo animals are provided fruit (e.g., apples, bananas, grapes, and papaya) and (or) vegetables (e.g., pumpkins, bell peppers, carrots, and leafy greens) either to help meet nutritional needs as part of the daily diet, or as a treat duri ng enrichment and keeper interactions (Dadone et al., 2016; Shapiro et al., 2018; Shim and Dierenfeld, 2017). However, more common 14 forage plants (e.g., corn, straw, hay, and oats) have also been found to accumulate NO 3 - in some instances (Church, 1993). In non - ruminants the presence of increased NO 3 - in feedstuffs is less of an issue, as they do not have the gut microbes required to reduce NO 3 - to NO 2 - that are present in ruminant animals. In order for non - ruminant species to suffer acute nitrate toxicity, an animal must drink from a nitrite - contaminated water source or directly ingest large amounts of nitrates (NRC, 2005). The extent to which NO 3 - in drinking water contribute to the detrimental effects on human and animal health is still under debate. Addiscott and Benjamin (2004) question the negative health effects associated with NO 3 - , and whether or not current recommendations for NO 3 - in drinking water for people should be raised due to lack of convincing evidence of harm to human health. As of this writing, the EPA still considers NO 3 - and NO 2 - contributing factors in baby and has not changed the MCL for NO 3 - nor NO 2 - concentrations in drinking water (St. Clair, 2019). The current MCL for NO 3 - is 10 mg/L and NO 2 - is 1 mg/L in drinking water (EPA, 2018). Current drinking water recommendations for livestock are 100 mg/L and 10 mg/L for NO 3 - and NO 2 - respectively (NRC, 1974, 2005). 1. 2 .9. Metals Copper, Fe, Pb, and Zn are metals that can be found in drinking water and may contribute to health issues in animals and humans. The degree to which these metals pose a risk to the health of animals consuming contaminated drinking water depends upon the type of metal present, its concentration, and the animal. Lead is recognized as a highly toxic substance even at relatively low concentrations, whereas, Fe concentrations above the SMCL recommendation of 0.3 mg/L in drinking water may or may not pose a health risk depending 15 upon the specie consuming the water. Each metal poses a different risk in drinking water and each has its own recommended safety concentrations, so considerations for each metal are d escribed in detail subsequently. 1. 2 .9.1. Copper and Molybdenum Copper is not naturally found in drinking water supplies, but when present imparts a metallic taste. It is present in drinking water as a result of the erosion of plumbing (DHEC, 2013). It t ypically poses little risk to human health when present in high amounts in drinking water, because humans stop drinking the water due to the taste before ever consuming toxic concentrations (DHEC, 2013). Copper and Mo have a close interrelationship in the body. High concentrations of Mo in the diet can lead to Cu deficiency, whereas low concentrations of Mo in the diet can lead to Cu toxicity (Church, 1993). Due to this relationship , the Mo status and specie of the animal play an important role in how susceptible an individual animal will be to increased Cu in the drinking water. Like Cu, Mo is typically not naturally found in drinking water supplies. It is used for many industrial and agricultural purposes across the United State s. Waste and runoff from these applications allow Mo to enter the water table, making its way into drinking water (WHO, 2011). The most common form of Mo used in industrial operations is molybdenum disulfide (MoS 2 ), which is not readily soluble in water, but is easily oxidized to more soluble and water - stable molybdates (WHO, 2011). It is recommended that the ratio of Cu - to - Mo in the diet of sheep be 6:1 or less, and not exceed 10:1, or Cu toxicosis may result (Villar et al., 2002). Whereas this is a di etary recommendation, it is important to take into account the possibility that high concentrations of Cu in drinking water can contribute to overall dietary Cu intake and an Cu status. Long term or toxic exposure to excess Cu in drinking water c an result in liver and kidney damage in humans, calves, and sheep with the latter 16 two having an increased risk of death (Church, 1993; Pond et al., 2005; EPA, 2018b). The current MCLG for Cu in drinking water for humans has an action level threshold 1.3 m g/L (EPA, 2018a). The upper limit for Cu in the drinking water of livestock and poultry is set at 0.5 mg/L (NRC, 1974). Currently, the EPA only has an HAs for Mo and it ranges from a 0.005 mg/L Reference Dose (RfD), to 0.08 mg/L 10 - day HAs for children, and at the high end 0.2 mg/L Drinking Water Equivalent Level (DWEL) (EPA, 2018a). 1. 2 .9.2. Lead 17 1. 2 .9.3. Zinc 1. 2 .9.4. Manganese 18 19 1. 2 .9.5. Iron In livestock, no safe upper limit wa s established for Fe in drinking water (Beede, 1991; NRC, 1974). An Fe concentration of 17 mg/L caused reduced growth, decreased milk production, and scouring in pastured cattle (NRC, 1974). A more recent study with dairy cows showed no observable decrea se in water intake at Fe concentrations up to 4 mg/L (Genther and Beede, 2013). 20 Increased concentrations of Fe in water can encourage the growth of iron - loving bacteria. 1. 2 .10. Water Quality Index (WQI) 21 Algorithm 1.1. O riginal Water Quality Index ( WQI ), w here C n is the rating scale, W n is the weighting factors, and M 1 is the coefficient for temperature, and M 2 is the coefficient for (Horton, 1965). (1.1.) 22 Algorithm 1.2. U pdated Water Quality Index ( WQI ) , w here w i is the weight of the parameter (a number between 0 to 1) and q i is the quality of the parameter (a number between 0 to100) (Brown et al., 1970). (1.2.) (1.3.) 23 Algorithm 1.3. F ormula for calculating the quality rating scale , w here C i is the measured concentration of the i th parameter and the S i is the standard value of the i th parameter (Akter et al., 2016) Algorithm 1.4. F ormula for calculating the relative weight (Akter et al., 2016). Algorithm 1.5. W eighted arithmetic Water Quality Index (WQI), w here w i is the relative weight of the i th parameter and q i is the quality rating scale of the i th parameter (Akter et al., 2016). 1. 3 Iron in the body 1. 3 .1 Importance (1.4.) (1.5.) 24 1. 3 .2 Metabolism 25 1. 3 .3 Transport 26 1. 3 .4. Storage 27 1. 3 .5. Regulation 28 29 1. 3 .6. Iron - Manganese Interactions 1. 3 .7. Disorders 30 31 1. 3 .8. Dietary Recommendations for Black Rhino 32 1. 4 . Black Rhino Biology 1. 4 .1. Wild Biology 33 1. 4 .2. Wild Diet and Eating Habits 34 1. 4 .3. Iron in the Wild Diet 35 1. 4 .4. Genetics of Iron Absorption 1. 4 .5. Gut Microbiome 36 1. 5 . Bla ck Rhino Husbandry 1. 5 .1 . Captive Diet and Eating Habits 37 1. 5 .2 . Iron in the Captive Diet 1. 5 .3. Iron Control Methods 38 1. 6 . Conclusions 39 CHAPTER 2 ASSESSMENT OF DRINKING WATER QUALITY AND RELATED HUSBANDRY PRACTICES IN NORTH AMERICAN ZOOS 40 2. 1 . Introduction quality drinking water can impact negatively the health of animals (NRC, 1974). For example, essential nutrients such as iron might be present in drinking water in great enough concentrations to contribute to health e ffects rhinoceros , if iron concentrations are not properly monitored and corrected. 41 2. 2 . Materials and Methods 2. 2 .1. Sample Groups A contact list of all zoo facilities accredited by the AZA was obtained directly from the Association. This list was used to randomly invite zoos to participate in the study. Initially the list included a total of 233 accredited facili ties; some were removed from consideration or disqualified (see reasons below), and the remainder were regrouped for random selection based on purposes of the study ( Figure F .1. ). One facility was removed due to loss of AZA accreditation and two others be cause they were only insect collections. Another 56 were removed because they were only aquariums, with no terrestrial mammalian species. The remaining 174 zoos were then divided into two groups, those with black rhinoceros (BR); n = 32] a nd those without BR - black (Non - BR); n = 142]. Once the BR and Non - BR groups were established, a random number generator selected a sample from 42 each group. Zoos to be selected and studied could participate either by: 1) completing the detaile d questionnaire plus providing the requested number of drinking water samples for water quality assessment (four samples per facility for BR zoos or one sample per facility for Non - BR zoos); or, 2) simply providing the requested drinking water sample(s) pe r group of zoo without completion of a questionnaire. There were 25 BR zoos (78% of total BR zoos) in the and water group. The remaining seven BR zoos constituted the sample group. For the Non - BR group, 100 zoos were selected randomly for the and water group with the remaining 42 being in the sample group. It was learned after group selection, that one BR zoo in the and water group did not actually have black rhino in their collection. The decision was made to place this zoo in the Non - BR and water group. This brought the BR and water group to 24, and the Non - BR group to 101 zoos. One other BR zoo recently had mo ved its last black rhino to another zoo for breeding program purposes. That zoo continued to maintain the exhibit where the last black rhino had lived but housed a different rhino species; that facility remained in the BR and water group for the study. The zoo collected the appropriate drinking water samples from the historically black rhino exhibit and completed a questionnaire while responding in reference and context to the black rhino. Once the initial BR and Non - BR ire and water and sample groups were established, all 174 eligible zoos were sent a letter of introduction with a pre - paid response postcard inviting each zoo to declare participation in our study. Zoos were provided multiple ways to respond to the invitation (e.g., the pre - paid return postcard, telephone, 43 or email). Three successive rounds of invitation letters and pre - paid response postcards were mailed ( Figure F .2. ). The first round of invitations resulted in 54 zoos responding to the invitation; 15 were BR zoos and 39 were Non - BR zoos. Two BR and two Non - BR zoos declined to participate; the rest agreed to participate. In the second round, another letter and postcard were sent to the 120 facilities that had not responded to the i nitial invitation. The second invitation resulted in 18 zoos responding with one BR zoo and 3 Non - BR zoos declining; 14 Non - BR facilities agreed to participate; no BR zoos agreed to participate in response to the second invitation. The final invitation l etter and postcard were sent to the remaining 102 zoos that had not responded to the first or second invitations. Fourteen zoos responded to the final letter. Four of five BR zoos agreed to participate, and eight of nine Non - BR zoos agreed to participate. The remaining 88 zoos never responded to any one of the three invitation letters and were designated as to Overall, a total of 76 zoos, 17 BR and 59 Non - BR, agreed to participate in the project at this stage of the selection proce ss. The 76 zoos that agreed to participate in the study were sent an email asking if they required approval by a committee at their respective facilities in order to proceed and participate; 34 of 76 zoos required in - house approval. Of those 34, five com mittees did not approve the project and the zoos were dropped from the study. The remaining 29 committees approved their participation in the study. Twenty - three zoos did not require in - house committee approval, and three withdrew from the st udy for other unknown reasons. Over the course of initiation and implementation of the study another 16 were removed due to lack of response to multiple correspondence attempts after initially agreeing to participate. Later in the study, after questionna ires and water sampling kits were mailed, two zoos (one that did not require approval, 44 and one that received approval) also were removed due to lack of responsiveness. These two zoos did not return water sampling kits or questionnaires and stopped respond ing to all correspondence attempts. This brought the total number of zoos participating in the study to 50 ( Figure F .3. ). 2. 2 .2. Questionnaires Two questionnaires were developed for the study. One questionnaire was given to Non - BR zoos randomly selected to complete the questionnaire. The second was given only to BR zoos randomly selected to complete a more extensive questionnaire. The Non - BR questionnaire contained twelve questions. The first ten (questions 1 through 10) pertained to water husbandry p ractices ( Figure C.1. ), whereas the last two (questions 11 and 12) were for our recordkeeping and not included in the analysis. These Non - BR questions were general and also were included as questions 1 through 10 in the BR questionnaire. This gave a total of 10 identical questions asked to all participating zoos selected to complete a questionnaire as part of our study. In addition to the first 10 questions answered by all zoos, BR zoos also were asked questions 11 through 28 written specifically abo ut BR husbandry and drinking water practices ( Figure C.2 . ). As with the Non - BR questionnaire, the final two questions in the BR questionnaire (questions 29 and 30) were for recordkeeping and were not included in the analysis. 2. 2 .3. Water Sampling Kits Once sample groups were established, drinking water sampling kits were created to facilitate water sample collection. To make the process as simple as possible for participating zoos, all required sampling equipment and pre - completed documents were provid ed with instructions in a prepaid shipping container mailed directly by the zoo to a commercial 45 laboratory for analysis. Four different kits were created for the different sampling groups as follows: 1) Non - BR without questionnaire, 2) Non - BR with questi onnaire, 3) BR without questionnaire, and 4) BR with questionnaire. Each kit contained drinking water sampling instructions, water sample submittal form(s) required by the laboratory, water sampling bottle(s), absorbent laboratory bench pads, 1 - gallon siz e zip - lock bag(s), and a prepaid shipping label. For zoos completing a questionnaire, the kit also included the appropriate questionnaire and a postage prepaid envelope to return the questionnaire directly to us, separate from the water sample(s). All of the kits were packaged in a U.S. Postal Service box that could be reused as the water sample shipping container. Examples of the sampling instructions, sample submittal forms, sampling bottles, and questionnaires are in Appendi x B . 2. 2 .3.1. Sample Collec tion from Non - Black Rhino Zoos All participating Non - BR zoos were asked to collect a single drinking water sample from the point. The is defined as the point within the zoo as close as possible to where the drinking water supply (e.g., well or off - site supply line) entered the zoo property. These samples were analyzed at a commercial laboratory [Cumberland Valley Analytical Services (CVAS ), Waynesboro , Pennsylvania ] using the Suitability which analyzes for a set of n on - microbial factors: pH, hardness, total dissolved solids (TDS), Ca, P, Mg, K, Na, Fe, Mn, Zn, Cu, chlorides, sulfate, and nitrate (CVAS, 2019). The Non - BR zoos that were part of the randomly selected subset also were asked to complete and return a quest ionnaire in addition to providing an drinking water sample. 