Citation
Persistence of Ingidenous E. coli in Manure and Manure Amended Soils

Material Information

Title:
Persistence of Ingidenous E. coli in Manure and Manure Amended Soils
Creator:
Cekic, Samantha King
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (108 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Food Science and Human Nutrition
Committee Chair:
SCHNEIDER,KEITH R
Committee Co-Chair:
DANYLUK,MICHELLE D
Committee Members:
TEPLITSKI,MAXIM
Graduation Date:
8/8/2015

Subjects

Subjects / Keywords:
Core samples ( jstor )
Experimentation ( jstor )
Manure ( jstor )
Microcosms ( jstor )
P values ( jstor )
Pathogens ( jstor )
Plant nutrition ( jstor )
Population density ( jstor )
Seasons ( jstor )
Soils ( jstor )
Food Science and Human Nutrition -- Dissertations, Academic -- UF
ecoli -- manure -- pathogen -- persistence
City of Live Oak ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Food Science and Human Nutrition thesis, M.S.

Notes

Abstract:
Incidences of produce-related foodborne illness have increased for several decades. Good agricultural practices (GAPs) and the upcoming Food Safety Modernization Act (FSMA) attempt to stem this increase using preventative rather than reactive measures, including field management. Manure is one potential source for foodborne pathogens, but research regarding manure-borne pathogen persistence in the field is minimal. The goal of this research is to determine the persistence of manure-borne generic E. coli in controlled laboratory and field conditions. Manure was applied at varying rates to fields in North and Central Florida during the fall and summer seasons, while microcosms of the manure and soil were prepared and maintained at 30C and 20C. The initial manure E. coli population density was approximately 4.2 log10 CFU/g. The population density was not significantly different between application rates after incorporation into the soil on day 0, but differences were seen later in the trials. The persistence of E. coli, water content, and pH were monitored until consecutive samples were devoid of recoverable E. coli. While the summer season fields and 30C microcosms concluded with consecutive negative samples on days 112 and 210, respectively, the fall season fields and 20C microcosms persisted until the end of the experiment, on days 280 and 420. Several samplings during the fall trial suggest that contamination events are responsible for the appearance of increased persistence in the fields. The results of these experiments show that E. coli populations initially decay at similar rates across all manure application rates, seasons, and locations in this experiment, between 0.02 and 0.04 log10 CFU/day. Additionally, the risk of E. coli associated with new contamination events, such as wild life intrusion, run-off, or other vectors, may be greater than the risk associated with its long-term survival in the field from manure used for soil amendments. Lastly, the increased persistence in the microcosms suggests that survival under laboratory conditions do not mimic real-world survival and may not be adequate for predicting E. coli population persistence in field conditions. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (M.S.)--University of Florida, 2015.
Local:
Adviser: SCHNEIDER,KEITH R.
Local:
Co-adviser: DANYLUK,MICHELLE D.
Statement of Responsibility:
by Samantha King Cekic.

Record Information

Source Institution:
UFRGP
Rights Management:
Copyright Cekic, Samantha King. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Classification:
LD1780 2015 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

PERSISTENCE OF MANURE INDIGENOUS E. coli IN MANURE AND MANURE AMENDED SOIL By SAMANTHA KING CEKIC A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUI REMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2015

PAGE 2

© 2015 Samantha King Cekic

PAGE 3

To my family

PAGE 4

4 ACKNOWLEDGMENTS First and foremost I would like to thank my husband , Cagri . Without his persistent and unwavering support I would have not dared to take on the challenge of graduate school, and certainly would not have succeeded. I would also like to tha nk my family, who always pushed me above and beyond what I thought I was capable of. I would like to thank Dr. Keith Schneider for giving me the opportunity to study in his lab. Without his financial support I would not have been able to pursue my degree . His mentoring and guidance allowed me to thrive during my studies at the University of Florida, and I could not have asked for a better advisor. I also want to thank my committee members, Dr. Michelle Danyluk and Dr. Max Teplitski for their advice on a nd support o f this project. I should also thank the Food Science and Human Nutrition Department, who gave me the opportunity to work as a teaching assistant in the department, giving me valuable experience and much needed financial support. My colleagues in the Aquatic Food Products Lab were critical in the success of this project. I want to thank my labmates, Kristina Underthun, Scott Gereffi, and Alan Gutierrez, who were always willing to help me, our lab technicians, Susie Richardson and Rachael Silver berg, who were always available to prepare the media needed for this project, and our post docs, Ash Sreedharan and You Li, who were ready with advice when I was lost. I must also thank Shuang Wu and Amber Ginn for helping me with the molecular aspects of my research. Finally, I must thank the USDA Special Crops Research Initiative (Grant 2011 51181 30767), which provided funding for this research, and the FSHN extension services, an invaluable asset to the success of this project.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURE S ................................ ................................ ................................ .......... 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 2 LITERATURE REVIEW ................................ ................................ .......................... 14 An Introduction to Foodborne Illness ................................ ................................ ...... 14 Fresh Produce Consumption and Safety ................................ ................................ 15 Manure ................................ ................................ ................................ .................... 16 Manure borne Pathogens ................................ ................................ ................. 17 Pathog en persistence in manure ................................ ................................ ...... 18 Regulations ................................ ................................ ................................ ............. 21 Analysis of Current Research ................................ ................................ ................. 23 Research Hypothesis and Objectives ................................ ................................ ..... 24 3 MATERIALS AND METHODS ................................ ................................ ................ 26 Presence of Indigenous E. coli and Total Coliforms in Dairy Bovine Manure and Soil ................................ ................................ ................................ ...................... 26 Soil and Manure Nutrient Analyses ................................ ................................ ......... 27 Meteorological Monitoring and Action Levels ................................ .......................... 27 PCR and Gel Electrophoresis protocols ................................ ................................ .. 28 Soil and Manure Microcosm Preparations and Sampling ................................ ....... 28 Field Trial Preparations and Sampling ................................ ................................ .... 30 Statistical Analyses ................................ ................................ ................................ . 32 4 RESULTS ................................ ................................ ................................ ............... 36 Presence of Indigenous E. coli and Total Coliforms in Dairy Bovine Manure and Soil ................................ ................................ ................................ ...................... 36 Laboratory Condition Microcosm Studies ................................ ............................... 36 Microcosms Held at 30ºC ................................ ................................ ................. 37 Microcosms Held at 20ºC ................................ ................................ ................. 38 Field Trial Studies ................................ ................................ ................................ ... 39 Field Trial One, Summer Season ................................ ................................ ..... 41 Field Trial Two, Fall Season ................................ ................................ ............. 47

PAGE 6

6 5 DISCUSSION AND CONCLUSIONS ................................ ................................ ...... 80 Presence of Indigenous E. coli and Total Coliforms in Dairy Bovine Manure and Soil ................................ ................................ ................................ ...................... 80 Laboratory Condition Microcosm Studies ................................ ............................... 81 Microcosms Held at 30ºC ................................ ................................ ................. 81 Microcosms Held at 20ºC ................................ ................................ ................. 83 Field Trial Studies ................................ ................................ ................................ ... 84 Field Tr ial One, Summer Season ................................ ................................ ..... 87 Field Trial Two, Fall Season ................................ ................................ ............. 91 APPENDIX A PFGE DENDROGRAM ................................ ................................ ........................... 97 B CONTINUED WORKS ................................ ................................ .......................... 102 LIST OF RE FERENCES ................................ ................................ ............................. 103 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 108

PAGE 7

7 LIST OF TABLES Table page 3 1 Primers for PCR protocols. ................................ ................................ ................. 34 3 2 PCR condition s . ................................ ................................ ................................ .. 34 4 1 Average E. coli (log 10 CFU/g medium) population density recovered from three manure types varying by processing step and soil varying by location. .... 55 4 2 Soil texture analysis of Live Oak and Citra soils. ................................ ................ 55 4 3 Average E. coli population density (log 10 CFU/g soil) and enrichments recovered from samples taken from both Live Oak and Citra microcosms. ........ 56 4 4 Average air temperature parameters (ºC) in Citra, FL and Live Oak, FL field locations over the duration of Trials 1 and 2. ................................ ...................... 67 4 5 Average soil temperature parameters (ºC) in Citra, FL and Live Oak, FL field locations over the d uration of Trials 1 and 2 . ................................ ...................... 67 4 6 Average weather parameters in Citra, FL and Live Oak, FL field locations over the duration of Trials 1 and 2. ................................ ................................ ..... 67 4 7 Average E. coli population density (log 10 CFU/g soil) and enrichments recovered from surface soil samples over time, taken from the summer season field in Live Oak, FL. ................................ ................................ .............. 68 4 8 Average E. coli population density (log 10 CFU/g soil) and enrichments recovered from core soil samples over time, taken from the summer season field in Live Oak, FL. ................................ ................................ ........................... 68 4 9 Average E. coli population density (log 10 CFU/g soil) recovered from surface and core soil samples over time, taken from the summer season field, trial one, in Live Oak, FL. ................................ ................................ .......................... 69 4 10 Average E. coli population density (log 10 CFU/g soil) and enrichments recovered from surface soil samples over time, taken fr om the summer season field in Citra, FL. ................................ ................................ ..................... 70 4 11 Average E. coli population density (log 10 CFU/g soil) and enrichments recovered from core soil samples over time, taken from the summer season field in Citra, FL. ................................ ................................ ................................ . 71 4 12 Average E. coli population density (log 10 CFU/g soil) recovered from surface and core soil samples over time, taken from the summer season field, trial one, in Citra, FL. ................................ ................................ ................................ . 72

PAGE 8

8 4 13 Average E. coli population density (log 10 CFU/g soil) and enrichments recovered from surface soil samples over time, taken from the fall season field in Live Oak, FL. ................................ ................................ ........................... 73 4 14 Average E. coli population density (log 10 CFU/g soil) and enrichments recovered from core soil samples over time, taken from the fall season field in Live Oak, FL. ................................ ................................ ................................ .. 74 4 15 Average E. coli population density (log10 CFU/g soil) recovered from surface and core soil samples over time, taken from the fall season field, trial two, in Live Oak, FL. ................................ ................................ ................................ ...... 75 4 16 Average E. coli population density (log 10 CFU/g soil) and enrichments recovered from surface soil samples over time, taken from the fall season field in Citra, FL. ................................ ................................ ................................ . 76 4 17 Average E. coli population density (log 10 CFU/g soil) and enrichments recovered from core soil samples over time, taken from the fall season field in Citra, FL. ................................ ................................ ................................ ......... 77 4 18 Average E. coli population density (log10 CFU/g soil) recovered from surface and core soil samples over time, taken from the fall season field, trial two, Citra, FL ................................ ................................ ................................ .............. 78 B 1 Average E. coli population density (log10 CFU/g soil) recovered from surface a nd core soil samples over time, taken from trial two fields after conclusion of thesis research, in Live Oak, FL ................................ ................................ ....... 102 B 2 Average E. coli population density (log10 CFU/g soil) recovered from surface and core soil samples over time, taken from trial two fields after conclusion of thesis research, in Citra, FL ................................ ................................ .............. 102

PAGE 9

9 LIST OF FIGURES Figure page 3 1 Diagram of field layout. ................................ ................................ ....................... 34 3 2 A map of Florida indicating the Live Oak field site and the Citra field site. ......... 35 4 1 Moisture leve l of microcosms by location and temperature ............................... . 57 4 2 pH lev els of microcosms by location and temperature. ................................ ...... 58 4 3 Moisture level of plots by application rate in Live Oak during the summer se ason, trial one ................................ ................................ ................................ . 59 4 4 pH levels of plots by application rate and sample type in Live Oak during the summer season, trial one ................................ ................................ ................... 60 4 5 Moisture level of plots by application rate in Citra during the summer season, trial one. ................................ ................................ ................................ .............. 61 4 6 pH levels of plots by application rate and sample type in Citra during the summer season. ................................ ................................ ................................ . 62 4 7 Moisture level of plots by application rate in Live Oak during the fall season, trial two. ................................ ................................ ................................ .............. 63 4 8 pH levels of plots by application rate and sample type in Live Oak during the fall season, trial two. ................................ ................................ ........................... 64 4 9 Moisture level of plots by application rate in Citra during the fall season, trial two. ................................ ................................ ................................ ..................... 65 4 10 pH levels of plots by application rate and sample type in Citra during the fall season, trial two. ................................ ................................ ................................ . 66 4 11 PCR results of Citra Trial 2 day 84. ................................ ................................ .... 79 4 12 PCR results of Citra Trial 2 day 105. ................................ ................................ .. 79 A 1 PFGE dendrogram of field experiment isolates. ................................ ............... 101

PAGE 10

10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science PERSISTENCE OF MANURE INDIGENOUS E. coli IN MANURE AND MANURE AMENDED SOIL By Samantha King Cekic August 2015 Chair: Keith R. Schneider Major: Food Science and Human Nutrition Incidences of produce related foodborne illness have increased for several decades. Good agricultural practices (GAPs) and the upcoming Food Safety Modernization Act (FSMA) attempt to stem this increase using preventative rather than reactive measures, including field management. Manure is one potential source for foodborne pathogens, but research regarding manure borne pathogen persistence in the field is minimal. The goal of this research is to determine the persistence of manure borne generic E. coli in controlled laboratory and field conditions. Manure was applied at varying rates to fields in North and Central Florida durin g the fall and summer seasons, while microcosms of the manure and soil were prepared and maintained at 30ºC and 20ºC. The initial manure E. coli population density was approximately 4.2 log 10 CFU/g. The population density was not significantly different between application rates after incorporation into the soil on day 0, but differences were seen later in the trials. The persistence of E. coli , water content, and pH were monitored until consecutive samples were devoid of recoverable E. coli.

PAGE 11

11 While the summer season fields and 30ºC microcosms concluded with consecutive negative samples on days 112 and 210, respectively, the fall season fields and 20ºC microcosms persisted until the end of the experiment, on days 280 and 420. Several samplings during the fall trial suggest that contamination events are responsible for the appearance of increased persistence in the fields. The results of these experiments show that E. coli populations initially decay at similar rates across all manure applicati on rates, seasons, and locations in this experiment , between 0.02 and 0.04 log 10 CFU/day . Additionally, the risk of E. coli associated with new contamination events, such as wild life intrusion, run off, or other vectors, may be greater than the risk asso ciated with its long term survival in the field from manure used for soil amendments. Lastly, the increased persistence in the microcosms suggests that survival under laboratory conditions do not mimic real world survival and may not be adequate for predi cting E. coli population persistence in field conditions.

PAGE 12

12 CHAPTER 1 INTRODUCTION Foodborne illness is a serious and growing issue in the United States ( US ) . From 1998 to 2014 the C enters for Disease Control and Prevention (CDC) estimates of foodborne illness increased from 9.6 million incidences to 48 million, 57,462 hospitalizations to 128,000, and 1,451 deaths to 3,000, in the US ( 8 , 4 5 ). These illnesses result in an economic bur den of between $51 billion to $77.7 billion annually, increasing as the prevalence of foodborne illness increases ( 46 ). Similarly, the rate of foodborne outbreaks has also increased. This is especially true of fresh produce related outbreaks, which have increased from 0.7% of all foodborne outbreaks between 1973 1997 to 14.5% between 1998 2008 ( 44 , 49 ). Th e most recent data implicates l eafy vegetables, vine stalk vegetable s, root vegetables, and sprouts as the most common cause of produce related foodbor ne outbreaks ( 44 ) . The most common pathogens implicated in produce outbreaks include Shigella , E. coli, Salmonella spp., Listeria monocytogenes, Campylobacter spp . and others ( 3 ). Many reservoirs of foodborne pathogens exist from the farm to fork . One of these reservoirs for fresh produce is manure used as a soil amendment. In the US, 5% of all cropland was fertilized with manure in 2006 ( 37 ). There are many reasons a conventional farmer may utilize manure to fertilize crops, but organic farmers are res tricted to non synthetic fertilizers, which may increase their use of manure ( 54 ). There are a variety of pathogenic microorganisms that are native to manure, including Salmonella spp ., E. coli, Campylobacter spp . , and others ( 41 ) . There are many differe nt factors that may influence the persistence of these pathogens in manure and manure amended soils.

PAGE 13

13 Due to the presence of pathogens in manure, th e US Food and Drug Administration (FDA) strongly encouraged the use of passive or active treatments such as c omposting, pasteurization, and others, to limit the viability of these microorganisms before application to the soil. However, the use of raw manure was still permissible and practiced, but it was suggested that the manure be incorporated into the soil pr ior to the planting of crops ( 15 ). Still, there were no legal obligations to ensure these practices. The National Organic Program (NOP) however, require s a waiting period of 90 to 120 days between manure application and crop harvest, depending on the amount of contact between the edible portion of the plant and manure, for organic farmers ( 57 ). In 2011, the Food Safety Modernization Act (FSMA) proposed a Produce Safety Rule that initially proposed regulations for bot h conventional and organic farmers was a waiting period of nine months between application of manure and harvesting of the crop ( 15 ). Recently, however, due to a lack of scientific data, the FDA opted to apply the current NOP standards of 90 and 120 days until more research is conducted ( 17 ). The goal of this research was to determine the persistence of manure borne E. coli and total coliforms under temperature controlled laboratory conditions and field conditions in Central and North Florida. The curren t data regarding pathogen persistence in manure is largely derived from laboratory studies. The survival of manure borne E. coli under field conditions may suggest a waiting period that results in maximum pathogen extinction while being conducive to reali stic farming practices.

PAGE 14

14 CHAPTER 2 LITERATURE REVIEW An Introduction to Foodborne Illness Despite the increasing knowledge of food safety and processing technology, foo dborne illness continues to be concern in the US. According to data collected between 1998 and 2008, there are approximately 9.6 million incidences of foodborne illnesses annually. In 2011 the Centers for Disease Control and Prevention (CDC) estimate d there were 9.4 million incidence s, 55,916 hospitalizations and 1,351 deaths associated with the consumption of food ( 44 ). Painter and others ( 44 ) examined 57,462 hospitalizations and 1,451 deaths associated with known causes and found 46% to be associated with produce consumption. Howe ver, the estimates from both studies only consider the illnesses that occur from the 31 known foodborne pathogens and do not include illnesses from unknown agents, such as undiscovered microbes, chemicals, or toxins that have not been proven to cause foodb orne illness, or known agents that were not conclusively implicated ( 44 , 4 5 ) . As of 2014, the CDC estimated that one in six Americans, approximately 48 million people will become ill from a foodborne illness, resulting in 128,000 hospitalizations and 3,000 deaths ( 8 ). Estimating the occurrence of foodborne illnesses can be difficult since many incidences are not reported, and the ( 6 ). The incidences of foodborne illnes s are both a health related and economic concern. Foodborne illness is estimated to cost between $1,068 and $1,686 per individual from lost wages, medical expenses, and other costs. This translates to between $51 billion and $77.7 billion annually ( 46 ). Th ese costs do not include the impact to an industry when the foodborne illnesses result in a publicized outbreak.

