Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Use of estrous expression within reproductive management and its association with conception and ovulation… Burnett, Tracy Anne 2019

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2020_may_burnett_tracy.pdf [ 2.48MB ]
Metadata
JSON: 24-1.0388204.json
JSON-LD: 24-1.0388204-ld.json
RDF/XML (Pretty): 24-1.0388204-rdf.xml
RDF/JSON: 24-1.0388204-rdf.json
Turtle: 24-1.0388204-turtle.txt
N-Triples: 24-1.0388204-rdf-ntriples.txt
Original Record: 24-1.0388204-source.json
Full Text
24-1.0388204-fulltext.txt
Citation
24-1.0388204.ris

Full Text

USE OF ESTROUS EXPRESSION WITHIN REPRODUCTIVE MANAGEMENT AND ITS ASSOCIATION WITH CONCEPTION AND OVULATION RATES IN DAIRY COWS by Tracy Anne Burnett  B.Sc., The University of British Columbia, 2011 M.Sc., The University of British Columbia, 2014  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF  THE REQUIREMENTS FOR THE DEGREE OF  Doctor of Philosophy in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Applied Animal Biology)    THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)  December 2019  ©Tracy Anne Burnett, 2019  ii The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled:  USE OF ESTROUS EXPRESSION WITHIN REPRODUCTIVE MANAGEMENT AND ITS ASSOCATION WITH CONCEPTION AND OVULATION RATES IN DAIRY COWS    submitted by Tracy Anne Burnett in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Applied Animal Biology  Examining Committee: Dr. Ronaldo L.A. Cerri, Applied Animal Biology Supervisor  Dr. Marina von Keyserlingk, Animal Welfare Program Supervisory Committee Member  Dr. Daniel Weary, Animal Welfare Program University Examiner Dr. Raymond Ng, Computer Science University Examiner       iii Abstract  Automated technologies have been developed to improve dairy cattle reproductive efficiency, yet there is still need for better understanding of how these technologies can be used and determine how the information collected relates to key events important for fertility. The aims of this thesis were to determine 1) if automated activity monitors (AAM) can predict estrus and be used within reproductive management, and 2) the interrelationship between estrous expression, ovulation and fertility. In Chapter 2, I provide evidence that AAM can be successfully incorporated into reproductive management without impeding the outcomes of an AI protocol and that increased estrous expression is associated with improved fertility. Chapter 3 investigates if changes in rumen-reticular temperature can be used to detect ovulation. Rumen-reticular temperature is shown to increase at the time of estrus and then declines near the time of ovulation, but the magnitude increase at estrus is dependent on the intensity of estrous expression as well as temperature and humidity at the onset of estrus. In Chapter 4 I demonstrate that the intensity of estrous expression is associated with the timing and failure of ovulation, where cows with lesser estrous expression had shorter intervals from estrus alert to ovulation and lower ovulation rates. Finally, in Chapter 5 I summarize work that investigated if the administration of GnRH at the moment of AI could increase fertility of cows with reduced estrous expression by increasing ovulation rates and modifying progesterone concentrations post-AI. The administration of GnRH increased fertility of cows with lesser estrous expression, but did not affect ovulation or progesterone profiles. Future research is merited to further understand the relationship between estrous expression and fertility of dairy cows.  iv Lay Summary   Modern dairy farms are adopting new technologies to detect estrus and ovulation in dairy cows to increase the efficiency of their reproductive programs. This thesis set out to determine if these types of technologies can contribute to existing reproductive management programs and also investigated the relationship between the intensity of estrus and the fertility of dairy cows. Reduced estrous expression intensity was associated with lower fertility and changes in the timing and failure of ovulation. Finally, this thesis investigated if fertility of cows with reduced estrous expression could be modulated by instigating ovulation using the hormone GnRH. GnRH administration increased fertility, but these findings were not due to changes in ovulation. Future research is merited to further understand the relationship between estrous expression and fertility of dairy cows.        v Preface  A version of the materials in Chapters 2 and 4 have been accepted for publication: Chapter 2 - Burnett, T.A., A.M.L. Madureira, B.F. Silper, A.C.C. Fernandes, and R.L.A. Cerri. 2017. Integrating an automated activity monitor into an artificial insemination program and the associated risk factors affecting reproductive performance of dairy cows. J. Dairy Sci. 100:50050-5018; Chapter 4 - Burnett, T.A., L.B. Polsky, M. Kaur, and R.L.A. Cerri. 2018. Effect of estrous expression on timing and failure of ovulation of Holstein dairy cows using automated activity monitors. J. Dairy Sci. 101:11310-11320. Versions of the materials in Chapter 3 and 5 will be submitted for review: Chapter 3 - Burnett, T.A., L.B. Polsky, M. Kaur, and R.L.A. Cerri. Rumen-reticular temperature during estrus and ovulation in dairy cows: Effects of estrous expression.; Chapter 5 - Burnett, T.A., A.M.L. Madureira, J.W. Bauer, and R.L.A. Cerri. The impact of estrous expression and GnRH administration at the time of artificial insemination on conception risk. For all chapters, T.A. Burnett was the lead investigator, responsible for all major areas of concept formulation, experimental design, material interpretation, data collection and processing, statistical analysis, and manuscript composition. R.L.A. Cerri acted in the supervisory role by helping in the idea formulation, statistical analysis and providing input and editing of manuscript drafts.  All other collaborative authors were in some way involved with data collection, processing, and analysis. All projects received UBC Animal Care approval as follows: Chapter 2 - certificate number: # A10-0290; Chapters 3 and 4 - certificate numbers: # A15-0089 and #A10-0290; Chapter 5 - certificate number: # A18-0039.  vi Table of Contents   Abstract ................................................................................................................................................... iii Lay Summary ......................................................................................................................................... iv Preface ....................................................................................................................................................... v Table of Contents .................................................................................................................................. vi List of Tables .......................................................................................................................................... ix List of Figures .......................................................................................................................................... x List of Abbreviations .......................................................................................................................... xii Acknowledgements ........................................................................................................................... xiii Chapter 1: Introduction ...................................................................................................................... 1 1.1 The estrous cycle ........................................................................................................................................ 4 1.1.1 Estrus ....................................................................................................................................................................... 5 1.1.2 Metestrus................................................................................................................................................................ 7 1.1.3 Diestrus ................................................................................................................................................................... 7 1.1.4 Proestrus ................................................................................................................................................................ 8 1.2 Endocrine regulation of the estrous cycle ........................................................................................ 8 1.2.1 Endocrine regulation of ovulation ............................................................................................................. 11 1.3 Behavioural signs of estrus ................................................................................................................. 14 1.4 Adoption of artificial insemination by the dairy industry ....................................................... 16 1.5 Wearable technologies for the detection of estrus ..................................................................... 19 1.5.1 Automated monitors measures and what they measure ................................................................. 21 1.5.2 What impacts the effectiveness of AAMs for estrus detection? ..................................................... 25 1.5.3 Can AAM be used effectively within reproductive programs? ....................................................... 30 1.6 What impacts estrous behaviour? .................................................................................................... 31 1.6.1 Impacts of estrus on fertility ........................................................................................................................ 38 1.7 Thesis objectives ..................................................................................................................................... 45 Chapter 2: Integrating an automated activity monitor into an artificial insemination program and the associated risk-factors affecting reproductive performance of dairy cows ......................................................................................................................................................... 47 2.1 Introduction .............................................................................................................................................. 47 2.2 Materials and methods ......................................................................................................................... 49 2.2.1 Animals and housing ....................................................................................................................................... 50 2.2.2 Study design ........................................................................................................................................................ 51 2.2.3 Cow-level scoring .............................................................................................................................................. 52 2.2.4 Ultrasonography, cyclicity and pregnancy diagnosis ........................................................................ 53 2.2.5 Expression of estrus ........................................................................................................................................ 54 2.2.6 Statistical analyses ........................................................................................................................................... 54  vii 2.3 Results ......................................................................................................................................................... 56 2.4 Discussion .................................................................................................................................................. 60 2.5 Conclusions ............................................................................................................................................... 66 Chapter 3: Rumen-reticular temperature during estrus and ovulation in dairy cows: effects of estrous expression ........................................................................................................... 84 3.1 Introduction .............................................................................................................................................. 84 3.2 Materials and methods ......................................................................................................................... 86 3.2.1 Animals and housing ....................................................................................................................................... 86 3.2.2 Study design ........................................................................................................................................................ 87 3.2.3 Determination of estrus, ovulation and ovulation times .................................................................. 88 3.2.4 Rumen-reticular temperature data ........................................................................................................... 89 3.2.5 Statistical analyses ........................................................................................................................................... 90 3.3 Results ......................................................................................................................................................... 91 3.3.1 Rumen-reticular temperature ..................................................................................................................... 91 3.3.2 Temperature alerts .......................................................................................................................................... 93 3.4 Discussion .................................................................................................................................................. 94 3.5 Conclusions ............................................................................................................................................... 98 Chapter 4: Effect of estrous expression on timing and failure of ovulation of Holstein dairy cows using automated activity monitors ..................................................................... 106 4.1 Introduction ............................................................................................................................................ 106 4.2 Materials and methods ....................................................................................................................... 108 4.2.1 Animals and housing .................................................................................................................................... 108 4.2.2 Study design ..................................................................................................................................................... 109 4.2.3 Determination of estrus, ovulation and ovulation times ............................................................... 110 4.2.4 Expression of estrus ..................................................................................................................................... 111 4.2.5 Health scoring ................................................................................................................................................. 112 4.2.6 Ultrasonography and pregnancy diagnosis ........................................................................................ 113 4.2.7 Statistical analyses ........................................................................................................................................ 113 4.3 Results ....................................................................................................................................................... 115 4.3.1 Intervals from activity alert to ovulation ............................................................................................. 115 4.3.2 Ovulation failure and false alerts ............................................................................................................ 116 4.3.3 Estrous expression and fertility ............................................................................................................... 118 4.4 Discussion ................................................................................................................................................ 119 4.5 Conclusions ............................................................................................................................................. 