2. 2 .3.2. Sample Collection from Black Rhino Zoos All participating BR zoos were asked to collect two drinking water samples from the 46 point, as well as, two drinking water sample s from within the black rhino exhibit; a total of f our drinking water samples. Two of the drinking water samples, one from the and one from within the exhibit, were analyzed by CVAS using the Suitability The remaining two dr inking water samples, one from the and one from within the exhibit, were analyzed by CVAS for total coliform and Escherichia coli bacteria. The BR zoos that were part of the randomly selected subset also were asked to complete and return a quest ionnaire in addition to providing four drinking water samples for laboratory analysis. 2. 2 .4. Water Quality Index (WQI) Calculations The results of each drinking water quality analysis were returned from the laboratory to us and the respective zoo upon c ompletion. The water quality analysis also was utilized to calculate four different water quality indexes (WQI) for each zoo. In general, the WQI is a mathematical algorithm derived as a method of assessing drinking water quality based upon standards of acceptability for each parameter or analyte set by government agencies (Brown et al., 1970; Horton, 1965). The different WQI calculations used in the current study were a specific set of non - microbial analytes: 1) low standard including 2) low standard using a set of 3) high standard including and 4) high standard using a set of as defined in Table F . 1 . The laboratory values for each analyte were used as the measured concentration ( C i ) i n the quality rating scale ( q i ) calculation ( Algorithm 2.1. ). (2.1.) Algorithm 2.1. F o rmula for calculating the quality rating scale. Where C i is the measured concentration of the i th parameter or analyte and the S i is the standard value of the i th parameter or analyte (Akter et al., 2016) 47 The standard value ( S i ) was used in the quality rating scale ( q i ) and relative weight ( w i ) ( Algorithm 2.2. ) calculations, being either the high or low standard value depending upon the WQI calculation performed. The standard values used for and standards are in Table F .2. with a more detailed table including the sources of the standards in Figure D.1 . The WQI formula used for all four calculations, regardless of S i used, was the weighted arithmetic calculation ( Algorithm 2.3. ). Total coliform (TC) and E. c oli values were not included as parameters in any of the WQI calculations because they have a standard value of zero by regulating bodies meaning that no amount of TC or E. coli in a sample is acceptable (EPA, 2018a; WHO, 2017). For the purposes of this s tudy, any concentration of TC or E. coli greater than zero was automatically used to designate water not suitable for drinking by animals. In addition, K and Mg were not included in any of the four WQI calculations because no standards are found, as neither is considered a threat to human or an imal health (BCME, 2017; EPA, 2018a; WHO, 2017). Also, P was not included because it is not defined as a threat to human or animal health (EPA, 2018; WHO, 2017). One standard value for P was found, but it was set in relation to algal blooms in water bodi es, not for water quality related human or animal health (BCME, 2017). (2.2.) (2.3.) Algorithm 2 . 2 . F ormula for calculating the relative weight (Akter et al., 2016). Algorithm 2.3. Weighted arithmetic WQI. Where w i is the relative weight of the i th parameter or analyte and q i is the quality rating scale of the i th parameter or analyte (Akter et al., 2016). 48 2. 2 .5. Exploratory Questions Developed for Statistical Analysis of the Information Thirteen statistical questions were developed to explore in the statistical analysis of information collected in the questionnaire. These statistical questions were not asked per se on the questionnaire, but rather were developed to assist in our exploratory work evaluating responses from the questionnaires: Question 1 What is the current state of drink ing water quality in non - black rhino zoos? Question 2 What is the current state of drinking water quality within black rhino exhibits in zoos? Question 3 Does having/utilizing a nutritionist or nutrition consultant affect whether or not the drinking water is tested for quality in zoos (non - black rhino and black rhino)? Question 4 Does having/utilizing a nutritionist or nutrition consultant affect whether or not the drinking water provided to the black rhino is tested specifically for iron concentration? Que stion 5 Are zoos with nutritionists/nutrition consultants more likely to be aware of the Nutrition Advisory Group ( NAG ) black rhino recommendations? Question 6 If zoos are aware of the Nutrition Advisory Group ( NAG ) black rhino recommendations do they formulate the black rhino diets based upon them? Question 7 Is there a significant difference between the quality of the origin point drinking water and the quality of the black rhino exhibit drinking water? Question 8 Does the age of the zoo affect the difference in drinking water quality between the origin point and black rhino exhibit? Question 9 Does the water source ( e.g., municipal, well [bore], or river ) affect the drinking water quality in zoos? Question 10 Does the size of the zoo have any effect on drinking water quality? 49 Question 11 Does replacing pipes within the zoo affect the difference in drinking water quality between the origin point and the black rhino exhibit? Question 12 Does the frequency of cleaning of the drinking water receptacle have any e ffect on drinking water quality within the black rhino exhibit? Question 13 Do zoos that test their drinking water for quality have better overall drinking water quality than zoos that do not test their drinking water? 2. 2 .6. Statistical Analysis Measures of central tendency and percentiles were used to evaluate questions 1 and 2 listed above . Cross - tabulation and Pearson Chi - square tests were used for questions 3 through 6. Because the data were not normally distributed, nonparametric analysis was used for the remainder of the questions (7 through 13; Linebach, Tesch, and Kovacsiss, 2014). A Sign test was used for question 7, Kruskal - Wallis H tests were used for questions 8 and 10, and Mann - Whitney U tests were used for questions 9 and 11 through 13 (Laerd Statistics, 2015). While not equivalent, a Sign test can be used for comparison of differences between paired observation data as an alternative analysis to a paired - sam ples t - test when the data are not normally distributed or Wilcoxon signed - ranked test when the data distribution is not symmetrical (Laerd Statistics, 2015). A Kruskal - Wallis H test can be considered a one - way ANOVA for non - parametric data, wh ereas the Ma nn - Whitney U test is an alternative to the independent - samples t - test for non - parametric data; al though neither are equivalent (Laerd Statistics, 2015). All effort was made to provide effect size values where possible (Tomczak and Tomczak, 2014). Descrip tive statistics also were used to describe drinking water quality based upon calculated WQI values. All statistical analysis was performed using SPSS 25.0 software (IBM Corp., 2017). Differences were declared at a significance level of P < 0.05. 50 2. 3 . Res ults and Discussion The following presentation of results and discussion is divided into three parts: 1) questionnaire responses; 2) results of WQI analysis; and, 3) examination of statistical questions utilizing WQI and questionnaire responses . 2. 3 .1. Que stionnaire Responses Questions 1 through 10 were asked to all Non - BR and BR zoos completing questionnaires ( Figure C.1. ) . In addition to the first 10 questions answered by all zoos, the BR zoos also were asked questions specifically about BR husbandry and drinking water practices. These questions were 11 through 28 in the BR questionnaire ( Figure C.2 . ). 2. 3 .1.1. General Questionnaire Responses (Questions 1 through 10) Water Quality and Source s . When asked if drinking water quality was ever co nsidered in the nutritional management of their animal collections, 23 of 39 (59%) zoos indicated Surprisingly, only seven (18%) of the same 39 zoos routinely analyzed the quality of their drinking water by laboratory analysis. Of the seven zoos t hat routinely analyzed drinking water, the majority (4 of 7) did so once per year, followed by monthly (2 of 7) and weekly (1 of 7). Eight of 39 (21%) had concerns about the quality of the drinking water in the zoo in the last 5 years. Of the 39 zoos sur veyed, the majority (92%) knew where the drinking water source entered the zoo property. Five of 39 (13%) treated the drinking water using one or more treatment methods (e.g., water softener, ion exchange, reverse osmosis, physical filtration, water recyc ling, or chlorination). Of these five, one zoo that treated the drinking water did not routinely analyze the water to monitor quality, and one other outsourced the monitoring and treatment of their drinking water to a contracted company. 51 A breakdown of t he primary drinking water sources used by all 39 zoos that completed questionnaires is in Figure F . 4 . The majority of zoos (85%) used municipal (city) water as the primary drinking water source. Well (bore) water was a distant second (8%). The remaining facilities utilized a combination of drinking water sources from municipal and well (bore) water (5%), or municipal and river water (3%). Although the majority (59%) of zoos indicated that drinking water quality was considered in the management of their animal collections, further questionnaire responses do not strongly support these claims. A fraction of the zoos (7 of 39) said they routinely analyze the drinking water provided to their animals to monitor the quality. With such a large portion (85%) o f zoos getting their drinking water from municipalities, it may be that they are concerned about drinking water quality, but assumed that the municipality is monitoring and treating the drinking water to a sufficient water quality standard for humans, whic h is, thus, acceptable for zoo animals. While it should be acceptable to trust the municipality to monitor and treat drinking water to a sufficient standard, recent events (e.g., high concentrations of lead in the drinking water in Flint, Michigan [CNN, 2 018]) might leave this to question. Contaminants and analytes may enter the drinking water via leaching or corrosion of the pipes carrying the drinking water from the municipal treatment facility to the zoo. Testing of the drinking water within the zoo a t least annually is recommended in order to monitor and, if necessary, treat the water before it is consumed by zoo animals. Diet Formulation and Length of Facility Operation. Questions not pertaining specifically to drinking water were asked to gain more perspectives about the 39 zoos. They were asked about diet formulation and who was responsible for the task. The majority (31%) of zoos worked 52 with an outside nutrition consultant to formulate diets. Some zoos formulated diets in - house either employing a full - time nutritionist (20%) or having a veterinarian(s) (26%) responsible for diet formulation. The remainder (23%) did not employ a nutritionist, consult a nutrition ist , nor did they have a veterinarian handle the diet formulation. It is unknown how or if these zoos had diet formulation and (or) monitoring of the diets for the animals in their care. The last two questions asked about the age and the size (measured in number of species kept) of the facilities. The majority of zoos had been in operatio n between 51 and 100 years (38%), followed by 36% for 50 years or less, and then 26% for over 100 years. Thirty - nine percent of zoos had over 201 species in their collections, followed by 33% with 101 to 200 species, and 28% with 100 species or less at th e time of the study. Most of the zoos in this study are large and old; with the majority (72%) having 101 species or more and 64% being more than 50 years old. The older the facility the greater the likelihood that plumbing for the drinking water also is old. Older pipes may corrode and contribute contaminants or analytes to the drinking water after the water enters the zoo. In a similar fashion , the more species a zoo has, the larger the zoo is likely to be spatially, and drinking water may have to trav el through more plumbing within the zoo to reach the exhibits. Due to this, both zoo age and zoo size might contribute to overall drinking water quality. Along the same lines, individuals tasked specifically with diet formulation might take drinking wate r quality into consideration and might monitor drinking water provided to the animals in their care. These relationships were investigated in this project. 2. 3 .1.2. Questionnaire Responses of Zoos with Black Rhino (Questions 11 through 28) In addition t o the questions all zoos answered, BR zoos also responded to questions more specifically aimed at BR husbandry and drinking water quality. 