PAGE 15

15 Foodborne illnesses constitute an outbreak when two or more people contract similar illnesses from the same food product. In 2009 there we re 675 foodborne outbreaks and 852 in 2010. Of these outbreaks, 38 were multi state. Bacteria were responsible for 52% of the outbreaks, followed by viruses at 42%, chemicals and toxins at 5%, and parasites at 0.2% ( 7 ). The top five pathogens responsibl e for foodborne illness incidences, hospitalizations, or deaths vary. Norovirus, Salmonella spp., Clostridium perfringens , Campylobacter spp. and Staphylococcus aureus are most often implicated in foodborne illness incidences, respectively. Most often th e pathogens Salmonella spp., Norovirus, Campylobacter spp., Toxoplasma gondii , and E. coli O157 are responsible for foodborne illness hospitalizations. Lastly, Salmonella spp., Toxoplasma gondii , Listeria monocytogenes , Norovirus, and Campylobacter spp. a re the most likely to result in death ( 44 ). Regardless of the pathogen, there are several factors that potentially play a role in the increase of foodborne illness. Large scale production and wide distribution of food, globalization of the food supply, a n increase in food consumption outside of the home, genetic changes in foodborne pathogens, and an increasing population of at risk consumers could all play a role in the trend ( 40 ). Fresh Produce Consumption and Safety Several programs have been initia ted on a federal level to raise awareness of the importance of a healthy diet and lifestyle, such as Choose MyPlate ( 58 ). As the public bec ame more concerned with health, the consumption of fresh fruits and vegetable s has increased . Between 1980 and 2009, fresh fruit consumption had increased by 20% and fresh vegetables by 22% ( 59 ). Also, the consumption of organic produce has increased by 7.7% in 2010, totaling to nearly 4% of all produce consumed ( 41 ). The prevalence of fresh produce related foodborne illness has also increased. From data

PAGE 16

16 collected between 1973 and 1997, produce implicated foodborne outbreaks had increased from 0.7% to 6% of all reported foodborne outbreaks. At that time, the most often implicated products were salads, lettuce, juice, melon, sprouts, and berries ( 49 ). This trend has continued, with data from 1998 2008 confirming that 14.5% of foodborne illness outbreaks were associated with produce. The most often implicated products at this time were leafy vegetables, vine stalk veg etables, root vegetables, and sprouts, respectively ( 44 ). Pathogens associated with raw produce include Shigella , enterotoxigenic and enterohemorrhagic E. coli, Salmonella spp., Listeria monocytogenes, Campylobacter spp . and others ( 4 ). Although there ar e several methods to reduce the populations of pathogens in fresh produce, there is no elimination method ( 27 ). This stresses the importance of hygienic growing conditions in the field. With the ratification of the FSMA, new rules to improve food safety were developed. One such rule is the P roduce S afety R ule, which attempts to improve the microbiological safety of produce in the field. Areas included in this rule are irrigation water quality, health and hygiene of employees, domestic and wild animal pr oximity, use of biological soil amendments, and standards for equipment, buildings, and tools ( 15 ). Manure Manure is animal excrement intended for use as an agricultural soil amendment. As of 2006, 5% of all cropland in the US was fertilized using man ure. The decision to livestock, regional specialization, commodity, geographical location, agronomic requirements, and size of the farm ( 37 ). In the case of organic a griculture, the use of manure is one of the n on synthetic fertilization methods approved for crops ( 57 ). With

PAGE 17

17 the increase of organic produce consumption, it may be suggested that organic farming, and therefore manure use, may also increase. From 2006 da ta analyzed by MacDonald and others ( 37 ), the most commonly used manures as agricultural soil amendments are dairy cattle, beef cattle, swine, and poultry, respectively. Excrement from other animals, such as horse and sheep, are also useable but are less frequently applied. Manure borne P athogens The most common manure borne pathogens include Salmonella spp ., E. coli, Campylobacter spp ., Yersinia enterocolitica, Listeria monocytogenes , Cryptosporidium parvum, and Giardia ( 42 ). The ability of pathogens in manure to contaminate produce in the field has been documented with varying results. Islam and others ( 33 ) demonstrated the ability of leaf lettuce and parsley to become contaminated with E. coli from inoculated manure, and persist for 77 and 177 days, r espectively. Furthermore, an experiment conducted by Solomon and others ( 52 ) showed that pathogens may be internalized by lettuce when grown in inoculated manure and irrigation water. Conversely , Johannessen and others ( 34 ) reported sporadic contaminatio n of lettuce when grown in inoculated manure. Ingham and others ( 31 ) reported contamination of carrots, radishes, and lettuce with indigenous E. coli when non composted manure was applied, persisting for a minimum of 69, 56, and 46 days, respectively. An imal borne isolates of E. coli O157 survive for a shorter period of time than human isolates under similar conditions in soil and manure, which stresses the importance of considering the origin of pathogens in a food system ( 18 ). While manure may be used i n its raw state, there are several treatment methods that may reduce the presence of pathogens. Manure treatments may be categorized into two methods, passive or active. Passive treatments utilize time and environmental

PAGE 18

18 factors such as temperature, UV ra diation, and moisture to kill indigenous bacteria in the manure. Active treatments include methods such as pasteurization, heat drying, alkali stabilization, anaerobic or aerobic digestion, or a combination of these. In the case of raw manure application , it is suggested that the manure be incorporated into the soil prior to the growing season. Those that use treated manure are also suggested to follow these protocols, along with additional practices to limit the possibility of recontamination or regrowt h of pathogens ( 15 ). With or without the use of treatments, pathogens may persist in manure. Pathogen persistence in manure There are many factors to consider when discussing pathogen persistence in manure. When manure is applied to the field, these fact ors form a complex relationship Many studies have been conducted in a controlled labora tory setting, but others have been conducted in the field. There are many types of soil which commonly vary based on geographical location. The types of soil are classified based on the ratio of sand:clay:silt. For instance, a 90% ratio or greater of s and and 10% ratio or less of clay constitutes sand, while a shift to 85% 70% sand and 10% 15% clay constitutes loamy sand. Minor changes in soil composition may change the ability of the soil to retain moisture and nutrients and subsequently the classific ation ( 47 ). The texture, nutrient content, and moisture of soil to which the manure is applied may favor specific pathogens. Cools and others ( 10 ) demonstrated that E. coli persistence was significantly greater in sandy soil than loamy sand at both 5 º C a nd

PAGE 19

19 25 º C when inoculated manure was applied. However, Enterococcus spp. persisted significantly longer in loamy sand than sandy soil at 25 º C. Furthermore, Ingham and others ( 31 ) determined that there was no significant difference in time from manure appli cation to the need for enrichment analysis for indigenous E. coli in loamy sand, silt loam, and silty clay loam soils. Nicholson and others ( 39 ) conducted an experiment that showed Salmonella spp., Listeria monocytogenes , Campylobacter jejuni , and E. coli O157 all persist longer in clay loam grassland soils than in sandy arable soils, though meteorological conditions may have influenced this field study. Franz and others ( 20 ) observed a difference in the initial decline of inoculated E. coli O157:H7 in sa ndy and loamy soil, but saw no difference in the persistence of the pathogen over time. The type of soil should be considered when studying any pathogen persistence in manure applied to the field, since a different preference for soil conditions may exist for various pathogens. The species of animal from which the manure is collected may impact the survival of pathogens. Also, indigenous pathogens may vary by animal. For instance, chickens do not naturally shed E. coli O157:H7 in their feces ( 9 , 14 , 28 ) . In the Nicholson and others ( 39 ) study, E. coli O 157, Salmonella spp., Campylobacter jejuni , and Listeria monocytogenes persisted for almost nine months when inoculated into dairy cattle manure and applied to land, as opposed to approximately one month or less in sheep, pig, beef cattle, and chicken manures. Hutchison and others ( 30 ) stated that the age of the animal is also a factor. They reported a higher prevalence of E. coli O157:H7 and Campylobacter jejuni in manure from calves, piglets, and lambs as opposed to mature animals. However, they found this only applied to young animals

PAGE 20

20 who were weaned. This study also showed a difference in the pathogens present in Further more , Franz and others ( 19 ) reported that high quality manure and fertilizer , such as pure manure and compost, resulted in a decreased persistence of E. coli O157:H7, as opposed to low quality sources like slurries or artificial fertilizers . These findings clearly show the importance of the anima l in the indigenous populations of bacteria and pathogen persistence in manure. While the source of manure can affect the persistence of pathogens, the use of composting methods and other treatments also play a role. Nicholson and others ( 39 ) indicated th at E. coli O 157, Salmonella spp., and Campylobacter jejuni survived for three months in stored slurries, while Listeria monocytogenes persisted for up to six months in the same conditions. Fremaux and others ( 21 ) reported similar findings for E. coli O26, a different strain of shiga toxin producing E. coli . Kudva and others ( 36 ) found that non aerated pig manure supported the persistence of E. coli O157:H7 for over one year, while the same manure, when aerated, supported persistence for only four months. From this research it is understood that the methods of composting, if any are used, impact the ability of pathogens to persist in manure. Competitive inhibition is a common theme in pathogen persistence that applies to many areas of study, including ma nure. Kim and others ( 35 ) observed the role of competitive inhibition in E. coli O157:H7 survival and regrowth in manure by autoclaving samples of manure and comparing the growth of the pathogen. Non autoclaved manure showed no growth, and autoclaved man ure saw a 4.4 5 .0 log CFU/g increase of the pathogen in the same conditions after inoculation. Also, the application rate of the manure and subsequent concentration of pathogens may factor into the ability of

PAGE 21

21 pathogens to persist. Zhai and others ( 61 ) an d Crane and others ( 12 ) observed no relationship between application rates of manure and survival of fecal coliforms in poultry manure. Gessel and others ( 23 ) tested the effect of the application rate of liquid pig manure on pathogen persistence. They found no correlation between application rate and S . Anatum , but suggested a potential positive correlation to enteric viruses. There are many field conditions that can impact the survival of pathogens. The season and subsequent temperatures and met eorological events are uncontrollable factors that can affect the pathogen persistence in manure. Sinton and others ( 48 ) observed that E. coli maintained populations of 10 3 to 10 4 CFU/g in spring, summer, and fall in fresh cow manure. However, S . enteric a only reached these concentrations in summer and fall. Campylobacter jejuni was in winter, reaching 10 2 CFU/g . Furthermore, this study showed that water content of the manure was the defining factor related t o growth of pathogens, and that rainfall events added moisture to desiccating manure. In the study conducted by Gessel and others ( 23 ), they cited an unusually fast decline of S. A natum in liquid pig manure. They hypothesized that the lack of rain during the study resulted in dry conditions that were not conducive to Salmonella survival. Cools and others ( 10 ) also tested the effect of temperature on pathogen persistence in various soils, and found that 5ºC resulted in greater persistence than 25 º C for E. coli and Enterococcus spp., regardless of soil type. Franz and others ( 20 ) reported no difference in the persistence of E. coli O157:H7 under conventional or organic field management practices . Regulations Currently the regulations regarding manure use as a soil amendment are limited. In 1998, the FDA released a guide entitled Guidance for Industry: Guide to Minimize

PAGE 22

22 Microbial Food Safety Hazards in Fresh Fruits and Vegetables ( 15 ) . This guide includes asp ects of hygienic and sanitary practices in the field, such as worker hygiene, equipment and facility sanitation, water quality, animal proximity to the field, and manure use as soil amendments. According to this document, farmers should maximize the time between manure application to the field and planting, but no specific amount of time is given. These practices are commonly referred to as good agricultural practices (GAPs). The National Organic Program (NOP), however, dictated a time frame for manure a pplication in 2000. As stated in the Code of Federal Regulations (CFR) , when using raw manure, the time between manure application and harvest of a plant that has edible portions in contact with the soil must be no fewer than 120 days. This timeframe is reduced to 90 days for plants that have edible portions that do not come into contact with the soil ( 57 ). At this time, GAPs were not a mandatory regulation and the rules dictated by the NOP only applied to farmers who were labeled organic. In 2011, FSMA changed the regulations of field conditions for fresh produce. The P roduce Safety R u le, included under FSMA, applied a time interval for all farmers intending to use raw manure on their fields. In the previously proposed rule § 112.56(a)(1)(i), farmers must wait for nine months between manure application to harvesting when the edible portion of the crop comes into contact with the soil. This rule would not be applied to crops that do not have an edible portion in contact with soil. For organic producer s, the NOP regulations would run concurrently to the FSMA regulations ( 16 ). Recently, however, the FDA opted to change this rule to follow the current NOP regulations until additional research in this area was made available ( 17 ) . The final Produce Safety Rule is scheduled for release on October 31 st , 2015.

PAGE 23

23 Analysis of Current Research As previously outlined, there are many complex factors to consider when reducing the microbiological hazards posed by the use of raw and composted manures. When proposing a time interval between manure application and harvesting, these factors must be considered. A study by Ingham and others ( 31 ) has challenged the NOP interval of 120 days. Their research suggests that a 100 day interval would not greatly increase the risk that the 120 day interval reduces for produce grown in Wisconsin. They further explained that the greatest reduction in pathogenic microorganisms noted in this experiment occur prior to this period. More rec ently, Pagadala and others ( 43 ) analyzed the p resence of generic and enterotoxigenic E . coli and Salmonella spp . on farms in the mid A tlantic region. Collecting samples of tomatoes, soil, compost, irrigation water, and other locations ensured a thorough observation of E. coli and Salmonella spp . in r eal farm practices. While the scientists were able to detect InvA, stx1 and stx2 genes, they were unable to recover isolates with these genes. Furthermore, a portion of all samples taken contained generic E. coli with exception to the compost samples. The relationships between factors that affect pathogen persistence in manure when applied to the field increase the need for further research. The various factors mentioned previously, such as the animal the manure is collected from, the season the manure is applied, the temperatures and meteorological conditions during the interval, the soil type the manure is applied to, indigenous microflora and competitive inhibition, and other factors may vary by geographical location. Further field research must be conducted in a variety of locations in order to ascertain a broad understanding of

PAGE 24

24 pathogen persistence in all agricultural locations. Future regulations should be guided based on these findings. Research Hypothesis and Objectives Understanding the persi stence of pathogenic microorganisms in manure and manure amended soil is a critical factor to preventing foodborne illness. This is especially true for raw produce, which has no kill step in processing. The many variables that impact pathogen persistence in field conditions are constantly in flux and are difficult to reproduce in laboratory conditions. There has not been sufficient research conducted in field settings to reliably state the survival of manure borne pathogens. The goal of this research was to determine the persistence of indigenous E. coli and total coliforms in dairy bovine manure applied to soil and held in both laboratory conditions and field conditions in north/central Florida. The laboratory study will examine the effects of temperatu re on persistence of E. coli and total coliforms in a controlled environment. The field study will give a realistic perspective on the effects of meteorological events, soil depth, solar radiation, temperature, soil type, and application rate on these coliforms in dairy bovine manure as they occur in n orth/ central Florida. While several of these variables are uncontrollable in field studies, they better represent the actual conditions in which pathogen persistence may occur in agriculture. It is hypothesized that E. coli will survive longer under fall weather conditions after heavy application of dairy bovine manure. Specifically, the objectives of this research were to:

PAGE 25

25 1. Determine indigenous populations of generic E. coli and total coliforms in dairy bovine manure and soils. Additionally, quantify the nutritive content of the manure and soils. 2. Determine the survival time of manure borne generic E. coli and total coliforms in dairy bovine manure and soil microcosms under heavy application rates in laboratory conditions at 30 º C and 20 º C. 3. Determine the su rvival time of manure borne generic E. coli and total coliforms in dairy bovine manure applied to fields in north/central Florida under light, medium, and heavy application rates during the s pring and f all seasons .

PAGE 26

26 CHAPTER 3 MATERIALS AND METHODS Presence of I ndigenous E. coli and Total C oliforms in Dairy Bovine Manure and S oil E. coli and total coliform populations were determined using standard plate count methods. Three random samples of three types of dairy bovine manure weighing approximately 1 kg were collected using a sanitized shovel from Shenandoah Dairy (Live Oak, FL) . Pre screened manure, post screened manure, and manure from a drying pile samples were collected. Similarly, three random soil samples were collected using a sanitized sho vel , weighing approximately 1 kg each, from the field s intended for the field trials at the North Florida Research and Education Center in Live Oak, FL and the Plant Science Research and Education Unit in Citra, FL bef ore manure was applied . Th ree samples from each site were composited. A 10 g sample was taken from the composite sample s and added to 90 mL of phosphate buffered saline ( PBS ) (Thermo Fisher Scientific Inc, Pittsburg, PA) in a f iltered Stomacher ® bag and macerated in a Smasher Lab Blender TM (Microbiology International, Frederick, MD) for 30 s. From the S tomacher ® bag, serial dilutions w ere performed using PBS and spread plated on ECC CHROMagar TM (CHROMagar Microbiology, Paris, France) in 1 mL and 0. 1 mL aliquots to obtain appropriate serial dilutions . ECC CHROMagar TM was incubated at 30 º C for 24 h . When the population of E. coli was below the limit of detection, ( i.e. , no growth on the plate ) samples were enriched. These samples were prepared by the addition of 10 g of soil in 90 ml lactose broth and macerated in a Smasher Lab Blender TM for 30 s . These bags were incubated at 37 º C for 24 h. After 24 h, all bags were tra nsferred to 9 mL of EC broth with Durham tubes in 1 mL aliquots and incubated at 42 º C for 24 h. After 24 h, all EC tubes were plated on L EMB agar to confirm E. coli , and the presence

PAGE 27

27 of gas was noted . The L EMB plates were incubated at 37 º C for 24 h and the growth of metallic green colonies was noted as a positive enrichment. All other growth was recorded as negativ e for E. coli . Soil and Manure Nutrient Analyse s In order to determine the nutrient content of the soil, soil samples were taken and submitted to the Extension Soil Testing Laboratory at the University of Florida for standard soil fertility tests. Soil was collected from the North Florida Research and Education Center in Live Oak, FL and the Plant Science Research and Education Unit remainder of this thesis (Fig ure 3 2) . The samples sent for analysis were taken from the aforementioned soil samples collected for indigenous E. coli testing in 50 g aliquots. In order to determine the nutrient content of the manure, manure samples were taken from the Shenandoah Dairy in Live Oak, FL. The previously mentioned manure samples were also submitted to the Extension Soil Testing Laboratory at the University of Florida for standard manure nutrient tests. Meteorological Monitoring and Action L evels Meteorological parameters were moni tored for each field location using the Florida Automated Weather Network (FAWN). The weather stations monitored soil temperature, air temperature, relative humidity, rainfall, average solar radiation, and wind speed and direction for the entirety of the field trials. Averages, minimums, and maximums for the previously mentioned parameters were documented in the event of rainfall, in addition to the date and day of the field trial. In the event of two inches of rainfall in a 48 h period, an additional sa mpling day was executed within seven days , barring extenuating circumstances or a scheduled sampling day within seven days .

PAGE 28

28 PCR and Gel Electrophoresis protocols Polymerase Chain Reaction ( PCR ) procedures w ere utilized throughout these experiments under varying circumstances, but the protocols remained constant. All PCR reactions were conducted using an Eppendorf Mastercycler Gradient thermal cycler (Eppendorf, Hauppauge, NY) . DNA samples were extracted usi ng a MO BIO PowerSoil extraction kit (MO BIO, Plainview, NY). The Master Mix was prepared using 25 uL AmpliTaq Gold ® Fast PCR Master Mix (Applied Biosystems, Foster City, CA), 21 uL distilled water, 1 uL eac h of forward and reverse primers ( Table 3 1 ) , and 2 uL of the DNA extract per reaction. Thes e volumes were pipetted into 96 well iCycler iQ ® PCR Plates (BIO RAD Laboratories Inc., Hercules, CA) . The PCR conditions were standa rd, with an annealing temperature of 53 º C (Table 3 2). After the PCR product was formed, samples were run on a 1% agarose gel, consisting of 100 mL of TAE Electrophoresis Buffer (Fermentas Life Sciences, Lithuania), 1 g agarose (Thermo Fisher Scientific In c . , Pittsburg, PA), and 2 uL ethidium bromide (Alfa Aesar, Ward Hill, MA) . Each gel was run with a 1 kb DNA ladder ( BIO RAD Laboratories Inc . , Hercules, CA), one positive control of E. coli K12, and one negative control of Salmonella spp. The first gel using these primers was also run with a water blank negative control . The gel s were run at 80 v for 45 min in an HE PLUS electrophoresis system (Hoefer Inc . , Holliston, MA) using TAE buffer and analyzed on a Gel Doc EZ imager (BIO RAD Laboratories Inc . , Hercules, CA). Soil and Manure Microcosm Preparations and Sampling In order to analyze the survival of E. coli and total coliforms under laboratory conditions, 12 microcosms of soil and manure were prepared. Six microcosms were prepa red using soil collected from Live Oak using a sanitized shovel, and six

PAGE 29

29 microcosms were prepared using soil collected from Citra similarly. Three microcosms from each location were stored in an incubator at 20 º C, the remaining three were stored at 30 º C. Samples were collected on days 0, 3, 5, 7, 10, 15, 30, and every 30 days thereafter, until extinction . To prepare the microcosms, manure was added to the soil in 250 g manure: 600 g of soil ratio, imitating a heavy application rate of 2 24.2 kg N/ ha in a space of 6.5 cm at 17.8 cm depth. This rate mimics the depth of incorporation that the rototillers used in the field trials would provide. The soil and manure was added to 25.4 cm x 35.6 cm sterile sampling bags (Thermo Fisher Scientific, Pittsburg, PA) and manually mixed for 1.5 min. During each sampling event, 10 g of mixture was weighed from the microcosms and diluted in 90 mL PBS and stomached for 30 s, as previously described. Serial d ilutions were prepared using 1 mL soil and PBS mixture into 9 mL of PBS, and 1 mL and 0.1 mL aliquots were pipetted on ECC CHROMagar TM plates . The plates were incubated at 30 º C for 24 h and subsequently counted. As the plate counts decreased, the dilutions decreased respectively. Once E. coli was no longer countab le, the samples were enriched at th e subsequent sampling times. Ten g of mixture was diluted in 90 mL lactose broth and stomached for 30 s. After incubation at 37 º C for 24 h, a 1 mL aliquot was transferred to 9 mL of EC broth with Durham tubes and incuba ted for 24 h at 42 º C. The presence of gas and/or cloudiness was noted. The broth was then streaked onto L EMB and incubated at 37 º C for 24 h and the growth of metallic green colonies was considered a positive enrichment; any other growth was considered a negative enrichment. After all microcosms held at the same temperature enriched