126 Chapter 5: The impact of GnRH administration at the time of artificial insemination on conception risk and its association with estrous expression ..................................... 134 5.1 Introduction ............................................................................................................................................ 134 5.2 Materials and methods ....................................................................................................................... 137 5.2.1 Animals and housing .................................................................................................................................... 137 5.2.2 Study design ..................................................................................................................................................... 138 5.2.3 Determination of estrus, ovulation and pregnancy ......................................................................... 139 5.2.4 Blood collection and progesterone analysis ....................................................................................... 140  viii 5.2.5 Statistical analyses ........................................................................................................................................ 141 5.3 Results ....................................................................................................................................................... 142 5.3.1 Pregnancy per AI and ovulation .............................................................................................................. 142 5.3.2 Progesterone .................................................................................................................................................... 144 5.4 Discussion ................................................................................................................................................ 144 5.5 Conclusions ............................................................................................................................................. 150 Chapter 6: General Discussion ..................................................................................................... 159 6.1 Thesis findings ....................................................................................................................................... 159 6.2 Implications, limitations and future directions ......................................................................... 163 6.3 General conclusions ............................................................................................................................. 169 References .......................................................................................................................................... 170        ix List of Tables  Table 2.1: Frequency distributions of explanatory variables for the entire study and stratified by farm. .................................................................................................................................................................................. 68 Table 2.2: Pregnancy per AI (P/AI) and DIM (mean ± SE) for treatments (ACT vs. TAI), cows complaint with the treatment1 (ACT vs. TAI), and AI categories (estrus vs. timed AI) for the entire study and stratified by farm. ...................................................................................................................... 70 Table 2.3: Pregnancy per AI (P/AI) and odds ratios (OR) with 95% confidence intervals (95% CI) for pregnancy outcomes for the entire study and for cows that were inseminated correctly according to their assigned treatment (compliant to treatment). ........................................................... 72 Table 2.4: Pregnancy per AI (P/AI) and odds ratios (OR) with 95% confidence intervals (95% CI) for pregnancy outcome interactions for the entire study and for cows that were inseminated correctly according to their assigned treatment (compliant to treatment) using a multivariable logistic regression. ....................................................................................................................................................... 74 Table 2.5: Pregnancy per AI (P/AI) and odds ratios (OR) with 95% confidence intervals (95% CI) for pregnancy outcomes stratified by farm....................................................................................................... 76 Table 2.6: Factors impacting the proportion and odds ratio (OR) with 95% confidence intervals (95% CI) of cows detected in estrus after a presynchronization protocol using an automated activity monitor. ........................................................................................................................................................... 78 Table 2.7: Factors impacting estrous expression parameters: peak activity1 and duration2 (LS means ± SE) as measured by an automated activity monitor. ................................................................................. 79 Table 3.1: Descriptive statistics of time intervals between the time of ovulation, automated activity monitor (AAM) alerts and temperature alerts for estrus an ovulation based on standard deviation changes in rumen-reticular temperature in dairy cows. On average, temperature alerts came after AAM alerts for estrus, but before ovulation. ................................................................. 99 Table 4.1: Descriptive statistics of ovulation timing variables. ....................................................................... 127 Table 4.2: Associations of intervals from activity alert to ovulation and estrous expression traits measured using two automated activity monitors. .................................................................................... 128 Table 4.3: Factors impacting ovulation failure and false alerts using automated activity monitoring systems. ......................................................................................................................................................................... 129 Table 4.4: Factors impacting pregnancy per AI including all true events and when excluding non-ovulated events. ......................................................................................................................................................... 131 Table 5.1: Cow characteristics (mean ± SD)  for all estrus events that were inseminated within the study, separated by farm. ...................................................................................................................................... 152  x List of Figures  Figure 1.1: Schematic drawing of follicular growth and the secretion of progesterone (P4; orange), follicle-stimulating hormone (FSH; navy blue) and luteinizing hormone (LH; green). Growing follicles are depicted as yellow circles, atretic follicles as red circles and ovulation as the half open circle. This figure is reprinted from Forde et al. (2011b). ................................................................. 5 Figure 1.2: Time intervals (hr  SD or SE) between key physiological events preceding ovulation of Holstein dairy cows. .................................................................................................................................................... 13 Figure 2.1: Schematic figure of the reproductive treatments used. Cows were assigned randomly to two treatments after a presynchronization protocol. Cows on the ACT treatment were bred by detection of estrus, and the remaining unbred cows were enrolled in an Ovsynch 56 protocol and bred by timed AI. Cows on the TAI treatment were all enrolled in an Ovsynch 56 protocol and bred by timed AI. ................................................................................................................................................. 80 Figure 2.2: Survival curves of DIM to conception until 300 DIM for: A) treatment (P > 0.05), B) cows compliant to both treatments (P > 0.05), C) AI category (insemination at estrus or timed AI) (P = 0.03), D) treatment by cyclicity interaction (P = 0.04), and E) farm by leg health interaction (P = 0.08). .............................................................................................................................................................................. 81 Figure 3.1: Area under the curve of rumen-reticular temperature relative to baseline at the time of estrus and ovulation for estrus events with high and low peak activity (panel A; P < 0.001) and long and short duration of estrus (panel B; P < 0.001). Superscripts of letters a-d denote significant differences (P < 0.05), while letters x-y denote tendencies (0.05  P < 0.10). High peak activity: estrous expression greater than the median of 80 index on the AAM. Low peak activity: estrous expression less than the median. Long duration: estrous expression lasting longer than the median of 12 hr. Short duration: estrous expression lasting less than the median. High THI: THI at the onset of estrus greater than 72. Low THI: THI at the onset of estrus less than 72. ................................................................................................................................................... 101 Figure 3.2: Maximum change in rumen-reticular temperature at the time of estrus and ovulation for estrus events with high and low peak activity (panel A; P < 0.01) and long and short duration of estrus (panel B; P < 0.01). Superscripts of letters a-d denote significant differences (P < 0.05), while letters x-y denote tendencies (0.05  P < 0.10). High peak activity: estrous expression greater than the median of 80 index on the AAM. Low peak activity: estrous expression less than the median. Long duration: estrous expression lasting longer than the median of 12 hr. Short duration: estrous expression lasting less than the median. High THI: THI at the onset of estrus greater than 72. Low THI: THI at the onset of estrus less than 72. ........................................ 102 Figure 3.3: Maximum negative change in rumen-reticular temperature at the time of estrus and ovulation for estrus events with high and low peak activity (panel A; P < 0.01) and long and short duration of estrus (panel B; P < 0.01). Superscripts of letters a-d denote significant differences (P < 0.05), while letters w-z denote tendencies (0.05  P < 0.10). High peak activity:  xi estrous expression greater than the median of 80 index on the AAM. Low peak activity: estrous expression less than the median. Long duration: estrous expression lasting longer than the median of 12 hr. Short duration: estrous expression lasting less than the median. High THI: THI at the onset of estrus greater than 72. Low THI: THI at the onset of estrus less than 72. .......... 103 Figure 3.4: Maximum postive (Panel A) and negative (Panel B) change in rumen-reticular temperature of at the time of estrus and ovulation for estrus events with varying body condition score. Superscripts of letters a-d denote significant differences (P < 0.05), while letters x-y denote tendencies (0.05  P < 0.10). Thin: BCS <2.75, Average: BCS = 2.75, Moderate: BCS > 2.75. ............................................................................................................................................. 104 Figure 3.5: Boxplots demonstrating the distribution of time intervals between automated activity monitor (AAM) alerts, alerts for estrus and ovulation based off of rumen-reticular temperature change, and the timing of ovulation. ................................................................................................................. 105 Figure 4.1: Frequency of estrus events for estrus alert to ovulation intervals relative to peak activity of estrus on the AAMC (panel A; P < 0.001) and the AAML (panel B; P < 0.01), and duration of estrus on the AAMC (panel C; P < 0.001). Peak activity was defined as the maximum activity during an episode of estrus. Duration of estrus was defined as the time the activity of the cow exceeded threshold values set by the AAM software. Peak activity and duration of estrus were classified above and below their respective medians relative to the AAM (80 index, 331 %-relative increase and 12 h for AAMC peak activity, AAML peak activity and AAMC duration, respectively). .............................................................................................................................................................. 132 Figure 5.1: The impact of GnRH administration at the time of artificial insemination on pregnancy per AI relative to the intensity of estrous expression. Superscripts of letters a-b denote differences (P < 0.05). High = estrous expression greater than the median of each farm. Low = estrous expression lesser than the median. ................................................................................................... 153 Figure 5.3: Survival curve analysis for ovulation for the interaction between GnRH administration at the moment of AI and estrous expression as measured using automated activity monitors (P = 0.17). High = estrous expression greater than the median of each farm. Low = estrous expression lesser than the median. ................................................................................................................... 155 Figure 5.4: The impact of GnRH administration at the time of artificial insemination on ovulation rates by 24 h (panel A), 48 h (panel B) and 7 d (panel C) post-alert relative to the intensity of estrous expression as measured using an automated activity monitor. Superscripts of letters a-d denote significant differences (P < 0.05), letters x-z denote tendency (0.05 < P < 0.10). High = estrous expression greater than the median of each farm. Low = estrous expression lesser than the median. .................................................................................................................................................................. 156 Figure 5.5: The interaction, with 95% confidence intervals, of the impact of GnRH administration at AI on pregnancy per AI dependant on the concentration of progesterone at estrus (P = 0.05). .......................................................................................................................................................................................... 158   xii List of Abbreviations  AAM – Automated activity monitor AI – Artificial insemination  AVPV – Anteroventral periventricular nucleus BCS - Body condition score  cAMP – Cyclic adenosine monophosphate DIM – Days in milk ECP – Estradiol-cypionate ER – Estrogen receptor FSH- Follicle-stimulating hormone  GnRH – Gonadotropin-releasing hormone  LH- Luteinizing hormone PR – Progesterone receptor PGF2 - Prostaglandin F2 TMR – Total mixed ration VMH - Ventromedial nucleus   xiii Acknowledgements  First and foremost, I would like to thank Ronaldo Cerri for teaching me all I know about research and dairy cattle, and becoming not only my supervisor but my friend throughout this process. His passion and encouragement has pushed me intellectually and nurtured all my aspirations within dairy cattle research – even if sometimes that meant he had to tell me to focus and to slow down. I would also like to thank my supervisory committee members, Nina von Keyserlingk and Murray Isman, for being so helping and understanding throughout my thesis writing process.  A special thanks to Dan Weary, Anne Marie de Passellé and Jeff Rushen for all their guidance and expertise throughout the years – I was extremely lucky to be surrounded by such great minds. I am tremendously thankful for Nelson Dinn, Barry Thompson, Brad Duncan, Ted Toenders, Bill Kramer and all the farmers who have come and gone, for making my time in Agassiz exciting, full of laughter and always passionate about cows – I’m still not sure how I will get by without my 5 uncles to look out for me ;) To my family, for their support throughout my PhD, particularly during times of high stress where all I needed was a little encouragement.  I have made so many amazing friends during my time as a graduate student, in both the Animal Welfare Program and on the Repro Team, it is impossible to list them all and is hard to imagine such a place could be filled by so many passionate and knowledgeable people. To the Repro Team members, both new and old – we still do not have sweaters – but at least we are amazing ;) Particularly I want to thank Augusto Madureira for his support over the years, both as a friend and colleague, may the infinite, never-ending stream of ideas live on. To Heather Neave and João Costa thank you for all the intellectual inputs and impromptu vacations all around the world! And for all my Blue House, and honorary Blue House, members - I will never forget the amazing times we spent together!   