53 Drinking Water Quality and Receptacles. Of the eight BR zoos that answered questionnaires, only two tested the drink ing water for overall quality, and those same two specifically tested the drinking water for iron (Fe) concentration. Receptacles used to provide drinking water to BR within the eight zoos included troughs exclusively (3 of 8) or automatic waterers (3 of 8). The remaining zoos (2 of 8) used trough waterers, but also provided a pool of water for recreation and drinking. The majority (5 of 8) of water receptacles consisted entirely of concrete. The remainder were made of metal (1 of 8) or multiple materials (2 of 8). Five of the eight facilities cleaned the drinking water receptacles daily and the remaining three cleaned less frequently than daily . Recreational Water Provided to the BR . Enrichment ice blocks were used by seven of eight zoos at least once a year. Three zoos provided ice block enrichment monthly; the rest provided ice blocks either yearly (2 of 7) or twice per year (2 of 7). All eight zoos provided their BR with at least one other source of recreational water. All provided wallows for the BR for cooling. The other forms of recreational water provided in conjunction with a wallow by some zoos included pools, hoses, and misters. The recreational water sources were availa ble through spring, summer, and fall in all eight zoos. Four of eight zoos also continued to allow access to the recreational water through winter at their locations. The zoos were asked if they knew the type of substrate used to form the wallow, in add ition to the water, two of eight zoos used red clay, one zoo used regular topsoil, and the remaining five zoos did not know what type of substrate was used in the wallow. For the type of substrate in the exterior area of the BR exhibit, all eight zoos had multiple types of substrate, and all contained some grass. In the interior of the BR exhibit, half of the eight BR zoos used bare 54 concrete with no other surface substrate, three of eight used concrete with some sort of bedding, and one zoo used a product called Flooring According to the product website, Sydney Flooring Solutions (China Spring, Texas) is a type of multiple layered durable rubber flooring for use over concrete in livestock barns and zoos (Sydney Flooring Solutions, 2020 ). The type of substrate and (or) flooring provided in interior, exterior, and recreational areas may not seem important to drinking water quality but , depending upon the type of substrate used , unknown contaminants might get into the drinking water provi ded the rhino . Unaccounted for water contaminants might negatively affect drinking water quality and may contribute additional nutrients to the overall diet of the rhino (e.g., added iron in the water from the red clay s ub s t rate in the wallow). Nutrition and Husbandry Practices in BR Zoos. In addition to drinking water and housing questions, zoos with BR also were asked about overall nutrition and husbandry. In order to gain an understanding of how nutrition is monitored , we asked these zoos how often di ets are analyzed for nutrient content. Half of the zoos analyzed feed ingredients yearly, one facility only analyzed new hay shipments, and another routinely analyzed dietary ingredients monthly, as well as with every new hay shipment. Two zoos indicated they analyzed feed items "as The majority (6 of 8) BR zoos were aware of the Nutrition Advisory Group (NAG) 2011 Workshop and its BR dietary recommendations for iron. Of the six zoos that were aware of the NAG recommendations, all six formula ted BR diets based on the recommendations to control and (or) limit iron intake. In addition, a majority (6 of 8) of the zoos also had their BR diets tested specifically for iron concentration. The majority (6 of 8) of BR zoos indicated they would be wil ling to share their current formulated diet composition and (or) nutrient analysis with 55 researchers. The last two questions asked of BR zoos pertained to BR body weight and body condition scoring (BCS) practices. Half (4 of 8) zoos weighed their BR month ly, two weighed twice a year, one weighed daily, and one never weighed their animals. To assess BCS, two of eight zoos assessed their animals daily, and two never assessed BCS. The remaining zoos assessed the BCS either monthly (1 of 8), quarterly (1 of 8), twice per year (1 of 8), or yearly (1 of 8). 2. 3 .2. Results of Water Quality Index Analysis For background and perspective, based upon the nature of the WQI algorithm, the closer the calculated WQI value is to zero , the better the quality of the drinking water being evaluated relative to a particular set of analyte standards defined by the person doing the calculation. Also, the WQI value being closer to zero, indicating better quality water, is true regardless of t he particular set of analytes used in the calculation. During our investigation, it was found that WQI calculations using higher standard values in the formula, that is higher (but acceptable in terms of water quality) analyte concentrations , were likely to have a greater WQI value in the end calculation in some cases, compared with the same calculation using lower drinking water standards (lower analyte concentrations) for the same data from laboratory analysis of a zoo sample. All calculated WQI values for origin point sampling for Non - BR and BR zoos are in Table F .3. All WQI values for BR exhibit sampling points are in Table F .4. For BR zoos, the WQI difference was calculated by subtracting the BR e xhibit WQI value from the o rigin WQI value ( Table F .5 ) . Due to the nature of the WQI calculation, as explained in section 2. 2 .4. above, it appears a higher standard value leads to a smaller weight ( w ) and quality rating scale ( q ). The final WQI 56 formula uses a summation of the weights as the denominator in th e formula ( Algorithm 2.3. ), with a high standard resulting in a smaller denominator than a low standard, leading to a higher WQI value that may or may not necessarily reflect the severity of the water quality in a drinking water sample. Further refinement of the WQI formula may be needed in future studies. The seemingly skewed WQI values are clearly seen in Table F .3. , looking specifically at zoo identification numbers 1007, 1026, 1038, and 2004. While these four zoos are good examples of the WQI calcula tion discrepancy, they are by no means the only zoos in this study to exhibit this apparent bias between WQI for a particular water sample when low (low analyte concentrations) or high (high analyte concentrations) standard values are used. Table F .3. pre sents the WQI values at the origin sampling point for all zoos (Non - BR and BR). The WQI values are ranked from highest to lowest using the low: all analytes formula. The 50 th and 90 th percentile values, 2.0 and 13.2 respectively, across all four analyte formulas were chosen to highlight values in the table to gain a general understanding of where each zoo ranked in terms of drinking water quality within the overall sampling group. Thr ee zoos had higher WQI values ( Table F .3. ) and will be highlighted and discussed in further detail. Zoo 1101 . The WQI values for zoo 1101 were by far the highest of any zoo in the study indicating this zoo had poor overall drinking water quality. The high WQI values were influenced heavily by manganese (Mn) , which was exceptionally high at 2.24 mg/L from the drinking water sample analysis; roughly 45 times the low standard of 0.05 mg/L and 5.6 times the high standard of 0.4 mg/L. Such a high concentration of Mn was the main contributing factor to the high WQI value. 57 58 The WQI values are ranked from highest to lowest using the Low: All Analytes formula. The 50 th and 90 th perc entile values, 2.0 and 10.2 respectively, across all four analyte formulas were chosen to highlight the WQI results within the table to gain a general understanding of where each zoo ranked. For the WQI for the BR exhibit, no Mn was present in any samples and only one zoo (2020) had Fe present in the exhibit drinking water sample. Consequently, zoo 2020 also had the highest WQI values. Although the measured Fe concentration (0.05 mg/L) was well below the standards, the presence of Fe in the WQI calculati on influenced putting zoo 2020 at the top of the list. Zoo 2020 also was the only zoo with a WQI value not within the 90 th percentile at the BR exhibit sampling point. Any change in rank order of WQI values among BR zoos between Table F .3. and Table F .4. can be attributed to a change in analytes present within the water samples between the origin and BR exhibit sampling points. For example, zoo 2004 is above zoo 2020 in Table F .3. but below it in Table F .4. ; this change can be attributed to the change in Fe presence between sampling points. Zoo 2004 had Fe present in the origin point sample but not in the BR exhibit sample, and zoo 2020 had no Fe present in the origin p oint sample but had Fe in the BR exhibit sample. Smaller changes in rank order can be contributed to slight changes in analyte 59 concentrations at the two different sampling points. While not included in WQI calculation or the water quality analysis, as stated above in section 2. 2 .4. , one BR zoo had total coliform (TC) present. Zoo 2012 had TC present at 73.8 colonies per 100 mL in the BR exhibit water sample, but not at the origin sampling point. The reason for this contamination is unknown. The water should be sampled and analyzed again to verify the results. Additional water samples should be taken at the same sampling points within some zoos to verify the initial laboratory results. Once verified, further action to correct the high analyte concent rations may need to be taken. Table F .5. shows the difference in WQI values between the origin and exhibit sampling points in BR zoos. A negative difference represents a decrease in drinking water quality from the origin point to the BR exhibit, a posit ive difference represents an increase in drinking water quality, and a zero difference suggests no change in drinking water quality. A negative zero ( - 0.0) value is the result of rounding minor changes between water quality (e.g., - 0.0067 rounded to - 0.0) . Zoo 2004 had a difference of 5.6 with a greater WQI value at the origin sampling point then the BR exhibit sampling point ( T able F .5. ). This change in water quality was due to the presence of Fe in the origin point sample and the absence of Fe in the B R exhibit sample. The opposite was presented for zoo 2020 with a WQI difference of - 4.7 with a greater WQI in the BR exhibit than at the origin sampling point. Again, this is likely due to the varied presence of Fe in the samples with no Fe in the orig in point sample but Fe present in the BR exhibit water sample. Additional water analysis would be recommended at both sampling points to verify the difference in Fe concentration. 60 2. 3 .2.1. WQI of Non - Black Rhino Zoos For all Non - BR zoos, the drinking water samples were taken at the origin point for WQI calculations. Because the results across all four standard WQI calculations had some exceptionally high values, the median, mode, minimum, and maximum values are reported in addition to the mea n and standard deviation ( Table F .6. ). Percentiles were calculated, including the 10th, 50th, and 90th percentiles, for the WQI values at the origin sampling point in all Non - BR zoos. ( Table F .7. ). At the 90 th percentile, for the All WQI calculation in Table F .7. , 90% of all WQI values for that category were at or below a value of 7.4. Descriptive statistics were used and provided to give a better overview of the WQI results for Non - BR zoos. The mean, median, mode, minimum, and maximum i n Table F .6. and the percentiles in Table F .7. provide a way for Non - BR zoos to compare the WQI value for their facility to that rest of the sampled group of Non - BR zoos. Allowing individual Non - BR zoos to identify which percentile their zoo categorized i nto, how close to the mean and median they were, and if their WQI value occurred frequently across their sampled group. The standard deviation values provided in Table F .6. indicate the data are spread out across a wide range , indicating a wide variabilit y in drinking water quality across the sampled Non - BR zoos. 2. 3 .2.2. WQI of Black Rhino Zoos For all BR zoos, drinking water samples were taken at the origin point and within the BR exhibit. A WQI calculation was performed based on the laboratory water a nalysis at each sampling location. One sample from one BR zoo had a measurable concentration of Total Coliforms in the laboratory analysis and was deemed not suitable for drinking. Even though the WQI values for both the origin ( Table F .8. ) and BR exhibi t ( Table F .9. ) sampling points had fewer and less extreme high values than in the non - BR zoos, f or consistency the median, mode, 61 minimum, and maximum values are still reported in addition to the mean and standard deviation. In addition to measures of cent ral tendency, percentiles also were calculated, including the 10th, 50th, and 90th percentiles, for the WQI values for both the origin ( Table F .10. ) and BR exhibit ( Table F .11. ) sampling points within BR zoos. Again, the mean, median, mode, minimum, and ma ximum in Table F .8. and Table F .9. as well as, the percentiles in Table F .10. and Table F .11. provide a way for BR zoos that participated in the study to compare the WQI value at both the origin and BR exhibit sampling point for their facility to that rest of the sampled group of BR zoos. Allowing individual BR zoos to identify which percentile t heir zoo categorized into, how close to the mean and median they were, and if their WQI value occurred frequently across their sampled group. The standard deviation values provided in Table F .8. and Table F .9. show the data for both the origin point and B R exhibit sampling points to be less spread out than was the case in Non - BR zoo origin poin t results , indicating less variability in drinking water quality across the sampled BR zoos at both sampling locations. 2. 3 .3. Examination of Statistical Questions U tilizing WQI and Questionnaire Responses The questions addressed in the statistical analysis were previous ly listed in the Material and Methods 2.3. Analysis of the key questions and interpretation of answers are addressed below. Question 3. Does having /utilizing a nutritionist or nutrition consultant affect whether or not the drinking water is tested for quality in zoos (non - black rhino and black rhino)? Due to having a small sample size and the data failing the third assumption (all cells should have expected counts greater than five) required for a Chi - square test of association, the Exact Test was instead used to analyze the data. 62 Of the 40 participating Non - BR zoos, 31 were from the randomly selected group of zoos invited to complete a questionnaire; the remaining 9 only submitted drinking water samples. Of the group of 31 zoos, only four employ ed a full - time nutritionist while 27 did not . Of the zoos that did not employ a nutritionist, only four (15%) routinely analyzed dri nking water. For zoos that did employ a nutritionist, surprisingly, none routinely analyzed the drinking water. There was no association between a Non - BR zoo employing a full - time nutritionist and that zoo routinely analyzing the quality of drinking wate r as assessed by a Fisher s exact test, ( P = 1.00). From the same group of 31 Non - BR zoos that complete d a questionnaire, 11 employed a nutrition consultant while 20 did not . Of the zoos that did not employ a nutrition consultant, only two (10%) routinely analyzed their drinking water. For zoos that did employ a nutrition consultant, surprisingly again, only two (18%) routinely analyzed the drinking water. There was no association between a zoo empl oying a nutrition consultant and the zoo routinely analyzing the drinking water for quality as assessed by a Fisher s exact test, ( P = 0.601). Of the 10 BR zoos, eight were from the randomly selected group of zoos invited to complete a questionnaire. Of the eight black rhino zoos, four employed a full - time nutritionist and four did not . Of the zoos that did not employ a nutritionist, two routinely analyzed the drinking water for quality. For zoos that did employ a nutritionist, one routinely analyzed th e drinking water. There was no association between a BR zoo employing a full - time nutritionist and the zoo routinely analyzing the drinking water for quality as assessed by a Fisher s exact test, ( P = 1.00). From the same group of eight BR zoos that comp leted a questionnaire , one employed a nutrition consultant and seven did not . Of the zoos that did not employ a nutrition consultant, 63 three routinely analyzed the drinking water. For the zoo that did employ a nutrition consultant, it did not routinely an alyze the drinking water. There was no association between a zoo employing a nutrition consultant and the zoo routinely analyzing the drinking water as assessed by a Fisher s exact test, ( P = 1.00). Question 4. Does having/utilizing a nutritionist or nutrition consultant affect whether or not the drinking water provided to the black rhino is tested specifically for iron concentration? Of the four BR zoos that did not employ a full - time nutritionist, one had analyzed the drinking water f or iron concentration. For the four zoos that did employ a nutritionist, again, one