PAGE 30

30 negative in two consecutive sampling events, DNA was extracted from the microcosms using the M O BIO PowerSoil extraction kit . The subsequent DNA samples were analyzed using PCR and gel electrophoresis protocols described previously. The absence of bands after gel electrophoresis indicated a true negative sample, at which time the trial was considered completed . When all enrichments from the same temperature were negative in two consecutive sampling events and the PCR was also negative, the trial was considered complete. Field Trial Preparations and Sampling Fields were prepared in Live Oak and Citra during two seasons. The sp ring trial was prepared in May 2014 at both locations and consisted of one 15 .2 m x 45.7 m block per location. The summer/fall trial was prepared in August 201 4 at both locations with one block per location, identical to those of the sprin g trial. T hese block s consisted of four rows of 3 .1 m x 9 .1 m plots in three replications, totaling 12 plots with 3 .1 m walking spaces between them (Fig ure 3 1) . The locations of the fields were within North and Central Florida, respectively (Fig ure 3 2) . Each plot was delineated with wooden stakes and twine. Post screening manure was collected from the dairy farm previously mentioned and applied to the soil at three rates, defined as heavy, medium and light. The application rates were identical in both locations, during both seasons, and were determined by the nutrient content of the manure and soil. The agronomic rate of application for tomato crops, henceforth termed as medium, was 1 12 .1 kg N/ ha . Light and heavy applications were 56.0 kg N/ ha and 224 .2 kg N/ ha , respectively. This translated to 50 .4 Mg manure/ ha , 100.9 Mg manure/ ha , and 201.8 Mg manure/ ha for light, medium, and heavy applications, respectively. The manure application methods used varied by location. In Live Oak , the manure was

PAGE 31

31 a pplied by filling and weighing 18.9 L buckets with manure and spreading it uniformly across each plot via dumping of the buckets and raking until the proper application rate was achieved. A 1.5 m Maletti rototiller (Kverneland Group Modena S.p.A, Modena, Italy) was then used to amend the manure into the soil at a depth of approximately 1 8 .0 cm , moving from light application to medium and lastly heavy, in order to minimize contamination between plots. In Citra , the manure was weighed in five gallon buckets , similarly to Live Oak, but was then dumped into a John Deere front loader model no. 426 (Deere & Co., Moline, IL) to spread into each plot. The manure was spread beginning with the light application rate, followed by medium and heavy to prevent contamin ation. The manure was further spread uniformly throughout each plot using rakes. A 1.8 m Maletti rototiller (Kverneland Group Modena S.p.A, Modena, Italy) amended the manure into the soil at a depth of approximately 1 8 .0 cm , moving from the light applica tion plots into the medium, then lastly heavy, to prevent contamination. These blocks were treated with Round Up ® ( Monsanto, St. Louis, MO ) herbicide as needed throughout both trials, in both locations. Once the manure was applied the fields were sampled. Sampling occurred on days 0, 1, 3, 7, 14, 28, 56, 84, and ever y four weeks thereafter until all samples taken were negative after enrichments and confirmed negative via PCR for two consecutive sampling periods. Additional sampli ng events were scheduled within seven days of a heavy rainfall event ( 5 .1 cm or more in 48 h), barring extenuating circumstances. Soil samples were collected from each plot and consisted of three surface samples and three 25 .4 cm core samples collected at random and compiled respective of plot and soil location (i.e. , surface or core). Surface samples were collected using s terile

PAGE 32

32 s pecimen c ontainers (Thermo Fisher Scientific Inc . , Pittsburg, PA) that were dragged across the surface of the plot for approxi mately 10 .2 cm and deposited in a labeled 1 7.8 cm x 3 0.5 cm sampling bag (Thermo Fisher Scientific Inc . , Pittsburg, PA). One specimen container was used per plot. Core samples were collected using a Model HC 4 5.7 cm Soil Sampler (Spectrum Technologies In c., Aurora, IL ), driven 25.4 cm into the soil and deposited in a labeled 17.8 cm x 30.5 cm sampling bag using a sterile tongue depressor. The soil sampler was sanitized between each plot by first rinsing in a 1 8.9 L bucket of water, then placed in a 18.9 L bucket of diluted peroxyacetic acid (PAA) at a concentration of 80 90 ppm, as determined by a Peracetic Acid Kit (LaMotte Company, Chestertown, MA), and lastly dried using paper towels. Gloves were refreshed between application rates and booties were re freshed between each plot. At the time of sampling the moisture of each plot was determined using a TDR 100 soil moisture meter (Spectrum Technologies, Pomona Park, FL) . The moisture was averaged from three readings taken at random within each plot. All sampling procedures were conducted beginning with control plots one, two, and three, respectively, followed by application rates light, medium, and lastly heavily, similarly, to prevent cross contamination. Samples were transported to the laboratory and processed within 24 h. Processing procedures were identical to those outlined previously, including plating methods, enrichment methods, and PCR technique. Once all samples from one field location were enri ched negatively for two consecutive sampling periods and confirmed negative with PCR, that field trial was considered complete. Statistical Analyses The statistical analysis software SAS 9.4 (SAS Institute Inc., Cary, NC) was utilized to analyze the data collected from the aforementioned experiments. Tukey -

PAGE 33

33 Kramer least square means, Tukey least square means, multivariate ANOVA, student T tests, and Tukey Kramer multiple comparisons methods were used in the program in order to determine the significance of E. coli population persistence across countable and uncountable variables. The initial rates of decay for the microcosms were determined by plotting the best fit linear model of the first 90 days of the experiment, due to the increased significance of th at time frame. The initial rates of decay for the field trials were determined by plotting the best fit linear model of the first 112 days of the experiment, due to this time frame being the common time frame between all trials and locations. For compari son of the microcosm initial decay rates to the field initial decay rates, the time frame plotted for the field trials was changed to day 84, in order to minimize the difference between microcosm and field trial time points for a more accurate comparison. Analysis of covariance (ANCOVA) was utilized to determine significance between initial death rates. The level of confidence for all statistical analyses in this project is =0.0 5.

PAGE 34

34 Table 3 1. Primers ( 38 ) utilized for PCR protocols for presence/absence test of E. coli in soil and manure samples (Integrated DNA Technologies, Coralville, IA). Primer Tm Band mw (kDa) UAL1939b ATG GAA TTT CGC CGA TTT TGC 54.7 º C 140 UAL2105b ATT GTT TGC CTC CCT GCT GC 58.8 º C Table 3 2. PCR conditions utilized for all PCR conducted over the course of this project . Temperature Time Cycles 95 º C 1 m 1 95 º C 15 s 30 53 º C 15 s 72 º C 30 s 72 º C 7 m 1 Heavy ( 3.1m x 9.1m ) 3.1 m Medium 3.1 m Light 3.1 m Control Walking space ( 3.1 m x 45.7 m ) Heavy Medium Light Control Walking space ( 3.1 m X 45.7 m ) Heavy Medium Light Control Figure 3 1. Diagram of field layout with manure application rates, not to scale.

PAGE 35

35 Figure 3 2. A map of Florida indicating the Live Oak field site ( 9 ) and the Citra field site ( 12 ) ( 11 ) .

PAGE 36

36 CHAPTER 4 RESULTS Presence of I ndigenous E. coli and Total C oliforms in Dairy Bovine Manure and S oil The presence and concentration of indigenous E. coli (log 10 CFU/g soil) and total coliforms in dairy bovine manure and the soil of both field locations were determined. Triplicate samples (n=3) of three varieties of dairy bovine manure were composited, respectively, and plated to determine the population density of indigenous E. coli (log 10 CFU/g soil) in those manures. The manure types varied by processing step, i.e. , before screening, after screening, and drying pile manures from the same dairy operation. Manure collected before screening had the highest densit y of E. coli , 6.20 log 10 CFU/g manure, while the drying pile manure had the lowest density, 2.40 log 10 CFU/g manure . The soil of both locations, however, was devoid of indigenous E. coli . The manure collected after screening, containing an E. coli popula tion density of approximately 4.41 log 10 CFU/g, was utilized for the following experiments due to feasibility of collection and application, and microbial load (Table 4 1 ). Additionally, a soil texture analysis was performed to determine the ratio of sand , silt, and clay in the Live Oak and Citra soils (Table 4 2). There was no significant difference between soil locations. Laboratory Condition Microcosm Studies The microcosm experiments allowed for a more controlled experiment than the field , limiting t he number of extrinsic factors . The sampling day was the most significant factor contributing to E. coli population density, followed by the combined effects of temperature and day (P value <0.0001). The effect of temperature was nearly as significant to that of the combined effects of temperature and day (P value = 0.0015). The location of the soil collected for the microcosms was significant on one occasion , as

PAGE 37

37 was the effect of temperature and location (P value <0.05). However, neither the combined ef fect of temperature and location nor the combined effect of temperature, location, and day were significant (P value 0.05). At the beginning of both temperature trials, there was no significant difference in E. coli population density at day 0, although significant differences were observed later in the trial (Table 4 3). Microcosms H eld at 30 º C All microcosms held at 30 º C were completed on day 240 (Table 4 3 ). The microcosms held at 30 º C had statistically similar population densities of E. coli as thos e held at 20 º C, initially. However, after day 1 5 the 30 º C microcosms consisting of Live Oak soil were no longer comparable to their 20 º C counterparts. This discrepancy was seen in the Citra soil microcosms a t day 30 . At these sampling days the 30 º C microcosms contained E. coli population densities similar to those recovered several months later in the 20 º C microcosms. With locational differences excluded, the 30 º C microcosms decreased in E. coli population density significantly throughout the cour se of the experiment (P value <0.05). When considering soil location, Citra soil microcosms initially increased significantly between days zero and three, followed by a plateau on days three, five, and seven. The initial population density decrease , from day zero through 90 , occurred at a rate of 0.0 5 log 10 CFU / day ( r 2 =0. 88 ) , followed by an insignificant decline between days 90 and 180, at which time the E. coli was recoverable solely by enrichment . Days 210 and 240 produced consecutive negative results , indicating the extinction of E. coli in the microcosms. The Live Oak soil microcosms saw a similar trend, increasing significantly from days zero to thre e, followed by a plateau from day three to seven. Day 10 decreased significantly from day seven and the decrease in E. coli population density continued significantly until day 9 0 .

PAGE 38

38 The initial rate of decay, from day zero to day 90, occurred at a rate of 0. 05 log 10 CFU / day ( r 2 =0. 87 ) . A fter which E. col i were recoverable only by enrichment until all samples were negative for E. coli at day 210. Days 210 and 240 were both negative for E. coli , indicating the completion of the microcosms. Upon conclusion of the experiment, it was determined that the Live Oak microcosms had a significantly higher E. coli population density, on average, than the Citra microcosms at 30 º C. Moisture and pH did not correlate with E. coli population density (Fig ure 4 1 , 4 2 ) . Microcosms H eld at 20 º C No microcosms held at 20 º C were completed, the experiment concluding on day 420. These microcosms had a significantly higher E. coli population density , on average, than that of the 30 º C microcosms (P value <0.05). As previously described, the 20 º C Live Oak and Citra soil micro cosms were, on average, higher in E. coli population density than their 30 º C counterparts after days 10 and 15, respectively. With locational differences excluded, the 20 º C microcosms decreased in E. coli population density significantly from day zero to day 270, after which the decrease was not significant (P value <0.05). When considering soil location, Live Oak soil microcosms increased significantly from day zero to day three, followed by a plateau on day five. The decrease between days five and seven was not significant, although the decrease at day seven was signific ant compared to day three. This step wise decline continued similarly until day 90 at a rate of 0. 03 log 10 CFU/ day ( r 2 =0. 53 ) , after which the E. coli population increased significantly from day 90 to day 120. An insignificant decrease occurred from day 1 20 to 180, and day 180 to 420 . Days 210 through 420 were dominated by enrichment procedures , the latter of which resulted in all negative

PAGE 39

39 samples . The Citra soil microcosms saw a similar trend, increasing significantly from day zero to day three. A plat eau occurred between days three through seven , followed by a decrease on day 10 that was significant to day five , but not to day seven . T he initial decrease in E. coli population density occurred from day zero to day 90 at a rate of 0.03 log 10 CFU/ day ( r 2 =0. 80 ) , continuing until day 120, after w hich an insignificant increase in E. coli population density occurred between day 120 and 150 . From day 150 to 270, significant decreases occurred every two to three sampling periods, at which time E. coli was only recoverable by enrichment until day 420 , when all samples were negative for E. coli (Table 4 3) . Upon conclusion of the experiment, it was determined that the Live Oak and Citra microcosms held at 20 º C were not significantly different in average E. coli population density, and similar levels of E. coli as the Citra microcosms at 30 º C. Field Trial Studies The field trial studies allowed for a realistic view of E. coli population density persistence, and many variables were monitored. The most significant factor affecting E. coli population density was sampling day followed by the sample type and application rate , which were similarly significant (P value <0.0001). Wh en averaged across locations and trials, the heavy application rate had the highest E. coli population density, significantly, followed by medium and light rates, respectively , as was expected . The surface soil samples were significantly higher in E. coli population density than their core counterparts. The combined effects of trial by day, location by day, and trial by location by day were all similarly significant (P value <0.0001). Trial and location alone were also significant contributors to E. coli population density (P value <0.05), as was their combined effects, the latter of which was more significant (P value = 0.0017). T he

PAGE 40

40 combined effects of treatment and type were not significant, nor were: trial and treatment, location and treatment, trial a nd type, location and type, or any other combination not yet mentioned (P value 0.05). Neither pH nor moisture level correlated significantly with E. coli population density (Figures 4 3, 4 4, 4 5, 4 6, 4 7, 4 8, 4 9, 4 10) . There was a significant diff erence in multiple weather and temperature parameters between the seasonal trials. The average air temperature ( º C), including average minimum and maximum temperatures, were significantly different between fall and summer seasons by location (P value <0.0 5) (Table 4 4 ) . Additionally, the average soil temperature ( º C), including average minimum and maximum temperatures, were also significantly different by location (P value <0.05) ( T able 4 5 ) . Upon monitoring the relative humidity (pct), average total rainfall (cm), and solar radiation ( w/m^2 ) during both trials, only solar radiation was significantly different by location (P value <0.05) (Table 4 6 ) . In order to compare the rate of decay between seasons and trials, the rate for each location was determ ined using the greatest common sampling day as the final plot point, day 112. During trial one, in Live Oak, the rates of decay for E. coli population density in surface samples were 0.02 log 10 CFU/ day ( r 2 =0. 53 ) , 0.02 log 10 CFU/ day ( r 2 =0. 68 ) , and 0.03 log 10 CFU/ day ( r 2 =0. 65 ) for light, medium, and heavy application rates, respectively. The rates of decay for the corresponding core samples were 0.02 log 10 CFU/ day ( r 2 =0. 54 ) , 0.02 log 10 CFU/ day ( r 2 =0. 69 ) , 0.03 log 10 CFU/ day ( r 2 =0. 66 ) , respectively. During trial one, in Citra, the rates of decay for E. coli population density in surface samples were 0.03 log 10 CFU/ day ( r 2 =0. 74 ) , 0.03 log 10 CFU/ day ( r 2 =0. 84 ) , and

PAGE 41

41 0.04 log 10 CFU/ day ( r 2 =0. 79 ) for light, medium, and heavy application rat es, respectively. The rates of decay for the corresponding core samples were 0.03 log 10 CFU/ day ( r 2 =0. 53 ) , 0.03 log 10 CFU/ day ( r 2 =0. 62 ) , and 0.03 log 10 CFU/ day ( r 2 =0. 55 ) , respectively. During trial two, in Live Oak, the rates of decay for E. coli population density in surface samples were 0.04 log 10 CFU/ day ( r 2 =0. 68 ) , 0.04 log 10 CFU/ day ( r 2 =0. 77 ) , and 0.04 log 10 CFU/ day ( r 2 =0. 68 ) for light, medium, and heavy application rates, respectively. The rates of decay for the corresponding core samples we re 0.03 log 10 CFU/ day ( r 2 =0. 71 ) , 0.0 4 log 10 CFU/ day ( r 2 =0. 74 ) , and 0.0 4 log 10 CFU/ day ( r 2 =0. 75 ) , respectively. During trial two, in Citra, the rates of decay for E. coli population density in surface samples were 0.0 4 log 10 CFU/ day ( r 2 =0. 7 1 ) , 0.03 log 10 CFU/ day ( r 2 =0. 77 ) , and 0.04 log 10 CFU/ day ( r 2 =0. 7 0 ) for light, medium, and heavy application rates, respectively. The rates of decay for the corresponding core samples were 0.03 log 10 CFU/ day ( r 2 =0. 67 ) , 0.03 log 10 CFU/ day ( r 2 =0. 78 ) , 0.04 log 10 CFU/ day ( r 2 =0. 73 ) , respectively. Field T rial O ne, S ummer S eason Observing only the effect of trial season, the decrease in E. coli population density during the summer season was significant from day zero to day 28, after which the decrease was not significant until the conclusion of the trial on days 127 and 140. Within trial one, there was no significant difference between the ave rage E. coli population density of the Live Oak field trial and the Citra field trial. Over the course of trial one ( summer season ) several temperature parameters were significantly different between the Live Oak and Citra field locations. The average min imum air temperature (ºC) was statistically different between locations (P value

PAGE 42

42 <0.05) , 20. 7 ºC in Live Oak and 18. 8 ºC in Citra, although average temperature and average maximum temperatures were not (Table 4 4 ). Similarly, the average minimum soil temp erature (ºC) was statistically different between locations (P value <0.05), 27.1 ºC in Live Oak and 24.3 ºC in Citra, while the average temperature and average maximum soil temperatures were not (Table 4 5 ). Between Live Oak and Citra during trial one , there were no significant differences in relative humidity (pct), total rainfall (cm), and solar radiation (w/m^2) (Table 4 6 ). The light application rate surface samples in Live Oak decreased insignificantly from day zero to one, significantly from day one to three, insignificantly from days three to seven, significantly from day seven to 14, and this pattern continues until day 56, w hen all light application rate surface samples wer e negative for E. co li . The medium application rate surface samples follow a similar progression, decreasing insignificantly from day zero to day three and significantly from day three to seven. On day 14, the samples increased significantly in E. coli p opulation density from day seven, containing a statistically similar level of E. coli as day three. Following this spike was a significant decrease on day 28, followed by negative enrichments from day 56 to 127 . The heavy surface samples differed from th e others in that they increased, although insignificantly, from day zero to day one, followed by an insignificant decrease on day three. A significant decrease occurred from day 14 to day 56, which was characterized by negative enrichments that continued to day 127 . When observing all surface samples across application rates, they were not statistically different from each other on days zero and one. However, on day three, heavy surface samples were significantly higher in E. coli than the light and medi um surface samples, which were not statistically

PAGE 43

43 different. This trend continued on day seven, but changed on day fourteen, where the medium and heavy application rates were significantly higher than the light. Through days 28 to 127 all application rate s were not significantly different (Table 4 7 ). The rate of decay for E. coli population density from day zero until the earliest common day of extinction, day 84 , was 0.03 log 10 CFU / day ( r 2 =0.58) , 0.0 3 log 10 CFU/ day ( r 2 =0. 73 ) , and 0.0 4 log 10 CFU per day ( r 2 =0. 68 ) , for light, medium, and heavy surface samples, respectively. The light core samples in Live Oak decrease d insignificantly from day zero to day three, dec reasing significantly from day three to seven , and then maintaining an insignifi cant decrease until day 56, after w hich all samples were negative. The medium c ore samples were similarly not significantly different from days zero to three , despite a small increase on day one. A significant decrease occurred from day three to day seve n, after which there was no significant change until day 28, which decreased significantly from day 14 and was characterized by enrichments. O n day 84 all samples were negatively enriched until day 127. The heavy core samples did not follow the same tren ds as the other application rates, in that there was a significant increase from day zero to day one, after which there was a significant decrease on day three, bringing the E. coli population density back to day zero levels. A significant decrease occurr ed from day 14 to 28, bringing the population density back to day seven levels. There was an insignificant decrease from day 28 to day 56, after which all samples were enriched negative until day 127. When observing all core samples across application ra tes, there was no significant difference in E. coli population density on day zero. However, on day one, the heavy application rate was significantly higher than the others. On day three