xiv And most importantly, I would like to thank the cows! Thank you for being such hilarious and curious creatures and bringing such joy to my life!  This may be my thesis, but I could never have done it without the help and encouragement from all the people that I have met along the way at UBC, in Agassiz, at conferences, and within the dairy industry.     1 Chapter 1: Introduction  In order for proficient outputs, modern dairies using year-round calving are required to have well-functioning reproductive programs that enable sufficient replacement rates, optimal milk yields and a consistent income during all targeted months of the year.  Poor reproductive management, and consequently low reproductive outcomes on farms, have been shown to be costly for producers (Plaizier et al., 1997; Inchaisri et al., 2010). Various economic models demonstrate that reproductive inefficiency and long delays in pregnancy can reduce farm profitability (Groenendaal et al., 2004; Meadows et al., 2005). Reduced reproductive performance has been associated with declines in economic returns due to reduced milk yield per cow per day, lower production of replacement heifers, and increased culling (Oltenacu et al., 1981). Given the profound impact that reproductive management has on overall farm profitability it is not surprising that this aspect of management is one of the top concerns for producers (Bauman et al., 2016; Denis-Robichaud et al., 2018a)  Economic returns are tightly bound to farm productivity (Krpálková et al., 2014). Shortly after calving, cows go through a period of maximal milk production, usually peaking between 60 and 100 days in milk (DIM), after which production will naturally decrease throughout the remainder of the lactation. In order to optimize the lactation curve, it is important that cows become pregnant in a timely manner to increase the amount of time a cow can be in peak milk production throughout her productive life (Weller and Folman, 1990); conception occurring late in lactation will result in a cow producing low amounts of milk for a longer time before being dried off before calving, yet still accumulating feed costs.  In contrast, calving to conception intervals that are too short are detrimental for milk  2 production (Weller and Folman, 1990) as the natural decline in milk production becomes exacerbated once they become pregnant (Bertilsson et al., 1997; Olori et al., 1997). The interval between calving and conception is important for optimizing lactational milk yield per cow.  Infertility has been directly associated with the culling of dairy cattle (Milian-Suazo et al., 1989; Bell et al., 2010; Nor et al., 2014), with anywhere from 16- 36% (Beaudeau et al., 1993; Hadley et al., 2006; Brickell and Wathes, 2011) of cows being culled due to reproductive reasons. In a large US study of 727 large herds followed over 6 years, non-pregnant cows were 5 to 7 times more likely to be culled than pregnant cows as DIM increased (De Vries et al., 2010). Moreover, De Vries et al. ( 2010) also reported that even in pregnant cows, culling risk was increased as days to conception increased, again suggesting that the time interval from calving to conception is an important factor associated with culling even for those cows that eventually become pregnant.  As a tool for better reproductive management, a large proportion of dairy farms in North America use hormone protocols to synchronize ovulation in dairy cows (Caraviello et al., 2006c; Denis-Robichaud et al., 2016). This practice has many benefits, but the use of reproductive hormones has been questioned, particularly in some parts of the European Union (EU) (Pieper et al., 2016). For example, consumer reluctance to purchase products which are produced from animals that are treated with antibiotics or hormones is one of the reasons, along with cost, that synchronization protocols are no longer commonly used on farms in the EU (Chanvallon et al., 2014). This is not surprising given public concerns regarding the use of exogenous hormones in food animal production systems.  This growing concern has resulted in veterinarians (Higgins et al., 2013), the dairy industry, and research  3 institutions having to better rationalize the use of pharmacological interventions in reproductive programs (Saint-Dizier and Chastant-Maillard, 2012). The movement to minimize the use of these types of interventions has placed tremendous pressure on the research community to identify effective and efficient approaches of inseminating cows with low, or at least strategic, use of hormone therapies.  Challenges associated with reproduction are clearly one of the main reasons for culling of dairy cattle and thus identifying ways to better manage dairy cattle reproduction has been a key theme of scientific research over the last decades. Fertility of dairy cattle is multifactorial and can be impacted by such things as nutrition, health, environmental conditions, genetics, as well as insemination management (see reviews by Lucy, 2001 and Walsh et al., 2011). The current introduction and thesis will touch on many of these aspects but will focus on reproductive physiology and management practices for the insemination of cows with specific reference to inseminations at the time of behavioural estrus.  To fully understand the complexity of this topic I will first begin by describing the fundamentals of the estrous cycle in dairy cattle, including how it is regulated. During the second part of this introduction I will discuss wearable technologies used for the detection of estrus, and what factors impact their efficacy and their incorporation within reproductive management. I will then briefly review the available literature on which factors impact the expression of estrous behaviour. Finally I will conclude with a discussion on the association of estrous expression with fertility in dairy cattle, reflecting on potential driving forces of this association.    4 1.1 The estrous cycle  The estrous cycle is the fundamental unit of female reproduction. In dairy cattle the estrous cycle begins at puberty when the animal is about 6 -12 mo old and is comprised of 4 distinct phases of ovarian activity: estrus, metestrus, diestrus and proestrus, collectively one cycle takes approximately 18 -24 d to complete. Some, however, distinguish the cycle simply by the absence and presence of a forming or functional corpus luteum thus breaking the cycle into two broader categories: the follicular phase and luteal phase.  During the bovine estrous cycle, follicles grow in a wave-like pattern of 2 to 3 waves depending on the age and breed of the animal, followed by ovulation and the formation of the corpus luteum; if pregnancy does not occur, the corpus luteum regresses and the estrous cycle will repeat itself. To provide the reader with a broad overview I will briefly describe the events that make up each of the phases of the estrous cycle, and then I will discuss in detail how each of these events are regulated. Discussion of the estrous cycle will be specific to bovines unless specifically identified. A depiction of the main hormones and ovarian dynamics of the bovine estrous cycle is depicted in Figure 1.1 and detailed further below.        5 Figure 1.1: Schematic drawing of follicular growth and the secretion of progesterone (P4; orange), follicle-stimulating hormone (FSH; navy blue) and luteinizing hormone (LH; green). Growing follicles are depicted as yellow circles, atretic follicles as red circles and ovulation as the half open circle. This figure is reprinted from Forde et al. (2011b).    1.1.1 Estrus  Animals osculate between phases of sexual receptivity and non-receptivity, dependant on where they are within the estrous cycle. Estrus is the only phase in which the animal is sexually receptive to a mate and is also the shortest phase, only lasting 12 – 18 hr on average in dairy cows (Roelofs et al., 2004; Valenza et al., 2012; Madureira et al., 2015a). The estrus phase is characterized by the absence of a corpus luteum (and thus low concentrations of progesterone) and the presence of a large dominant pre-ovulatory follicle. The pre-ovulatory follicle contains the oocyte and high concentrations of estradiol, which is a hormone highly researched for its role in the characteristic estrual behaviours that occur  6 during this phase of sexual receptivity; although there are also works that point to GnRH as potentially having a more direct role in the manifestation of estrual behaviours, as detailed below. Along with behavioural signs of estrus, other signs of estrus include increased production of cervical mucus with a reduced viscosity (Rutllant et al., 2005) and electrical resistance (Rorie et al., 2002), swelling of the vulva (Diskin and Sreenan, 2000a), increased uterine tone (Bonafos et al., 1995) and increased body temperature (Suthar et al., 2011).  Estradiol begins to increase once the corpus luteum of the previous cycle regresses sufficiently; progesterone begins to decrease approximately 60 h hr prior to standing estrus, while estradiol begins to increase about 40.0 h prior (Dieleman et al., 1986). Plasma estradiol peaks near the beginning of the onset of estrus and then decreases (Chenault et al., 1975); this is paralleled within the fluid of the pre-ovulatory follicle, where estradiol concentrations decrease rapidly 6 hr after the luteinizing hormone  (LH) surge (Dieleman et al., 1983b). This decrease in estradiol is thought to be related to the negative impacts the LH surge has on androstenedione within the follicle (Dieleman et al., 1983a). Although production of estradiol occurs in the granulosa cells of antral follicles through the stimulatory effects of follicle-stimulating hormone (FSH) (Campbell et al., 1995), granulosa cells lack the ability to produce androgens. When stimulated by LH, the theca cells produce androgens, specifically androstenedione, that are then used a precursors for estradiol within the granulosa cells (Hansel and Convey, 1983). Consequently, reduction in androstenediones due to the LH surge results in a drop in estradiol before ovulation (Dieleman et al., 1983a).     7 1.1.2 Metestrus  This phase begins with ovulation which occurs in response to a surge of LH; releasing the oocyte into the oviduct. Ovulation occurs approximately 26 to 30 hr after the onset of estrus (Walker et al., 1996; Roelofs et al., 2005b; a). Within this phase, the granulosa and theca cells of the of the ovulated follicle luteinize, forming the corpus luteum, and begin to produce progesterone. Progesterone concentrations have been reported to increase approximately 24 hr after the LH surge (Dieleman et al., 1986). Progesterone concentrations increase as the corpus luteum grows, and a new wave of follicles is initiated from the secretion of FSH by the anterior pituitary. Metestrus typically lasts for 3 – 4 d.     1.1.3 Diestrus  This is the longest of the phases, lasting approximately 12 d. Within this phase the corpus luteum becomes fully mature, producing progesterone, and is now responsive to prostaglandin F2 (PGF2). During diestrus, waves of follicles continue to cycle, but the presence of a negative feedback loop prevents another ovulation from occurring. Diestrus will continue if pregnancy occurs, but will progress to proestrus (in the response to the secretion of PGF2 from the uterus, causing the corpus luteum to regress) in the event that pregnancy does not occur. During diestrus, progesterone has a negative feedback on estradiol and LH. Although new follicles will grow during this phase, they only produce low levels of estradiol as they are thought to lack the ability to produce estradiol, coupled with the lack of LH, which is necessary to stimulate thecal androgen synthesis, as described above (Dieleman et al.,  8 1983b, 1986); thus, the negative feedback of progesterone on LH will directly reduce the amount of estradiol produced by the follicle and inevitably hinder ovulation.   1.1.4 Proestrus  Proestrus is characterized by the regression of the corpus luteum and development of the pre-ovulatory follicle; proestrus ends at the manifestation of estrual behaviours where the estrus phase will begin once again. The regression of the corpus luteum, as measured by blood progesterone, has been reported to take approximately 20 hr (Dieleman et al., 1986).  Under synergistic action of FSH and LH, the dominant follicle will increase in size, as it grows the internal concentration of estradiol will also increase, acting on a positive feedback mechanism to continue the growth of the follicle. On average, proestrus lasts between 3-4 d.  1.2 Endocrine regulation of the estrous cycle  The estrous cycle is primarily regulated by 6 reproductive hormones: estradiol, GnRH, progesterone, FSH, LH, and PGF2. These hormones act on a cohort of different organs, including the hypothalamus, anterior pituitary gland, ovary, and uterus. Each of these organs play a unique role in the progression of the estrous cycle. In this section I will review how the estrous cycle is regulated. The Ventromedial Nucleus (VMH) and the Medial Preoptic Area (MPA) , which includes the anteroventral periventricular nucleus (AVPV), are the main nuclei in the  9 hypothalamus which control reproductive behaviour and govern the estrous cycle (Woelders et al., 2014). These nuclei respond to stimuli from the environment or different types of chemosensory cues, such as pheromones; in fact, these areas contain the most dense populations of estrogen receptors (ERs) (Simerly, 1998). These regions integrate hormonal and behavioural stimuli (i.e. chemosensory and tactile information). The AVPV functions in mediating hormonal feedback on gonadotropin secretion. Projections from the AVPV provide direct inputs to gonadotropin-releasing hormone (GnRH)-containing neurons found in the preoptic region, and are thought to also influence LH secretion (Simerly, 1998). There is feedback between estradiol and GnRH, albeit indirect, as GnRH neurons do not express ERs or progesterone receptors (PRs) (Simerly et al., 1996). Furthermore, the neurons of the AVPV express both ERs and PRs suggesting that the AVPV is a location that is able to integrate both the effect of estradiol and progesterone on gonadotropin secretion (Simerly, 1998). Using rats, Simerly et al. (1996) demonstrated that the expression of ERs and PRs on the AVPV are influenced by circulating concentrations of both estradiol and progesterone, exhibiting the presence of feedback mechanisms; these findings have also been demonstrated in other locations of the hypothalamus (Simerly, 1998). Once GnRH is secreted from specialized neurons in the hypothalamus, it travels via the hypophyseal portal blood system to the pituitary gland where it binds to GnRH receptors on the gonadotroph cells (Forde et al., 2011b). Upon stimulation from GnRH, gonadotroph cells secrete LH and FSH from their cytoplasm (Weck et al., 1998). Dependant on the stage of the estrous cycle, a GnRH surge can be induced by increased estradiol concentrations produced by the pre-ovulatory follicle; this GnRH surge will induce an LH surge, and with sufficient frequency and amplitude of LH ovulation may occur (Forde et al., 2011b).  Shortly  10 after the LH surge,  estrous behaviours have been shown to decline rapidly (van Eerdenburg et al., 2000); this is suggested to be advantageous as it ensures that the cow is only receptive to a mate when chances of successful conception are highest (van Eerdenburg et al., 2000).  Estradiol stimulates a multitude of mechanisms in the brain which can be grouped as those that are genomic and those that are non-genomic and growth factor-dependant (Woelders et al., 2014). Mechanisms that are genomic are associated with the activation of ERs which stimulate the transcription of certain products, such as oxytocin and PRs (Vasudevan and Pfaff, 2008) by binding to specific DNA sequences and inducing gene expression (Panin et al., 2015). In fact, binding of estradiol with ERs has been shown to increase its own transcription, as seen by the positive feedback mechanism which controls estradiol production (Jungblut et al., 1976). The non-genomic actions of estradiol are complimentary to the genomic actions and occur in the cytoplasm or plasma membrane of neurons or target cells. The non-genomic mechanisms of estradiol are associated with the activation of second messenger systems such as cyclic adenosine monophosphate (cAMP) and intracellular calcium pathways (Vasudevan and Pfaff, 2008; Woelders et al., 2014); these non-genomic actions are much more rapid than those of the genomic (Vasudevan and Pfaff, 2008).  