analyzed the drinking water specifically for iron concentration. There was no association between a zoo employing a nutritionist and the zoo having ever analyzed the drin king water provided to the black rhino specifically for iron concentration as assessed by a exact test, ( P = 1.00). Of the seven BR zoos that did not employ a nutrition consultant, two analyzed the drinking water specifically for iron concentration. For the one zoo that d id employ a nutrition consultant, again, it had never analyzed the drinking water for iron concentration. There was no assoc iation between a zoo employing a nutrition consultant and the zoo having ever analyzed the drinking water for iron concentration as assessed by a Fisher s exact test, ( P = 1.00). Question 5. Are zoos with nutritionists/nutrition consultants more likely to be aware of the Nutrition Advisory Group (NAG) black rhino recommendations? From the same group of eight BR zoos invited to complete the questionnaire, four employed a full - time nutritionist and f our zoos did not . All four of the zoos that did not employ a nutritionist were aware of the NAG nutritional recommendations for formulating black rhino 64 diets. For the zoos that did employ a nutritionist, two were aware of the NAG dietary recommendations. There was no a ssociation between a zoo employing a nutritionist and whether or not that zoo was more likely to be aware of the NAG dietary recommendations for black rhino when assessed by a Fisher s exact test, ( P = 0.429). Again, seven BR zoos did not , and one did emp loy a nutrition consultant. Of the zoos that did not employ a nutrition consultant, five were aware of the NAG dietary recommendations for formulating black rhino diets. The one zoo that did employ a nutrition consultant also was aware of the NAG dietary recommendations. There was no association between a zoo employing a nutrition consultant and whether or not that zoo was more likely to be aware of the NAG dietary recommendations for black rhino as assessed by a Fisher s exact test, ( P = 1.00). Question 6. If zoos are aware of the Nutrition Advisory Group ( NAG ) black rhino recommendations do they formulate the black rhino diets based on them? Of the eight BR zoos randomly selected to complete a questionnaire, six were aware of the NAG dietary recommendations, and all six formulated the BR diets based upon the NAG recommendations. The other two zoos that completed the questionnaire were not awa re of the NAG recommendations, and therefore, the question related to formulating the black rhino diets based upon the recommendations did not apply. There was an association between a zoo being aware of the NAG dietary recommendations and formulating bla ck rhino diets based upon those recommendations as assessed by a Fisher s exact test, (P = 0.036). Question 7. Is there a significant difference between the origin point drinking water quality and the black rhino exhibit drinking water quality ? Due to the distribution of the data being neither normally nor symmetrically distributed, a n 65 exact Sign test was used to determine if there was a difference in drinking water quality between the origin point and within the BR exhibit across the four different ana lyte formulas to compute the WQI in the 10 BR zoos. The exact Sign test compares the change in sign when the difference is calculated. Results are presented as medians unless otherwise stated. There was no median change in drinking water quality between the origin point and within the BR exhibit across the four different analyte formulas ( Table F .12. ). As noted above, the exact Sign test compares the change in sign when the difference is calculated. In this case WQI within the BR exhibit is subtracted from the WQI at the origin point. The change in sign indicates whether the water quality is worse at the origin point (+), worse within the BR exhibit ( - ), or whether there was no change in water quality as signified by the respective WQI between the two sampling points ( Table F .13. ). Based upon the results , t here was no change in drinking water quality between the two sampling points regardless of set of analytes used to calculate the WQI value . No change between the two sampling points suggests that the pipes carrying the water between the locations were not contributing any additional analytes to the drinking water. For example, had there been a significant increase in the WQI value from the origin point to the BR exhibit it would be advisable to examine, and possibly replace, the pipes for corrosion, damage, or contamination. Continued sampling at both points, and comparison between the results, would help zoo personnel monitor and respond t o any changes in drinking water pipe status in the future. Qu estion 8. Does the age of the zoo affect the difference in drinking water quality between the origin point and the water source within black rhino exhibit? Due to the non - linear relationship b etween the age of the BR zoos and the difference in WQI 66 value between the two sampling points , linear regression could not be used to analyze the data; instead a Kruskal - Wallis H test was used in this analysis ( Table F .14. ). Zoo age categories were in three groups: less than 50 years (n = 2), 51 to 100 years (n = 2), and greater than 100 years (n = 4). The difference in drinking water quality between the origin and BR exhibit was calculated by subtractin g the WQI: BR Exhibit from the WQI: Origin, creating a new variable Values are mean ranks unless otherwise stated. Distributions of WQI: Difference were not similar for all zoo age groups, as assessed by visual inspection of box plots ( Figures F.5 ., F.6 ., F.7 ., and F.8 . ) for each WQI calculation (low: all analytes, low: select analytes, high: all analytes, and high: select analytes). The mean rank of WQI: Difference was not different between zoo age groups across all four analyte formulas. While that is the case, the c hange in drinking water quality between the origin point and BR exhibit decreased as the mean rank value increased ; meaning the quality of the drinking water was lower as the mean rank increased. Based on the P - values in Table F .14. , the age of the zoo ha d no effect on the difference in drinking water quality between the o rigin and BR e xhibit sampling points. The change in mean rank does indicate that the lowest quality drinking water is in the age category than or equal to 50 years (6.00), the n than 100 years (4.13), and the highest quality drinking water is in the age category - 100 years (3.75) , supporting the P - value results . The idea behind this analysis was older zoos might have older plumbing and pipe systems, which might affect drinking water quality. Monitoring drinking water quality is still important regardless of the age of the zoo, as shown here, with younger zoos having the lowest quality drinking water across zoo age categories. 67 Question 9. Does the water source (e.g., municipal, well [bore], or river) affect the drinking water quality in zoos? A Mann - Whitney U test, using an exact sampling distribution for U (Dineen and Blakesley, 1973), was used to determine whether there were differences in drink ing water quality (WQI) at the origin point among zoos using different drinking water sources (municipal, well [bore], or river) across all four WQI computations. Pyramid chart distributions for the WQI at the origin point for zoos that did and zoos tha t did not use municipal water as a drinking water source were not similar ( Figures F.9 . , F.10 . , F.11 . , and F.12 . ). Because of this, inferences cannot be made about the difference in medians between groups. Instead we looked at the differences in the dist ributions and mean ranks. Recall that a WQI value closer to zero indicates better drinking water quality. The WQI values at the origin point for zoos that did not use municipal water were greater than for zoos that did use municipal water as a drinking w ater source for three out of the four analyte formulas ( Table F .15. ). Only the all calculation method did not have a difference in drinking water quality among zoos that did and zoos that did not use municipal drinking water as their prim ary water source. Overall, the drinking water quality at the origin point for zoos that used municipal water was better than that of zoos that did not use municipal water as their primary drinking source as shown by the change in mean rank value, with a s maller mean rank indicating better drinking water quality. Pyramid chart d istributions for the WQI at the origin point for zoos that did and zoos that did not use well (bore) water as a drinking water source were not similar ( Figures F.13 . , F.14 . , F.15 . , and F.16 . ). Because of this, inferences cannot be made about the difference in medians 68 between groups. Instead we looked at the differences in the distributions and mean ranks. M edian WQI values at the origin point for zoos that did use well (bore) water as a drinking water source compared with zoos that did not use well (bore) water were not different across all four analyte formulas or WQI computations ( Table F .16. ). Therefore , there was no difference in drinking water quality between zoos that did and zoos that did not use well (bore) water as the primary drinking water source. Pyramid chart distributions for the WQI at the origin point for zoos that did or did not use river water as a drinking water source were not similar ( Figures F.17 . , F.18 . , F.19 . , and F.20 . ). Because of this, inferences cannot be made about the difference in medians between groups. Instead we looked at the differences in the distributions and mean ran ks. Median WQI value at the origin point for zoos that did use river water as a drinking water source compared with zoos that did not use river water as a drinking water source was not different across all four analyte formulas ( Table F .17. ). Again, there was no difference in drinking water quality between zoos that did and zoos that did not use river water as the primary drinking water source. Question 10. Does the size of the zoo have any effect on drinking water quality? A Kruskal - Wallis H test was used to determine if the size of zoos affects the quality of the drinking water at the origin point. The zoo size groups were: less than or equal to 100 animal (n = 11), 101 to 200 species (n = 13), and greater tha n 200 species (n = 15). Values are mean ranks unless otherwise stated. Distributions of WQI: Difference were not similar for all zoo groups, as assessed by visual inspection of box plots ( Figures F.21 ., F.22 ., F.23 ., and F.24 . ) for each calculatio n (low: all analytes, low: select analytes, high: all analytes, and high: select analytes) of WQI. Drinking water quality at the origin point decreased as the mean 69 rank value increased meaning that the quality of the drinking water is lower as the median rank increases ( Table F .18. ). Therefore, the size of the zoo had no effect on overall origin point drinking water quality. Question 11. Does replacing pipes within the zoo affect the d ifference in drinking water quality between the origin point and the black rhino exhibit? Of the zoos that responded to the questionnaire, nine had replaced and 39 had not replaced any drinking water pipes within the last 5 years. A Mann - Whitney U test was run to determine if there was a difference in drinking water quality between the origin and the BR exhibit sampling points for zoos that had replaced their drinking water pipes versus zoos that had not replaced their drinking water pipes within the las t 5 years across all four WQI analyte formulas. For all four formulas, pyramid chart distributions for zoos that had replaced their drinking water pipes compared with zoos that had not replaced pipes were not similarly shaped ( Figure F.25 . , F.26 . , F.27 . , and F.28 . ). There was no difference in the drinking water quality for zoos that had or had not replaced pipes within the last five years between the origin point and black rhino exhibit, regardless of analyte formula (WQI) used ( Table F .19. ). Question 12 . Does the frequency of cleaning of the drinking water receptacle have any effect on drinking water quality within the black rhino exhibit? A Mann - Whitney U test was used to determine if the frequency at which the drinking water receptacles are cleaned has an effect on the drinking water quality within the BR exhibit. The cleaning frequency of the drinking water receptacles was broken into tw o categories; (n = 5) and less than (n = 3). Distributions for cleaning frequency were not similar, as assessed by visual inspection of the pyramid chart ( Figure F.29 . , F.30 . , F.31 . , and F.32 . ). There 70 was no difference in the WQI value wi thin the BR exhibit in relation to the drinking water receptacle cleaning frequency, regardless of formula used ( Table F .20. ). Although the data showed the frequency in which drinking water receptacles are cleaned does not have an effect on drinking water quality, it is worth noting the water samples taken for analysis in this study were all taken at one point in time. Question 13. Do zoos that test their drinking water for quality have better overall drinking water quality than zoos that do not test th eir drinking water? Only 7 of 39 zoos in the study routinely analyzed the drinking water for quality. A Mann - Whitney U test was used to determine if routine analysis of the quality of the drinking water affects overall drinking water quality as measured at the origin sampling point across all four analyte formulas and WQI computations. Pyramid chart distributions for the routine analysis of the drinking water were not similarly shaped ( Figures F.33 . , F.34 . , F.35 . , and F.36 . ). Because of this, i nferences cannot be made about the difference in medians among groups. Instead we look ed at the differences in the distributions and mean ranks. There was no difference in the quality of the drinking water at the origin point as indicated by the WQI valu es in relation to whether or not the facility routinely analyzed the drinking water for quality, regardless of analyte formula used ( Table F .21. ). Based upon the results of the Mann - Whitney U test , routine analysis of the quality of the drinking water doe s not a ffect the actual drinking water quality at the sampling origin point. 2. 