PAGE 44

44 all application rates were statistically similar. On day seven the medium application rate was significantly higher than the light. On day 14 the heavy application rate was significantly higher than the light application rate. From day 28 to day 127 there was no significant difference between application rate core samp les (Table 4 8 ) . The rate of decay for E. coli population density from day zero until the earliest common day of extinction, day 84 , was 0.0 3 log 10 CFU / day ( r 2 =0.5 5 ) , 0.03 log 10 CFU/ day ( r 2 =0. 7 3 ) , and 0.0 4 log 10 CFU per day ( r 2 =0. 68 ) , for light, medium, and heavy core samples, respectively. During the summer season in Live Oak, when averaging sample type, the heavy application rate plots were significantly higher in E. coli population density than the other application rates (P value <0.05). When averagi ng application rates, surface samples were significantly higher in E. coli population density than the core samples at this location (P value <0.05). When sample types were averaged, there was a very significant difference in E. coli population density from day zero to day 28 in all application rates (P value <0.0001). On days seven, 14, and 112 one control plot was positive for E. coli (Table 4 9 ). Upon statistical analysis of all samples, there was no significant difference betwee n positively or negatively enriched samples. The light surface samples in Citra increased significantly from day zero to day one, followed by a significant decrease on day three that brought the E. coli population density back to day zero levels. There wa s a significant decrease on day 28, bringing the population density back to day seven levels. Another significant decrease occurred on day 56, which was characterized by enrichments, and was followed by negative enrichments on day 72 until day 140. The m edium surface samples trended similarly,

PAGE 45

45 having a significant increase from day zero to day one. From day seven until day 28 the decrease in E. coli population density was insignificant , with d ay 56 characterized by negative enrichments which continued un til day 140 . The heavy surface samples also underwent a significant increase from day zero to day one. The following decrease s were not significant until day 28. On day 56 there was a significant decrease from day 28, characterized by enrichments that c ontinued until day 84, after which all heavy surface sa mples were negative (Table 4 10 ) . When observing all surface samples across application rates, there was no significant difference on day zero. On day one, the heavy application E. coli population de nsity was significantly higher than the medium rate. On day three, the heavy application rate was significantly higher than the light rate. On day seven, the light application rate was lower than both the medium and heavy rates, which continued into day 14. On day 28, the medium rate was significantly higher than the others. From day 56 to day 140 there was no difference in E. coli population density between application rates (Table 4 9 ) . The rate of decay for E. coli population density from day zero u ntil the earliest common day of extinction, day 112, was 0.03 log 10 CFU / day ( r 2 =0.74 ) , 0.03 log 10 CFU/ day ( r 2 =0. 84 ) , and 0.0 4 log 10 CFU per day ( r 2 =0. 79 ) , for light, medium, and heavy surface samples, respectively. The light core samples in Citra increased significantly from day zero to day one, followed by a significant decrease on day three. The significant decrease continued into day seven, at which time the E. coli population density was similar to that on day zero. Following this decrease, an insignificant increase occurred on day 14. A significant decrease on day 56 brought the samples to negative enrichments. Day 7 2 resulted in positive enrichments . From day 84 until day 140 all enrichments were negative . The

PAGE 46

46 medium core samples also inc reased significantly from day zero to day one, which was also followed by a significant decrease on day three. A significant decrease on day 28 brought the population density down to day seven levels. A significant decrease occurred on day 56, which was characterized by negative enrichments that continued until day 140. The heavy core samples also increased significantly from day zero to day one, however, the decrease on day three was insignificant. Day seven underwent a significant decrease from day th ree . An insignificant decreased continued through days 14 and 56, with day 56 characterized by negative enrichments that continued until day 140 . When observing core samples across all application rates, both the light and medium rates were significantly higher in E. coli population density than the heavy rate on day zero. On day one, however, all application rates were similar. On day three the heavy rate was significantly higher than the light rate, while on day seven both the medium and heavy rates w ere significantly higher than the light. This trend continued on day 14. On day 28 the medium rate was significantly higher than both light and heavy application rates. From day 56 until day 140 there was no significant difference between application ra tes ( T able 4 1 1 ). The rate of decay for E. coli population density from day zero until the earliest common day of extinction, day 112, was 0.03 log 10 CFU / day ( r 2 =0. 53 ) , 0.03 log 10 CFU/ day ( r 2 =0. 62 ) , and 0.0 3 log 10 CFU per day ( r 2 =0. 55 ) , for light, medium, and heavy core samples, respectively . During the summer season in Citra, when averaging sample type, the light application rate samples were significantly lower in E. coli population density than the other application rates (P value <0 .05). When averaging all application rates, the surface samples were significantly higher in E. coli population density than the core

PAGE 47

47 samples (P value <0.05). When sample types were averaged, there was a very significant difference in E. coli population density across all application types from days zero to 28 (P value <0.0001). The control plots were not positive for E. coli at any time during this trial. Upon statistical analysis, there was no significant difference between samples that were positivel y or negatively enriched (Table 4 12). Field T rial T wo, F all S eason Observing only the effect of trial season, the decrease in E. coli population density during the fall season was significant from day zero to day 55. After which there was a similar step wise decrease as seen in the 20 º C microcosms, having significant decreases every other sampling period until day 140. After day 140 s ignificant decreases in E. coli population density were only observed every two sampling periods until the experiment concluded without E. coli extinction on day 280. Furthermore, the E. coli population density was greater, on average, in trial two than t rial one (P value <0.05). During the course of trial two, the average E. coli population density was significantly greater in the Live Oak field trial than the Citra field trial. Over the course of trial two , fall season, several temperature parameters w ere significantly different between the Live Oak and Citra field locations. The average air temperature (ºC) was statistically different between locations (P value <0.05) , although average minimum and maximum temperatures were not (Table 4 4 ). Similarly, the average soil temperature (ºC) was statistically different between locations (P value <0.05), while the average maximum and minimum soil temperatures were not (Table 4 5 ). Between Live Oak and Citra during trial two , there were no significant differen ces in relative humidity (pct), total rainfall (cm), and solar radiation (w/m^2) (Table 4 6 ).

PAGE 48

48 The light surface samples in Live Oak increased significantly from day zero to day one, followed by a significant decrease on day three. An insignificant increase occurred on day seven and day 14, followed by a significant decrease on day 22. An insignificant decrease occurred on day 28 that was characterized by enrichments. Day 55 until day 168 were characterized by all negative enrichme nts, with day 196 seeing a return to positive enrichments. Day 252 resulted in all negative enrichments, but on the final sampling day, day 280, there was a reemergence of E. coli in the light plots . The medium surface samples saw an insignificant increa se from day zero to day one, followed by an insignificant decrease on day three. The decrease on day seven was significant, but was followed by an insignificant increase on day 14. The decrease on day 22 was insignificant. Day 28 de creased significantly from day 22 , with an insignificant decrease on day 55 characterized by all negative enrichments. From day 55 to day 112 all enrichments were negative, with a return to plate counts on day 140. Days 168 and 196 saw an insignificant decrease characterized by a reduction in positive enrichments, until day s 224 through 280 when all enrichments were negative. The heavy surface samples follow a similar trend, with a significant increase from day zero to day one, followed by an insignificant decrease on day th ree. A significant decrease occurred on day seven, but a significant increase returned the E. coli population density to day three levels on day 14. A significant decrease occurred from day 28 to 55, the latter of which was characterized by enrichments. On day 84 all enrichments were negative, but a return to positive enrichments occurred from day 112 to day 224, with all negative enrichments returning on day 252 . However, one heavy surface sample enriched positive for E. coli on day 280. When observin g surface

PAGE 49

49 samples across all application rates, day zero had similar E. coli population densities. Day one saw a heavy application rate that was significantly higher than the medium rate. On day three the light rate was significantly lower than the other s, but on day seven there was no significant difference between them. This changed again on day 14, where the heavy rate was higher than the light. On day 22 the medium application rate was significantly higher than the heavy, which was in turn significa ntly higher than the light. From day 28 through day 252 there was no significant difference between application rates (Table 4 13) . The rate of decay for E. coli population density from day zero until the earliest common day of extinction, day 84, was 0. 05 log 10 CFU / day ( r 2 =0.72 ) , 0.06 log 10 CFU/ day ( r 2 =0. 81 ) , and 0.0 6 log 10 CFU per day ( r 2 =0. 75 ) , for light, medium, and heavy surface samples, respectively. The light core samples in Live Oak increased insignificantly from day zero to day one, followed by a significant decrease on days three and seven. On day 14 there was a significant increase that brought the E. coli population density back to day one level s. A significant decrease occurred on days 22, 28, and 55, the latter of which was characterized by enrichments. Days 84 through 252 were all negative enrichments , followed by a reemergence of E. coli on day 280 . The medium application rate followed a s imilar progression, with an insignificant increase from day zero until day three, after which the E. coli population density decreased significantly on day seven. A significant decrease occurred on day 22, which continued into day 28 and 55, both of which were characterized by enrichments. Days 84 through 168 obtained all negative enrichments, followed by day 196 which returned to positive enrichments. The two remaining sampling periods were all negative enrichments. The heavy application rate increased

PAGE 50

50 significantly from day zero to day one. Day three decreased insignificantly, followed by a significant decrease on day seven. The E . coli population density decreased significantly on day 28, and was characterized by both plate counts and enrichments. Day 55 was characterized by all enrichments, with day 84 obtaining all negative enrichments. While day 112 was also characterized by all negative enrichments, positive enrichment results returned on day 140, and returned to negative on day 168. Both day 196 and 224 had one positive enrichment result, which returned to negative on day s 252 and 280 . When observing core samples across all application rates there was no significant difference between rates on day zero. On day one, however, the heavy applica tion rate was significantly higher than the others. On day three this significance decreased, where the light rate was significantly lower than both the medium and heavy rates, this trend continuing on day seven. From day 14 onward, all application rates were statistically similar by day for E. coli population density (Table 4 1 4 ) . The rate of decay for E. coli population density from day zero until the earliest common day of extinction, day 84, was 0.04 log 10 CFU / day ( r 2 =0.73 ) , 0.05 log 10 CFU/ day ( r 2 =0. 78 ) , and 0.0 6 log 10 CFU per day ( r 2 =0. 79 ) , for light, medium, and heavy core samples, respectively. During the fall season in Live Oak, when averaging sample type, the light application rate samples were significantly lower in E. coli population density than the other application rates (P value <0.05). When averaging all application rates, the surface samples were significantly higher in E. coli population density than the core samples (P value <0.05). When sample types were averaged, there was a very significant difference in E. coli population density across all application types from days

PAGE 51

51 zero to 28 (P value <0.0001). On days 140 one control plot was positive for E. coli , on day 196 two control plots were positive, and on day 280 all control plots w ere positive. Upon statistical analysis, there was no significant difference between samples that were positively or negatively enriched (Table 4 15). The light surface samples in Citra increased significantly from day zero to day one, followed by a smal l fluctuation of insignificant decreases and increases on days three and seven. The first significant decrease occurred on day 14 and continued into day 28. The E. coli population density increased insignificantly on day 44, and decreased insignificantly on day 55. Day 84 was significantly lower than day 55 and was characterized by all negative enrichments. An insignificant fluctuation occurred thereafter, with positive enrichments on days 105, 154, 196, and 224, the remaining days consisting of all neg ative enrichments. The medium surface samples also increased from day zero to day one, but it was not significant. An insignificant decrease and increase occurred on days three and seven, respectively. The first significant decrease was seen on day 14, and continued into day 28. An insignificant decrease occurred through days 44, 55, and 105, the latter two of which were characterized by enrichments. An insignificant increase occurred on days 112 and 140, followed by a decrease on day 154. The E. coli population density increased insignificantly on day 168 and plateaued on day 196, after which all samples were negative. The heavy application rate samples progressed similarly, beginning with a significant increase from day zero to one. An insignifican t increase continued into day three, followed by an insignificant decrease on day seven. The first significant decrease occurred on day 14 and continued into day 28. Days 44, 55, and 84 decreased insignificantly, the latter two

PAGE 52

52 of which were characterize d by enrichments. Day 105 increased slightly , and then decreased through days 112 and 140. A significant increase occurred on day 196, followed by a decrease and plateau on days 224 and 252, respectively. All surface samples were negative for E. coli on day 280. When observing surface samples across all application rates, there was no significant difference in E. coli population density between rates on day zero. On day one the heavy rate was significantly higher than the medium rate, and significantly higher than both the light and medium rates on day three. From day seven until day 252 there was no significant difference in E. coli population density between surface sample application rates, by day (Table 4 1 6 ) . The rate of decay for E. coli population density from day zero until the final sampling day 280, was 0. 013 log 10 CFU / day ( r 2 =0. 50 ) , 0. 013 log 10 CFU/ day ( r 2 =0. 60 ) , and 0. 0 12 log 10 CFU per day ( r 2 =0. 51 ) , for light, medium, and heavy surface samples, respectively. The light core samples on Citra increased significantly from day zero to day one, followed by an insignificant decrease on days three and seven. The first significant decrease occurred on day 14, after which there was an insignificant decrease through days 28, 44, and 54, all o f which were characterized by enrichments. Days 84 through 112 were all negative enrichments, with a return to positive enrichments on day 140 . The light core samples were dominated by negative enrichments thereafter until day 252, when positive enrichme nts returned. On day 280 all light core samples were negative for E. coli . The medium core samples also increased from day zero to one, but insignificantly. A fluctuation of insignificant decreases then increases occurred on days three and seven, respec tively. T he first significant decrease occurred on day 28. Days 44 through 140 were characterized by insignificant fluctuations caused by positive and

PAGE 53

53 negative enrichments until day 154, when all enrichments were negative. Positive enrichments returned on days 168, 196, and 224, followed by all negative enrichments on day 252 and 280 . The heavy core samples increased insignificantly from day zero until day three, after which they decreased insignificantly on day seven. The first significant decrease in E. coli population density occurred on day 14 and continued on day 28. Days 44 through 140 fluctuated insignificantly due to positive and negative enrichments, followed by all negative enrichments on day 154. There was a return to positive enrichments f rom day 168 until day 252 and 280 . When observing core samples across all application types, the heavy rate was significantly higher than the light on day zero. However, there was no significant difference between the rates on day one. On day three the heavy rate was significantly higher than the others. The heavy rate was then significantly higher than the light rate on day seven and 14. From day 28 until day 280 there was no significant difference in E. coli population density between core sample app lication rates, by day (Table 4 1 7 ) . The rate of decay for E . coli population density from day zero until the latest common negative sampling, day 280, was 0. 009 log 10 CFU / day ( r 2 =0. 45 ) , 0. 011 log 10 CFU/ day ( r 2 =0. 59 ) , and 0. 0 12 log 10 CFU per day ( r 2 =0. 51 ) , for light, medium, and heavy surface samples, respectiv ely During the fall season in Citra, when averaging sample type, the heavy application rate samples were significantly higher in E. coli population density than the other application rates (P value <0.05). When averaging all application rates, the surface samples were significantly higher in E. coli population density than the core samples (P value <0.05). When sample types were averaged, there was a significant difference in E. coli population density across all application types from days zero to 55

PAGE 54

54 (P value <0.05). On days 105 and 168 one control plot was positive for E. coli. Upon statistical analysis, there was no significant differ ence between samples that were positively or negatively enriched (Table 4 18). Over the course of all trials, positive PCR results for the presence of E. coli were obtained from samples that were previously enriched negative for E. coli only during the Ci tra summer trial, trial 2. These results were obtained from a heavy top sample on day 84 (Fig ure 4 11) and a control top sample on day 105 (Fig ure 4 12).

PAGE 55

55 Table 4 1 . Average E. coli (log 10 CFU/g medium) population density recovered from three manure types varying by processing step and soil varying by location. Population Density of E. coli (log 10 CFU/g soil) a Manure Type Soil Location Pre screening Post Screening Drying Live Oak Citra 6.2 0 ± 0. 42 4.41 ± 0. 55 2.40 b 0.00 ± 0. 00 0.00 ± 0. 00 a Values are mean ± standard deviation of counts taken from duplicate experiments (n= 2). b Value obtained from one experiment consisting of three compiled samples. Table 4 2. Soil particle analysis of Live Oak and Citra soils. Soil Texture (%) Location Soil Name Clay Silt Loam Live Oak Gainesville Sandy Loam 2.67 2.00 95.33 Citra Candler Sand 2.67 3.00 94.33

PAGE 56

56 Table 4 3 . Average E. coli population density (log 10 CFU/g soil) and enrichments recovered from samples taken from both Live Oak and Citra microcosms held at 20 º C and 30 º C. Population Density of E. coli (log 10 CFU/g soil) a Temperature 20 º C 30 º C Location Day Live Oak Enrch Citra Enrch Live Oak Enrch Citra Enrch 0 3.73 ± 0.07 hij 0,0 3.77 ± 0. 26 ghij 0,0 3.69 ± 0.17 hij 0,0 3.55 ± 0. 05 ij 0,0 3 5.13 ± 0.31 ab 0,0 4.94 ± 0. 42 abc 0,0 5.27 ± 0.14 a 0,0 4.67 ± 0. 38 bcd 0,0 5 4.85 ± 0.27 abc 0,0 4.89 ± 0. 73 abc 0,0 5.21 ± 0.18 a 0,0 4.76 ± 0. 26 abcd 0,0 7 4.62 ± 0.36 bcde 0,0 4.28 ± 0. 08 defg 0,0 4.98 ± 0.07 abc 0,0 4.50 ± 0. 44 cdef 0,0 10 4.10 ± 0.18 efgh 0,0 4.05 ± 0. 36 fghi 0,0 4.30 ± 0.22 defg 0,0 3.86 ± 0. 47 ghij 0,0 15 2.51 ± 0.23 lmno 0,0 3.49 ± 0. 07 jk 0,0 3.67 ± 0.11 hij 0,0 3.35 ± 0. 16 jk 0,0 30 2.59 ± 0.11 lmn 0,0 3.34 ± 0. 65 jk 0,0 3.00 ± 0.22 lk 0,0 2.75 ± 0. 25 lm 0,0 60 2. 69 ± 0.25 lm 0,0 2.05 ± 0. 52 on 0,0 1.26 ± 0.20 qr 0,0 0.67 ± 0.24 stu 2,2 90 2 .26 ± 0.39 mno 0,0 1.99 ± 0. 77 op 0,0 0.67 ± 0.24 stu 2,2 0.50 ± 0. 00 tuv 3,3 120 1 .49 ± 0.16 qp 0,0 1.40 ± 0. 17 q 0,0 0. 33 ± 0. 24 tuv 2,3 0.5 0 ± 0. 00 tuv 3,3 150 1.33 ± 0.35 qp 0,0 1.20 ± 0. 17 qrs 0,0 0.33 ± 0.24 tuv 2,3 0.50 ± 0. 00 tuv 3,3 180 1.06 ± 0.60 qrs 1,1 1.32 ± 0.55 q 0,0 0.33 ± 0.24 tuv 2,3 0.50 ± 0. 00 tuv 3,3 210 0.77 ± 0.46 rst 2,2 0.50 ± 0. 00 tuv 3,3 0.00 ± 0.00 v 0,3 0.00 ± 0.00 v 0,0 240 0.77 ± 0.46 rst 2,2 0.67 ± 0. 29 stu 2,2 0.00 ± 0.00 v 0,3 0.00 ± 0.00 v 0,0 270 0.50 ± 0.00 tuv 3,3 0. 50 ± 0.00 tuv 3,3 b b b b 300 0.33 ± 0.24 tuv 2,3 0.17 ± 0. 24 uv 1,3 b b b b 330 0.33 ± 0.24 tuv 2,3 0.50 ± 0. 00 tuv 3,3 b b b b 360 0.33 ± 0.24 tuv 2,3 0.50 ± 0. 00 tuv 3,3 b b b b 390 0.33 ± 0.24 tuv 2,3 0.33 ± 0. 24 tuv 1,3 b b b b 420 0.00 ± 0.00 v 0,3 0.00 ± 0.00 v 0,3 b b b b a Values are mean ± standard deviation of counts taken from triplicate samples (n=3). b Not measured a v Denote significant differences between counts across all rows and columns. =0.05, P value 0.05 Enrichment column denotes positive enrichments, total number of enrichments

PAGE 57

57 A B Figure 4 1. Moisture level, measured as percentage volumetric water content, of microcosms by location and temperature during A) the first thirty days of the experiment and B) the entirety of the experiment . Error bars represent standard deviation of triplicate sampl es (n=3) . 0 5 10 15 20 25 30 0 5 10 15 20 25 30 VWC (%) Day Live Oak 30°C Citra 30°C Live Oak 20°C Citra 20°C 0 5 10 15 20 25 30 0 50 100 150 200 250 300 350 VWC (%) Day Live Oak 30°C Citra 30°C Live Oak 20°C Citra 20°C