Although much research has suggested estradiol is the regulatory hormone for estrual behaviours (Lyimo et al., 2000; Roelofs et al., 2004), there is a body of work suggesting the true regulatory hormone is GnRH (van Eerdenburg, 2008). During proestrus ERs in the hypothalamus are actually down regulated and are completely absent at the time of estrus (van Eerdenburg et al., 2000). In fact, estradiol concentrations are already high from at least 15 hr prior to the LH surge and onset of estrual behaviours surge (Cook et al., 1986; Dieleman et al., 1986; van Eerdenburg, 2008), suggesting that there may be other more important  11 factors that initiate estrous behaviour. GnRH has been mentioned as a regulator of sexual receptivity in other species (Jennes et al., 1997; Caraty et al., 2002), but not yet well researched in cattle. When investigating the relationship of GnRH with estrous expression intensity, using LH as a proxy, van Eerdenburg (2008) reported a much closer synchrony in changes in GnRH with estrous expression than with changes in estradiol concentrations. This is supported by others that could not find a relationship between estradiol concentrations and estrous expression intensity (Cook et al., 1986; Coe and Allrich, 1989; Madureira et al., 2015a). Future research  is required to elucidate the relationship between estradiol, GnRH and estrous expression intensity.  1.2.1 Endocrine regulation of ovulation   As the dominant follicle grows, the continuously increasing amount of estradiol eventually leads to the GnRH and LH surges necessary for ovulation. The LH surge occurs at about 18 hr or between 18 and 21 h following estradiol treatment in pre-pubertal heifers (Swanson and McCarthy, 1978) or beef cows (Gazal et al., 1998), respectively. Surges of LH are highly correlated to surges of GnRH, both in magnitude and in frequency (Gazal et al., 1998). Estradiol has also been associated with the magnitude of the LH surge (Swanson and McCarthy, 1978; Lammoglia et al., 1998). For ovulation to occur, the LH surge must be of sufficient frequency and amplitude; pulses of the pre-ovulatory LH surge will occur approximately every 40-70 mins (Roche, 1996) and last 8 to 10 hr (Chenault et al., 1975; Bernard et al., 1983; Dieleman et al., 1986). A detailed schematic of the timing of events surrounding ovulation is presented in Figure 1.2.   12 The LH surge has been shown to be inhibited by progesterone, but stimulated by increased estradiol concentrations (Kesner et al., 1982) due to the stimulation of GnRH production (Hansel and Convey, 1983). The ability of the pituitary gland to release LH in response to GnRH is greatest during estrus and increased with continued exposure to GnRH (Hansel and Convey, 1983). This is thought to be driven by a priming effect of GnRH on the anterior pituitary gland which increases the amount of LH and FSH released by a standard dose of GnRH (Crighton and Foster, 1977); the priming effect is increased in the presence of estradiol but inhibited in the presence of progesterone (Padmanabhan et al., 1982). The increased sensitivity of the pituitary gland to every sequential pulse of GnRH leads to a GnRH surge followed by the LH surge. 13 Figure 1.2: Time intervals (hr  SD or SE) between key physiological events preceding ovulation of Holstein dairy cows.  * denotes publications which detected estrus using automated activity monitors, while all else detected estrus visually. 1Valenza et al., 2012; 2 Chenault et al., 1975; 3 Stevenson et al., 1998; 4 Rajamahendran et al., 1989; 5 Walker et al., 1996; 6 Hockey et al., 2010; 7 Aungier et al., 2015; 8 Bloch et al., 2006; 9 Dieleman et al., 1983b 14   1.3 Behavioural signs of estrus  Estrual behaviours can be separated into primary and secondary signs. Standing immobile while being mounted, is the primary estrous behaviour often referred to as the gold standard (Van Eerdenburg et al., 1996) given that this behaviour is not performed outside of the estrus phase (Woelders et al., 2014). Other secondary signs of estrus include mounting, chin resting and sniffing of other cows (Kerbrat and Disenhaus, 2004; Roelofs et al., 2005b), increased walking and agnostic interactions (Hurnik et al., 1975), and decreased resting and eating times (Hurnik et al., 1975). Chin resting and sniffing have been found to occur the most, while standing to be mounted is the least frequent behaviour; reported to only occur in 35 – 60% of estrus events (Van Eerdenburg et al., 1996; Roelofs et al., 2005b). The frequency of behaviours vary considerably between animals as well (Kerbrat and Disenhaus, 2004; Roelofs et al., 2005b), especially standing to be mounted with some reporting a range of 7 – 10 mounts per estrus event (Walker et al., 1996; Lopez et al., 2004; Rivera et al., 2010). The length of standing estrus (time from first standing mount to the last) also varies, ranging from 5.5 – 9.5 h (At-Taras and Spahr, 2001; Lopez et al., 2004). However, the expression of secondary signs starts before standing estrus and ends after (Reith and Hoy, 2018). For example, Dobson et al. (2018) reported the length of estrus based on secondary signs was 14 h (ranging from 8.5 to 29 h) where as the primary behaviour – standing to be mounted - only lasted 5 h (0.25 to 18.25 h).  Estrous behaviour is affected entirely by the hormone environment. For example, Vailes et al. (1992) artificially manipulated progesterone and estradiol in ovariectomized cows and showed that animals that had high estradiol and low progesterone engaged in  15   estrual behaviours and consistently stood to be mounted compared with high progesterone and basal estradiol, which exhibited the least number of estrous behaviours. Historically, some have argued that estrous behaviour occurs on an “all or nothing” basis, requiring a threshold amount of estradiol that triggers the suite of behaviours, after which any additional estradiol will not have any affect (Allrich, 1994). However, more recent work has shown that the threshold level of estradiol required to trigger estrus varies from one cow to another (Reames et al., 2011). A number of studies across species using ovariectomized females provide evidence that increasing concentrations of estradiol as well as increased frequency of doses affects both behaviour and intensity of estrous expression (rats, Hardy and DeBold, 1971; ewes, Fabre-Nys et al., 1993; goats, Sutherland and Lindsay, 1991). In contrast the available evidence for dairy and beef cattle is less consistent: some studies reporting that estrus intensity and duration are correlated with estradiol concentration (Van Eerdenburg et al., 1996; Lopez et al., 2004; Lammoglia et al., 1998) but not in others (Madureira et al., 2015a). The nature of this relationship was also explored by Fabre-Nys et al. (1993) who provided evidence in ewes and goats that the duration of estrual behaviours may be associated with the length of time in which higher concentrations of estradiol is circulating the body, and not just maximum concentration. Progesterone has been suggested to have a role in priming the hypothalamus to be receptive to estradiol and thus involved in the occurrence of estrus (Woelders et al., 2014). States of prolonged high estradiol concentrations can result in animals becoming unresponsive to estradiol, and this state of refractoriness can be remedied with treatment of progesterone (Carrick and Shelton, 1969). This phenomenon is hypothesized to be the main  16   reason why early postpartum cows are known to have silent estrus (failure to show primary or secondary behaviours of estrus) for their first ovulations and is the primary argument in favor of progesterone priming (Woelders et al., 2014). Progesterone priming has been demonstrated necessary for some species to demonstrate estrus (i.e. ewes (Fabre-Nys et al., 1993), but not in others (i.e. goats, Sutherland and Lindsay, 1991; cattle, Davidge et al., 1987). Melampy et al. (1957) demonstrated that after injecting a small dose of estradiol (200 g), ovariectomized cows became restless but did not perform mounting behaviours. By pre-treating the cows with progesterone (1 mg i.m. 12 h before) prior to the administration of estradiol, all cows became receptive to a bull. Despite these advances, the mechanism behind progesterone priming is not well understood.  There is some thought that progesterone increases the number of ERs on the hypothalamus and increases estradiol production (Rhodes et al., 2003). Treatment of non-cycling postpartum cows with progesterone has shown to increase pulsatile release of LH (Rhodes et al., 2002), although, Kyle et al. (1992) reported no advantages in terms of days to first postpartum ovulation or proportion of cows expressing estrus at first, second or third ovulations postpartum when treating cows with progesterone from d10 to d15 of lactation regardless of cyclicity.   1.4 Adoption of artificial insemination by the dairy industry  Artificial insemination using frozen semen was first introduced in 1947 (Stevenson and Britt, 2017), which enabled farmers to move away from having to manage bulls as well as allowing for large genetic gains over a shorter period of time.  However, this move also  17   required the adoption of determining how to manage the timing of inseminations to ensure conception. Although historically behavioural signs of estrous expression was used as a marker for ovulation, estrus has become increasingly difficult to detect in dairy cows. For example, mounting behaviour in the late 1970’s and 80’s was reported upwards of 20 standing mounts per estrus (Hurnik et al., 1975; Esslemont and Bryant, 1976; Pennington et al., 1986), this has now declined to less than 5 standing mounts per estrus (At-Taras and Spahr, 2001; Rivera et al., 2010). The length of estrus has also reduced over the last 50 years, declining from 18 h to less than 8 h (Reames et al., 2011). It appears that in addition to estrous expression intensity, the proportion of animals displaying estrus has declined from approximately 80% to 50% (Dobson et al., 2008). Beginning in the 1960s, in addition to herd size increasing exponentially, milk production per cow has increased; these changes have also resulted in a growing percentage of cows being housed continuously indoors, often on concrete, and increased the cow to employee ratio (Stevenson and Britt, 2017). Housing cows on flooring surfaces that are wet and slippery have been related with increased lameness (Solano et al., 2015) and knee lesions (Zaffino Heyerhoff et al., 2014), both of which have negative associations with reproductive success (Lee et al., 1989; Barkema et al., 1994; Burnett et al., 2017b). All together these factors make the traditional method of visual detection of estrus for reproductive management harder to carry out effectively, pushing for new developments in reproductive management. Although hormones, such as progestins and prostaglandins, were previously known to aid in synchronizing estrus, the introduction of the Ovsynch protocol in 1995 (Pursley et  18   al., 1997; see review by Stevenson and Britt, 2017) that allowed for the timed artificial insemination (AI) dramatically changed how reproduction is managed on dairy farms.  Briefly, this protocol uses three hormone injections, one GnRH injection followed 7 d later by a PGF2 injection, followed 36 hr later by another GnRH injection, and culminated with AI performed 16 – 20 hr after (Pursley et al., 1997b). This protocol, and all its current variations, allows for the synchronization of ovulation such that cows can be inseminated at a fixed time, eliminating the need for estrus detection (for more details on synchronization protocols and their history see reviews by Bisinotto et al., 2014 and Wiltbank and Pursley, 2014). The use of timed AI protocols reduced labour requirements associated with visual estrus detection while enabling them to increase the rate at which cows were inseminated (Chebel et al., 2010) and reduce the variability of parturition to first service (Stevenson, 2001). The adoption of these protocols in North America has been widespread; 84% of Canadian farms (Denis-Robichaud et al., 2016) and 75% of farm in the US (Caraviello et al., 2006c) reportedly use reproductive hormones for inseminations through estrus or ovulation synchronization. As discussed above, the public have grave concerns regarding hormone use in the food animal industries (Frewer et al., 2013); and, when made aware, their demands have resulted in reduced use (i.e. the European Union) of these types of hormone technologies (Chanvallon et al., 2014). This increased pressure has now put into question the use of hormonal reproductive management tools in North America resulting in the need to develop alternative tools to aid reproduction management on dairy farms.   19   1.5 Wearable technologies for the detection of estrus   Given that behavioural estrus is increasingly hard to detect visually (Roelofs et al., 2010), herd sizes are increasing (Barkema et al., 2015) and labour costs are rising (Denis-Robichaud et al., 2018a), alternative strategies to improve reproductive management on farms are needed. Examples of low cost tools that aid in estrus detection that do not rely on electricity, the Internet or some form of networking, include tail chalk and or rump patches. These types of simple tools measure mounting activity by one cow onto another that causes rubbing off or activating some sort of dye or scratch card. Both types of tools have been shown to be able to detect cows in estrus (Pennington and Callahan, 1986; Xu et al., 1998; Rivera et al., 2004) at efficiencies similar to visual observation (Sawyer et al., 1986; Holman et al., 2011b); however, there is some evidence that they may have a high number of false positives under certain situations (Pennington and Callahan, 1986; Sawyer et al., 1986; Holman et al., 2011b) and may be easily lost (Williamson et al., 1972; Gwazdauskas et al., 1990). In 2004, tail chalk was reported to be the most common estrus detection aid used on large farms (Caraviello et al., 2006a); similar results were found in a survey of a few select farms (n = 16) in 2010 (Ferguson and Skidmore, 2013). It is suspected their use may have declined due to the daily hands-on care needed to keep them reliable (Foote, 1975) and likely the development of new technologies. Electronic mount detectors have also been developed and record when mounts occur; standing estrus is defined as 3 or more mounts of greater than 2 s in a 4 hr period (Diskin and Sreenan, 2000a; Peralta et al., 2005). These types of monitors have been shown to have comparable (Cavalieri et al., 2003b; Peralta et al., 2005; Palmer et al., 2010) or better (At- 20   Taras and Spahr, 2001) detection efficiency when compared with visual observation and the use of tail chalk (Palmer et al., 2010). However, the characteristics of estrous behavior (e.g., number of mounts, duration of each mount) seem to be underestimated by this type of technology and have reduced agreement with visual detection (Cavalieri et al., 2003b; Palmer et al., 2010).  