4 . Conclusions. Using the weighted arithmetic WQI algorithm with two different sets of analyte standards and two different groups of analyte selections (analyte concentrations) , this study calculated four 71 WQI values for each zoo allowing for a thorough assessment of the drinking water quality in each zoo. Half of the zoos had a WQI at or below 2.0, with the majority (90%) below 13.2 across all four analyte formulas. This was t he first known use of a WQI to assess drinking water quality in zoos and suggests the possibility for further use of WQI calculations in assessing water quality by zoo staff and future researchers. This study also sought to find possible associations amon g drinking water quality husbandry practices and zoo characteristics. The questionnaire responses were interesting. The majority (59%) of zoos responded that drinking water quality was taken into consideration in the nutritional management of their anim al collections, yet surprisingly few (18%) routinely analyzed the drinking water and even fewer (13%) used some form of water treatment. Yet, 21% of zoos that responded had concerns about the quality of their drinking water within the last 5 years. All z oos should be routinely analyzing the drinking water for basic quality at the minimum once per year and more frequently if necessary. This will allow zoos to know the quality of their drinking water, as well as, allow them to monitor, and correct, any cha nges that might occur. When it came specifically to monitoring drinking water provided to the black rhino populations within participating zoos, only two out of eight zoos routinely analyze the water to monitor iron concentration. Considering the recomme ndation from the leading workshop on Iron Overload Disorder (IOD) in black rhino suggests testing the drinking water for iron concentration at least once per year, this is a lower number in our study than anticipated . When asked , six of the eight BR zoos that responded to questionnaires responded that they were aware of the workshop recommendations, and all six that were aware said they formulate the black rhino diets based upon the recommendations. With six of eight zoos 72 adhering to the dietary recommendations offered in the workshop publication, it is surprising these zoos do not also follow the drinking water testing recommendations, as well. Again, in addition to routine analysis of general drinking water quality zoos with black r hino should follow the advice of the IOD Workshop and monitor iron concentration in the drinking water provided to their black rhino collections. Only three zoos in our study sample (n = 50 ) had WQI values at the origin point above the 90 th percentile cuto ff value of 13.2. Of the three zoos above the 90 th percentile, two exclusively used well (bore) water as the primary drinking water source. It might be advisable for zoos using well (bore) water to monitor the drinking water quality more than once per ye ar to more vigilantly monitor changes in the drinking water quality so action can be taken to correct high analyte concentrations when necessary. In the case of the two zoos (1101 and 1012 in Table F .3. ) with the highest WQI values, manganese (Mn) was the problem analyte that should be monitored. Comparing questionnaire responses to water quality values expressed as WQI found very little relationship between zoo husbandry practices, size, age, or drinking water source with one exception. Based upon the an alysis, zoos that used municipal water as their drinking water source had lower WQI values. The lower the WQI the better the drinking water quality, which means zoos that used municipal water had better drinking water quality than zoos that used a differe nt source for drinking water. The use of the weighted arithmetic WQI formula in this project, to compare drinking water quality using both a high and a low standard, also led to the discovery of an unusual characteristic in how the WQI formula works. Use of a high drinking water stand ard results in a 73 smaller denominator, which in turn leads to a larger WQI value than if a low drinking water standard was used on the same data points, as shown in Table F .3. Further research is needed to confirm this characteristic and, if necessary, der ive an alternative mathematical representation for the algorithm. Otherwise, a very sound understanding of the behavior of the algorithm and the resulting WQI values is needed for interpretation of WQI from field and research samples. Future research into drinking water quality in zoos needs to be done, preferably with an even larger sample size than was used in this study. Frankly, the direct impact on health of zoo animals, except the information known about iron on black rhino, of any of the analytes m easured in this study is largely unknown and simply surmised or inferred from knowledge in other animal species. Furthermore, the impact of high concentrations of other analytes and other drinking water contaminants not measured in this study (e.g., lead) on zoo animal health is unknown. There is a wide variety of species kept in zoos around the world. These species vary greatly in size, dietary needs, and health requirements. More research into relationships between drinking water quality and animal he alth is needed. 74 APPENDICES 75 APPENDIX A RANDOMIZED NUMBER GENERATOR OUTPUT 76 Figure A.1. Random number generator output to select the Black Rhino Zoo subsample invited to complete a questionnaire and submit drinking water samples. Twenty - five in the outline are included within the questionnaire subsample, and the remaining BR zoos only submit ted a drinking water sample. Figure A.2. Random number generator output to select the Non - BR zoo subsample 77 Figure A.2. Random number generator output to select the Non - Black Rhino zoo subsample invited to complete a questionnaire and submit drinking water samples. One hundred in the outline are included within the questionnaire subsample, and the remaining Non - Black Rhino zoos only submitted a drinking wat er sample. 78 APPENDIX B ZOO SAMPLING KIT DOCUMENTS 79 MSU Zoo Drinking Water Project INSTRUCTIONS PAGE 1 Thanks for agreeing to participate in this project! Task 1: Water Sample Collection A. Please review and follow the accompanying water sampling instructions. B. Use the white Priority Mail shipping box that you received with the empty bottle to send the filled bottle to the laboratory (shipping label accompanying). C. Complete the green highlighted sections of the Water Sampl e submittal form(s) and insert the form(s) into the box with the full water bottle(s). D. Cross out States Postal Service and PRIORITY and Affix the UPS Next Day Air label over the top of your address on the box, please seal the box with shipping tape (not scotch tape), and send immediately via UPS on Monday, Tuesday, or Wednesday. Thanks! Figure B.1. Cover page for all zoo water sample only sampling kits. Figure B.1. Cover page for all zoo water sample only sampling kits. 80 MSU Zoo Drinking Water Project INSTRUCTIONS PAGE 1 Thanks for agreeing to participate in this project! Two tasks need to be performed as soon as possible. Task 1: Water Sample Collection A. Please review and follow the accompanying water sampling instructions. B. Use the white Priority Mail shipping box that you rec eived with the empty bottle to send the filled bottle to the laboratory (shipping label accompanying). C. Complete the green highlighted sections of the Water Sample submittal form(s) and insert the form(s) into the box with the full water bottle(s). D. Cross ou t States Postal Service and PRIORITY and Affix the UPS Next Day Air label over the top of your address on the box, please seal the box with shipping tape (not scotch tape), and send immediately via UPS on Monday, Tuesday, or Wednesday. Task 2: Complete the Questionnaire ASAP! A. Appropriate zoo staff (e.g., curator, nutritionist, veterinarian, animal keepers) please complete questionnaire within 1 week. B. Return it in the accompanying postage paid envelope to Christine Homminga. Thanks! Figure B. 2 . Cover page for all zoo water sample and questionnaire subsample sampling kits. 81 Water Samp le Collection: Please collect a drinking water sample using the water sample collection instructions on the back of this page. One livestock suitability as close as possible to the origin (e.g., well or off - site supply line) of the drinking water source. If you have any questions or need any clarification of the water sampling instructions; please email Christine at homming2@msu.edu Water Sampling Instructions: Livestock Suitability Sample Container: Sample from the cold - water faucet/hose/spigot. Do not take sample from standing water Run cold water for 2 - 3 minutes. Rinse bottle 5 times, by filling to the top and dumping the water out. After the 5 th rinse; fill the sample bottle to the shoulder. Tigh ten the lid completely to avoid spillage during shipping. Wrap water sample with one of the blue absorbent pads provided. Place water sample in Ziploc bag and seal. Figure B. 3 . Origin point water sample collection instruction sheet for Non - Black Rhino zoo sampling kits. 82 Place water sample into re - used white shipping box; using the other blue absorbent pad as cushioning. Fill out highlighted sections on the water sample submittal form, and place in shipping box. Affix provided shipping label over top old address and postage mark, and mail within 24 hours of collection Monday through Wednesday Make sure to pack age water bottles so that there is adequate cushioning to prevent breakage during shipment . Do not send out samples on Thursdays, Fridays, Saturdays or Holidays because we will not receive them in time to be analyzed. Samples need to be received within 24 hours of collection for testing. Do not freeze samples. If possible, refrigerate samples prior to shipping. Figure B. 3 . 83 Water Sample Collection: Please collect four drinking water samples total using the attached water sample collection instructions. Two (one livestock suitability and one bacterial) as close as possible to the origin (e.g., well or off - site supply line) of the drinking water source. Two (one livestock suitability and one bacterial) from within the black rhino exhibit , as close as possible to the location the drinking water enters the exhibit. Water Sampling Instructions: Livestock Suitability Sample Container: Sample from the cold - water faucet/hose/spigot. Do not take sample from standing water Run cold water for 2 - 3 minutes. Rinse bottle 5 times, by filling to the top and dumping the water out. After the 5 th rinse; fill the sample bottle to the shoulder. Tigh ten the lid completely to avoid spillage during shipping. Wrap water sample with one of the blue absorbent pads provided. Figure B. 4 . Origin point and Exhibit water sample collection instruction sheet for Black Rhino zoo sampling kits. 84 Place water sample in Ziploc bag and seal. Place water sample into re - used white shipping box; using the other blue absorbent pad as cushioning. Fill out highlighted sections on the water sample submittal form, and place in shipping box. Affix provided shipping label over top old address and postage mark, and mail within 24 hours of collection Monday through Wednesday Make sure to package water bottles so that there is adequate cushioning to prevent breakage during shipment . Do not send out samples on Thursdays, Fridays, Saturdays or Holidays because we will not receive them in time to be analyzed. Samples need to be received within 24 hours of collection for testing. Do not freeze samples. If possible, refrigerate samples prior to shipping. Water Sampling Instructions: Bacterial Sample Container: Do not rinse bottle prior to collection because this will remove the sodium thiosulfate (used to remove residual chlorine). These are sterile, sealed containers. To avoid contamination, do not touch the inside of the bottle, cap, or threads. Figure B. 4 . 85 Sample from the cold - wa ter faucet/hose/spigot. Do not take sample from standing water source (trough, bucket, etc.) Run cold water for 2 - 3 minutes. DO NOT RINSE! Fill the sample bottle to just above the 100ml line. Tighten the lid completely to avoid spillage during shipping. Wrap water sample with one of the blue absorbent pads provided. Place water sample in Ziploc bag and seal. Place water sample into re - used white shipping box; using the other blue absorbent pad as cushioning. Fill out highlighted sections on the water sample submittal form, and place in shipping box. Affix provided shipping label over top old address and postage mark, and mail within 24 hours of collection Monday through Wednesday Make sure to package water bottles so that there is adequate cush ioning to prevent breakage during shipment . Do not send out samples on Thursdays, Fridays, Saturdays or Holidays because we will not receive them in time to be analyzed. Samples need to be received within 24 hours of collection for testing. Do not freeze samples. If possible, refrigerate samples prior to shipping. Figure B.4. 86 Figure B. 5 . Laboratory water sample submittal forms required for each water sample submitted by a zoo. 87 APPENDIX C QUESTIONNAIRE 88 1. Is drinking water quality ever considered in the nutritional management of your animal collection? Yes No 2. Does your facility routinely analyze the drinking water provided to your animal collection for quality? Yes No a. If yes, how frequently? 3. City (Municipal) Well (bore) Recycled Other (please explain): 4. Has your facility had any concerns about drinking water quality during the last 5 years? Yes No a. If yes, please explain briefly. 5. Do you or some other employee know the location where the main drinking water source comes into the zoo? Yes No 6. Has your facility replaced a ny or all of the drinking water pipes within the last 5 years? Yes No 7. Does your facility treat the drinking water to improve quality? Yes No a. If yes, what type of drinking water treatment is your facility currently employing? 8. Does your facility employ a full - time nutritionist or a nutrition consultant to formulate zoo animal diets? 9. How old is the original portion of your facility? 10. How many species does your facility have in its collection? a. If greater than 500, please specify total number of species below: Figure C.1. Questions 1 through 10 provided to all zoos (Black Rhino and Non - Black Rhino) in the questionnaire subsample group. 89 11. Has your facility ever tested the drinking water provided to black rhino for water quality? Yes No 12. Has your facility ever tested the drinking water provided to the black rhino specifically for iron concentration? Yes No 13. How many drinking water receptacles are provided in the black rhino exhibit and holding area(s)? 14. What type of drinking water receptacles are provided in the black rhino exhibit and holding area(s)? 15. What type of material is/are the drinking water receptacle(s) made of? 16. How frequently is/are the drinking water receptacle(s) cleaned in the black rhino exhibit and holding area(s)? 17. Does your facility provide ice block enrichment to your black rhino at any time throughout the year? Yes No a. If yes, how often are the black rhino offered ice block enrichment? b. What is the water source used to create the ice blocks, if known? 18. Do the black rhino at your facility have access to a recreational water source (e.g., wallow, pool, hose, sprinkler/m ister, etc.)? Yes No a. If yes, what type of recreational water source is provided to the black rhino? a. If yes, what is the water source used to supply the recreational water, if known? b. If yes, what type of clay/soil is in your wallow, if known? c. If yes, in which seasons do the black rhino have access to the recreational water source? (check all that apply): Spring Summer Fall Winter 19. What type of ground cover is currently in the exterior area of the black rhino exhibit? Figure C.2. Questions 11 through 19 of 28 provided only to Black Rhino zoos in the questionnaire subsample group. 90 20. What type of ground cover is currently in the interior area of the black rhino exhibit and holding area(s)? 21. How frequently does your facility analyze the nutrient contents of diets (e.g., crude protein, fiber, and minerals)? 