PAGE 58

58 A B Figure 4 2. pH levels of microcosms by location and temperature over A) the first thirty days of the experiment and B) the entirety of the experiment . Error bars represent standard deviation of triplicate samples (n=3) . 6 6.5 7 7.5 8 0 5 10 15 20 25 30 pH Day Live Oak 30°C Citra 30°C Live Oak 20°C Citra 20°C 6 6.5 7 7.5 8 0 100 200 300 400 500 pH Day Live Oak 30°C Citra 30°C Live Oak 20°C Citra 20°C

PAGE 59

59 A B Figure 4 3. Moisture level, measured as percentage volumetric water content, of plots by application rate in Live Oak during the summer season, trial one over the first A) 28 days and B) the entirety of the trial . Error bars represent standard deviation of tripl icate samples (n=3). 0 10 20 30 40 50 60 70 80 0 5 10 15 20 25 VWC (%) Day Light application Medium application Heavy application 0 10 20 30 40 50 60 70 80 0 20 40 60 80 100 120 140 VWC (%) Day Light application Medium application Heavy application

PAGE 60

60 A B Figure 4 4. pH levels of plots by application rate and sample type in Live Oak during the summer season, trial one over A) the first 28 days and B) the entirety of the trial Error bars represent standard deviation of triplic ate samples (n=3) . 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 0 5 10 15 20 25 Ph Day Light Core Medium Core Heavy Core Light Surface Medium Surface Heavy Surface 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 0 50 100 150 Ph Day Light Core Medium Core Heavy Core Light Surface Medium Surface Heavy Core

PAGE 61

61 A B Figure 4 5. Moisture level, measured as percentage volumetric water content, of plots by application rate in Citra during the summer season, trial one over A) the first 28 days and B) the entirety of the trial . Error bars represent standard deviation of triplicate samples (n=3). 0 1 2 3 4 5 6 7 8 9 10 0 10 20 VWC(%) Day Light application Medium application Heavy application 0 10 20 30 40 50 60 70 80 0 50 100 150 VWC(%) Day Light application Medium application Heavy application

PAGE 62

62 A B Figure 4 6. pH levels of plots by application rate and sample type in Citra during the summer season, trial one over A) the first 28 days and B) the entirety of the trial. Error bars repres ent standard deviation of triplicate samples (n=3). 4 4.5 5 5.5 6 6.5 7 7.5 8 0 5 10 15 20 25 pH Day Light Core Medium Core Heavy Core Light Surface Medium Surface Heavy Surface 4 4.5 5 5.5 6 6.5 7 7.5 8 0 50 100 150 pH Day Light Core Medium Core Heavy Core Light Surface Medium Surface Heavy Surface

PAGE 63

63 A B Figure 4 7. Moisture level, measured as percentage volumetric water content, of plots by application rate in Live Oak during the fall season, trial two over A) the first 28 days and B) the entire ty of the trial . Error bars represent standard deviation of triplicate samples (n=3). 0 2 4 6 8 10 12 14 16 0 5 10 15 20 25 VWC (%) Day Light Medium Heavy 0 2 4 6 8 10 12 14 16 0 50 100 150 200 250 300 VWC (%) Day Light Medium Heavy

PAGE 64

64 A B Figure 4 8. pH levels of plots by application rate and sample type in Live Oak during the fall season, trial two over A) the first 28 days and B) the entirety of the trial. Error bars represent standard deviation of triplicate samples (n=3). 4 4.5 5 5.5 6 6.5 7 0 5 10 15 20 25 pH Day Light Core Medium Core Heavy Core Light Surface Medium Surface Heavy Surface 4 4.5 5 5.5 6 6.5 7 7.5 0 50 100 150 200 250 300 pH Day Light Core Medium Core Heavy Core Light Surface Medium Surface Heavy Surface

PAGE 65

65 A B Figure 4 9. Moisture level, measured as percentage volumetric water content, of plots by application rate in Citra during the fall season, trial two over A) the fir st 28 days and B) the entirety of the trial . Error bars represent standard deviation of triplicate samples (n=3) . 0 5 10 15 20 25 30 0 5 10 15 20 25 VWC (%) Day Light Medium Heavy 0 5 10 15 20 25 30 0 50 100 150 200 250 300 VWC (%) Day Light Medium Heavy

PAGE 66

66 A B Figure 4 10. pH levels of plots by application rate and sample type in Citra during the fall season, trial two over A) the first 28 days and B) the entirety of the trial . Error bars represent standard deviation of triplicate samples (n=3) . 4 4.5 5 5.5 6 6.5 7 0 5 10 15 20 25 pH Day Light Core Medium Core Heavy Core Light Surface Medium Surface Heavy Surface 4 4.5 5 5.5 6 6.5 7 0 50 100 150 200 250 300 pH Day Light Core Medium Core Heavy Core Light Surface Medium Surface

PAGE 67

67 Table 4 4 . Average air temperature parameters ( º C ) in Citra, FL and Live Oak, FL field locations over the duration of Trials 1 and 2, i.e. , summer and fall, respectively. Air Temperature Parameters Trial Location Temp avg ( º C ) Temp min avg ( º C ) Temp max avg ( º C ) Summer (one) LO 26.13 ± 1.94 a 20.66 ± 2.57 a 33.30 ± 2. 31 a CA 26.04 ± 1.81 a 18.79 ± 6.04 b 33.11 ± 2. 23 a Fall (two) LO 17. 97 ± 6. 54 b 12.08 ± 7.3 8 c 2 5 . 21 ± 6. 70 b CA 19.13 ± 5. 94 c 12.42 ± 7. 27 c 2 6.29 ± 5. 95 b a c, Values are not significantly different within columns =0.05 , P value 0.05 LO and CA signify locations Live Oak and Citra, respectively Table 4 5. Average soil temperature parameters ( º C ) in Citra, FL and Live Oak, FL field locations over the duration of Trials 1 and 2 , i.e. , summer and fall, respectively. Soil Temperature Parameters Trial Location Soil Temp avg ( º C ) Soil Temp min avg ( º C ) Soil Temp max avg ( º C ) Summer (one) LO 30.06 ± 2.21 a 27.13 ± 2.31 a 33.49 ± 2.43 a CA 29.60 ± 1.56 a 24.32 ± 7.36 b 33.02 ± 1.93 a Fall (two) LO 20. 94 ± 6. 48 b 18.41 ± 6. 47 c 23. 88 ± 6. 68 b CA 22. 44 ± 5.42 b 19.31 ± 6. 73 c 24. 54 ± 5. 78 b a c , Values are not significantly different within columns =0.05 , P value 0.05 LO and CA signify locations Live Oak and Citra, respectively Table 4 6. Average weather parameters in Citra, FL and Live Oak, FL field locations over the duration of Trials 1 and 2 , i.e. , summer and fall, respectively. Weather Parameters Trial Location Relative Humidity avg ( pct ) Rainfall total avg ( cm ) Sol ar Radiation avg (w/m^2) Summer (one) LO 80.70 ± 7.77 a 0.57 ± 1.45 a 229.78 ± 52.87 a CA 81.07 ± 7.53 a 0.36 ± 0.90 a 224.22 ± 56.90 a Fall (two) LO 80. 63 ± 10.58 a 0. 35 ± 1. 09 a 163 . 69 ± 69.89 b CA 79.7 0 ± 9. 63 a 0. 26 ± 0. 86 a 163.22 ± 71.02 b a ,b Values are not significantly different within columns =0.05 , P value 0.05 LO and CA signify locations Live Oak and Citra, respectively

PAGE 68

68 Table 4 7. Average E. coli population density (log 10 CFU/g soil) and enrichments recovered from surface soil samples over time, taken from the summer season field in Live Oak, FL. Population Density of E. coli (log 10 CFU/g soil) a Application Rate Day Light Enrch Medium Enrch Heavy Enrch 0 3.63 ± 0.27 ab 0,0 3. 17 ± 0. 41 abc 0,0 3.87 ± 0. 01 a 0,0 1 3.17 ± 0.17 abc 0,0 2.96 ± 0. 40 bcd 0,0 3. 51 ± 0. 07 ab 0,0 3 1.92 ± 0. 79 fg 0 , 0 2.24 ± 0. 41 def 0,0 3.14 ± 0.2 1 abc 0,0 7 1.17 ± 0. 29 ghij 1, 1 1.18 ± 0. 49 ghij 1,1 2.17 ± 0.2 8 ef 0,0 14 1.10 ± 0. 14 hijk 2,2 2.12 ± 0. 59 ef 0,0 2.60 ± 0. 37 cdef 0,0 28 0.50 ± 0. 00 jkl 3,3 1.09 ± 1 .2 1 hijk 1,2 1.26 ± 1.07 ghij 2,2 56 0.00 ± 0. 00 l 0,3 0.00 ± 0. 00 l 0,3 0.00 ± 0. 00 l 0,3 84 0.00 ± 0. 00 l 0,3 0.00 ± 0. 00 l 0,3 0.00 ± 0. 00 l 0,3 112 0.17 ± 0.2 4 l 1,3 0.00 ± 0.00 l 0,3 0.00 ± 0.00 l 0,3 127 0.00 ± 0.00 l 0,3 0.00 ± 0.00 l 0,3 0.00 ± 0.00 l 0,3 a Values are mean ± standard deviation of counts taken from triplicate samples (n=3). a l =0.05, P value 0.05 Enrichment column denotes positive enrichments, total number of enrichments . 0.00 ± 0.00 values indicate all samples are enriched, and all enrichments are negative with PCR confirmation. Table 4 8. Average E. coli population density (log 10 CFU/g soil) and enrichments recovered from core soil samples over time, taken from the summer season field in Live Oak, FL. Population Density of E. coli (log 10 CFU/g soil) a Application Rate Day Light Enrch Medium Enrch Heavy Enrch 0 2.97 ± 0.47 b 0,0 2.47 ± 0. 47 bc 0,0 2.81 ± 0. 24 b 0,0 1 2.54 ± 1.31 bc 0,0 2.87 ± 0. 37 b 0,0 3. 80 ± 0. 93 a 0,0 3 2.22 ± 0. 88 bc 0,0 2.40 ± 0. 07 bc 0,0 2.59 ± 0. 14 bc 0,0 7 0.50 ± 0.00 fgh 3,3 1.42 ± 0. 59 de 0,0 1.14 ± 0. 91 def 2,2 14 0.67 ± 0. 70 efgh 2,2 1.25 ± 0.55 def 0,0 1.83 ± 0.26 cd 0,0 28 0.77 ± 0. 43 efgh 0,1 0.33 ± 0.24 gh 2,3 0.67 ± 0.24 efgh 2,2 56 0.00 ± 0. 00 h 0,3 0.17 ± 0. 24 h 1,3 0.17 ± 0. 24 h 1,3 84 0.00 ± 0. 00 h 0,3 0.00 ± 0. 00 h 0,0 0.00 ± 0. 00 h 0,0 112 0.00 ± 0. 00 h 0,3 0.00 ± 0.00 h 0,0 0.00 ± 0.00 h 0,0 127 0.00 ± 0.00 h 0,3 0.00 ± 0.00 h 0,0 0.00 ± 0.00 h 0,0 a Values are mean ± standard deviation of counts taken from triplicate samples (n=3). a h =0.05, P value 0.05 Enrichment column denotes positive enrichments, total number of enrichments . 0.00 ± 0.00 values indicate all samples are enriched , and all enrichments are negative with PCR confirmation.

PAGE 69

69 a Values are mean ± standard deviation of counts taken from triplicate samples (n=3). a n =0.05, P value 0.05. 0.00 ± 0.00 values indicate all samples are enriched, and all enrichments are negative with PCR confirmation. Table 4 9. Average E. coli population density (log 10 CFU/g soil) recovered from surface and core soil samples over time, taken from the summer season field, trial one, in Live Oak, FL. Population Density of E. coli (log 10 CFU/g soil) a Sample Type Surface Core Application Rates Day Light Medium Heavy Light Medium Heavy 0 3.63 ± 0.27 ab 3. 17 ± 0. 41 abcd 3.87 ± 0. 01 a 2.97 ± 0.47 bcde 2.47 ± 0. 47 defg 2.81 ± 0. 24 cdef 1 3.17 ± 0.17 abcd 2.96 ± 0. 40 bcde 3. 50 ± 0. 07 abc 2.54 ± 1.31 defg 2.87 ± 0. 37 bcdef 3. 80 ± 0. 93 a 3 1.92 ± 0. 79 ghi 2.24 ± 0. 41 efg 3.14 ± 0.2 1 abcd 2.22 ± 0. 88 efg 2.40 ± 0. 07 defg 2.59 ± 0. 14 defg 7 1.17 ± 0. 29 ijkl 1.18 ± 0. 49 ijkl 2.17 ± 0.2 8 fgh 1.52 ± 1.11 lmn 1.42 ± 0. 59 hijk 1.14 ± 0. 91 ijkl 14 1.10 ± 0. 14 jklm 2.12 ± 0. 59 fgh 2.60 ± 0. 37 defg 0.67 ± 0. 70 klmn 1.25 ± 0.55 ijkl 1.83 ± 0.26 ghij 28 0.50 ± 0. 00 lmn 1.09 ± 1 .2 1 jklm 1.26 ± 1.07 ijkl 0.77 ± 0. 43 klmn 0.33 ± 0.24 mn 0.67 ± 0.24 klmn 56 0.00 ± 0. 00 n 0.00 ± 0. 00 n 0.00 ± 0. 00 n 0.00 ± 0. 00 n 0.17 ± 0. 24 n 0.17 ± 0. 24 n 84 0.00 ± 0. 00 n 0.00 ± 0. 00 n 0.00 ± 0. 00 n 0.00 ± 0. 00 n 0.00 ± 0. 00 n 0.00 ± 0. 00 n 112 0.17 ± 0.2 4 n 0.00 ± 0.00 n 0.00 ± 0.00 n 0.00 ± 0. 00 n 0.00 ± 0.00 n 0.00 ± 0.00 n 127 0.00 ± 0.00 n 0.00 ± 0.00 n 0.00 ± 0.00 n 0.00 ± 0.00 n 0.00 ± 0.00 n 0.00 ± 0.00 n

PAGE 70

70 Table 4 10 . Average E. coli population density (log 10 CFU/g soil) and enrichments recovered from surface soil samples over time, taken from the summer season field in Cit ra, FL . Population Density of E. coli (log 10 CFU/g soil) a Application Rate Day Light Enrch Medium Enrch Heavy Enrch 0 2.13 ± 0.47 def 0,0 2.47 ± 0. 15 d 0,0 2.26 ± 0. 20 de 0,0 1 3.91 ± 1.31 ab 0,0 3.60 ± 0. 85 bc 0,0 4.61 ± 0. 08 a 0,0 3 2.22 ± 0. 88 de 0,0 2.76 ± 0. 67 cd 0,0 3.58 ± 0. 71 bc 0,0 7 1.52 ± 1.11 ef 1,1 2.42 ± 0. 27 d 0,0 2.93 ± 0. 33 cd 0,0 14 2.24 ± 0. 70 de 0,0 1.95 ± 0. 57 def 0,0 2.81 ± 0. 85 cd 0,0 28 1.35 ± 0. 43 f 1,1 1.95 ± 0.47 def 0,0 2.36 ± 0.38 de 0,0 56 0.17 ± 0. 24 g 1,3 0.00 ± 0. 00 g 0,3 0.17 ± 0. 24 g 1,3 72 0.00 ± 0. 00 g 0,3 0.00 ± 0. 00 g 0,3 0.17 ± 0. 24 g 1,3 84 0.00 ± 0. 00 g 0,3 0.00 ± 0. 00 g 0,3 0.17 ± 0. 24 g 1,3 112 0.00 ± 0.00 g 0,3 0.00 ± 0.00 g 0,3 0.00 ± 0.00 g 0,3 140 0.00 ± 0.00 g 0,3 0.00 ± 0.00 g 0,3 0.00 ± 0.00 g 0,3 a Values are mean ± standard deviation of counts taken from triplicate samples (n=3). a g =0.05, P value 0.05 Enrichment column denotes positive enrichments, total number of enrichments 0.00 ± 0.00 values indicate all samples are enriched, and all enrichments are negative with PCR confirmation.

PAGE 71

71 Table 4 1 1 . Average E. coli population density (log 10 CFU/g soil) and enrichments recovered from core soil samples over time, taken from the summer season field in Citra, FL . Population Density of E. coli (log 10 CFU/g soil) a Application Rate Day Light Enrch Medium Enrch Heavy Enrch 0 0.93 ± 0.56 e 1,1 0 . 93 ± 0. 56 e 1,1 0.50 ± 0. 00 f 3,3 1 3.70 ± 0.23 ab 0,0 4.06 ± 0. 53 a 0,0 4.40 ± 0. 41 a 0,0 3 2.63 ± 0. 21 c 0,0 2.93 ± 0. 21 bc 0,0 3.71 ± 0. 63 ab 0,0 7 0.99 ± 0.40 e 1,1 2.10 ± 0. 48 cd 0,0 2.62 ± 0. 46 c 0,0 14 1.44 ± 0. 75 de 1,1 2.47 ± 0. 63 c 0,0 2.56 ± 0. 33 c 0,0 28 0.84 ± 0. 86 e 1,2 1.56 ± 1.26 d 0,0 0.83 ± 0.24 e 1,1 56 0.00 ± 0.00 f 0,3 0.00 ± 0.00 f 0,3 0.00 ± 0.00 f 0,3 72 0.17 ± 0.24 f 1,3 0.00 ± 0.00 f 0,3 0.00 ± 0.00 f 0,3 84 0.00 ± 0.00 f 0,3 0.00 ± 0.00 f 0,3 0.00 ± 0.00 f 0,3 112 0.00 ± 0.00 f 0,3 0.00 ± 0.00 f 0,3 0.00 ± 0.00 f 0,3 140 0.00 ± 0.00 f 0,3 0.00 ± 0.00 f 0,3 0.00 ± 0.00 f 0,3 a Values are mean ± standard deviation of counts taken from triplicate samples (n=3). a f =0.05, P value 0.05 . Enrichment column denotes positive enrichments, total number of enrichments 0.00 ± 0.00 values indicate all samples are enriched, and all enrichments are negative with PCR confirmation.

PAGE 72

72 Table 4 12 . Average E. coli population density (log 10 CFU/g soil) recovered from surface and core soil samples over time, taken from the summer season field, trial one, in Citra, FL. Population Density of E. coli (log 10 CFU/g soil) a Sample Type Surface Core Application Rates Day Light Medium Heavy Light Medium Heavy 0 2.13 ± 0.47 efghi 2.47 ± 0. 15 egf 2.26 ± 0. 20 efghi 0.93 ± 0.56 klm 0 . 93 ± 0. 56 klm 0.50 ± 0. 00 lmn 1 3.91 ± 1.31 ab 3.59 ± 0. 85 bcd 4.61 ± 0. 08 a 3.70 ± 0.23 bc 4.06 ± 0. 53 ab 4.40 ± 0. 4 ab 3 2.22 ± 0. 88 efghi 2.76 ± 0. 67 defg 3.58 ± 0. 71 bcd 2.63 ± 0. 21 efg 2.93 ± 0. 21 cde 3.71 ± 0. 63 bc 7 1.52 ± 1.11 hijk 2.42 ± 0. 27 efg 2.93 ± 0. 33 cdd 0.99 ± 0.40 klm 2.10 ± 0. 48 efghij 2.62 ± 0. 46 efg 14 2.24 ± 0. 70 efghi 1.95 ± 0. 57 ghij 2.81 ± 0. 85 def 1.44 ± 0. 75 ijk 2.47 ± 0. 63 egf 2.56 ± 0. 33 egf 28 1.35 ± 0. 43 jkl 1.95 ± 0.47 fghij 2.36 ± 0.38 efgh 0.84 ± 0. 86 klmn 1.56 ± 1.26 hijk 0.83 ± 0.24 klmn 56 0.17 ± 0. 24 mn 0.00 ± 0. 00 n 0.17 ± 0. 24 mn 0.00 ± 0.00 n 0.00 ± 0.00 n 0.00 ± 0.00 n 72 0.00 ± 0. 00 n 0.00 ± 0. 00 n 0.17 ± 0. 24 mn 0.17 ± 0.24 mn 0.00 ± 0.00 n 0.00 ± 0.00 n 84 0.00 ± 0. 00 n 0.00 ± 0. 00 n 0.17 ± 0. 24 mn 0.00 ± 0.00 n 0.00 ± 0.00 n 0.00 ± 0.00 n 112 0.00 ± 0.00 n 0.00 ± 0.00 n 0.00 ± 0.00 n 0.00 ± 0.00 n 0.00 ± 0.00 n 0.00 ± 0.00 n 140 0.00 ± 0.00 n 0.00 ± 0.00 n 0.00 ± 0.00 n 0.00 ± 0.00 n 0.00 ± 0.00 n 0.00 ± 0.00 n a Values are mean ± standard deviation of counts taken from triplicate samples (n=3). a n =0.05, P value 0.05. 0.00 ± 0.00 values indicate all samples are enriched, and all enrichments are negati ve with PCR confirmation.