Research has also been carried out on the use of electrical resistance of vaginal mucus, with resistance reducing on the day of estrus and increasing during the luteal phase (Lewis et al., 1989). These changes in electrical resistance are thought to be due to the hydration of the vulvar tissue, seen as edema (Senger, 1994). The efficiency and accuracy of using electrical resistance of vaginal mucus has been shown to vary from 65 to 82% and  57 to 82%, respectively (Lehrer et al., 1992), however, the technology has not be greatly adopted. Although these types of estrus detection aids are generally quite efficient, they still require one-to-one contact of the farmer with the cow, whether to physically take a reading (e.g., electrical resistance of vaginal mucus) or to replace and assess the device (e.g., tail chalk). Although the two latter examples have an electrical component, they are not fully automated. Automated technologies are ideally allocated to an animal only once and have the ability to store and transmit data to a main server where the data can be filtered using an algorithm. The intent being that an alert can be produced, usually based on deviations from that cow’s own baseline, that informs the stockperson that the cow can be AI.      21   1.5.1 Automated monitors measures and what they measure  Although originally developed in the 1980’s and 90’s, AAMs have started to gain traction on modern dairy farms (Stevenson and Britt, 2017). As with the introduction of most technologies, the adoption of AAMs requires some initial outlay in cost and labour, but once implemented, it is suggested that less labour is required on the cow-level, however, there is little to no research to prove or disprove these claims (Hostiou et al., 2017). In a recent on-line survey within the USA, 41.3% of 109 respondents stated that they had already adopted AAMs for estrus detection and, after mastitis, detection of estrus was the second most useful technology (Borchers and Bewley, 2015). In contrast, a survey in Canada, reported that only 23% of surveyed farms (n=832) used AAM systems for estrus detection; but this was driven almost exclusively by free-stall farms as the percentage increased to 56% when tie-stall herds were excluded (Denis-Robichaud et al., 2016). Interesting is that these reports are much higher than the 8% of AAM farms surveyed a decade before by Caraviello et al. (2006). There has been a variety of AAMs commercially produced for the dairy industry (see review by Roelofs and van Erp-van der Kooij, 2015). These technologies primarily rely on the following: general activity, steps, lying time, lying bouts, neck movements, rumination, and changes in temperature (Roelofs and van Erp-van der Kooij, 2015). Pedometers have been shown in many studies to be able to detect cows in estrus with good efficiencies (Firk et al., 2002) and positive predictive values (Madureira et al., 2015a), and have been useful for the prediction of ovulation (Roelofs et al., 2005a). One experiment that compared steps at estrus, using pedometers, with a visual behavioural score reported steps to be correlated with an overall behavioural score, with chin resting and successful mounts of other cows,  22   but not with sniffing, attempted mounts or with standing to be mounted (Van Vliet and Van Eerdenburg, 1996). Pedometer readings have also been shown to be correlated with mounting, standing to be mounted, chin resting, sniffing, licking the back of other cows, and butting, but not licking the front of other cows and disoriented mounting (Pennington et al., 1986). Generally, pedometers have been found equally efficient at detecting cows in estrus as visual estrus detection, where in order for visual detection to be superior to a pedometer, visual detection must be performed at least 4 to 6 times per day (Pennington et al., 1986). Many commercially available pedometers are able to measure both steps and lying behaviours. As cows are known to walk more during estrus, it is not surprising that lying time and lying bouts have also been shown to decrease on the day of estrus (Silper et al., 2017; Zebari et al., 2018). Using pedometers, nulliparous heifers have shown to decrease lying time by 36% (Silper et al., 2015). Lying time and lying bouts were also reported to be reduced during estrus at the end of a timed AI protocol using estradiol, where both measures were reduced by approximately 35% (Silper et al., 2017).  Similarly, in lactating cows, lying bouts have shown to decrease by 28% and lying time by 32%  during spontaneous estrus events (Zebari et al., 2018). This latter study also noted a decrease in feed intake, feeding duration, and number of visits to feed bins, with a negative correlation between the number of steps and feed intake, demonstrating that when walking behaviours are increased other daily activities (e.g., lying and eating) are decreased. Using what they termed lying-balance, a function of whether a cow is lying more or standing more, Jónsson et al. (2011) demonstrated that by using both lying behaviour and steps in an algorithm they were able to reduce the amount of false positives, which led to an increase it specificity.   23   Collar-mounted AAMs have had good success with estrus detection efficiencies with reports of 73 to 90% (Hockey et al., 2010; Aungier et al., 2012, 2015). Many collar-mounted AAMs not only measure activity but also include measures of rumination, a behaviour reported to decrease at the time of estrus (Reith et al., 2014; Dolecheck et al., 2015; Reith and Hoy, 2018). Cows will reduce their rumination by 19% (~90 min/d) during estrus;  in one study, 86% of cows decreased rumination during estrus (Reith et al., 2014).  The use of both activity and rumination was shown to increase the accuracy of a collar-mounted AAM (Kamphuis et al., 2012). Contrarily, Talukder et al. (2015) reported that with cows housed on pasture, an activity alert from a collar-mounted AAM generated more accurate alerts when only using activity data than either both activity and rumination or rumination alone.  The use of collar-mounted AAMs that measure both activity and rumination have been reported to have comparable efficiency and accuracy of estrus detection to twice daily visual observation by some (Michaelis et al., 2014) but not others (tail paint comparison; Kamphuis et al., 2012). These discrepancies have been suggested to, in part, be a consequence of cows in pasture systems having to walk greater and more variable distances (Kamphuis et al., 2012) which may result in a greater amount of false alerts.  Measurements of body temperature as a means to detect estrus have received much interest given that vaginal temperature initially increases during estrus and then declines around ovulation (Wrenn et al., 1958; Suthar et al., 2011). This rise in vaginal temperature has also been associated with the occurrence of the LH surge (Mosher et al., 1990; Fisher et al., 2008). In addition to vaginal temperature, this rise and fall in temperature has been reported in be present in the external auditory canal  (Randi et al., 2018), and also on the  24   skin located on the underside of the tail (Miura et al., 2017). Miura et al. (2017) made use of a wearable temperature monitoring device placed at the lower surface of the ventral tail base to measure temperature changes relative to the average of the last 3 d, at the same time each day. These authors reported a transient decrease in temperature change approximately 48 hr before ovulation, corresponding to the decrease of P4, then temperature increased to its maximum 24 hr before ovulation which corresponded to the LH surge, and then went back to basal levels around the time of ovulation (Miura et al., 2017). Changes in vaginal temperature at estrus are believed to be due to increased blood flow to the area due to increased concentrations of estradiol in the blood (Abrams et al., 1973). A finding supported by the fact that changes in vaginal temperature at estrus were observed in tie-stall housed cows, suggesting that changes in vaginal temperature occur at estrus even in the absence of physical exertion (Suthar et al., 2011). Using vaginal temperature with a threshold change of 0.4 0C above basal temperature for a minimum of 3 h, Kyle et al. (1998) reported achieving a higher sensitivity than using visual observations for the detection of estrus. Another study reported 69% accuracy and 81% efficiency of estrus detection using vaginal temperature with a threshold change of 0.3 0C above basal temperature for a minimum of 3 hr, a finding similar to using a pedometer to measure steps, and better than casual visual observation that had no strict observation schedule (Redden et al., 1993).  Skin temperature using infrared thermography has also been explored as a possible physiological measure to use for the detection of estrus. It appears that vulvar skin temperature has higher sensitivity and specificity than muzzle, ear, and eye temperature (Talukder et al., 2015). Full automation of the measurement of vulvar skin temperature using  25   infrared technology would be very difficult and likely not practical, since measurements can be obscured by fecal contamination, skin colour of the vulva, and the physical barrier of the tail covering the vulva (Talukder et al., 2014). Overall, infrared technologies have not been very successful for the detection of estrus (Redden et al., 1993; Talukder et al., 2014, 2015). However, other automatic methods of measuring body temperature are being developed, including measuring leg surface temperature (Kou et al., 2017), ear canal temperature (Randi et al., 2018), subcutaneous skin temperature (Morais et al., 2006; Lee et al., 2016) and rumen-reticular temperature (Bewley et al., 2008; Adams et al., 2013). Overall, although changes in body temperature have been associated with the occurrence of estrus, there is little research on how the intensity of estrous expression is associated with these changes; this is a topic that will be investigated further in Chapter 3 of this thesis.   1.5.2 What impacts the effectiveness of AAMs for estrus detection?  Although the largest predictor of true alerts is estrous behaviour (a large change in the performance of a specific behaviour would be easier to detect), much remains unknown regarding factors that affect the number of false alerts. Interestingly, Holman et al. (2011) described that out of 26 false alerts from a pedometer, 16 of them occurred in 7 cows, ranging from 4 to 2 false events per cow. Detailed observations showed that all false alerts occurred when at least one other cow was in estrus in the same pen, suggesting that these animals may be more likely to participate in a sexually active group even when not in estrus themselves. Improved understanding of individual variation in behaviour among cows both  26   in estrus and those not in estrus (but housed in the same pen) may help understand why certain cows are more prone to false alerts than others.  Different attributes of cows are associated with increased false positives while others have been associated with increased true positive alerts. Insemination of animals that are incorrectly diagnosed in estrus has negative impacts on reproductive programs due to inappropriate allocation of labour and costs, and increased chance of termination of any pre-existing pregnancies (Sturman et al., 2000). There is limited research on the specific cow-level factors which impact the accuracy of AAMs, but stage of lactation, lameness, uterine disease, milk yield and components, and BCS have been reported, although these factors are not consistent throughout studies.  Cows late in lactation have been reported be have more false alerts when using pedometers (Holman et al., 2011b), but others have found no impact of stage of lactation on true alerts using a collar-mounted AAM (Talukder et al., 2015). Contrarily, Aungier et al. (2012) reported that the odds of correctly identifying estrus events occurring at second or subsequent pre-ovulatory cycles were 8.2 times greater than for the first estrus events after calving while using a collar-mounted AAM; this potentially may be more a function of silent heats postpartum and not necessarily associated with stage of lactation, although the two factors are highly associated (Mwaanga and Janowski, 2000).  Using a collar-mounted AAM that measured both activity and rumination, mildly and severely lame cows were reported to have lower odds of generating a true estrus alert than sound animals (Talukder et al., 2015). However, others found no impact of lameness on false  27   (Holman et al., 2011b) or true alerts (Aungier et al., 2012). Additionally, cows with the presence of uterine disease at the time of estrus were less likely to have true estrus alerts than those with the absence of uterine disease (Aungier et al., 2012). Milk yield and percentage of milk protein have been reported to impact alerts, with animals producing less milk (Aungier et al., 2012), or a higher percentage of protein in milk (Talukder et al., 2015), more likely to have true estrus alerts. Cows with a higher percentage of protein in their milk have been argued by some to be in a better energy balance status than those with a lower percentage of protein and this may explain why they are more likely to be found in alert (Talukder et al., 2015). The odds of a true alert being correctly identified was improved by 1.4 for every increase in 0.25 BCS unit (Aungier et al., 2012), but an absence of  association of BCS with true (Talukder et al., 2015) and false alerts (Holman et al., 2011b) has also been reported; none of these three studies reported an impact of parity on the occurrence of true and false alerts. Anovulatory status, presence of purulent vaginal discharge at 5 wk postpartum, increased parity, and cows that were both lame and with low BCS had reduced odds of being found in estrus between 55 and 80 DIM when using an AAM (LeRoy et al., 2018). However, it should be noted that it is difficult to decipher if these changes are due to the efficiency of the AAM or decreased/ absent estrous behaviour.  The housing environment can have an impact on how effective AAMs can be at detecting estrus. In a study where they compared cows housed indoors in free-stalls and those on pasture, using electronic mount detectors, tail chalk, and visual observation, no difference in estrus detection efficiency or accuracy within housing type was found but all three types of estrus detection had higher efficiency when used on pasture compared with  28   indoors (Palmer et al., 2010). The exception being accuracy of detection which was only different for the electronic mount detector, and found to be less accurate indoors compared with when it was used on pasture. Contrarily, a different study using a leg- and a neck-mounted AAM, found no difference in specificity, sensitivity or positive predictive value between cows on pasture or those housed indoors (Roelofs et al., 2017). Reasons for these contrary results could be that Palmer et al. (2010) measured mounting activity while Roelofs et al. (2017) measured increased movement which was automatically corrected for herd-level activity by the AAM. Behavioural measures have been reported to vary more within a farm than between farms (i.e., lying time; Ito et al., 2009), thus it is not surprising that there is commonly an effect of farm when assessing the efficacy of AAMs across farms. For example, when comparing the accuracy between a mount detector and pedometers across three farms, Cavalieri et al. (2003a) reported that for 2 farms, accuracy was similar between the two estrus detection tools, but that on the third farm the accuracy for the mount detector was higher than for the pedometer. Another study found comparable specificity of a collar-mounted AAM between two herds, but one herd had a higher sensitivity (Hockey et al., 2010). It is not well understood why some farms are able to achieve greater success than others; a topic in need of additional research. Previous research has shown that false estrus alerts, measured using pedometers, have less steps per estrus event compared with true positive alerts (477  322 steps vs. 841  259 steps). Similarly, Aungier et al. (2012) reported that when using a collar-mounted AAM, the odds of a correct alert increased by 20% and 91% with every 1-unit increase in  29   peak activity and 2-h increase duration, respectively. Differences in the activity profiles of true and false alerts has potential for helping producers make decisions regarding when to breed. For example, cut points such as by requiring any alerted events to have a duration greater than 6 hr Aungier et al. (2012; collar-mounted AAM) can increase the efficiency of estrus detection from 72 to 87.5% and reduce the false positive rate from 32% to 21.3%, suggesting that by simply removing events that are more likely to be false alerts performance of the monitor can be improved. Alternatively, improvements may be achieved by simply comparing their duration with their intensity. The duration and intensity of estrus events have been shown to be highly correlated using many different AAMs (collar-mounted, Aungier et al., 2012; leg-mounted, Madureira et al., 2015a), thus by discounting events that have high intensity but short duration, or vice versa, the number of false alerts can be reduced; however, research is required to determine the efficacy of this practice. Although AAMs are able to detect cows in estrus at acceptable levels of efficiency and accuracy, a combination of estrus detection tools may still be warranted given the positive findings when using more than one tool (Firk et al., 2002; Holman et al., 2011b; Michaelis et al., 2014). Furthermore, studies where cows wore multiple detection tools, it is consistently reported that almost without exception in each study some cows are reported to having been detected by one or multiple AAMs but not others (Peralta et al., 2005; Holman et al., 2011b; Dolecheck et al., 2015). This difference in detection may be due to the effectiveness of the technology, but more likely due to differences between cows in their expression of certain behaviours during estrus; and given that some devices place more emphasis on certain behaviours certain cows may be more likely to be identified by one type  30   of technology than another. Although most AAM are based on behavioural changes relative to a cows’ own baseline, future research should focus on determining which types of behaviours are more or less likely to cause alerts, especially on AAMs which measure general movements in comparison to specific behaviours such as steps or lying. This information will aid in improving our understanding of estrous behaviours and also enable technologies to become more accurate.   