22. nutritional recommendations reported by the 2011 Nutrition Advisory Group workshop (published in the Journal of Zoo and Wildlife Medicine)? Yes No a. If yes, ar e the black rhino diets formulated according to the recommendations published in the workshop report? Yes No 23. Has your facility ever tested the diet provided to the black rhino specifically for iron content? Yes No 24. Would your facility be willi ng to share the current rhino diet formulations and/or analyses? Yes No 25. How many male black rhino are currently at your facility? (please circle correct number): 0 1 2 3 4 5 6 7 8 26. How many female black rhino are currently at your facility? (please circle correct number): 0 1 2 3 4 5 6 7 8 27. How frequently are the black rhino at your facility weighed? 28. How frequently does your facility assess body condition score on the black rhino ? Figure C.2. 91 APPENDIX D ANALYTE STANDARDS 92 Figure D.1. Standard values and sources used for both the low and high standard Water Quality Index (WQI) calculations. 93 APPENDIX E QUESTIONS USED FOR STATISTICAL ANALYSIS 94 Question 1 What is the current state of drinking water quality in non - black rhino zoos? Question 2 What is the current state of drinking water quality within black rhino exhibits in zoos? Question 3 Does having/utilizing a nutritionist or nutrition consultant affect whether or not the drinking water is tested for quality in zoos (non - black rhino and black rhino)? Question 4 Does having/utilizing a nutritionist or nutrition consultant affect whether or not the drinking water provided to the black rhino is tested specifically for iron concentration? Question 5 Are zoos with nutritionists/nutrition consultants more likely to be aware of the Nutrition Advisory Group ( NAG ) black rhino recommendations? Question 6 If zoos are aware of the Nutrition Advisory Group ( NAG ) black rhino recommendations do they formulate the black rhino diets based upon them? Question 7 Is there a significant difference between the quality of the origin point drinking water and the quality of the black rhino exhibit drinking water? Question 8 Does the age of the zoo affect the difference in drinking water quality between the origin point and black rhino exhibit? Question 9 Does the water source ( e .g., municipal, well [bore], or river ) affect the drinking water quality in zoos? Question 10 Does the size of the zoo have any effect on drinking water quality? Question 11 Does replacing pipes within the zoo affect the difference in drinking water qualit y between the origin point and the black rhino exhibit? Question 12 Does the frequency of cleaning of the drinking water receptacle have any e ffect on drinking water quality within the black rhino exhibit? 95 Question 13 Do zoos that test their drinking water for quality have better overall drinking water quality than zoos that do not test their drinking water? 96 APPENDIX F FIGURES AND TABLES 97 Figure F .1. Organizational chart showing definition and partitioning of candidate zoos in the study design for participation . 98 Figure F .2. Organizational flow chart of study invitation responses. 99 Figure F .3. Organizational chart showing final disposition and fate of zoos initially agreeing to participate based on confirmation to one of the three invitations. Fifty total zoos participated in the study (10 in the Black Rhino [ BR ] group and 40 in the Non - Black Rhino [Non - BR] group). 100 Figure F.4. Pie chart showing the overal l breakdown of primary drinking water sources used by all 39 zoos to complete questionnaires, including both Non - Black Rhino and Black Rhino facilities. 101 Figure F.5. Box plot chart showing the distributions for zoo age across Black Rhino zoo Water Quality Index (WQI) plots show the distributions between zoo age group were not similar in shape. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 102 Figure F.6. Box plot chart showing the distributions for zoo age across Black Rhino zoo Water Quality Index (WQI) distributions between zoo age group were not similar in shape. Due to the dist ributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 103 Figure F.7. Box plot chart showing the distributions for zoo age across Black Rhino zoo Water Quality Index (WQI) distributions between zoo age group were not similar in shape. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 104 Figure F.8. Box plot chart showing the distributions for zoo age across Black Rhino zoo Water Quality Index (WQI) distributions between zoo age group were not similar in shape. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 105 Figure F.9 . Pyramid chart showing the distribution of zoos that did and did not use municipal water as their primary drinking water source. All Water Quality Index (WQI) values are from la. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 106 Figure F.10 . Pyramid chart show ing the distribution of zoos that did and did not use municipal water as their primary drinking water source. All Water Quality Index (WQI) values are from Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value present ed was an asymptotic value and not an exact p - value. 107 Figure F.11 . Pyramid chart showing the distribution of zoos that did and did not use municipal water as their primary drinking water source. All Water Quality Index (WQI) values are from ula. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 108 Figure F.12 . Pyramid chart showing the distribution of zoos that did and did not use municipal water as their primary drinking water source. All Water Quality Index (WQI) values are from ormula. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 109 Figure F.13 . Pyramid chart showing the distribution of zoos that did and did not use well(bore) water as their primary drinking water source. All Water Quality Index (WQI) values are from ula. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 110 Figure F.14 . Pyramid chart showing the distribution of zoos that did and did not use well(bore) water as their primary drinking water source. All Water Quality Index (WQI) values are from ormula. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 111 Figure F.15 . Pyramid chart showing the distribution of zoos that did and did not use well(bore) water as their primary drinking water source. All Water Quality Index (WQI) values are from mula. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 112 Figure F.16 . Pyramid chart showing the distribution of zoos that did and did not use well(bore) water as their primary drinking water source. All Water Quality Index (WQI) values are from formula. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 113 Figure F.17 . Pyramid chart showing the distribution of zoos that did and did not use river water as their primary drinking water source. All Water Quality Index (WQI) values are from the Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 114 Figure F.18 . Pyramid chart showing the distribution of zoos that did and did not use river water as their primary drinking water source. All Water Quality Index (WQI) values are from the a. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 115 Figure F.19 . Pyramid chart showing the distribution of zoos that did and did not use river water as their primary drinking water source. All Water Quality Index (WQI) values are from the Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 116 Figure F.20 . Pyramid chart showing the distribution of zoos that did and did not use river water as their primary drinking water source. All Water Quality Index (WQI) values are from the la. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 117 Figure F.21 . Box plot showing the distribution of zoo size across categories in number of species. All Water Quality Index (WQI) values are from the origin sampling point and were Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. Due to the few exceptionally high WQI values making it difficult to see whether or not the box plots are similarly shaped, these values were removed in order to better assess the box plot distributions. All analysis was performed, and all results are presented with the exceptionally high values included in the data set. 118 Figure F.22 . Box plot showing the distribution of zoo size across categories in number of species. All Water Quality Index (WQI) values are from the origin sampling point and were Black Rhino and Non - Bl ack Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. Due to the few exceptionally high WQI values making it difficult to see whether or not the box plots are similarly shaped, these values were removed in order to better assess the box plot distributions. All analysis was performed, and all results are presented with the exceptionally high values included in the data set. 119 Figure F.23 . Box plot showing the distribution of zoo size across categories in number of species. All Water Quality Index (WQI) values are from the origin sampling point and were Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. Due to the few exceptionally high WQI values making it difficult to see whether or not the box plots are similarly shaped, these values were removed in order to better assess the box plot distributions. A ll analysis was performed, and all results are presented with the exceptionally high values included in the data set. 120 Figure F.24 . Box plot showing the distribution of zoo size across categories in number of species. All Water Quality Index (WQI) values are from the origin sampling point and were Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. Due to the few exceptionally high WQI values making it difficult to see whether or not the box plots are similarly shaped, these values were removed in order to better assess the box plot distribu tions. All analysis was performed, and all results are presented with the exceptionally high values included in the data set. 121 Figure F.25 . Pyramid chart showing the distribution of zoos that had and had not replaced drinking water pipes within the last 5 years. All Water Quality Index (WQI) values are from the Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 122 Figure F.26 . Pyramid chart showing the distribution of zoos that had and had not replaced drinking water pipes within the last 5 years. All Water Quality Index (WQI) values are from the a. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 123 Figure F.27 . Pyramid chart showing the distribution of zoos that had and had not replaced drinking water pipes within the last 5 years. All Water Quality Index (WQI) values are from the Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 124 Figure F.28 . Pyramid chart showing the distribution of zoos that had and had not replaced drinking water pipes within the last 5 years. All Water Quality Index (WQI) values are from the a. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 125 Figure F.29 . Pyramid chart showing the distribution of zoos that cleaned the drinking water receptacles provided to the black rhino daily or less than daily. All Water Quality Index (WQI) values are from the Black Rhino exhibit sampling point and were calculated using Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 126 Figure F.30 . Pyramid chart showing the distribution of zoos that cleaned the drinking water receptacles provided to the black rhino daily or less than daily. All Water Quality Index (WQI) values are from the Black Rhino exhibit sampling point and were calculated using Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. Figure F.31 . Pyramid chart showing the distribution of zoos that cleaned the drinking water receptacles provided to the black rhino daily or less than daily. All Water Quality In dex (WQI) values are from the Black Rhino Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 127 Figure F.32 . Pyramid chart showing the distribution of zoos that cleaned the drinking w ater receptacles provided to the black rhino daily or less than daily. All Water Quality Index (WQI) values are from the Black Rhino Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. Figure F.33 . Pyramid chart showing the distribution of zoos that did and did not routinely analyze the drinking water provided to their animal collections. All Water Quality Index (WQI) formula. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 128 Figure F.34 . Pyramid chart show ing the distribution of zoos that did and did not routinely analyze the drinking water provided to their animal collections. All Water Quality Index (WQI) formu la. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. Figure F.35 . Pyramid chart showing the distribution of zoos that did and did not routinely analyze the drinking water provided to their animal collections. All Water Quality Index (WQI) formula. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 129 Figure F.36 . Pyramid chart showing the distribution of zoos that did and did not routinely analyze the drinking water provided to their animal collections. All Water Quality Index (WQI) formula. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Due to the distributions not being similarly shaped, the p - value presented was an asymptotic value and not an exact p - value. 130 Table F.1 . Analytes included in each of the four Water Quality Index (WQI) calculations performed for each participating zoo. The four WQI calculations were as follows 1) Low all analytes, 2) Low select analyte, 3) High all analytes, and 4) High select analytes. WQI Calculations Low High All Analytes Select Analytes All Analytes Select Analytes pH pH pH pH NO 3 - Cl - NO 3 - Cl - Cl - SO 4 Cl - SO 4 SO 4 Ca SO 4 Ca TDS Fe TDS Fe Hardness Mn Hardness Mn Ca Na Ca Na Cu Cu Fe Fe Mn Mn Na Na Zn Zn 131 Analytes Standard Low High pH 6.5 8.5 Nitrate as Nitrogen (mg/L) 10 10 Nitrate as NO 3 - (mg/L) 45 50 Hardness (mg/L) 100 300 Total Dissolved Solids (mg/L) 500 600 Chloride (mg/L) 250 250 Sulfate (mg/L) 250 500 Calcium (mg/L) 100 300 Phosphorus (mg/L) 0.01 0.01 Magnesium (mg/L) Potassium (mg/L) Sodium (mg/L) 20 200 Iron (mg/L) 0.2 0.3 Manganese (mg/L) 0.05 0.4 Zinc (mg/L) 3.0 5.0 Copper (mg/L) 1.0 2.0 Total Coliform 0 0 Table F .2. List of low and high standards used in the calculation of Water Quality Index (WQI) values for each analyte included in the WQI calculations. Phosphorus and magnesium do not have a standard value and were not included in any of the WQI calculations nor s tatistical analysis. 132 Table F .3. Origin point Water Quality Index (WQI) values for all zoos (Non - Black Rhino and Black Rhino ) and all four analytes. Ranked in order of highest to lowest WQI value for the Low: All Analytes formula. (a) indicat es a WQI value greater than or equal to 2.0, the highest 50 th percentile value across all analyte formulas . (b) indicat es a WQI value greater than or equal to 13. 