PAGE 73

73 Table 4 1 3 . Average E. coli population density (log 10 CFU/g soil) and enrichments recovered from surface soil samples over time, taken from the fall season field in Live Oak, FL. Population Density of E. coli (log 10 CFU/g soil) a Application Rate Day Light Enrch Medium Enrch Heavy Enrch 0 3.22 ± 0.34 fg 0,0 3.84 ± 0.29 cdef 0,0 3.88 ± 0.80 cdef 0,0 1 4.66 ± 0.08 abc 0,0 4.30 ± 0. 31 bc de 0,0 5.28 ± 0. 24 a 0,0 3 3.08 ± 0. 84 fg 0,0 4.95 ± 0.25 ab 0,0 4.76 ± 0. 74 abc 0,0 7 3.15 ± 0.68 fg 0,0 3.18 ± 0. 17 fg 0,0 3.36 ± 0. 30 fg 0,0 14 3.58 ± 0. 15 ef 0,0 3.72 ± 0. 60 def 0,0 4.60 ± 0. 49 abcd 0,0 22 1.36 ± 1.29 h 0,1 3.56 ± 0. 97 ef 0,0 2.49 ± 0. 46 g 0,0 28 0.50 ± 0. 00 hi 3,3 0.89 ± 0.93 hi 1,2 0.67 ± 0.47 hi 0,1 55 0.00 ± 0. 00 i 0,3 0.17 ± 0. 24 i 1,3 0.17 ± 0. 24 i 1,3 84 0.00 ± 0. 00 i 0,3 0.00 ± 0. 00 i 0,3 0.00 ± 0. 00 i 0,3 112 0.00 ± 0.00 i 0,3 0.00 ± 0.00 i 0,3 0.33 ± 0. 24 i 2,3 140 0.00 ± 0.00 i 0,3 0. 93 ± 0.0 6 hi 2,2 0. 33 ± 0. 24 i 2,3 168 0.00 ± 0. 00 i 0,3 0.33 ± 0. 24 i 2,3 0.33 ± 0. 24 i 2,3 196 0.33 ± 0. 24 i 2,3 0.17 ± 0. 24 i 1,3 0.17 ± 0. 24 i 1,3 224 0.00 ± 0.00 i 0,3 0.00 ± 0.00 i 0,3 0.17 ± 0. 24 i 1,3 252 0.00 ± 0.00 i 0,3 0.00 ± 0.00 i 0,3 0. 00 ± 0.00 i 0,3 280 0.17 ± 0. 24 i 1,3 0.00 ± 0.00 i 0,3 0. 43 ± 0. 75 i 0,2 a Values are mean ± standard deviation of counts taken from triplicate samples (n=3). a i =0.05, P value 0.05 . Enrichment column denotes positive enrichments, total number of enrichments . 0.00 ± 0.00 values indicate all samples are enriched, and all enrichments are negative with PCR confirmation.

PAGE 74

74 Table 4 1 4 . Average E. coli population density (log 10 CFU/g soil) and enrichments recovered from core soil samples over time, taken from the fall season field in Live Oak, FL. Population Density of E. coli (log 10 CFU/g soil) a Application Rate Day Light Enrch Medium Enrch Heavy Enrch 0 2.96 ± 0.16 cdefg 0,0 3.14 ± 1.25 cdef 0,0 3.75 ± 0. 43 bcd 0,0 1 3.42 ± 1.17 bcde 0,0 3.74 ± 0. 60 bcd 0,0 4.95 ± 0. 47 a 0,0 3 2.66 ± 0. 32 efg 0,0 3.79 ± 0. 71 bc 0,0 4.24 ± 0. 13 ab 0,0 7 1.30 ± 1.48 ijk 0,1 2.91 ± 1.01 defg 0,0 2.86 ± 0. 21 defg 0,0 14 3.01 ± 0. 71 cdef 0,0 2.87 ± 0. 57 defg 0,0 3.29 ± 0. 18 cdef 0,0 22 2.08 ± 0.77 ghi 0, 0 1.54 ± 0. 23 hij 0,0 2.45 ± 0. 40 fgh 0,0 28 0.93 ± 0. 33 jkl 1,1 0.50 ± 0.41 kl 1,2 0.74 ± 1.05 jkl 0,2 55 0.17 ± 0. 24 lm 1,3 0.33 ± 0. 24 lm 2,3 0.33 ± 0. 24 lm 2,3 84 0.00 ± 0. 00 m 0,3 0.00 ± 0. 00 m 0,3 0.00 ± 0. 00 m 0,3 112 0.00 ± 0.00 m 0,3 0.00 ± 0.00 m 0,3 0.00 ± 0.00 m 0,3 140 0.00 ± 0.00 m 0,3 0.00 ± 0.00 m 0,3 0. 33 ± 0. 47 lm 0,2 168 0.00 ± 0.00 m 0,3 0.00 ± 0.00 m 0,3 0.00 ± 0.00 m 0,3 196 0.00 ± 0.00 m 0,3 0.17 ± 0.24 lm 1,3 0.17 ± 0. 24 lm 1,3 224 0.00 ± 0. 00 m 0,3 0.00 ± 0.00 m 0,3 0.17 ± 0. 24 lm 1,3 252 0.00 ± 0.00 m 0,3 0.00 ± 0.00 m 0,3 0. 00 ± 0.00 m 0,3 280 0.67 ± 0.47 kl 0,1 0.00 ± 0.00 m 0,3 0.00 ± 0.00 m 0,3 a Values are mean ± standard deviation of counts taken from triplicate samples (n=3). a m =0.05, P value 0.05 . Enrichment column denotes positive enrichments, total number of enrichments . 0.00 ± 0.00 values indicate all samples are enriched, and all enrichments are negative with PCR confirmation.

PAGE 75

75 Table 4 15. Average E. coli population density (log10 CFU/g soil) recovered from surface and core soil samples over time, taken from the fall season field, trial two, in Live Oak, FL. Population Density of E. coli (log 10 CFU/g soil) a Sample Type Surface Core Application Rates Day Light Medium Heavy Light Medium Heavy 0 3.22 ± 0.34 hijklm 3.84 ± 0.29 cdefgh 3.88 ± 0. 80 cdefgh 2.96 ± 0.16 hijklmn 3.14 ± 1.25 hijklm 3.75 ± 0. 43 defghij 1 4.66 ± 0.08 abcd 4.30 ± 0. 31 bc def 5.28 ± 0. 24 a 3.42 ± 1.17 fghijkl 3.74 ± 0. 60 defghij 4.95 ± 0. 47 ab 3 3.08 ± 0. 84 hijklm 4.95 ± 0. 25 defghi 4.76 ± 0. 74 abc 2.66 ± 0. 32 klmn 3.79 ± 0. 71 defghi 4.24 ± 0. 13 bcdefg 7 3.15 ± 0.68 hijklm 3.18 ± 0. 17 hijklm 3.36 ± 0. 30 ghijklm 1.30 ± 1.48 pqr 2.91 ± 1.01 ijklmn 2.86 ± 0. 21 jklmn 14 3.58 ± 0. 15 fghijk 3.72 ± 0. 60 efghij 4.60 ± 0. 49 abcde 3.01 ± 0. 71 hijklm 2.87 ± 0. 57 ijklmn 3.29 ± 0. 18 hijklm 22 1.36 ± 1.29 pqr 3.56 ± 0. 97 fghijk 2.49 ± 0. 46 lmn 2.08 ± 0.77 nop 1.54 ± 0. 23 opq 2.45 ± 0. 40 mno 28 0.50 ± 0. 00 rst 0.89 ± 0.93 qrst 0.67 ± 0.47 qrst 0.93 ± 0. 33 qrs 0.50 ± 0.41 rst 0.74 ± 1.05 qrst 55 0.00 ± 0. 00 t 0.17 ± 0. 24 t 0.17 ± 0. 24 st 0.17 ± 0. 24 st 0.33 ± 0. 24 st 0.33 ± 0. 24 st 84 0.00 ± 0. 00 t 0.00 ± 0. 00 t 0.00 ± 0. 00 t 0.00 ± 0. 00 t 0.00 ± 0. 00 t 0.00 ± 0. 00 t 112 0.00 ± 0.00 t 0.00 ± 0.00 t 0.33 ± 0. 24 st 0.00 ± 0.00 t 0.00 ± 0.00 t 0.00 ± 0.00 t 140 0.00 ± 0.00 t 0. 93 ± 0.0 6 qrst 0. 33 ± 0. 24 st 0.00 ± 0.00 t 0.00 ± 0.00 t 0. 33 ± 0. 47 st 168 0.00 ± 0. 00 t 0.33 ± 0. 24 st 0.33 ± 0. 24 st 0.00 ± 0.00 t 0.00 ± 0.00 t 0.00 ± 0.00 t 196 0.33 ± 0. 24 st 0.17 ± 0. 24 st 0.17 ± 0. 24 st 0.00 ± 0.00 t 0.17 ± 0.47 st 0.17 ± 0. 24 st 224 0.00 ± 0.00 t 0.00 ± 0.00 t 0.17 ± 0. 24 st 0.00 ± 0. 00 t 0.00 ± 0.00 t 0.17 ± 0. 24 st 252 0.00 ± 0.00 t 0.00 ± 0.00 t 0. 00 ± 0.00 t 0.00 ± 0.00 t 0.00 ± 0.00 t 0. 00 ± 0.00 t 280 0.17 ± 0. 24 st 0.00 ± 0.00 t 0. 43 ± 0. 75 st 0.67 ± 0.47 qrst 0.00 ± 0.00 t 0.00 ± 0.00 t a Values are mean ± standard deviation of counts taken from triplicate samples (n=3). a t =0.05, P value 0.05. 0.00 ± 0.00 values indicate all samples are enriched, and all enrichments are negative with PCR confirmation.

PAGE 76

76 Table 4 1 6 . Average E. coli population density (log 10 CFU/g soil) and enrichments recovered from surface soil samples over time, taken from the fall season field in Citra, FL . Population Density of E. coli (log 10 CFU/g soil) a Application Rate Day Light Enrch Medium Enrch Heavy Enrch 0 3.22 ± 0.26 de 0,0 3.69 ± 0.07 cd 0,0 3.82 ± 0. 08 cd 0,0 1 4.45 ± 0.65 abc 0,0 3.88 ± 0. 07 cd 0,0 4.84 ± 0. 10 ab 0,0 3 4.23 ± 0. 65 bc 0,0 3.68 ± 0. 95 cd 0,0 5.25 ± 0. 33 a 0,0 7 4.34 ± 0.60 bc 0,0 4.20 ± 0.40 bc 0,0 4.88 ± 0. 47 ab 0,0 14 1.82 ± 0. 54 fg 0,0 2.48 ± 0. 73 ef 0,0 2.49 ± 0. 39 ef 0,0 28 0.83 ± 0.46 hijk 2,2 1.35 ± 0. 65 ghij 1,1 1.50 ± 1.11 gh 1,1 44 1.00 ± 0. 71 ghijk 2,2 0.93 ± 0.33 ghijk 1,1 1.07 ± 0.76 ghijk 0,1 55 0.67 ± 0. 24 hijk 2,2 0.49 ± 0. 70 jk 0,2 0.83 ± 0. 62 hijk 0,1 84 0.00 ± 0. 00 l 0,3 0.33 ± 0. 47 kl 0,2 0.33 ± 0. 47 kl 0,2 105 0. 33 ± 0. 24 k l 2,3 0. 00 ± 0. 00 l 0,3 0.93 ± 0. 40 ghijk 1,1 112 0.00 ± 0.00 l 0,3 0.50 ± 0. 50 jk 1,2 0.50 ± 0. 50 jk l 1,2 140 0.00 ± 0.00 l 0,3 0. 87 ± 0. 33 hijkl 2,2 0. 17 ± 0. 24 kl 1,3 154 0. 33 ± 0. 24 kl 2,3 0. 17 ± 0. 24 kl 1,3 0. 17 ± 0. 24 kl 1,3 168 0.00 ± 0.00 l 0,3 0.50 ± 0.00 jk l 3,3 0. 57 ± 0. 98 ijkl 2,2 196 0. 85 ± 0. 87 hijkl 1,2 0. 50 ± 0. 50 jkl 1,2 0.33 ± 0. 24 kl 2,3 224 0.33 ± 0. 47 kl 2,2 0.00 ± 0.00 l 0,3 0.17 ± 0. 24 kl 1,3 252 0.00 ± 0.00 l 0,3 0.00 ± 0.00 l 0,3 0.17 ± 0. 24 kl 1,3 280 0.00 ± 0.00 l 0,3 0.00 ± 0.00 l 0,3 0.00 ± 0.00 l 0,3 a Values are mean ± standard deviation of counts taken from triplicate samples (n=3). a l =0.05, P value 0.05 . Enrichment column denotes positive enrichments, total number of enrichments . 0.00 ± 0.00 values indicate all samples are enriched, and all enrichments are negative with PCR confirmation.

PAGE 77

77 Table 4 1 7 . Average E. coli population density (log 10 CFU/g soil) and enrichments recovered from core soil samples over time, taken from the fall season field in Citra, FL . Population Density of E. coli (log 10 CFU/g soil) a Application Rate Day Light Enrch Medium Enrch Heavy Enrch 0 2.11 ± 0.62 ef 0,0 2.78 ± 0.51 cde 0,0 3.03 ± 0. 46 cd 0,0 1 3.48 ± 0.72 abc 0,0 3.49 ± 0. 22 abc 0,0 3.59 ± 0. 32 abc 0,0 3 3.37 ± 0. 64 bc 0,0 3.37 ± 0. 35 bc 0,0 4.36 ± 0. 45 a 0,0 7 3.14 ± 0.94 c 0,0 3.64 ± 0.23 abcd 0,0 4.13 ± 0. 48 ab 0,0 14 1.47 ± 1 . 09 fg 0,1 2.35 ± 0. 92 de 0,0 2.79 ± 0. 33 cde 0,0 28 0.66 ± 0.75 hi 1,2 1.22 ± 0. 55 fgh 1,1 1.12 ± 0. 56 gh 1,1 44 0.33 ± 0. 24 hi 2,3 0.43 ± 0.61 hi 0,2 0.50 ± 0.00 hi 3,3 55 0.33 ± 0. 24 hi 2,3 0.66 ± 0. 61 ghi 1,2 0.60 ± 0. 54 ghi 1,2 84 0.00 ± 0. 00 i 0,3 0.33 ± 0. 47 hi 0,2 0.17 ± 0. 24 i 1,3 105 0.00 ± 0. 00 i 0,3 0.17 ± 0. 24 i 1,3 0.33 ± 0. 24 hi 2,3 112 0.00 ± 0.00 i 0,3 0.17 ± 0. 24 i 1,3 0.17 ± 0. 24 i 1,3 140 0.17 ± 0. 24 i 1,3 0. 73 ± 0. 33 ghi 1,2 0. 66 ± 0. 61 ghi 1,2 154 0.00 ± 0.00 i 0,3 0.00 ± 0.00 i 0,3 0.00 ± 0.00 i 0,3 168 0.00 ± 0.00 i 0,3 0. 17 ± 0. 24 i 1,3 0. 17 ± 0. 24 i 1,3 196 0.00 ± 0.00 i 0,3 0. 17 ± 0. 24 i 1,3 0.33 ± 0. 24 hi 2,3 224 0.00 ± 0.00 i 0,3 0.17 ± 0. 24 i 1,3 0.17 ± 0. 24 i 1,3 252 0. 43 ± 0. 61 hi 0,2 0.00 ± 0.00 i 0,3 0.77 ± 1.34 ghi 0,2 280 0.00 ± 0.00 i 0,3 0.00 ± 0.00 i 0,3 0.00 ± 0.00 i 0,3 a Values are mean ± standard deviation of counts taken from triplicate samples (n=3). a i =0.05, P value 0.05 . Enrichment column denotes positive enrichments, total number of enrichments . 0.00 ± 0.00 values indicate all samples are enriched, and all enrichments are negative with PCR confirmation.

PAGE 78

78 . Table 4 18. Average E. coli population density (log10 CFU/g soil) recovered from surface and core soil samples over time, taken from the fall season field, trial two, Citra, FL Population Density of E. coli (log 10 CFU/g soil) a Sample Type Surface Core Application Rates Day Light Medium Heavy Light Medium Heavy 0 3.22 ± 0.26 fghij 3.69 ± 0.07 cdefg 3.82 ± 0.08 cdef 2.11 ± 0.62 klmn 2.78 ± 0.51 hijk 3.02 ± 0.46 fghij 1 4.45 ± 0.65 abc 3.88 ± 0.07 cdef 4.84 ± 0.10 ab 3.48 ± 0.72 defgh 3.49 ± 0.22 defgh 3.59 ± 0.32 cdefgh 3 4.23 ± 0.65 bcde 3.68 ± 0.95 cdefg 5.25 ± 0.33 a 3.37 ± 0.64 efghi 3.37 ± 0.35 efghi 4.35 ± 0.45 abcd 7 4.34 ± 0.60 bcd 4.20 ± 0.40 bcde 4.88 ± 0.47 ab 3.14 ± 0.94 fghij 3.65 ± 0.23 cdefgh 4.13 ± 0.48 bcde 14 1.82 ± 0.54 lmno 2.48 ± 0.73 ijkl 2.49 ± 0.39 ijkl 1.47 ± 1.09 mnopq 2.35 ± 0.92 jklm 2.79 ± 0.33 ghijk 28 0.83 ± 0.46 pqrstu 1.35 ± 0.65 nopqr 1.50 ± 1.11 mnop 0.66 ± 0.75 pqrstu 1.22 ± 0.55 nopqrs 1.12 ± 0.56 opqrs 44 1.00 ± 0.71 opqrst 0.93 ± 0.33 opqrst 1.07 ± 0.76 opqrst 0.33 ± 0.24 stu 0.43 ± 0.61 stu 0.50 ± 0.00 rstu 55 0.67 ± 0.24 pqrstu 0.49 ± 0.70 rstu 0.83 ± 0.62 pqrstu 0.33 ± 0.24 stu 0.66 ± 0.61 pqrstu 0.60 ± 0.54 pqrstu 84 0.00 ± 0.00 u 0.33 ± 0.47 stu 0.33 ± 0.47 stu 0.00 ± 0.00 u 0.33 ± 0.47 stu 0.17 ± 0.24 tu 105 0.33 ± 0.24 stu 0.17 ± 0.24 tu 0.93 ± 0.40 opqrst 0.00 ± 0.00 u 0.17 ± 0.24 tu 0.33 ± 0.24 stu 112 0.00 ± 0.00 u 0.50 ± 0.50 rstu 0.50 ± 0.50 rstu 0.00 ± 0.00 u 0.17 ± 0.24 tu 0.17 ± 0.24 tu 140 0.00 ± 0.00 u 0.87 ± 0.33 pqrstu 0.17 ± 0.24 tu 0.17 ± 0.24 tu 0.73 ± 0.33 pqrstu 0.66 ± 0.61 pqrstu 154 0.33 ± 0.24 stu 0.17 ± 0.24 tu 0.17 ± 0.24 tu 0.00 ± 0.00 u 0.00 ± 0.00 u 0.00 ± 0.00 u 168 0.00 ± 0.00 u 0.50 ± 0.00 rstu 0.57 ± 0.98 qrstu 0.00 ± 0.00 u 0.17 ± 0.24 tu 0.17 ± 0.24 tu 196 0.85 ± 0.87 pqrstu 0.50 ± 0.50 rstu 0.33 ± 0.24 stu 0.00 ± 0.00 u 0.17 ± 0. 24 tu 0.33 ± 0.24 stu 224 0.33 ± 0.47 stu 0.00 ± 0.00 u 0.17 ± 0.24 tu 0.00 ± 0.00 u 0.17 ± 0.24 tu 0.17 ± 0.24 tu 252 0.00 ± 0.00 u 0.00 ± 0.00 u 0.17 ± 0.24 tu 0.43 ± 0.61 stu 0.00 ± 0.00 u 0.77 ± 1.34 pqrstu 280 0.00 ± 0.00 u 0.00 ± 0.00 u 0.00 ± 0.00 u 0.00 ± 0.00 u 0.00 ± 0.00 u 0.00 ± 0.00 u a Values are mean ± standard deviation of counts taken from triplicate samples (n=3). a u Values are =0.05, P value 0.05. 0.00 ± 0.00 values indicate all samples are enriched, and all enrichments are negative with PCR confirmation

PAGE 79

79 Figure 4 11. PCR results of Citra Trial 2 day 84, showi ng a positive result from heav y top sample plot two extraction. C+ indicates the positive control, C indicates the negative control. Figure 4 1 2 . PCR results of Citra Trial 2 day 105, sho wing a positive resu lt from the control top sample plot three extraction. C+ indicates the positive control, C indicates the negative control.