1.5.3 Can AAM be used effectively within reproductive programs?  In a survey of Canadian farms, producers were asked to identify their main (>50%) reproductive management for first AI and this management practice was compared to their herd’s 21d-pregnancy rate, 21d-insemination rate, and conception risk (Denis-Robichaud et al., 2016). This comprehensive study noted no difference in pregnancy rates between herds with varying reproductive management strategies, but insemination rates were higher on farms that used timed AI and a combination of practices, in comparison with visual estrus detection alone, but not different than farms that used AAMs for the detection of estrus (Denis-Robichaud et al., 2016). Furthermore, conception risk was found to be higher for farms that mainly used visual estrus detection for first AI in comparison with those that used timed AI, but again no differences were found with those using AAMs or a combination of practices (Denis-Robichaud et al., 2016). When comparing the use of AAMs to using only  visual estrus detection, Michaelis et al. (2014) reported no differences in median days to first AI, median days open, conception rate at first AI or overall, but did find that the proportion  31   of pregnant animals at 200 DIM was higher for cows detected for estrus using an AAM rather than visual estrus detection. Over the years there has been mixed results in fertility outcomes when inseminations occur at estrus vs. at timed AI. Although pregnancy rates are often greater when using timed AI protocols when compared breeding by estrus detection (Cartmill et al., 2001; Cerri et al., 2004), this may simply be driven by the fact that timed AI protocols cause all cows to be submitted to AI in a timely fashion. There is little consensus among the various reports regarding comparison of conception rates of inseminations using estrus detection; some argue that they are greater (DeJarnette et al., 2001; Tenhagen et al., 2004a; Stevenson and Phatak, 2005; Bruno et al., 2013), lower (Santos et al., 2017), or similar (Pursley et al., 1997b; DeJarnette et al., 2001; Neves et al., 2012) to Ovsynch-like protocols.   Although research has been able to demonstrate that AAMs are able to detect estrus and are generally beneficial in comparison to visual detect, there was interest to understand if AAMs could be incorporated into reproductive management either alongside or in replace of timed AI protocols without reducing the efficiency of the programs; this topic was investigated within Chapter 2 of this thesis.   1.6 What impacts estrous behaviour?  As mentioned above, one of the key factors that could be driving these differences in the efficiency and accuracy of the AAM could be the variation in estrous behaviour by the female. The main factors that have been found to impact estrous behaviour have been summarized below.   32   At estrus, increased parity has been shown to decrease the amount of physical activity measured as steps (López-Gatius et al., 2005a; b; Dolecheck et al., 2015), neck movement (Reith et al., 2014), standing to be mounted measured with an electronic mount detector (Peralta et al., 2005) and have a less drastic decreases in rumination (Reith et al., 2014). Contrarily, others have found no impact of parity on steps (Pennington et al., 1986) or rumination and feeding behaviour (Dolecheck et al., 2015). Interestingly, when using multiple AAMs on the same cow, Dolecheck et al. (2015) reported that parity differences associated with physical activity at estrus were only found with monitors measuring step/hr and not found with monitors measuring head and neck movements. This may suggest that although multiparous cows have less steps per estrus event in comparison with primiparous cows, they may still be similar in terms of other behaviours monitored using accelerometers around the neck.  Heats stress has been shown to reduce the number of standing mounts (de Silva et al., 1981), decrease duration of estrus, and increase interval from luteolysis to estrus (Hein and Allrich, 1992) when measured using either visual detection or AAM (López-Gatius et al., 2005b; reviewed by Polsky et al., 2017). In addition, cows subjected to heat stress have lower levels of estradiol at estrus and delayed luteolysis and ovulation (Wilson et al., 1998). Reduced estrous expression during heat stress has been suggested to be an adaptive response to limit the heat production of the body (Hansen and Aréchiga, 1999), but also has been associated as a stress response. Cows under heat stress have increased cortisol concentrations (Wise et al., 1988; Elvinger et al., 1992) and decreased LH pulses (Wise et al., 1988). Stress can be linked to impairments of reproductive function through the activation  33   of the hypothalamic-pituitary-gonadal (HPG) axis through inhibition of GnRH secretion.  This may attribute to reduced estrous expression (van Eerdenburg, 2008) and interferes with GnRH induced LH release, and, to a lesser extent, alters the secretion of sex steroids (Rivier and Rivest, 1991) and renders target tissues less responsive to gonadal hormones (Charmandari et al., 2005). Impacts of stress on estrous expression and reproduction are also impacted by pain and health disorders such as lameness. Lameness is a disease which causes pain in dairy cattle (Whay et al., 1998, 2005) and has been used as a natural model for a chronic stressor (Walker et al., 2008). Using a visual behavioural score for estrus detection, multiple studies have observed lame cows to have less intense estrous expression than sound animals (Walker et al., 2010; Morris et al., 2011). Using the same system, severely lame cows had less intense estrous expression than cows that were moderately lame and those that were non-lame, but no difference in the incidence of displaying estrus was found between lameness classifications (Walker et al., 2008). Similarly, it was reported that lame cross-bred dairy cows stood less times to be mounted and a higher proportion of lame cows had a shorter duration (8.5 – 13 h) estrus than non-lame cows; no differences were found for other secondary estrual signs such as sniffing, chin resting, or mounting of other cows (Sood and Nanda, 2006). In addition to a reduced estrous behaviour score, lame cows have been reported to have a shorter interval from PG to the start of estrus and to first standing to be mounted and were more likely to fail to ovulate (Morris et al., 2011). Contrarily, when comparing mildly lame cows with healthy cows, a lack of correlation has been reported between estrous expression score and claw score, locomotion score or foot angle (Gómez et  34   al., 2003). Although the estrous behaviour score is generally reduced for lame cows, specific behaviours that change are not consistent between trials.  Unfortunately, there is a dearth of research examining the effects of lameness on estrous expression measured using AAMs. When measuring vulvar skin temperature with infrared thermography, cows with a higher locomotion score had a shorter duration of estrus compared to sound animals; the authors suggest that lame cows may have a reduced capability of physical movement which may dampen the increase in core body temperature during estrus, leading to a shorter duration above the set thresholds (Talukder et al., 2014). Another study reported that cows were less likely to be found in estrus using an AAM when they were both lame and with low BCS, but did not report the intensity of the estrous expression in the animals that were alerted on the monitor (LeRoy et al., 2018). Body condition score has also been commonly associated with differences in estrous expression. Generally, it is agreed that cows that are thin and have a low body condition score have lower estrous expression when using AAMs (Pennington et al., 1986; Madureira et al., 2015a). Future research focusing on determining how the health status of cows (e.g.:  lameness and BCS) impacts estrous expression when using AAMs is needed. Increased milk production has been reported to decrease mounting activity of cows during estrus measured by mounting behaviour (Lopez et al., 2004; Rivera et al., 2010). Increased clearance of hormones, specifically progesterone and estradiol, have also been thought to cause a decline in expression of estrual behaviours (Wiltbank et al., 2006). Blood flow to the liver has been shown to be chronically elevated in high-producing dairy cows due to their high feed intake (Sangsritavong et al., 2002). The association between increased feed  35   intake and clearance of hormones has also been supported by other research (Rabiee et al., 2001). Alternatively, when measuring estrous behaviour with both a collar- and leg-mounted AAM, targeting secondary signs of estrus, a lack of relationship between milk production and estrous expression has been reported (Madureira et al., 2015a). In fact, in one experiment that measured neck movements and rumination, cows that produced more milk had a greater decrease in rumination at estrus (thus increased estrous expression) and no difference in activity (Reith et al., 2014). However, these authors did note that the absolute change in neck activity was greater for low producing cows. Pennington et al. (1986) measured estrous expression using visual observation as well as pedometers, and reported that cows that produced less milk per Kg of body weight tended to have less estrous behaviours (including mounts and some secondary behaviours), but there was no association of milk production with steps. Contrarily, López-Gatius et al. (2005b) reported decreased walking activity using a pedometer in cows that produced more milk, where with every 1 kg increase in milk production per day there was a decrease in 1.6% walking activity at estrus. Further research is needed to determine the impacts of milk production on estrous expression, specifically in regards to the impacts on primary and secondary signs of estrus. Cows with greater days in milk have been found to have higher increases in physical activity at estrus (Pennington et al., 1986; Dolecheck et al., 2015) and more drastic decreases in lying behaviour (Dolecheck et al., 2015). Using a collar-mounted AAM, Aungier et al. (2012) found that first postpartum estrus events were of both lower intensity and shorter duration than subsequent estrus events. Similarly, Boyle et al. (2007) reported that with increasing number of estrus events postpartum the duration of estrus increased, although,  36   there was no increase in the number of mounts per estrus as measured using an electronic mount detector. In contrast, others have reported higher mounting activity in cows < 80 DIM when using an electronic mount detectors (Peralta et al., 2005) and no impact of days in milk on walking activity using an AAM (López-Gatius et al., 2005b). Environmental conditions in which the cow is housed has been shown to impact estrous expression. Vailes and Britt (1990) demonstrated that mounting activity was 3- to 15- fold greater on a dirt surface in comparison with a concrete surface. When cows had a choice between mounting an estrual cow on dirt or on concrete they spent 66% of their time in the dirt arena and only 32% on concrete. Interestingly, within that experiment, although mounting activity increased on dirt, secondary estrual signs remained generally unchanged between flooring types (Vailes and Britt, 1990). Another study which allowed cows either access to free-stalls with a concrete floor or to both the free-stall area in addition to a dirt lot found that when cows were housed solely on concrete flooring they had a reduced number of standing mounts, mounting of other cows, and reduced duration of estrus when compared to cows which also had access to a dirt lot (Rodtian et al., 1996). Contrarily, when comparing concrete floors with rubber matted floors, no difference in the proportion of cows that expressed estrus nor the intensity or duration of estrous expression, as measured using an electronic mount detector, was found (Boyle et al., 2007).  However, the results of these last two studies should be viewed with caution as both were pseudoreplciated given that the authors used the individual cows as the experimental unit even though treatment was applied at the pen level.   37   Using visual observation, cows were found to have increased standing to be mounted behavior and to have a drastic reduction in slipping while mounting when housed on rubber matted floors in contrast with concrete slatted floor (Platz et al., 2008). When comparing cows housed within a free-stall setting with those housed on pasture it was found that estrual cows housed on pasture received more mounts (Palmer et al., 2012), a higher proportion of the cows displayed estrus and a lower proportion displayed estrus at a very low intensity (Palmer et al., 2010); however, there was no differences in secondary signs such as sniffing and chin resting or the duration of estrus as measured using an electronic mount detector (Palmer et al., 2010, 2012). The decrease in mounting behaviour found when cows are housed on concrete floors has been suggested to be due to the footing being less secure and thus reducing the motivation of other animals to mount. Palmer et al. (2010) discusses that since mounting of cows in estrus is an ‘altruistic’ behaviour (aka it is at no benefit to the mounting cow), as the potential cost involved in performing the behaviour (i.e.: slipping and falling) increase the frequency of these behaviour decrease. There have been some reports that the time of day has an impact on estrous behavior. Doing twice daily visual observations, de Silva et al. (1981) reported that cows that were detected in estrus in the morning accepted more standing mounts than those that were detected in the evening. Additionally, Peralta et al. (2005) reported that when using twice daily visual observation a higher proportion of animals were detected in estrus in the morning than in the evening. In that study, the animals were wearing a mount detector as well as a collar-mounted accelerometer and it seems that the circadian differences in estrus detection rate was driven by general movement and not standing behaviours. Another study  38   using a collar-mounted AAM reported that estrus alerts were most common at night and during the early morning (Reith et al., 2014); however, when comparing the circadian rhythm of activity and rumination during the estrus period to a non-estrual reference period, the pattern of activity and rumination between the two periods were the same.  In summary, behaviours performed at estrus are altered by the physiology of the animal (e.g., age, milk production, lameness, stage of lactation), as well as the housing environment. Some of the behaviours will be impacted by factors similar to those that affect the efficiency and accuracy of AAMs to detect cows in estrus (e.g., stage of lactation, housing environment) while others are not (e.g., age, time of day, heat stress).    1.6.1 Impacts of estrus on fertility  The occurrence of estrus as well as the intensity of estrous expression has been shown to have positive impacts on fertility. The occurrence of estrus, detected using tail chalk, at the end of timed AI protocols that use estradiol-cypionate (ECP) increases conception rates (Cerri et al., 2004; Galvão et al., 2004; Pereira et al., 2014) and reduces pregnancy losses (Galvão et al., 2004; Pereira et al., 2014, 2016). Similarly, the occurrence of estrus in recipient cows appears to increase pregnancy per embryo transfer when using either ECP-based (Pereira et al., 2016) or GnRH- based protocols (Jinks et al., 2013); and has been associated with decreased pregnancy losses after embryo transfer (Pereira et al., 2016). Although the occurrence of estrus when using ECP-based protocols is more common than when using those based on GnRH (Souza et al., 2007; Pereira et al., 2013), when estrus does  39   occur when using GnRH-based programs conception rates are increased (Pereira et al., 2013). In fact, in one study which compared two different timed AI protocols both with and without the use of ECP, reported that irrespective of the type of protocol or the use of ECP, the occurrence of estrus increased conception rates, although they did not find an impact on pregnancy losses (Hillegass et al., 2008).  