2 , the highest 90 th percentile value across all analyte formulas . Zoo ID WQI Calculations Origin Point Low High All Analytes Select Analytes All Analytes Select Analytes 1101 3358.2 b 3553.3 b 207.9 b 236.3 b 1012 195.8 b 207.0 b 38.9 b 44.1 b 1038 46.9 b 48.6 b 81.2 b 91.2 b 1026 7.5 a 7.9 a 13.1 a 14.8 b 1007 7.4 a 7.6 a 11.5 a 12.9 a 2004 7.3 a 7.7 a 11.6 a 13.2 b 1052 3.2 a 1.2 3.9 a 1.8 1136 2.9 a 0.9 3.6 a 1.7 1089 2.7 a 1.3 3.0 a 1.9 1004 2.4 a 1.1 2.8 a 1.7 1105 2.2 a 1.4 2.4 a 1.9 1109 2.2 a 0.7 2.9 a 1.6 1037 1.8 1.8 2.0 a 2.1 a 1096 1.8 1.7 2.0 a 1.8 133 Table F .3. Zoo ID WQI Calculations Origin Point Low High All Analytes Select Analytes All Analytes Select Analytes 1017 1.6 1.5 2.0 a 1.9 1042 1.6 1.2 2.6 a 1.8 1029 1.5 1.4 2.0 a 1.8 1057 1.5 1.3 2.0 a 1.9 2016 1.5 1.4 2.4 a 2.2 a 2021 1.5 1.6 1.8 2.1 a 1128 1.4 1.3 1.9 1.8 1016 1.2 1.0 2.0 a 1.7 1084 1.2 1.1 1.7 1.8 1107 1.2 1.0 2.8 a 2.1 a 1126 1.2 0.8 2.7 a 1.7 1139 1.2 1.0 2.3 a 1.8 1021 1.1 1.1 1.8 2.1 a 1045 1.1 1.2 1.8 2.0 a 1068 1.1 1.0 2.0 a 2.0 a 1116 1.1 1.1 1.9 2.0 a 1130 1.1 0.9 1.8 1.7 1132 1.1 1.2 1.8 2.1 a 134 Table F.3. Zoo ID WQI Calculations Origin Point Low High All Analytes Select Analytes All Analytes Select Analytes 1143 1.1 1.0 2.0 a 1.6 2003 1.1 1.0 2.0 a 2.1 a 1019 1.0 1.0 1.6 1.7 1034 1.0 0.8 1.6 1.6 1066 1.0 0.9 1.8 1.8 1074 1.0 0.9 1.7 1.7 2008 1.0 1.0 1.9 2.0 a 1036 0.9 0.9 1.8 2.0 a 1091 0.9 0.9 1.6 1.7 1131 0.9 0.9 1.7 1.7 2019 0.9 0.7 1.6 1.6 1035 0.8 0.8 1.6 1.8 1127 0.8 0.8 1.6 1.7 2011 0.8 0.8 1.5 1.7 2012 0.8 0.8 1.7 1.8 2020 0.8 0.8 1.6 1.8 135 Table F .3. Zoo ID WQI Calculations Origin Point Low High All Analytes Select Analytes All Analytes Select Analytes 1069 0.7 0.8 1.5 1.7 2006 0.7 0.8 1.5 1.7 Table F.4. Exhibit Water Quality Index (WQI) values for all Black Rhino zoos and all four analytes. Ranked in order of highest to lowest WQI value for the Low: All Analytes formula. (a) indicates a WQI value greater than or equal to 2.0, the highest 50 th percentile value across all analyte formulas . (b) indicates a WQI value greater than or equal to 10.2, the highest 90 th percentile value across all analyte formulas . Zoo ID WQI Calculations Black Rhino Exhibit Low High All Analytes Select Analytes All Analytes Select Analytes 2020 5.5 a 5.8 a 9.7 a 11.1 b 2004 1.7 1.8 1.8 2.0 a 2016 1.5 1.4 2.4 a 2.3 a 2021 1.4 1.5 1.9 2.0 a 2011 1.1 0.8 1.8 1.7 2003 1.0 1.0 2.0 a 2.1 a 136 Table F.4. Zoo ID WQI Calculations Black Rhino Exhibit Low High All Analytes Select Analytes All Analytes Select Analytes 2008 1.0 1.0 1.9 1.9 2012 0.8 0.8 1.6 1.8 2006 0.7 0.7 1.5 1.7 2019 0.7 0.7 1.5 1.7 137 Table F .5. Difference in drinking water quality between the o rigin and Black Rhino e xhibit sampling points, as shown by a change in calculated Water Quality Index (WQI) value between the two points across all four analyte formulas . The Difference was calculated by subtracting the Black Rhino e xhibit WQI value from the o rigin WQI value (O rigin - Black Rhino Exhibit=Difference). ( a ) Indicates a decrease in drinking water quality from the o rigin to the Black Rhino e xhibit sample points. Negative zero ( - 0.0) being possible due to the rounding of minor changes between water quality at the two s ampling points (e.g., - 0.0067 rounding down to - 0.0). Zoo ID WQI Values Low: All Analytes Origin Exhibit Difference 2003 1.1 1.0 0.0 2004 7.3 1.7 5.6 2006 0.7 0.7 0.0 2008 1.0 1.0 0.0 2011 0.8 1.1 - 0.3 a 2012 0.8 0.8 0.0 2016 1.5 1.5 - 0.0 a 2019 0.9 0.7 0.1 2020 0.8 5.5 - 4.7 a 2021 1.5 1.4 0.1 Low: Select Analytes Origin Exhibit Difference 2003 1.0 1.0 0.0 2004 7.7 1.8 5.9 2006 0.8 0.7 0.0 2008 1.0 1.0 0.0 2011 0.8 0.8 0.0 2012 0.8 0.8 - 0.0 a 138 Table F .5. Origin Exhibit Difference 2016 1.4 1.4 - 0.0 a 2019 0.7 0.7 - 0.0 a 2020 0.8 5.8 - 4.9 a 2021 1.6 1.5 0.1 High: All Analytes Origin Exhibit Difference 2003 2.0 2.0 0.0 2004 11.6 1.8 9.8 2006 1.5 1.5 0.0 2008 1.9 1.9 0.0 2011 1.5 1.8 - 0.3 a 2012 1.7 1.6 0.1 2016 2.4 2.4 - 0.0 a 2019 1.6 1.5 0.1 2020 1.6 9.7 - 8.2 a 2021 1.8 1.9 - 0.0 a High: Select Analytes Origin Exhibit Difference 2003 2.1 2.1 0.0 2004 13.2 2.0 11.2 2006 1.7 1.7 0.0 139 Table F .5. Origin Exhibit Difference 2008 2.0 1.9 0.0 2011 1.7 1.7 0.0 2012 1.8 1.8 - 0.0 a 2016 2.2 2.3 - 0.0 a 2019 1.6 1.7 - 0.0 a 2020 1.8 11.1 - 9.3 a 2021 2.1 2.0 0.1 Table F .6. Measures of Central Tendency for Non - Black Rhino zoo Water Quality Index (WQI) values at the o rigin sampling point calculated using the four different analyte formulas. WQI: Origin Low High All Analytes Select Analytes All Anal ytes Select Analytes N 40 40 40 40 Mean 91.6 96.6 10.7 11.6 Median 1.2 1.1 2.0 1.8 Mode 1.1 0.9 2.0 1.7 Std. Deviation 530.7 561.6 34.8 39.6 Min 0.7 0.7 1.5 1.6 Max 3358.2 3553.3 207.9 236.3 140 Table F .7. Percentiles for Water Quality Index (WQI) values at the o rigin point for Non - Black Rhino zoos, calculated using the four different analyte formulas. WQI: Origin Low High All Analytes Select Analytes All Anal ytes Select Analytes N 40 40 40 40 Percentiles 10th 0. 9 0.8 1.6 1.7 50th 1.2 1. 1 2.0 1.8 90th 7. 5 7. 9 12.9 14.6 Table F .8. Measures of Central Tendency for Black Rhino zoo Water Quality Index (WQI) values at the o rigin sampling point calculated using the four different analyte formulas. (a) Multiple modes exist for the data; smallest value shown. WQI: Origin Low High All Analytes Select Analytes All Anal ytes Select Analytes N 10 10 10 10 Mean 1.6 1.7 2.8 3.0 Median 1.0 0.9 1.8 1.9 Mode 0.8 0.8 1.5 a 1.7 a Std. Deviation 2.0 2.1 3.1 3.6 Min 0.7 0.7 1.5 1.6 Max 7.3 7.7 11.6 13.2 141 Table F .9. Measures of Central Tendency for Black Rhino zoo Water Quality Index (WQI) values within the Black Rhino e xhibit calculated using the four different analyte formulas. (a) Multiple modes exist for the data; smallest value shown. WQI: Black Rhino Exhibit Low High All Analytes Select Analytes All Anal ytes Select Analytes N 10 10 10 10 Mean 1.5 1.6 2.6 2.8 Median 1.1 1.0 1.9 2.0 Mode 0.7 a 0.7 a 1.5 a 1.7 Std. Deviation 1.4 1.5 2.5 2.9 Min 0.7 0.7 1.5 1.7 Max 5.5 5.8 9.7 11.1 Table F .10. Percentiles for Black Rhino zoo Water Quality Index (WQI) values at the o rigin sampling point calculated using the four different analyte formulas. WQI: Origin Low High All Analytes Select Analytes All Anal ytes Select Analytes N 10 10 10 10 Percentiles 10th 0.7 0.7 1.5 1. 7 50th 1.0 0.9 1. 8 2.0 90th 6.7 7.1 10.7 12.1 142 Table F .11. Percentiles for Black Rhino zoo Water Quality Index (WQI) values within the Black Rhino e xhibit calculated using the four different analyte formulas. WQI: Black Rhino Exhibit Low High All Analytes Select Analytes All Anal ytes Select Analytes N 10 10 10 10 Percentiles 10th 0.7 0.7 1.5 1.7 50th 1.1 1.0 1.9 2.0 90th 5.1 5.4 9.0 10.2 Table F .12.: Sign Test summary table for the water quality difference between the o rigin and Black Rhino e xhibit sampling points for Black Rhino zoos calculated using the four different analyte formulas. The Difference was calculated by subtracting the Black Rhino e xhibit Water Quality Index (WQI) value from the o rigin WQI value (Origin - Black Rhino Exhibit=Difference). WQI Difference Between the Origin and Black Rhino Exhibit Low High All Analytes Select Analytes All Anal ytes Select Analytes P 0.688 0.625 1.000 1.000 Effect Size 0.13 0.16 0.00 0.00 Origin Median 0.95 0.90 1.75 1.90 Exhibit Median 1.05 1.00 1.85 1.95 Difference Median 0.00 0.00 0.00 0.00 143 Table F .13.: Sign Test summary table for the sign change of the drinking water quality between the origin and Black Rhino exhibit sampling points for Black Rhino zoos calculated using the four different analyte formulas. The change in sign indicates whether the water quality is worse at the o rigin point (+), worse within the Black Rhino exhibit ( - ), or whether there was no change in water quality between the two sampling points. Water Quality Ind ex (WQI) Change Between the Origin and Black Rhino Exhibit Low High All Analytes Select Analytes All Analytes Select A nalytes Positive (+) 4 3 3 3 Negative ( - ) 2 1 3 3 No Change 4 6 4 4 144 Table F .14.: Kruskal - Wallis H Test summary table for the effect zoo age has on the water quality difference between the origin point and Black Rhino exhibit sampling points for Black Rhino zoos calculated using the four different analyte formulas. The change in drink ing water quality between the origin point and Black Rhino exhibit decreases as the mean rank value increases; meaning the quality of the drinking water is lower as the me an rank increases. Zoo Age Effect on the Water Quality Index (WQI) Difference Between the Origin and Black Rhino Exhibit Low High All Analytes Select Analytes All Anal ytes Select Analytes P 0.314 0.248 0.088 0.248 2.318 2.789 4.860 2.789 Effect Size 0.13 0.10 0.10 0.10 Rank 6.00 6.00 7.50 6.00 51 - 100 years Mean Rank 3.75 2.50 2.75 2.50 Mean Rank 4.13 4.75 3.88 4.75 145 Table F .15. Summary table showing the p - value, Mann - Whitney U statistic, z - value, effect size, and mean ranks for the difference in Water Quality Index (WQI) values sampled at the origin point for zoos that did a nd did not use municipal water as their primary drinking water source. The WQI values for all four analyte formulas are shown in the table. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Difference in WQI Origin Between Zoos That Did and Did not Use a Municipal Water Source Low High All Analytes Select Analytes All Analytes Select Analytes P 0.027 0.004 0.126 0.009 U 13.50 5.00 24.00 8.00 z - 2.144 - 2.599 - 1.591 - 2.460 Effect Size - 0.34 - 0.42 - 0.25 - 0.39 Mean Rank for zoos that do use Municipal water 18.88 18.64 19.17 18.72 Mean Rank for zoos that do not use Municipal water 33.50 36.33 30.00 35.33 146 Table F .16. Summary table showing the p - value, Mann - Whitney U statistic, z - value, effect size, and mean ranks for the difference in WQI values sampled at the origin point for zoos that did and did not use well (bore) water as their primary drinking water source. The Water Quality Index (WQI) values for all four analyte formulas are shown in the table. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Differe nce in WQI Origin Between Zoos That Did and Did not Use a Well(bore) Water Source Low High All Analytes Select Analytes All Analytes Select Analytes P 0.526 0.314 0.823 0.192 U 100.5 110.0 90.50 117.0 z 0.654 1.057 0.232 1.364 Effect Size 0.10 0.17 0.04 0.22 Mean Rank for zoos that do use Well(bore) water 23.10 25.00 21.10 26.40 Mean Rank for zoos that do not use Well(bore) water 19.54 19.26 19.84 19.06 147 Table F .17. Summary table showing the p - value, Mann - Whitney U statistic, z - value, effect size, and mean ranks for the difference in Water Quality Index (WQI) values sampled at the orig in point for zoos that did and did not use river water as their primary drinking water source. The WQI values for all four analyte formulas are shown in the table. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Difference in WQI Origin Between Zoos That Did and Did not Use a River Water Source Low High All Analytes Select Analytes All Analytes Select Analytes P 0.564 0.821 0.356 0.410 U 10.00 15.50 4.50 7.50 z - 0.803 0.00 - 1.296 - 1.037 Effect Size - 0.13 - 0.05 - 0.21 - 0.17 Mean Rank for zoos that do use River water 11.00 16.50 5.50 8.50 Mean Rank for zoos that do not use River water 20.24 20.09 20.38 20.30 148 Table F .18. Summary table showing the P - value, Kruskal - Wallis H statistic reported as the X 2 , effect size, and mean ranks for the different zoo size categories reported in numbers of species. The p - value reported is asymptotic and not exact. The Water Quality Index (WQI) values for all four analyte formulas are shown in the table. Both Black R hino and Non - Black Rhino zoos are included in this WQI grouping. Zoo Size Effect on the WQI at the Origin Point Low High All Analytes Select Analytes All Analytes Select Analytes P 0.265 0.189 0.365 0.316 2 2.655 3.336 2.014 2.302 Effect Size 0.10 0.10 0.10 0.10 100 Species Mean Rank 18.73 16.32 18.64 16.14 101 - 200 Species Mean Rank 24.12 24.46 23.62 23.12 > 200 Species Mean Rank 17.37 18.83 17.87 20.13 Table F .19. Summary table showing the p - value, Mann - Whitney U statistic, z - value, and effect size for the difference in Water Quality Index (WQI) values between the o rigin point and the Black Rhino e xhibit for zoos that have replaced their drinking water pipes within the last 5 years versus zoos that have not replace their drinking water pipes with the last 5 years. The WQI values for all four analyte formulas are shown in the table. Only Black Rhino zoos are included in this WQI grouping. WQI Difference Between Zoos That Ha d and Ha d Not Replaced Drinking Water Pipes Low High All Analytes Select Analytes All Analytes Select Analytes P 0.114 0.200 0.114 0.200 U 2.0 3.0 2.0 3.0 z - 1.845 - 1.654 - 1.845 - 1.654 Effect Size - 0.65 - 0.58 - 0.65 - 0.85 149 Table F .20. Summary table showing the p - value, Mann - Whitney U statistic, z - value, and effect size for the difference in Water Quality Index (WQI) values sampled within the Black Rhino exhibit for zoos that cleaned the drinking water receptacles provided to the black r hino daily or less than daily . The WQI values for all four analyte formulas are shown in the table. Only Black Rhino zoos are included in this WQI grouping. Drinking Water Receptacle Cleaning Frequency Effect on WQI in Black Rhino Zoos Low High All Analytes Select Analytes All Analytes Select Analytes P 0.786 1.00 0.393 0.786 U 8.5 7.0 10.5 9.0 z 0.300 - 0.151 0.905 0.45 Effect Size 0.11 - 0.05 0.32 0.16 Table F .21. Summary table showing the p - value, Mann - Whitney U statistic, z - value, and effect size for the difference in Water Quality Index (WQI) values sampled at the origin point for zoos that did and did not routinely analyze the drinking water provided to their an imal collections. The WQI values for all four analyte formulas are shown in the table. Both Black Rhino and Non - Black Rhino zoos are included in this WQI grouping. Difference in WQI Origin Between Zoos That Did and Did not Routinely Analyze the Drinking Water Low High All Analytes Select Analytes All Analytes Select Analytes P 0.629 1.00 0.530 0.378 U 98.0 113.0 94.0 136.5 z - 0.515 0.037 - 0.663 0.910 Effect Size - 0.08 0.01 - 0.11 0.15 150 LITERATURE CITED 151 LITERATURE CITED Addiscott, T. M., & Benjamin, N. (2004). Nitrate and human health. Soil Use and Management , 20 (2), 98 104. https://doi.org/10.1079/Sum2004256 Akter, T., Jhohura, F. T., Akter, F., Chowdhury, T. R., Mistry, S. K., Dey, D., Milan, K.B., Islam, M.A., & Rahman , M. (2016). Water Quality Index for measuring drinking water quality in rural Bangladesh: A cross - sectional study. Journal of Health, Population and Nutrition , 35 (4), 1 12. https://doi.org/10.1186/s41043 - 016 - 0041 - 5 Alberts, B., Bray, D., Hopkin, K., Johns on, A., Lewis, J., Raff, M., Roberts, K . , & Walter, P. (2013). Essential Cell Biology (4th ed.). New York, NY: Garland Science. Batal, A., Fairchild, B., Ritz, C., & Vendrell, P. (2005). The effect of water manganese on broiler performance. Poultry Science , 84 (Supplement 1), 13 14. Beede, D. K. (1991). Mineral and w ater n utrition. Veterinary Clinics of North America - Food Animal Practice , 7 (2), 373 390. Beede, D. K. (2005). The m ost e ssential e ssential n utrient: w ater. In Western Dairy Management Conference (p. 22). Reno, Nevada. Beede, D. K. (2009). Solving b ad w ater p roblems for t hirsty c ows. In Western Dairy Management Conference (p. 9). Reno, Nevada. Beutler, E., West, C., Speir, J. A., Wilson, I. A., & Worley, M. (2001). The HFE g ene of b rowsi ng and g razing r hinoceroses: a p ossible s ite of a daptation to a l ow - i ron d iet. Blood Cells, Molecules, and Diseases , 27 (1), 342 350. https://doi.org/10.1006/bcmd.2001.0386 Bjø rklund, G., Aaseth, J., Skalny, A. V., Suliburska, J., Skalnaya, M. G., Nikonorov, A. A., & Tinkov, A. A. (2017). Interactions of iron with manganese, zinc, chromium, and selenium