PAGE 80

80 CHAPTER 5 DISCUSSION AND CONCLUSIONS From 1998 to 2014, the occurrence of foodborne illness has increased five fold (8 , 4 4 , 45 ). This increase has also been seen in the occurrence of produce related foodborne outbreaks , increasing by 13.6% from 1973 to 2008 ( 4 4 , 49 ). Preventing the spread of pathogens after harvest is difficult, considering there is no kill step involved in fresh produce processing ( 4 , 27 ). In order to take a preventative rather than reactive approach to fresh produce safety, programs such as NOP, GAP, and as of 2011, the Produce Safety Rule , aim to prevent contamination and proliferation in the field. Few studie s have observed the survival of manure borne pathogens in field conditions. Variables such as temperature, solar radiation, soil type, moisture, and many others fluctuate in real field conditions on a daily basis and may affect the persistence of pathogen s . The overall goal of this research was to determine the persistence of manure borne E. coli , an indicator organism of fecal contamination and possible human pathogen , in field conditions in north and central Florida. Presence of I ndigenous E. coli and T otal C oliforms in Dairy Bovine Manure and S oil Although the greatest population of E. coli was found in pre screened manure, the collect ion and application of this manure for field studies was not feasible, and therefore screened manure was selected because it had the next highest population. De watered m anure from the drying pile would most likely be use d in a real farm application . S creene d manure with higher water content, provided a much heav ier microbial load , better suited for this study . This higher load allowed for the assumption that any E. coli recovered during the se experiments was from the manure amendment and not native to the s oil.

PAGE 81

81 Laboratory Condition Microcosm Studies Regardless of microcosm temperature or soil location, there was no significant correlation between E. coli population density and neither moisture level nor pH. Previous studies have noted a significant correlat ion between such variables and persistence of various microorganisms , and this understand ing has been accepted for decades ( 26 , 50 , 55 ). However, i t is important to consider that upon statistical analysis of the significance of day and temperature, pH and moisture may be insignificant in comparison . While studies focusing solely on these factors may find significance, this study and others like it further exemplify the complex nature of E. coli and other microorganisms persistence under a variety of facto rs, of which moisture and pH would not be significant ( 20 ). Previous experiments monitoring the persistence of E. coli in bovine manure have reported varying rates . Wang and others ( 60 ) reported a survival of 42 49 days at 37 º C, and 49 56 days at 22 º C. Even at 5 º C, they did not observe the persistence seen in this experiment ( 60 ) . Similarly, Kudva and others ( 36 ) reported a maximum persistence of 100 days in bovine manure when frozen at 4 º C, with all other temperatures resulting in reduced survival. H owever, Jiang and others (2002 ) recovered E. coli from manure using enrichment methods at 21 º C until day 210. This experiment shows increased survival of E. coli as compared to the literature. Microcosms H eld at 30 º C The microcosms held at 30 º C were used to mimic high heat conditions in Florida, such as thos e experienced in summer months. The appropriateness of this temperature is further supported by the field trials, which averaged approximately 30 º C in the soil during the course of trial one (Table 4 5 ). When held at this temperature,

PAGE 82

82 both Live Oak and Citra soil mi crocosms concluded on day 240. Microcosms held at 30 º C consisting of Live Oak soil and manure concluded on day 240. The first enrichment was on day 90, where two microcosms were enriched. The final plate count took place on day 90, the sample taken from one microcosm. The first negative enrichment was from one bag on day 120, the remaining bags were all enrichments. Microcosms held at 30 º C consisting of Citra soil and manure concluded o n day 240. The first enrichment was on day 60, from one microcosm, while the remaining two microcosms were the final plate counts. All of the microcosms were enriched on day 90, and the first negative enrichment took place on day 210, when all of the mic rocosms were negative. Upon conclusion of the trial, it was determined that the Live Oak microcosms contained a higher E. coli population density, on average, than the Citra microcosms. However, the starting concentrations of E. coli were not significan tly different between the Live Oak and Citra microcosms and the rate s of decay between soil locations were also not significantly different . The rate of decay during the initial 90 days in E. coli population density for the Live Oak microcosms was 0. 05 lo g 10 CFU/day ( r 2 =0. 87 ) and the Citra microcosms E. coli population density also decayed at a rate of 0. 05 log 10 CFU/day ( r 2 =0. 88 ). When analyzing the soil particulates, there was not a significant difference between the soil in the Live Oak and Citra microcosms (Table 4 2 ). Additionally, there was no significant correlation between moisture level and pH of the microcosms and the E. coli population density. Upon observing the significant increase in E. coli population density from day zero to day three, the E. coli present in the Live Oak microcosms

PAGE 83

83 increased to levels significantly higher than those observed in the Citra microcosms . This likely occurred due to decreased stress or increased available nutrients. This hypothetical decreased stress could be due to a decrease in competitive inhibition, an increase in organic matter and nutrients that happened at random, or other such traits that are native to the soil. Indeed, upon statistical analysis it was determined that soil location was a significant factor regarding E. coli population density in this experiment , but what trait of the soil made it significant is not certain. Microcosms H eld at 20 º C The microcosms held at 20 º C were used to mimic conditions in Florida similar to those experi enced in the spring or fall months. This choice of temperature is further supported by the field trials, which averaged approximately 20 º C in the soil over the course of trial two (Table 4 5 ). Microcosms held at 20 º C consisting of Live Oak s oil and manure concluded on day 420, although consecutive negative samples were not obtained . The first enrichment was conducted on one microcosm on day 180. The final plate count was taken from one microcosm on day 210. All microcosms were being enrich ed on day 240. Microcosms held at 20 º C consisting of Citra soil and manure also concluded on day 420, although consecutive negative samples were not obtained . The first enrichment was performed on all three microcosms on day 210. The final plate count w as taken from one microcosm on day 244. The E. coli population density persisted significantly longer in the 20 º C trial than in the 3 0 º C , and this is most likely due to the increased temperature of the latter. As stated previously, temperature was determ ined to be a significant factor indicating E. coli population density (P value <0.0015), particularly when combined with the effects of day (P value <0.0001). The

PAGE 84

84 ability of E . coli to persist for a longer time in cooler temperatures has been mentioned on numerous occasions in the literature ( 10 , 48 ). Upon conclusion of the trial, it was determined that the Live Oak microcosms and Citra microcosms contained a similar E. coli population density, on average , during the 20 º C experiment . The initial starting concentrations were not statistically different (Table 4 3 .), and the rates of decay were also insignificantly different , with the Live Oak microcosms declining at 0. 03 log 10 CFU/ day ( r 2 = 0 . 53 ) in the first 90 days , and the Citra micr ocosms also declining at 0. 03 log 10 CFU/ day ( r 2 = 0 . 80 ). While this rate of decay trend is nearly half of that observed in the 30 º C microcosms, this is very likely due to the difference in temperature. Field Trial Studies When observing the weather and temperature parameters between field trials, all temperature parameters were significantly lower in trial two than in trial one (Tabl es 4 4, 4 5 ). The persistence of E. coli was significantly greater in trial two than in tri al one, and it is highly probable that the significant temperature difference played an important role. Indeed, the persistence of E. coli increasing at lower temperatures in soil has been documented on numerous occasions ( 10, 32 , 48 ). However, the only weather parameter , unrelated to temperature, with a significant difference between trials one and two was solar radiation, which decreased in trial two (Table 4 6 ). Solar radiation has previously been described as a viable method of E. coli eradication in dairy cow feed lots, and very likely played a role in the reduction of E. coli during trial one ( 2 ) . Over the course of both trials, in both locations, there have been instances of E. coli population density spikes. Across all fields there were notable increase s in

PAGE 85

85 E . coli population density between days zero and one (Tables 4 15, 4 16, 4 17, 4 18 ). This may be contributed to the lag and exponential growth phase s , which are well documented in most microorganisms ( 54 ). Even more, some instances of population density increases occurred in plots that had previously been determined as E. coli negative during consecutive sampling periods. This gives rise to the possibility of a new contamination event such as runoff contaminati on after a heavy rain fall or wild life intrusio n in the field plots . This possibility is further supported by the occasional presence of E. coli in control plots, which have been shown to be devoid of indigenous E. coli populations initially. While the significance between app lication rates varied by trial, the surface samples contained a significantly higher average E. coli population density than core samples, without exception. This is most likely due to the methods utilized during the field set up. When the rototiller was used to incorporate the manure into the soil, a method used by many farmers in the US, the incorporation was likely not uniform. It i s reasonable to assume that the significance seen between sample types is due to a greater amount of manure present in the top s amples than in the core samples, and therefore a higher population density of E. coli . Furthermore, regardless of trial or location, there was no significant correlation between E. coli population density and neither moisture nor pH (Figures 4 1, 4 2, 4 3, 4 4, 4 5, 4 6, 4 7, 4 8, 4 9, 4 10) . This is likely due to the reasons stated previously, wherein other factors such as manure application rate, sample type, and other variables also affected moisture, pH, and E. coli population density simulta neously . For example, it is well known that soil type and

PAGE 86

86 organic matter play an important role in moisture retention, and therefore these factors overshadow the role of moisture alone ( 25 ) . When comparing the rates of decay across all field experiments, there was little variation regardless of sample type and application rate. This was consistent with previous studies by Crane and others ( 12) and Gessel and others ( 23 ), who concluded that manure application rate does not affect persistence of bacterial p athogens. When considering guidelines and regulations such as FSMA and NOP, these results are encouraging. These findings suggest that any waiting period required between manure application and harvest will be equally effective regardless of season or lo cation in central and north Florida, the efficacy of such a regulation is only affected by the initial microbial load and subsequent contamination events. However, the importance and regularity of such contamination events must be considered, as several i nstances during the se field trials suggest that contamination events play a crucial role in the persistence of E. coli in the field. Even more, previous studies observed significantly shorter survival times of E. coli in bovine manure. For example, Kudva and others ( 36 ) reported a survival of 40 days for E. coli in bovine manure that was monitored in the environment. While the persistence of E. coli , particularly in field settings is important to understand, the clinical significance of this research is not clear. There are many pathways in which pathogenic E. coli may travel from manure to a human host, such as waterborne, foodborne, or airborne methods ( 22 ). Gerba and Smith ( 22 ) state that ther e is no minimum infectious dose for enteric pathogens. This makes quantitative risk assessment of manure borne E. coli difficult. Furthermore, the risk of infection can vary

PAGE 87

87 by crop. For example, Franz and others ( 19 ) showed that 2.8% of lettuce exposed to contaminated manure was likely to result in one head of lettuce/hectare being contaminated with E. coli O157:H7, that likelihood decreasing to 0.1% for two or more heads of lettuce/hectare. However, such small numbers should not be overlooked. Accord ing to Danyluk and Schaffner ( 13 ), a 0.1% prevalence of E. coli O157:H7 at levels of 1 log CFU/g in spinach fields could result in an outbreak like that experienced in 2006. While, within this experiment, counts less than 1 log 10 CFU/g may not be statis tically significant from E. coli negative samples, these counts may still be clinically significant. Further risk assessments should be conducted in order to better understand the clinical significance of low level persistence of E. coli before suggesting a waiting period for manure use in agriculture. Field T rial O ne, S ummer S eason The first field trials began in early May, 2014, and were concluded in mid September, 2014. The fields prepared in both Live Oak and Citra locations were used to show real life field conditions during summer months in North Central Florida. The Live Oak and Citra fields monitored during trial one concluded within one week of each other , and the population density of E. coli was not significantly different between locat ions . While the average temperatures of the air and soil were not significantly different, the average minimum temperature of both the air and soil was significantly lower in Citra than Live Oak (Table s 4 4, 4 5 ), which was the only significantly differen t weather related parameter between locations. During the course of trial one, PCR did not identify E. coli in any negative enrichment samples. When comparing rates of decay between studies , there was no significant difference observed between field tria ls and corresponding microcosm trials, regardless

PAGE 88

88 of location. This suggests that rates of decay determined by microcosm studies may be useful for field persistence determination under similar temperature conditions. However, as described previously, the effect of contamination events may play a large role in the apparent persistence of E. coli in field conditions, and cannot be accounted for when applying rates of decay. It is important to understand th e limits of these comparisons a s the sampling dates were not the same between microcosm and field trials . The initial population density of E. coli after manure incorporation varied by application rate, as was expected , but was not significantly different (Table 4 15 ) . The highest population density at t he time of application was the heavy manure applicatio n rate, top samples, averaging 3 .87 log 10 CFU/g. The lowest population density was the medium manure application rate, core samples, averaging 2 .47 log 10 CFU/g. However, throughout the trial, the heavy application rate was significantly higher in E. coli population density, on average , than all other sample application rates. Further more , the surface samples had a significantly higher average E. coli population density than the core samples . The initial E. coli population on the day of application was lower than expected, the original manure containing an E. coli population of approximately 5.81 CFU/g . The initial decrease in population density may be explained in part by the mixing of a dense population medium (i.e. , manure) with a medium containing no E. coli (i.e. , soil). Additionally, the shock of incorporation into a ne w environment may cause death to injured cells. However, on day one, several applicatio n rates and sample types increased, some significantly, from day zero. This was likely the effect of acclimatization and the lag/ exponential growth phases described previously, and were expected. Throughout the experiment, there was some fluctuation in p opulation

PAGE 89

89 density. While this trial was predominantly characterized by a constant decrease in E. coli , several sampling days observed an increase from the previous sampling day. When observing these fluctuations, they appear to be random and likely cause d by random variations within the plot when sampled (e.g., a concentrated portion of manure in one sample). On day 84, all application rate plots were negative for E. coli enrichment and were confirmed as such with PCR. However, the following sampling period, day 112, one light and one core plot were enriched for E. coli and a colony was isolated. Using PFGE, the colonies isolated proved to be identical (Figure A 1) . While it is possible that this E. coli had originated in the light plot, the consecutive negative enrichments and PCR confirmations would suggest that this E. coli was newly introduced. Th is probability is further supported by the presence of identical E. coli in the neighboring control plot. This instance highlights the possibility of new contamination events in real field settings. Upon observing the rates of decay determined by plotting the E. coli population density from day zero until day 112, no difference was seen between surface samples and corresp onding core samples by application rate (i.e., 0.02 log 10 CFU/day for light and medium samples, and 0.03 log 10 CFU/day for heavy samples) . Additionally , no significant difference in rate of decay was seen between sample types of varying application rates. These findings suggest that the factors responsible for the death of E. coli in this trial act equally on all samples, regardless of application rate and sample type. While certain factors, such as solar radiation, were hypothesized to increase the rate of decay in surface samples, these findings suggest otherwise. It is reasonable to

PAGE 90

90 assume that other factors, such as temperature, which may affect all samples and application rates equally, have a greater effect on the rate of decay. For the purpose o f comparing summer season trial one to the 30 º C microcosms, the rate of decay of the prior was determined by plotting the E. coli population density of the core and surface samples, averaged, from day zero to day 84, the closest sampling period within the initial decrease observed in the microcosms. A rate of decay of 0.04 log 10 CFU/day (r 2 =0.71) was observed in the Live Oak site, which was slightly lower than that observed in the Live Oak microcosms at 30 º C. Despite this small decrease in rate of decay, the Live Oak site persisted for less time than the microcosms. This could be due to the decrease in initial E. coli population density concentration. This discrepancy could be due to an initial E. coli death on day zero that occurred in the field due to increased stress, such as solar radiation and temperature fluctuation. D uring the set up of the field, the manure was left in the elements for several hours before incorporation into the soil, while the microcosms did not endure such stressors. The initial population density of E. coli after manure incorporation varied by application rate, as was expected (Table 4 16) . The highest population density at the time of application was the heavy applicatio n rate, top samples, averaging 2 .26 CFU/g. The lowest population density was the light application rate, core samples, averaging 1 .15 CFU/g. This initial population density trend was observed throughout the trial, with the heavy application rate, top sampl es, having the significantly highest E . coli population density on average , while the light application rate, core samples, had the significantly lowest average E. coli population density. A lower than expected initial

PAGE 91

91 population density occurred after th e appl ication of manure to soil, with a subsequent increase, as described previously. Several instances showed potential for E. coli contamination at this site. One control plot was found to contain E. coli on days seven and 14. This suggests either a ru n off event or contamination event that led to the movement of E. coli to the previously barren control plots . Indeed, it has been shown on numerous occasions that E. coli is readily able to move in the environment, especially in the presence of water or other vectors ( 1, 14 , 24 , 29 , 49 , 51 , 53 , 56 ) . Unlike Live Oak, Citra concluded with consecutive negative sampling periods without incident on days 112 and 140. When comparing the rates of decay, there was no difference between the surface and core samples of the light and medium application rates of 0.03 log 10 CFU/day. However, there was an insignificant increase from 0.03 to 0.04 log 10 CFU/day in the heavy application rate surface samples. Having been an insignificant increase, it is reasonable t o assume this change was random and not an effect of a difference in persistence factors , as described previously. The rate of decay for the E. coli population density, averaged between surface and core samples, for the heavy application rate was plotted f or comparison to the 30 º C Citra microcosms. Similarly, the rate of decay was 0.04 log 10 CFU/day (r 2 =0.70), which was also seen in the Live Oak site. As explained previously, the initial E. coli population density in the Citra site was lower than that obs erved in the microcosms, likely for similar reasons explained in the Live Oak site discussion. Field T rial Two , F all S eason The second field trials began in mid August, 2014, and were concluded in late May, 2015. E. coli was still recoverable in both field locations at the time of completion .

PAGE 92

92 The E. coli population density mo nitored during field trial two was significantly higher on average than trial one (P value <0.05) . This may be attributed to the continued presence of E. coli , although i t is not possible to exclude the possibility of the plots becoming recontamin at ed during the course of the trial . The fields prepared in both Live Oak and Citra locations were used to show real life field conditions during fall months in North Central Flo rida. During trial two, Live Oak had a significantly higher E. coli population density than C itra, while Citra was not signif icantly different from either location in trial one (P value <0.05) . While neither field study concluded with two consecutive n egative samplings, th e significant ly higher populations seen in Live Oak are most likely due to an increase in population de nsities seen during the initial sampling periods after day zero . During this trial, Citra maintained a significantly higher average air and soil temperature than Live Oak, with all other temperature and weather parameters not being significantly different (Table s 4 4, 4 5 ). It is reasonable to suggest that these temperature differences may have been a factor in the increased population density seen in Live Oak as opposed to Citra in trial two. PCR was able to identify the presence of E. coli on two occasions during this trial in Citra, which will be discussed further (Figures 4 11, 4 12) . However, there was no sig nificant difference in the rate of decay between locations. Unlike trial one, this trial was characterized by repeated re emergence of E. coli in previously negative plots , regardless of application rate . However, this trend was observed more frequently in top soil samples. When analyzing the PFGE dendrogram, there were only two outlying clades that would suggest the isolates did not originate

PAGE 93

93 from manure. Numerous presumed contamination isolates clustered closely with presumed manure indigenous isolate s. While this may suggest that the isolates were not from contamination events, it is also possible that, because the remaining clade had a broad genetic variety with in, that new contamination or recontamination could result in isolates that were not dis tinguishable from the others (Fig ure A 1). I t is reasonable to suggest that many of the E. coli reemergence events were due to new contamination events, as described previously. Even more, multiple outbreaks of E. coli have been caused by contamination e vents by outside sources ( 5 , 29 ). When comparing the field persistence of E. coli popu lation density to that of the 20 º C microcosms, there is a significant difference. T he rate of decay is approximately double that observed in the microcosms, regardless of location. Unlike the summer trial and corresponding microcosms, the rate of decay calculated from the 20 º C microcosms are not suitable for approximating persistence in similar field conditions. I t should be emphasized that this comparison is not ideal because the sampling days were not identical. As described in trial one, there was some initial variation in E. coli population density by application rate, but the differences were insignificant (Table 4 17) . The light core sample had the lowest concentration of E. coli with 2.96 log 10 CFU/g, while the heavy surface sample contained the highest concentration with 3.88 log 10 CFU/g. The variation was consistent with expectations. Also , similar to trial one , there was an increase in population density from day zero to day one in all samples, which w as discussed previously.