The amount of activity that occurs at the time of estrus has also been shown to impact fertility. When using visual estrus detection, increased estrous expression, based on standing to be mounted behaviour, was associated with increased conception rates (Gwazdauskas et al., 1981; Reimers et al., 1985). Recently, when using AAMs, increased walking activity (López-Gatius et al., 2005b; Madureira et al., 2015a), decreased lying time and lying bouts (Silper et al., 2017), and increased activity using neck movement (Madureira et al., 2015a) have all been associated with increased pregnancy per AI. However, these findings are not universal as Aungier et al. (2012) failed to find any associations of the duration or intensity of estrus alerts using a collar-mounted AAM. Similarly, Yániz et al. (2006) did not find an impact of increased walking activity at estrus between calving and 50 DIM on the odds of cows becoming pregnant before 90 DIM. Within that study, they averaged the increase in walking activity of multiple estrus events rather than using specific events when the inseminations occurred. To date there is little research on the repeatability of estrus intensity within cows. It could be predicted that by taking the average of multiple events they may have dampened the impact that estrous expression has on the success at insemination but more work is needed to elucidate this. The direct connection between  40   estrous expression and fertility is unknown, but aspects that may be associated with the benefits of estrous expression are discussed hereafter.  Progesterone concentrations have been linked with both benefits for estrous expression as well as the fertility of cows. Progesterone before, during and after AI have been shown to influence conception rates and to be associated with estrous expression. As mentioned above, ideally, progesterone concentrations should be high during diestrus, decrease during proestrus, reaching a nadir at estrus, and then slowly climbing after ovulation; any deviation from this pattern can have repercussions on the fertility of the cow.  As discussed above, progesterone concentrations within diestrus are thought to have a priming effect on the hypothalamus, making the hypothalamus more responsive to estradiol and the occurrence of estrus (Woelders et al., 2014), thus may also be involved in the intensity of estrous expression. By inducing different lengths of the luteal phase preceding estrus, recent work reported that those with a short luteal phase (5 d) had a tendency to have reduced estrous expression in comparison with those with a normal length luteal phase (14 d). Interestingly this difference was not seen when controlled for BCS, suggesting that BCS may have had a greater impact on estrous expression than the length of the luteal phase (Denis-Robichaud et al., 2018c). These same authors found no impact of the length of the luteal phase on the expression of estrogen receptor alpha in the endometrium at the time of the PGF2 prior to breeding. In a study using ovariectomized cows, Davidge et al. (1987) administered four levels of progesterone for 5 d, followed 72 hr later with an estradiol treatment and reported that with higher concentrations of P4, there were lower amounts of estrual behaviours displayed at estrus. Within lactating dairy cows it has been  41   reported that estrus synchronization with or without an intravaginal progesterone insert did not impact walking activity at estrus (López-Gatius et al., 2005b).   In the days before estrus, decreased progesterone has been shown to decrease conception rates (Bisinotto et al., 2010; Chebel et al., 2010) likely because of its importance for the growing follicles. Some studies have demonstrated that with decreasing concentrations of progesterone during diestrus, a higher percentage of cows will have double ovulations (Cunha et al., 2008; Cerri et al., 2011). The negative feedback of progesterone on FSH will decrease the amount of circulating FSH around the time of follicle selection, resulting in the selection of only one dominant follicle (Wiltbank et al., 2012). Bisinotto et al. (2010) also demonstrated that by manipulating cows to either ovulate a follicle that grew under high or low progesterone, follicles which grew under high circulating concentrations of progesterone were more fertile. Concentrations of progesterone prior to estrus have also been shown to impact embryo quality.  In a study using superovulation for embryo transfer, more embryos of high enough quality for embryo transfer in animals which had higher concentrations of progesterone during diestrus were noted (Rivera et al., 2011). Low progesterone concentrations at the time of AI have been shown to be beneficial for fertility and to be necessary for the expression of estrus. Although it has been well established that progesterone is low at estrus (i.e. Rajamahendran et al., 1989; Allrich, 1994), there is limited research on how progesterone concentrations may modulate the intensity of estrous expression in cows that actively demonstrate estrual behaviours. Cows that express estrus at the end of a timed AI protocol have been shown to have lower concentrations of progesterone in comparison to those that lacked estrous expression (Pereira et al., 2016).  42   But again not all studies agree; using a leg-mounted AAM, Madureira et al. (2015b) failed to find a correlation between estrous expression intensity and progesterone concentration at estrus.  Incomplete luteolysis has been shown to be detrimental to inseminations carried out at both timed AI (Souza et al., 2007; Brusveen et al., 2009) as well as at detected estrus (de Silva et al., 1981; Ghanem et al., 2006). Negative impacts of progesterone on fertility are believed to be associated with altered uterine or oviductal contractility impeding sperm and oocyte transport ultimately resulting in reduced fertilization rates (Hunter, 2005). Increased progesterone at estrus is also associated with reduced cleavage and blastocyst development rates (Silva and Knight, 2000), as well as thinning of the endometrium, which is negatively associated with embryonic development (Souza et al., 2011). Progesterone concentrations at the time of AI as well as estrous expression have also been associated with ovulation rates. Cows which show estrus, as measured using tail chalk, at the end of timed AI protocols using ECP, have been reported to have increased ovulation rates (Galvão et al., 2004; Pereira et al., 2014). It has also been shown that on the day of estrus, larger decreases in the number of lying bouts or lying time per day, relative to baseline, are associated with increased ovulation rates (Silper et al., 2017), as well as in cows with increased estrous expression measured using a collar- and leg-mounted AAM (Madureira et al., 2015b). Given that reductions in ovulation rate directly impacts fertility, research is required to understand the contribution of ovulation failure and ovulation timing to the poor fertility of cows with low estrous expression; this topic will be the focus of Chapter 4 within this thesis.  43   In addition to changes in progesterone around estrus, there is also evidence that increased estradiol at the time of estrus could have benefits on fertility. Cows subjected to a synchronization protocol that exhibited estrus had higher concentrations of estradiol and that the increase in estradiol after luteinisation was steeper than in cows that did not exhibit estrus (Perry et al., 2014). Using a estrous behaviour score by visual detection, one study found a strong correlation between the timing of maximum estradiol concentrations and maximal estrous expression (Lyimo et al., 2000). However, when comparing estradiol concentrations at estrus with secondary estrous behaviours measured using AAM, the relationship between estrous expression and estradiol concentration has been shown to be weak (Madureira et al., 2015b; a). Nevertheless, estradiol concentrations have been associated with factors that are associated with increased fertilization and embryonic survival. Such factors include production of specific oviductal glycoproteins which increase capacitation, motility and viability of sperm (Buhi, 2002), as well as changes in the composition of proteins produced within the lumen of the uterus (Bartol et al., 1981). Estradiol has also been reported to increase the sensitivity of granulosa cells to FSH and LH, ultimately increasing progesterone production by granulosa cells post-ovulation (Welsh et al., 1983). Beef cows with greater estradiol concentrations at estrus were more likely to have a fertilized embryo and had increased concentrations of progesterone 7 d post-AI in comparison with cows with lower concentrations of estradiol at estrus (Jinks et al., 2013).  Progesterone concentrations following ovulation have also been associated with the preceding estrous expression and fertility. Cows detected in estrus after protocols using ECP for timed AI (Pereira et al., 2014, 2016) or embryo transfer (Pereira et al., 2016) have been  44   found to have higher circulating progesterone concentrations 7 d post-estrus. Conception rates are higher in cows with greater concentrations of progesterone 7 d post-AI (Stronge et al., 2005; Pereira et al., 2014, 2016) as well as those with a quicker increase in progesterone post-ovulation (Larson et al., 1997). However, Pereira et al. (2016) reported that concentrations of progesterone had a quadratic association with conception at timed AI, but found no impact for embryo transfer. Similarly, cows with lower progesterone ( 1.0 and < 3.5 ng/ml) concentrations at 11 d post-AI were found to have lower conception rates at first service than those with higher progesterone (Herlihy et al., 2013). Circulating concentrations and the rate of increase in progesterone at the beginning of the luteal phase has been associated with increased embryo survival (Morris and Diskin, 2008) and with changes in the uterine environment, such as the secretion of certain endometrial proteins, required for the elongation of the conceptus (Clemente et al., 2009; Forde et al., 2011a). Furthermore, 16 d after insemination, embryos which developed under a late post-ovulatory rise in progesterone were found to produce very low concentrations of interferon-tau, which is the hormone that is required for maternal recognition of pregnancy (Mann et al., 1999). Progesterone is not only required for the maintenance of pregnancy but also for its establishment. Progesterone has a key role in creating local immunosuppression in the uterus to allow for the presence of the embryo without the activation of the maternal immune system (Arck et al., 2007). All together progesterone within the early luteal phase has marked benefits for embryo survival and development. Overall, the direct link between estrous expression and fertility is not well understood, but it is suspected to be associated with the hormonal milieu surrounding  45   estrus, as both progesterone and estradiol play important roles for the development of the oocyte and the receptivity of uterus while concomitantly being associated with differences in the occurrence and expression of estrus. Although these relationships are present, more research is needed to fully understand their causality. In Chapter 5 of this thesis, we will investigate a potential mitigation to reduced fertility in cows with lesser estrous expression.    1.7 Thesis objectives   The overall aims of this thesis were to determine if monitors can be used to predict estrus, and if they thus can be used within reproductive management, and then to determine the interrelationship between the intensity of estrous expression, ovulation and fertility. The specific objectives of my thesis were: 1) investigate if AAM could be effectively used to detect cows in estrus within a reproductive program without compromising the reproductive efficiency of the farm; 2) to determine if changes in rumen-reticular temperature, in contrast to only movement, could be used as a novel factor to determine ovulation 3) determine if the differences in fertility associated with estrous expression could be related to changes in ovulation timing and failure; 4) determine if the lower fertility of cows with reduced estrous expression could be mitigated by artificially stimulating ovulation at the time of AI.  I hypothesized that AAM would be able to detect cows in estrus at acceptable rates, such that reproductive efficiency would not be reduced, yet the requirement of cows to complete a timed AI protocol would be lessened (Chapter 2). In addition to changes in movement, I believed that changes rumen-reticular temperature could be used to predict  46   ovulation in dairy cows, irrespective of estrous expression (Chapter 3). Additionally, I hypothesized that changes in fertility due to estrous expression would be associated with changes in ovulation timing and increased ovulation failure (Chapter 4) and that this decrease in fertility could be mitigated by stimulating cows to ovulate with the use of GnRH at the moment of AI (Chapter 5).    47   Chapter 2: Integrating an automated activity monitor into an artificial insemination program and the associated risk-factors affecting reproductive performance of dairy cows 1 2.1 Introduction The dairy industry significantly relies on timed artificial insemination (AI) protocols to synchronize ovulation or estrus for postpartum AI. Surveys have indicated that approximately 75% and 21% of herds implement an estrus or ovulation synchronization program for the first postpartum AI in the US and Canada, respectively (Caraviello et al., 2006b; Denis-Robichaud et al., 2016). Because of increasing evidence of poor detection of estrus resulting from poor or unidentified expression of estrus (Stevenson, 2001), use of synchronization programs has significantly improved AI submission rates (Chebel et al., 2010) and reduced the duration and variability of the interval from calving to first service (Stevenson, 2001). Although evidence that breeding cows at the time of estrus may result in greater (DeJarnette et al., 2001; Tenhagen et al., 2004a; Stevenson and Phatak, 2005), or similar (DeJarnette et al., 2001; Santos et al., 2004), conception rates compared with Ovsynch-like timed AI protocols, overall pregnancy rates are often greater (Cartmill et al., 2001; Cerri et al., 2004) for timed AI protocols because all cows are submitted to AI. Synchronization of ovulation allows for management practices to reduce time needed for                                                              1 A version of this chapter has been accepted for publication: Burnett, T.A., A.M.L. Madureira, B.F. Silper, A.C.C. Fernandes, and R.L.A. Cerri. 2017. Integrating an automated activity monitor into an artificial insemination program and the associated risk-factors affecting reproductive performance of dairy cows. J. Dairy Sci. 100:5005-5018.  48   visual observation of estrus and creates a fixed schedule of AI allowing better planning and use of labor resources. Compliance to injection schedules has been  identified as a problem in carrying out protocols successfully (Stevenson and Phatak, 2005). In spite of the success observed in timed AI programs, there has been increasing concern about the extensive use of hormone therapies in animal production. The dairy industry is not an exception to this trend (Pieper et al., 2016), and more interest from commercial farms and research institutions have been placed to better rationalize the use of pharmacological interventions in reproductive programs (Saint-Dizier and Chastant-Maillard, 2012). The challenge, nonetheless, is to achieve overall herd fertility similar to currently adopted reproductive programs.  In recent years, automated estrus-detection systems, such as pedometers and accelerometers, have become more reliable, with evidence that they are able to correctly identify cattle in estrus (Roelofs et al., 2005a; Hockey et al., 2010; Løvendahl and Chagunda, 2010), and properly indicate insemination times by predicting the timing of ovulation (Roelofs et al., 2005a; Stevenson et al., 2014). Previous studies have indicated that if detection of estrus can be performed more frequently and during night hours, it reduces the proportion of cattle with unobserved estrus episodes (Hall et al., 1959; Van Vliet and Van Eerdenburg, 1996; Roelofs et al., 2005a). Furthermore, Chebel and Santos (2010) demonstrated that visual detection of estrus assisted by tail chalk removal in addition to a timed AI protocol did not result in significantly different pregnancies per AI compared with cows that were subjected to timed AI protocols alone. In fact, a recent survey across Canada demonstrated that dairy producers that adopted automated activity monitors (AAM) found  49   an increase in pregnancy risk from 14.9 to 17.0 between the year before and year after adoption of the AAM (Neves and LeBlanc, 2015). Summarizing, the incorporation of detection for estrus (“cherry-picking”) in timed AI based reproductive programs has been common practice for many years, but the introduction of AAM has the potential to further minimize pharmacological interventions for efficient breeding programs.  