as related to prophylaxis and treatment of iron deficiency. Journal of Trace Elements in Medicine and Biology , 41 (8610), 41 53. https://doi.org/10.1 016/j.jtemb.2017.02.005 Boyd, C. E. (2015). Water Quality: An Introduction (2nd ed.). Switzerland: Springer. Retrieved from http://dx.doi.org/10.1007/978 - 3 - 319 - 17446 - 4 British Columbia Ministry of Environment. (2017). Source Drinking Water Quality Guidelin es: Guideline Summary . Victoria, B.C. Brown, R. M., McClelland, N. I., Deininger, R. A., & O Connor, M. F. (1973). A w ater q uality i ndex c rashing the p sychological b arrier. In S. H. Jenkins (Ed.), Advances in Water Pollution Research (pp. 787 797). Oxfor d, U.K.: Pergamon Press Limited. https://doi.org/10.1016/b978 - 0 - 08 - 017005 - 3.50067 - 0 152 Brown, R. M., McClelland, N. I., Deininger, R. A., & Tozer, R. G. (1970). A water quality index - do we dare? Water Sewage Works , 117 (10), 339 343. Church, D. C. (1993). The Ruminant Animal: Digestive Physiology and Nutrition . (D. C. Church, Ed.) (2nd ed.). Waveland Press. Clauss, M., Dierenfeld, E., Goff, J., Klasing, K., Koutsos, L., Lavin, S., Livingston, S., Niels e n, B., Schlegel, M., Sullivan, K. , Valdes, E., & Ward, A. (2012). IOD in r hinos n utrition g roup r eport: r eport from the n utrition w orking g roup of the i nternational w orkshop on i ron o verload d iso r der in b rowsing r hinoceros (February 2011). Journal of Zoo and Wildlife Medicine , 43 (3), S108 S113. Retrieved from https://www - jstor - org.proxy1.cl.msu.edu/stable/pdf/41681918.pdf?refreqid=excelsior%3A7fd311d700b5b10d a56d3acffaab6f98 Clauss, M., & Paglia, D. E. (2012). Iron storage disorders in captive wild mammals: the comparative evidence. Jo urnal of Zoo and Wildlife Medicine , 43 (3), S6 S18. Retrieved from https://www.jstor.org/stable/41681907 CNN. (2018). Flint w ater c risis f ast f acts. Retrieved March 19, 2019, from https://www.cnn.com/2016/03/04/us/flint - water - crisis - fast - facts/index.html Co stagliola, A., Roperto, F., Benedetto, D., Anastasio, A., Marrone, R., Perillo, A., Russo, V., Papparella, S., & Paciello, O. (2014). Outbreak of fatal nitrate toxicosis associated with consumption of fennels ( Foeniculum vulgare ) in cattle farmed in Campan ia region (southern Italy). Environmental Science and Pollution Research , 21 (9), 6252 6257. https://doi.org/10.1007/s11356 - 014 - 2520 - 9 Cumberland Valley Analytical Services. (2019). Water Evaluation Services. Retrieved March 18, 2019, from https://www.forag elab.com/Lab - Services/Water Dadone, L. I., Schilz, A., Friedman, S. G., Bredahl, J., Foxworth, S., & Chastain, B. (2016). Training giraffe ( Giraffa camelopardalis reticulata ) for front foot radiographs and hoof care. Zoo Biology , 35 (3), 228 236. https://do i.org/10.1002/zoo.21279 Department of Health and Environmental Control. (2013). Drinking w ater: c ommon w ater q uality p roblems and t heir t reatment. Columbia, SC: South Carolina Department of Health and Environmental Control. Retrieved from https://scdhec.gov/sites/default/files/Library/CR - 006453.pdf Dierenfeld, E. S. (1996). Rhino SSP Husbandry Manual. Dinneen, L. C., & Blakesley, B. C. (1973). Algorithm AS 62: A g enerator for the s ampling d istribution of the Mann - Whitney U s tatistic. Appli ed Statistics, 22(2), 269 273. Retrieved from https://www.jstor.org/stable/2346934 Dorst, J., & Dandelot, P. (1969). A Field Guide to the Larger Mammals of Africa . Boston: 153 Houghton Mifflin Company. Doster, E., Chislock, M. F., Roberts, J. F., Kottwitz, J. J., & Wilson, A. E. (2014). Recognition of an important water quality issue at zoos: prevalence and potential threat of toxic cyanobacteria. Journal of Zoo and Wildlife Medicine , 45 (1), 165 168. Retrieved from https://doi.org/10.1638/2013 - 0040R1.1 Dryden, G. (2008). Animal Science Nutrition (1st ed.). Wallingford, UK: CABI. Genther, O. N., & Beede, D. K. (2013). Preference and drinking behavior of lactating dairy cows offered water with different concentrations, valences, and sources of iron. Journal of Dai ry Science , 96 (2), 1164 1176. https://doi.org/10.3168/jds.2012 - 5877 Gibson, K. M., Nguyen, B. N., Neumann, L. M., Miller, M., Buss, P., Daniels, S., Ahn, M.J., Crandall, K.A., & Pukazhenthi, B. (2019). Gut microbiome differences between wild and captive bl ack rhinoceros - implications for rhino health. Scientific Reports , 9 (1), 7570. https://doi.org/10.1038/s41598 - 019 - 43875 - 3 Gray, N. F. (2008). Drinking Water Quality: Problems and Solutions (2nd ed.). Cambridge University Press. Helary, S. F., Shaw, J. A., Brown, D., Clauss, M., & Owen - Smith, N. (2012). Black r hinoceros ( Diceros bicornis ) n atural d iets: c omparing i ron l evels a cross s eason and g eographical l ocations. Journal of Zoo and Wildlife Medicine , 43 (3), S48 S54. Retrieved from https://www.jstor.org/stable/41681911 Horton, R. K. (1965). An i ndex - n umber s ystem f or r ating w ater q uality. In J. Water Pollu. Cont. Fed. (Vol. 37, pp. 300 305). Retrieved from https://www.jstor.org/stable/25035251 Huuskonen, A., Tuomisto, L., & Kauppinen , R. (2011). Effect of drinking water temperature on water intake and performance of dairy calves. Journal of Dairy Science , 94 (5), 2475 2480. https://doi.org/10.3168/jds.2010 - 3723 IBM Corp. (2017). IBM SPSS s tatistics for m ac. Armonk, NY: IBM Corp. Johnson, D. L., Ambrose, S. H., Bassett, T. J., Bowen, M. L., Crummey, D. E., Isaacson, J. S., Johnson, D.N., Lamb, P., Saul, M., & Winter - Nelson, A. E. (1997). M eanings of e nvironmental t erms. Journal of Environment Quality , 26 (3), 581 589. https://doi.org/10.2134/jeq1997.00472425002600030002x Johnson, S. P., Venn - Watson, S. K., Cassle, S. E., Smith, C. R., Jensen, E. D., & Ridgway, S. H. (2009). Use of p hlebotomy t reatment in Atlantic b ottlenose d olphins with i ron o verload. Journal of the Am erican Veterinary Medicinal Association , 235 (2), 194 200. Kaneko, J. J., Harvey, J. W., & Bruss, M. L. (2008). Clinical Biochemistry of Domestic Animals . 154 (Jiro Jerry Kaneko, J. W. Harvey, & M. L. Bruss, Eds.) (6th ed.). Amsterdam: Academic Press. https://d oi.org/https://doi.org/10.1016/B978 - 0 - 12 - 370491 - 7.X0001 - 3 Kurnit, J. (2009). Diceros bicornis . Retrieved March 26, 2019, from https://animaldiversity.org/accounts/Diceros_bicornis/ Laer d Statistics (2015). Statistical tutorials and software guides. Retrieved from https://statistics.laerd.com/ Lardner, H. A., Braul, L., Schwartzkopf - Genswein, K., Schwean - Lardner, K., Damiran, D., & Darambazar, E. (2013). Consumption and drinking behavior of beef cattle offered a choice of several water types. Livestock Science , 157 (2 3), 577 585. https://doi.org/10.1016/j.livsci.2013.08.016 Lavin, S. R. (2012). Plant p henolics and t heir p otential r ole in m itigating i ron o verload d isorder in w ild a nimals. Journal of Zoo and Wildlife Medicine , 43 (3), S74 S82. Retrieved from https://www.jstor.org/stable/41681915 Linebach, J. A., Tesch, B. P., & Kovacsiss, L. M. (2014). Nonparametric statistics for applied research. Nonparametric Statistics for Applied Research. https://doi.org/10.1007/978 - 1 - 4614 - 9041 - 8 Malyszko, J. (2009). Hemojuvelin: t he h epcidin s tory c ontinues. Kidney Blood Pressure Research , 32 (2), 71 76. https://doi.org/10.1159/000208988 McKen zie, V. J., Song, S. J., Delsuc, F., Prest, T. L., Oliverio, A. M., Korpita, T. M., Alexiev, A., Amato, K. R., Metcalf, J. L., Kowalewski, M., Avenant, N. L., Link, A., Di Fiore, A., Seguin - Orlando, A., Feh, C., Orlando, L. Mendelson, J. R., Sanders, J., & Knight, R. (2017). The effects of captivity on the mammalian gut microbiome. Integrative and Comparative Biology , 57 (4), 690 704. https://doi.org/10.1093/icb/icx090 Muya, S. M., & Oguge, N. O. (2000). Effects of b rowse a vailability and q uality on b lack r hino ( Diceros bicornis michaeli Groves 1967) d iet in N airobi National Park, Kenya. African Journal of Ecology , 38 (1), 62 71. https://doi.org/10.1046/j.1365 - 2028.2000.0 0213.x Mylniczenko, N. D., Sullivan, K. E., Corcoran, M. E., Fleming, G. J., & Valdes, E. V. (2012). Management s trategies of i ron a ccumulation in a c aptive p opulation of b lack r hinoceroses ( Diceros bicornis minor ). Journal of Zoo and Wildlife Medicine , 43 (3), S83 S91. Retrieved from https://www.jstor.org/stable/41681916 National Research Council (U.S.). (1974). Nutrients and toxic substances in water for livestock and poultry: a report . (T. J. Cunha, J. P. Bowland, C. W. Deyoe , W. H. Hale, J. E. Halver, E. C. Naber, W. H. Pfander, Eds.). Washington, D.C.: National Academy of Sciences. Retrieved from http://hdl.handle.net/2027/uc1.$b504402 National Research Council (U.S.). (2005). Mineral tolerance of animals (2nd rev.). Washi ngton, 155 D.C.: National Academies Press. National Research Council (U.S.). (2007). Nutrient requirements of horses (6th rev.). Washington, D.C.: National Academies Press. Nelson, D. L., & Cox, M. M. (2013). Lehninger principles of biochemistry (6th ed.). New York, NY: W.H. Freeman and Company. Paglia, D. E., & Dennis, P. (1999). Role of c hronic i ron o verload in m ultiple d isorders of c aptive b lack r hinoceroses ( Diceros bicornis ). In Proceedings of the American Association of Zoo Veterinarians Annual Meeting (p p. 163 171). Pond, W. G., Church, D. C., Pond, K. R., & Schoknecht, P. A. (2005). Basic Animal Feeding and Nutrition (5th ed.). New York, NY: Wiley. Ross, A. C., Caballero, B., Cousins, R. J., Tucker, K. L., & Ziegler, T. R. (2014). Modern nutrition in hea lth and disease . (A. C. Ross, B. Caballero, R. J. Cousins, K. L. Tucker, & T. R. Ziegler, Eds.) (11th ed.). Philadelphia: Wolters Kluwer Health/Lippincott Williams & Wilkins. Shapiro, M. E., Shapiro, H. G., & Ehmke, E. E. (2018). Behavioral responses of th ree lemur species to different food enrichment devices. Zoo Biology , 37 (3), 146 155. https://doi.org/10.1002/zoo.21414 Shim, H., & Dierenfeld, E. S. (2017). Added dietary vegetables and fruits improved coat quality of capybara in Seoul Zoo, Republic of Kor ea: A case study. Zoo Biology , 36 (1), 50 55. https://doi.org/10.1002/zoo.21334 Skinner, J. D., & Chimimba, C. T. (2005). The Mammals of the Southern African Subregion (3rd Editio n ). Cape Town, South Africa: Cambridge University Press. https://doi.org/10.1192/bjp.111.479.1009 - a St. Clair, K. (2019). Nitrite r ule for t ransient w ater s ystems. Retrieved March 23, 2019, from https://www.epa.gov/sites/production/files/2018 - 05/documents/nitrate_and_nitrite_rule_training_presentation.pdf Stein, L. Y., & Klotz, M. G. (2016). The nitrogen cycle. Current Biology , 26 (3), R94 R98. https://doi.org/https://doi.org/10.1016/j.cub.2015.12.021 Taste and Smell. (2012). Retrieved March 27, 2019, from http://www.brainfacts.org/Thinking - Sensing - and - Behaving/Taste/2012/Taste - and - Smell Tomczak, M., & Tomczak, E. (2014). The need to report effect size estimates revisited. An overview of some recommended measures of effect size. TRENDS in Sport Sciences (Vol. 1). 156 Tricker, A. R., & Preussmann, R. (1991). Carcinogenic N - nitrosamines in the diet: occurrence, formation, mechanisms and carcinogenic potential. Mutation Research , 259 (3 4), 277 289. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/2017213 Tyagi, S., Singh, P., Sharma, B., & Singh, R. (2014). Assessment of w ater q uality for d rinking p urpose in d istrict Pauri of Uttarakhand, India. Applied Ecology and Environmental Sciences , 2 (4), 94 99. https://do i.org/10.12691/aees - 2 - 4 - 2 U.S. Centers for Disease Control and Prevention. (2014). Safe w ater s ystem: c hlorination. Retrieved March 27, 2019, from https://www.cdc.gov/safewater/chlorine - residual - testing.html U.S. Environmental Protection Agency. (2004). Drinking Water Health Advisory for Manganese . Washington, D.C. https://doi.org/EPA - 822 - R - 04 - 003 U.S. Environmental Protection Agency. (2015). Drinkin g Water Health Advisory for the Cyanobacterial Toxin Cylindrospermopsin . Washington, D.C. https://doi.org/doi:10.2514/6.1979 - 763 U.S. Environmental Protection Agency. (2017). Secondary d rinking w ater s tandards: g u idance for n uisance c hemicals. Retrieved Ma rch 27, 2019, from https://www.epa.gov/dwstandardsregulations/secondary - drinking - water - standards - guidance - nuisance - chemicals U.S. Environmental Protection Agency. (2018a). 2018 Edition of the Drinking Water Standards and Health Advisories Tables . Washingto n, D.C. https://doi.org/EPA 822 - S - 12 - 001 U.S. Environmental Protection Agency. (2018b). National Primary Drinking Water Regulations. U.S. Environmental Protection Agency. (2019a). EPA Drinking Water Health Advisories for Cyanotoxins. U.S. Environmental Pro tection Agency. (2019b). Learn about Cyanobacteria and Cyanotoxins. U.S. Geological Survey. (2016a). Water Hardness and Alkalinity. Retrieved March 27, 2019, from https://water.usgs.gov/owq/hardness - alkalinity.html U.S. Geological Survey. (2016b). Water Pr operties: Turbidity. Retrieved March 27, 2019, from https://www.usgs.gov/special - topic/water - science - school/science/turbidity - and - water?qt - science_center_objects=0#qt - science_center_objects Van Sant, S. (2018). Critically e ndangered b lack r hinos d ie a fter t ransportation t o Kenya National Park. Retrieved March 23, 2019, from https://www.npr.org/2018/07/13/628862085/critically - endangered - black - rhinos - die - after - transportation - to - kenya - national - pa 157 Vigil, K. M. (2003). Clean Water: An Introduction to Water Quality and Water Pollution Control (2nd ed.). Corvallis, Oregon: Oregon St ate University Press. Villar, D., Carson, T. L., Janke, B. H., Pallares, F. J., Frenandez, G., & Kinker, J. A. (2002). Retrospective study of chronic copper poisoning in sheep. An. Vet. (Murcia) , 18 , 53 60. Washington State Department of Health. (2019). Bl ue - Green Algae. Retrieved March 27, 2019, from https://www.doh.wa.gov/CommunityandEnvironment/Contaminants/BlueGreenAlgae Wilks, D. L., Coppock, C. E., Lanham, J. K., Brooks, K. N., Baker, C. C., Bryson, W. L., Elmore, R.G., & Stermer, R. A. (1990). Respon ses of l actating Holstein c ows to c hilled d rinking - w ater in h igh a mbient - t emperatures. Journal of Dairy Science , 73 (4), 1091 1099. Retrieved from https://doi.org/10.3168/jds.S0022 - 0302(90)78768 - 1 World Health Organization. (2003). Algae and c yanobacteria i n f resh w ater. In Guidelines for Safe Recreational Water Environments, Volume 1: Coastal and Fresh Waters (pp. 136 158). Geneva, Switzerland: World Health Organization. Retrieved from https://ebookcentral - proquest - com.proxy1.cl.msu.edu/lib/michstate - ebooks /detail.action?docID=284656 World Health Organization. (2011a). Manganese in Drinking - water . Geneva, Switzerland. World Health Organization. (2011b). Molybdenum in Drinking - water . Geneva, Switzerland. World Health Organization. (2017). Guidelines for Drinking - water Quality (4th ed.). Geneva, Switzerland: World Health Organization. Retrieved from https://apps.who.int/iris/bitstream/handle/10665/254637/9789241549950 - eng.pdf?sequence=1 World Wildlife Fund. (2019). Fact s heet: b lack r hino. Retrieved March 27, 2019, from https://www.worldwildlife.org/species/black - rhino Yisa, J., & Jimoh, T. (2010). Analytical s tudies on w ater q uality i ndex of River Landzu J. Yisa and T. Jimoh. American Journal of Applied Sci ences , 7 (4), 453 458.