PAGE 94

94 Over the course of the trial, there were various fluctuations of E. coli population density over time, but one notable change was an increase observed on day 14 in all samples, with exception of the medium core samples . Some of these increases were significant . This spike was noted two days after a rain event of 3.23 cm, and was accompa nied by an increase in soil moistu re ( F igure 4 7 ). While moisture has been documented as a significant contributor to microorganism persistence, this trend was only seen on this occasion during the course of both field trials. Further more , as previously stated, moisture did not correlate significantly with E. coli population density. While this spike may be attributed to increased moisture, it is also possible that new contamination or run off may have resulted in increased population densities ( 24 , 56 ). However, it is also plausible th at moisture did indeed play a role in the increase in E . coli population density, but this trend was only observed once because the event occurred early in the trial, while the majority of E. coli was still viable and able to grow exponentially in favorable conditions . During the course of trial two, Live Oak saw multiple occurrences of E. coli re emergence in previously negative plots, as compared to Citra. As previously discussed , these fluctuations may be due to sampling variation and recovery of previously injured cells. However, it is reasonable to suggest that new contamination events are also a likely cause of these reappearances. For example, over the course of this trial, the light surface samples were negative for five consecutive samplin g periods before recovering E. coli (Table 4 11) . Similar trends of reemergence were seen in all samples in Live Oak , which further supports the possibility of recontamination ( Table 4 -

PAGE 95

95 17 ). Furthermore, E. coli was recovered in control plots on multiple days at varying frequencies, further suggesting the occurrence of new contamination. The rate of decay for the E. coli population density, averaged between surface and core samples, for the heavy application rate was plotted for comparison to the 2 0ºC Liv e Oak microcosms. The rate of decay was found to be 0.0 6 log 10 CFU/day (r 2 =0.7 7 ), which was double the rate observed in the Live Oak 2 0ºC microcosms . As explained previously , there are many stressors present in the environment that can cause an increased rate of cell death . While neither the field experiment nor the microcosm experiment under these conditions have concluded, it is likely that the field experiment will conclude prior to day 420. When considering the likelihood of contamination, as descri bed earlier, it is reasonable to assume that the remaining persistence observed in the field is due to contamination, rather than persistence of the initial E. coli populations. As noted in the other trials, Citra also experienced initial variation that was not significant between application rates. Also, the same increase from day zero to one was noticed, with significant differences seen in some samples, but not others (Table 4 18 ). R andom fluctuations in E. coli population density occurred throughout the experiment, regardless of application rate or type, but the predominant trend was one of decline. While this trial also saw reemergence of E. coli in previously negative plots, the p eriods of consecutive negative sampling were not as distinct . Additionally , the heavy surface samples were never completely negative for E. coli during the course of this experiment (Table 4 13) . While these findings may be interpreted to show increased persistence of manure indigenous E. coli , the previous trials and locations suggest occurrences of new contamination. Coupled with the observance of positive control

PAGE 96

96 plots on days 105 and 168, it is reasonable to suggest that new contamination events may mimic persistence. On days 84 and 105, PCR identified E. coli in a heavy application plot from a top sample and a control plot from a top sample , r espectively, that had enriched negative for E. coli (Figures 4 11, 4 12) . This may suggest the presence of VBNC or severely injured cells that are unable to grow in enrichment but are still intact. However, since this was not a common event, it is reasonable to suggest that E. coli existing in a VBNC state is not a significant concern in field conditions. Th e rate of decay for the E. coli population density was similarly determined as stated in previous sections. The rate of decay was found to be 0.0 5 log 10 CFU/day (r 2 =0. 80 ), which was also higher than that observed in the Citra 20 º C microcosms. This trend is similar to that observed in the Live Oak field and 20 º C microcosm comparison . As stated previously, it is likely that the field experiment will conclude prior to the microcosm experiment. Furthermore, the Citra field experiment may appear to persist, but actually be the result of new contamination.

PAGE 97

97 APPENDIX A PFGE DENDROGRAM A

PAGE 98

98 B

PAGE 99

99 C

PAGE 100

100 D

PAGE 101

101 E Figure A 1. PFGE dendrogram of field experiment isolates, separated into sections A, B, C, and D due to size constraints.

PAGE 102

102 APPENDIX B CONTINUED WORKS Table B 1. Average E. coli population density (log10 CFU/g soil) recovered from surface and cor e soil samples over time, taken from trial two fields after conclusion of thesis research , in L ive Oak, FL Population Density of E. coli (log 10 CFU/g soil) a Sample Type Surface Core Application Rates Day Light Medium Heavy Light Medium Heavy 308 0.00 ± 0.00 a 0.00 ± 0.00 a 0.00 ± 0.00 a 0.00 ± 0.00 a 0.00 ± 0.00 a 0.00 ± 0.00 a a Values are mean ± standard deviation of counts taken from triplicate samples (n=3). a =0.05, P value 0.05. 0.00 ± 0.00 values indicate all samples are enriched, and all enrichments are negative with PCR confirmation. Table B 2. Average E. coli population density (log10 CFU/g soil) recovered from surface and cor e soil samples over time, taken from trial two fields after conclusion of thesis research , in Citra , FL Population Density of E. coli (log 10 CFU/g soil) a Sample Type Surface Core Application Rates Day Light Medium Heavy Light Medium Heavy 308 0.00 ± 0.00 a 0.00 ± 0.00 a 0.00 ± 0.00 a 0.00 ± 0.00 a 0.00 ± 0.00 a 0.00 ± 0.00 a a Values are mean ± standard deviation of counts taken from triplicate samples (n=3). a =0.05, P value 0.05. 0.00 ± 0.00 values indicate all samples are enriched, and all enrichments are negative with PCR confirmation.

PAGE 103

103 LIST OF REFER ENCES 1. Barreiro , C . , H. Albano, J. Silva, and P. Teixeira. 2013. Role of Flies as Vectors of Foodborne Pathogens in Rural Areas . ISRN Microbiol . 20 (1): 31 3 7. 2. Berry , E . D . and J. E. Wells. 2012. Soil Solarization Reduces Escherichia coli O157:H7 and Total Escherichia coli on Cattle Feedlot Pen Surfaces J. Food Prot. 75(1):7 13. 3. Beuchat , L . R. 1996. Pathogenic Microorganisms Associated with Fresh Produce. J . Food Prot . 2:115 2 16. 4. Burnett , S . L . and L. R. Beuchat. 200 1 . Human pathogens associated with raw produce and unpasteurized juices, and difficulties with decontamination. J . Ind . Microbiol . Biotechnol . 2 7:104 110 . 5. California Food Emergency Response Team. 2007. Investigation of an Escherichia coli O157:H7 Outbreak Associated with Dole Pre Packaged Spinach. CDHS 1 50. 6. Centers for Disease Control and Prevention . 201 2 . Surveillance of foodborne disease outbreaks United States 2009 2010. M orb. M ortal. W kly. R ep. 62:41 4 7. 7. Centers for Disease Control and Prevention. 2013 . CDC Estimates of Foodborne Illness in the United States . Available at : http://www.cdc.gov/foodborneburden/2011 foodborne estimates.html . Accessed 8 November 2013. 8. Centers for Disease Control and Prevention. 201 4 . Estimating Foodborne Disease: An Overview. Available at : http://www.cdc.gov/foodborneburden/estimates overview.ht ml . Accessed 23 January 2015. 9. Chapman , P . A . , C. A. Siddons, A. T. Gerdan Malo, and M. A. Harkin. 1997. A 1 year study of Escherichia coli O157 in cattle, sheep, pigs and poultry. Epidemio . l Infect . 119(2):245 50. 10. Cools , D . , R. Merckx, K. Vlassak, and J. Verhaegen. 2001. Survival of E. coli and Enterococcus spp. derived from pig slurry in soils of different texture. Appl . Soil Ecol . 17(1):53 62. 11. Universit y of Florida. 2012 . County extension offices . Available at : http://sfyl.ifas.ufl.edu/map/index.shtml . Accessed 10 June 2015. 12. Crane , S . R . , P. W. Westerman, and M. R. Overcash. 198 0 . Die off of Fecal Indicator Organisms Following Land Application of Poultry Manure . J . Environ . Qual . 9 (3):5 31 5 37. 13. Danyluk , M . D . and D. W. Schaffner. 2011 . Quantitative Assessment of the Microbial Risk of Leafy Greens from Farm to Consumption: Preliminary Framework, Data, and Risk Estimates. J. Food Prot. 74(5):700 8.

PAGE 104

104 14. Doane , C . A . , P. Pangloli, H. A. Ri chards, J. R. Mount, D. A. Golden, and F. A. Draughon. 2007. Occurrence of Escherichia coli O157:H7 in diverse farm environments. J . Food Prot . 70(1):6 10. 15. Food and Drug Administration. 1998. Guidance for Industry: Guide to Minimize Microbial Food Safety Hazards for Fresh Fruits and Vegetables . Available at : http://www.fda.gov/Food/GuidanceRegulation/Guid anceDocumentsRegulatoryInfo rmation/ProducePlantProducts/ucm064574.htm#iii . Accessed 15 November 2013. 16. F ood and Drug Administration. 2013. F SMA Proposed Rule for Produce Safety . Available at : http://www.fda.gov/Food/guidanceregulation/FSMA/ucm334114.htm . Accessed 4 October 2013. 17. Food and Drug Administration. 2015. FSMA Proposed Rule for Produce Safety: Stan dards for the Growing, Harvesting, Packing, and Holding of Produce for Human Consumption. Available at : http://www.fda.gov/Food/GuidanceRegulation/FSMA/ucm334114.htm . Accessed 25 January 2015. 18. Franz , E . , A. van Hoek, E. Bouw, and H. Aarts. 2011. E. coli O157 strain variability in manure amended soil survival in relation to strain origin, virulence profile and carbon nutrition profile . App l. Environ . Microbiol . 77(22):8 088 8096 . 19. Franz , E . , A. V. Semenov, and A. H. C. Van Bruggen. 2008 a . Modelling the contamination of lettuce with Escherichia coli O157:H7 from manure amended soil and the effect of intervention strategies. J. Appl. Microbiol. 105(5):1569 15 84. 20. Franz , E . , A. V. Semenov, A. J. Termorshuizen, O. J. De Vos, J. G. Bokhorst , and A. H. C. Van Bruggen. 2008 b . Manure amended soil characteristics affecting the survival of E. coli O157:H7 in 36 Dutch soils. Appl. Environ. Microbiol. 10(2):313 3 27. 21. Fremaux , B . , C. P rigent Combaret, M. Delignette Muller, M. Dothal, and C. Vernozy Rozand. 2007. Persistence of Shiga toxin producing Escherichia coli O26 in cow slurry. Lett. Appl. Microbiol. 45(1):55 61. 22. Gerba , C . P . and J. E. Smith . 2005. Sources of Pathogenic Microorganisms and Their Fate during Land Application of Wastes. J . Environ . Qual . 34(1):42 4 8. 23. Gessel , P . D . , N. C. Hansen, S. M. Goyal, L. J. Johnston, and J. Webb. 2004. Persistence of zoonotic pathogens in surface soil tre ated with different rates of liquid pig manure. Appl . Soil Ecol . 25(3):237 2 43. 24. Guber , A . K . , J. S. Karns, Y. A. Pachepsky, A. M. Sadeghi, J. S. Van Kesse l , and T. H. Dao. 2007. Comparison of release and transport of manure borne Escherichia coli and ente rococci under grass buffer conditions . Lett. Appl. Microbiol. 44(2):161 16 7. 25. Gupta , S . C . and W. E. Larson. 1979. Estimating soil water retention characteristics from pa rticle size distribution, organic matter percent, and bulk density. Water Resour . Res . 15(6):1633 163 5.

PAGE 105

105 26. Hagedorn, C., D. T. Hansen, and G. H. Simonson. 1978. Survival and Movement of Fecal Indicator Bacteria in Soil under Conditions of Saturated Flow. J. Environ. Qual. 7(1):55 59. 27. Harris , L . J . , J. N. Farber, L. R. Beuchat, M. E. Parish, T . V. Suslow, E. H. Garrett , and F. F. Busta. 2003. Outbreaks Associated with Fresh Produce: Incidence, Growth, and Survival of Pathogens in Fresh and Fresh Cut Produce. Compr . Rev . Food Sci. F . 2: 78 141. 28. Himathongkham , S . and H. Riemann. 1999. Destruction of Salmonella typhimurium , Escherichia coli O157:H7 and Listeria monocytogenes in chicken manure by drying and/or gassing with ammonia. FEMS Microbiol . Lett . 171(2):179 1 82. 29. Hrudey , S . E . , P. Payment, P. M. Huck, R. W. Gillham, and E. J. Hrudey . 2003. A fatal waterborne disease epidemic in Walkerton, Ontario: comparison with other waterborne outbreaks in the developed world. Water Sci . and Tech . 47(3):7 14. 30. Hutchison , M . L . , L. D. Walters, S. M. Avery, F. Munro, and A. Moore. 2005. Analyses of L ivestock Production, Waste Storage, and Pathogen Levels and Prevalences in Farm Manures . Appl . Environ . Microbiol . 71(3):1231 12 36. 31. Ingham , S . C . , J. A. Losinski, M. P. Andrews, J. E. Breuer , J. R. Breuer, T. M. Wood, and T. H. Wright. 2004. Escherichia coli Contamination of Vegetables Grown in Soils Fertilized with Noncomposted Bovine Manure: Garden Scale Studies . Appl. Environ. Microbiol . 70 (11): 6421 6427. 32. Ishii , S . , W. B. Ksoll, R. E. Hicks, and M. J. Sadowsky . 2006. Presence and Growth of Naturalized Escherichia coli in Temperate Soils from Lake Superior Watersheds. Appl . Environ . Microbiol . 72(1):612 6 21. 33. Islam , M . , M. P. Doyle, S. C. Phatak, P. Millner, and X. Jiang. 2004. Persistence of Enterohemorrhagic Es cherichia coli O157:H7 in Soil and on Leaf Lettuce and Parsley Grown in Fields Treated with Contaminated Manure Composts or Irrigation Water. J . Food Prot . 67( 7 ): 1365 13 70. 34. Johannessen , G . S . , G. B. Bengtsson, B. T. Heier, S. Bredholt, Y. Wasteson., and L . M. Rørvik . 2005. Potential Uptake of Escherichia coli O157:H7 from Organic Manure into Crisphead Lettuce. Appl . Environ . Microbiol . 71(5):2221 22 25. 35. Kim , J . , F. Luo , and X. Jiang. 2009. Factors Impacting the Regrowth of Escherichia coli O157:H7 in Dairy Manure Compost. J. Food Prot . 72(7):1576 15 84. 36. Kudva , I . T . , K. Blanch, and C. J. Hodve. 1998. Analysis of Escherichia coli O157:H7 survival in ovine or bovine manure and manure slurry. Appl . Environ . Microbiol . 64 (9): 3166 31 74. 37. MacDona ld , J . , M. Ribaudo, M. Livingston, J. Beckman, and W. Huang. 2009. Manure Use for Fertilizer and for Energy: Report to Congress. AP 0371 53. 38. Maheux , A . F . , F. J. Picard, M. Boissinot, L. Bissonette, S. Paradis, and M. G. Bergeron. 2009. Analytical compar ison of nine PCR primer sets designed to

PAGE 106

106 detect the presence of Escherichia coli / Shigella in water samples. Water . Res . 43(12):3019 30 28. 39. Nicholson , F . A . , S. J. Groves, and B. J. Chambers. 2005. Pathogen survival during livestock manure storage and following land application. Bioresour. Technol . 96(2):135 1 43. 40. Nyachuba , D . G. 2010. Foodborne illness: is it on the rise? Nutr. Rev. 68(5):257 2 69. 41. Organic Trade Association. 2011. T he Organic Trade Association's 2011 Organic Industry Survey. 42. Pachepsky , Y . A . , A. M. Sadeghi, S. A. Bradford, D. R. Shelton , A. K. Guber, and T. Dao. 2006. Transport and fate of manure borne pathogens: Modeling perspective. Agric.Water Manage . 86(1 2):81 92. 43. Pagadala , S . , S. Marine, S. Micallef, F. Wang, D. Pahl, M. Melendez, W. Kline, R. Oni, C. Walsh, K, Everts, and R. Buchanan. 2015. Assessment of region, farming system, irrigation source and sampling time as food safety risk factors for tomatoes. Int . J . of Food Microbiol . 1 96 : 98 108. 44. Painter , J . A . , R. M. Hoekstra, T. Ayers, R. V. Tauxe, C. R. Braden, F. J. Angulo, and P. M. Griffin. 2013. Attribution of Foodborne Illnesses, Hospitalizations, and Deaths to Food Commodities by using Outbreak Data, 199 8 2008. Emerg . Infec . Dis . 19(3):407 415 . 45. Scallan , E., P. M. Griffin, F. J. Angulo, R. V. Tauxe, and R. M. Hoekstra. 2011. Foodborne illness acquired in the United States unspecified agents. Emerg. Infect. Dis. 17(1):16 22. 46. Scharff , R . L. 2012. Economic Burden from Health Losses Due to Foodborne Illness in the United States. J. Food Prot . 75(1):123 1 31. 47. Shirazi , M . and L. Boersma. 1984. A Unifying Quantitative Analysis of Soil Texture. Soil Sci . Soc . Am . J . 48 : 142 1 47. 48. Sinton , L . W . , R. R. Braithwaite, C. H. Hall, and M. L. Mackenzie. 2007. Survival of Indicator and Pathogenic Bacteria in Bovine Feces on Pasture. Appl . Environ . Microbiol . 73(24):7917 79 25. 49. Sivapalasingam , S . , C. R. Friedman, L. Cohen, and R. V. Tauxe. 2004. Fresh Produce: a growing cau se of outbreaks of foodborne illness in the United States, 1973 through 1997. J . Food Prot . 67(10):2342 23 53. 50. Sjogren, R. E. 1994. Prolonged survival of an environmental Escherichia coli in laboratory soil microcosms . Water Air Soil Poll. 75(3 4):389 403. 51. Smith , M . S . , G. W. Thomas, R. E. White, and D. Ritonga. 1985. Transport of Escherichia coli Through Intact and Disturbed Soil Columns. J . Environ . Qual . 14(1):87 91. 52. Solomon , E . B . , S. Yaron, and K. R. Matthews. 2002. Transmission of Escherichia coli O157:H7 from Contaminated Manure and Irrigation Water to Lettuce Plant

PAGE 107

107 Tissue and Its Subsequent Internalization . Appl . Environ . Microbiol . 68(1):397 400. 53. Soupir , M . L . , S. Mostaghimi, E. R. Yagow, C. Hagedorn, and D. H. Vaughan DH. 2006. Transport Of Fe cal Bacteria From Poultry Litter and Cattle Manures Applied to Pastureland. Water Air Soil Poll . 169(1 4):125 1 36. 54. Swinnen , I . A . M . , K. Bernaerts, E. J. J. Dens, A. H. Geeraerd, and J. F. Van Impe. 2004. Predictive modelling of the microbial lag phase: a review. Int. J. Food Microbiol. 94(2):137 1 59. 55. Tate, R. L. 1978. Cultural and Environmental Factors Affecting the Longevity of Escherichia coli in Histosols. Appl. Environ. Microbiol. 35( 5):925 929. 56. Unc , A . and M. J. Goss. 2004. Transport of bacteria from manure and protection of water resources. Appl. Soil Ecol . 25(1):1 18. 57. U . S . Department of Agricultur e . 2000. National Organic Program. C.F.R. 7(205):203. 58. U . S . Department of Agriculture. 2011. Choose My Plate . Available at: http://www.choosemyplate.gov/ . Accessed 22 June 2015. 59. U . S . Department of Commerce. 2012. Per Capita Utilization of Selected Commercially Produced Fruits and Vegetables. Available at : https://www.census.gov/compendia/statab/2012/tables/12s0218.pdf . Acces sed 10 November 2013. 60. Wang , G., T. Zhao, and M. P. Doyle. 1996. Fate of enterohemorrhagic Escherichia coli O157:H7 in bovine feces. Appl. Environ. Microbiol. 62(7):2568 2570. 61. Zhai , Q . , M. S. Coyne, and R. I. Barnhisel. 1995. Mortality rates of fecal bact eria in subsoil amended with poultry manure. Bioresour . Technol . 54(2):165 1 69.

PAGE 108

108 BIOGRAPHICAL SKETCH Samantha was born and raised in Spartanburg, SC. In 2013 she graduated from f ood s cience and t echnol ogy, with a specification in human nutrition. Shortly after graduation, she moved to Gainesville, FL f ood s cience with Dr. Keith Schneider. She intends to graduate in August 2015. After graduation Samantha plans on moving to Turkey with her husband. She enjoys studio art, cooking, and writing in her spare time.