Although strong evidence exists that AAM are capable of detecting cows in estrus, factors that may impact their efficiency are still unclear. Lameness has been shown to decrease walking activity during estrus (Walker et al., 2008) and decrease the number of standing mounts in a given estrus episode (Diskin and Sreenan, 2000b). In addition, BCS has also been reported to decrease the expression of estrus (Roelofs et al., 2010; Madureira et al., 2015a). Although impacts of physical health have been reported using visual observation of estrous behaviors, it is still unclear if factors of physical health also impact the expression of estrus as measured by AAM.  The objective of this study was to investigate the effect of integrating the use of AAM in conjunction with a timed AI protocol on submission rates and pregnancy/AI compared with solely using a timed AI protocol for first AI. Furthermore, we investigated the effects of parity, BCS, milk production, and gait and hock lesion scores on the expression of estrus, as measured by AAM, and the previously mentioned reproductive program treatments.   2.2 Materials and methods   This experiment was conducted between September 2012 and July 2014 at the University of British Columbia’s Dairy Education and Research Centre (Farm A), Agassiz, BC  50   as well as a local commercial dairy farm (Farm B) in Dewdney, BC. All procedures were approved by the Animal Care Committee of the University of British Columbia. The cattle used in this experiment were cared for as outlined by the guidelines provided by the Canadian Council of Animal Care (2009).  2.2.1 Animals and housing  A total of 918 high producing Holstein dairy cows were enrolled in this study from 2 different herds (Farm A: n = 466; Farm B: n = 452). The rolling herd average size of each farm was 260 and 350 cows for Farms A and B, respectively. Cows produced a 12,195 ± 2,145 (Farm A) and 12,965 ± 2,215 (Farm B) kg of milk (mean ± SD 305-d mature-equivalent yield) and had a range of BCS from 2 to 4 on both farms at 40 ± 7 DIM. Cows from Farm A were housed in a naturally ventilated wood-framed barn with a free-stall design, equipped with deep sand-bedded stalls. Cows were milked twice daily at 0500 h and 1500 h with automatic milking machines. Cows from Farm B were housed in a naturally ventilated wood-framed barn equipped with fans and a free-stall design; stalls were equipped with mattresses and bedded with sawdust. Farm B milked 3 times daily at 0400 h, 1200 h, and 1600 h with automatic milking machines. Fresh TMR was delivered twice daily on both farms at approximately 0700 h and 1600 h. The TMR was formulated following the Nutrient Requirements of Dairy Cattle (NRC) guidelines (NRC, 2001) to meet or exceed the requirements of a 620 kg Holstein cow producing 40 kg/d of 3.5% fat corrected milk. All cows had ad libitum access to both TMR and water.  51   2.2.2 Study design  Cows were assigned randomly to either of 2 treatments based on their ear tag number: 1) TAI: all first AIs performed by timed AI and 2) ACT: first AI based upon the detection of estrus by AAM after presynchronization, whereas all remaining non-inseminated cows were enrolled in a timed AI protocol. All cows were enrolled in a presynchronization protocol at 43 ± 7 DIM (Farm A) and 40 ± 7 DIM (Farm B). The presynchronization protocol consisted of two injections of PGF2α (dinoprost tromethamine, Lutalyse; Zoetis, Florham Park, NJ; 25 mg i.m.) 2 wk apart. Cows from the TAI treatment were automatically enrolled into the Ovsynch-56 protocol 12 d later. The Ovsynch-56 protocol consisted of one injection of GnRH (gonadorelin hydrochloride, Factrel; Zoetis Inc.; 100 μg i.m.), followed 1 wk later by an injection of PGF2α, then 56 h later by a GnRH injection, and finally AI was performed 16 h after the final GnRH injection. For the ACT treatment, once the presynchronization protocol was complete, cows were bred upon detection of estrus by the AAM; if cows were not detected in estrus within 12 d after the second PGF2α of the presynchronization they were also enrolled into the Ovsynch-56 protocol. A diagram of each reproductive program is included in Figure 2.1. The AAM of both farms were checked twice daily for cows in estrus, and breeding was carried out using the a.m./p.m. rule. To understand the practicality of the reproductive programs, treatments included all cows that were enrolled into the entire study, regardless of when they were inseminated. In addition, cows that complied to their specific reproductive program (i.e.: were inseminated appropriately according to their treatment) were tested separately.  52   Cows on both farms were equipped with AAM within 2 wk of parturition. Physical activity was continuously monitored on Farm A using SCR Heatime tags (Heatime, SCR Engineering, Israel) and on Farm B using Afimilk pedometers (AfiAct Pedometer Plus, Afimilk, Kibbutz Afikim, Israel). The threshold activity to be considered an estrus event was set at an activity index of 35 and at a relative increase in activity of 180% for Farm A and B, respectively; an index of 35 approximately equates to a 6 SD increase in activity compared with baseline for the SCR system. Activity data from Farm A were collected and used to determine the effects of BCS, gait and hock scores on the expression of estrus. All estrus episodes (n = 666 from 340 cows) between calving until first AI were used in the analysis. Independent of treatment, the timing of insemination was classified to categories as either occurring at estrus (Estrus) or at the end of a timed AI protocol (Timed AI).  2.2.3 Cow-level scoring  All cows had their body condition, lameness, and hocks scored twice, 2 wk apart, at the time of the presynchronization injections. Body condition was scored on a 5-point scale from thin (1) to obese (5) as outlined by Edmonson et al. (1989). Cows were later categorized as thin (≤ 2.75) or moderate (> 2.75). Lameness was scored on a 5-point scale from normal (1) to severely lame (5) as outlined by Flower and Weary (2009). Cows were later categorized as sound (≤ 2) and lame (> 2). Hock lesions were scored on a 4-point scale modified from the Hock Assessment Chart for Cattle developed by Cornell Cooperative Extension (http://hdl.handle.net/1813/36913) as follows: (1) no swelling and without hair missing, (2) minor swelling without hair missing, (3) minor to moderate swelling with bald  53   area and (4) severe swelling with loss of hair, with or without broken or scabbed skin. Hock lesions were later categorized as normal (≤ 2) and swollen (> 2). As hock lesions and lameness were found to be correlated within this study, hock lesion and lameness scores were combined into one score (leg health). Two categories of leg health were created: (1) adequate: cows with both normal hocks and that were sound, (2) compromised: cows with either swollen hocks or that were lame, or both. All remaining transition health and production information was collected by the dairy herd personnel with the assistance of the herd veterinarian, and confirmed and recorded by the project leader using the on-farm Dairy Comp 305 software (Valley Agricultural Software, Tulare, CA).   2.2.4 Ultrasonography, cyclicity and pregnancy diagnosis  Ovaries were examined by a portable ultrasound (Ibex Pro; E.I. Medical Imaging, Loveland, CO) using a 7.5 MHz linear-array rectal transducer, twice, 2 wk apart, at the time of the presynchronization injections, starting at 43 ± 7 DIM (Farm A) and 40 ± 7 DIM (Farm B). Presence and diameter of the largest follicles and corpora lutea were measured and recorded. Cows were classified as cyclic if there was at least one corpus luteum present at one of the exams and anovular if there was no corpus luteum present at either exam. Pregnancy diagnosis was carried out by ultrasonography by the herd veterinarians at 36 ± 7 days post-AI for the detection of an embryonic vesicle with a viable embryo (presence of heartbeat).   54   2.2.5 Expression of estrus  Data from the collar mounted AAM (Heatime, SCR Engineers, Netanya, Israel) were collected from cows on Farm A to determine the effects of treatment, parity, BCS, lameness and milk production on the expression of estrus; data representing 666 estrus episodes were collected from 340 cows. The effect of hock lesions was excluded as there was not enough variation in hock lesions on Farm A. Expression of estrus was quantified using two criteria: 1) peak activity and 2) duration, as previously performed in Madureira et al. (2015). Peak activity was defined as the maximum activity index during an estrus episode. Duration of an estrus episode was defined as the amount of time the cow spent with an index greater than the threshold. The threshold activity to be considered an estrus event was set at an index level of 35 (roughly 6 SD in relation to baseline activity). The timing at which each estrus episode occurred was classified as follows: (1) Early: estrus episodes occurring before the first PGF2α of the presynchronization protocol; (2) PGF1: after the first PGF2α and before the second PGF2α of the presynchronization protocol; (3) PGF2: after the second PGF2α of the presynchronization protocol but before the start of the Ovsynch protocol; (4) OVS: between the first injection of GnRH until the timed AI of the Ovsynch protocol. Specific estrus-expression traits could not be collected on Farm B.   2.2.6 Statistical analyses  All analyses for this experiment were performed using SAS (version 9.4; SAS Inst. Inc., Cary, NC) with cow as the experimental unit. Before all analyses, data were tested for  55   normality using the UNIVARIATE procedure and probability distribution plots; any variables deemed not normal were transformed to fit normality and subsequently back transformed for geometric means. Treatment, compliant to treatment, and AI category as well as peak activity and duration were always tested in separate models throughout the entire statistical analysis because of collinearity. The effects of treatment, AI category, farm, parity, cyclicity, BCS, leg health, and milk production on the dichotomous variables pregnancy outcome and detection of estrus were tested with logistic regression using the LOGISTIC procedure with backwards elimination using Wald’s statistic criterion when P < 0.15. Summaries of frequency comparisons for pregnancy per AI and AI category were carried out using the FREQ procedure. For the analysis of the expression of estrus, the peak activity and duration variables were considered dependent variables and tested for the effects of parity, cyclicity, BCS, leg health, milk production, and the timing of the estrus using ANOVA for repeated measures with cow as the subject using the MIXED procedure. Frequency distributions of all explanatory variables are summarized for the entire study and by farm in Table 2.1. Effects of treatment, compliant to treatment, and AI category on hazard risk of pregnancy by 300 DIM were analyzed with PHREG procedure of SAS. Cox-proportional hazard regression models included days open as the outcome variable and treatment, compliant to treatment, AI category, farm, parity, cyclicity, BCS, leg health, and milk production as explanatory variables. Observations were right-censored at culling or at 300 DIM if pregnancy had not been previously confirmed. Only variables with P ≤ 0.15 were kept  56   in the final models. Survival curves were drawn from the proportion of non-pregnant cows at each time point given by the LIFETEST procedure.  2.3 Results  Overall pregnancy per AI did not differ between treatments (30.8 vs. 33.5% for ACT and TAI treatments; P = 0.39). Compliance within the TAI treatment was 83.2%, where 16.8% of cows on the TAI treatment were bred by estrus at some point after the second PGF2α and before the end of the Ovsynch protocol, instead of at the scheduled timed AI; compliance on Farm A was 87.3% and on Farm B was 77.8%. With inclusion of only cows that were inseminated compliant to assigned treatment (n = 779), pregnancy per AI was still not different between treatments for the entire study (30.8% vs. 35.9% for ACT and TAI treatments; P = 0.13). This lack of an effect was consistent for Farm B (31.1% vs. 31.3% for ACT and TAI treatments; P = 0.97), but not for Farm A (30.5% vs. 40.0% for ACT and TAI treatments; P = 0.04).  Estrus was detected for 50.5% of cows; detection of estrus was determined as the proportion of cows in the ACT treatment bred by estrus between the end of presynchronization and the beginning of the Ovsynch protocol. There was a substantial difference in the detection of estrus between farms, as Farm A had 65.1% of cows detected in estrus, whereas Farm B detected 34.4% of their cows in estrus after the presynchronization (P < 0.01). When only cyclic cows were analyzed, detection of estrus was 52.2% for the entire study and 66.0%, and 35.3% for Farm A and Farm B, respectively. Although cyclicity differed by farm (87.8% vs. 71.7% for Farm A and Farm B, respectively; P  57   < 0.001), it does not seem to be the sole reason for this difference in detection of estrus. Pregnancy was impacted by AI category, where the odds of a cow bred at timed AI to become pregnant were 1.48 times greater than the odds of a cow that was bred at estrus (P = 0.02). In contrast, cows bred at timed AI were bred at significantly later DIM than those bred at estrus (75.7 ± 0.2 vs. 63.2 ± 0.3 DIM; Table 2.2). A summary of pregnancy per AI found for treatment and AI category (insemination at estrus or timed AI) is presented in Table 2.2 for the entire study and by farm.  Pregnancy success at first postpartum AI was affected by parity (P < 0.01), cyclicity (P = 0.03), BCS (P < 0.01), milk production (P = 0.03), and a tendency for leg health (P = 0.10); these results are summarized in Table 2.3 for the entire study and for cows that were compliant to the treatment. Additional interactions were found for treatment by parity (P = 0.02), treatment by cyclicity (P = 0.07) and leg health by parity (P = 0.10). Pregnancy outcomes were not affected by farm (P = 0.90) or treatment (P = 0.33). In contrast, when only including cows that were intended to be treated there was a tendency for an effect of treatment on pregnancy outcomes (P = 0.09; Table 2.3). Within the treatment by parity interaction, primiparous cows on the ACT treatment had 2.36 greater odds of pregnancy than multiparous cows on the ACT treatment; alternatively, there was no difference in odds of pregnancy between primiparous and multiparous cows enrolled in the TAI treatment (P = 0.02; Table 2.4). There was a tendency for primiparous cows with adequate leg health to have 1.83 greater odds of pregnancy than primiparous; however, there was no impact of leg health on the odds of pregnancy in multiparous cows (P = 0.10; Table 2.4). There was also a tendency for a treatment by cyclicity interaction, as cyclicity status did not impact the odds  58   of pregnancy on the TAI treatment, but cyclic cows enrolled in the ACT treatment had 2.46 greater odds of pregnancy than anovulatory cows enrolled in the ACT treatment (P = 0.07; Table 2.4). Pregnancy outcome interactions for the entire study in addition to those that were compliant to the treatment are summarized in Table 2.4.  Analyzing pregnancy outcomes stratified by farm enabled us to see distinct differences in factors that affected pregnancy success. On Farm A, parity (P = 0.02), cyclicity (P = 0.05), and BCS (P < 0.01) affected successful pregnancy outcomes, whereas on Farm B, BCS (P = 0.02), milk production (P = 0.02), and leg health (P = 0.06) affected successful pregnancy outcomes (Table 2.5).  Factors that impacted the odds of cows being detected in estrus after the presynchronization protocol (Table 2.6) were farm (P < 0.001), parity (P < 0.01) and cyclicity (P = 0.08). The proportion of cows detected in estrus was different for cows that had poor and adequate leg health, where fewer cows with poor leg health were detected in estrus (40.8% vs. 62.0%; P < 0.001). Frequency distributions and odds ratios for the detection of estrus are summarized in Table 2.6.  Factors affecting the hazard of pregnancy by 300 DIM were farm, parity, BCS, as well as a treatment by cyclicity interaction and a leg health by farm interaction. Farm B had an increased hazard of pregnancy (Hazard ratio [HR] = 1.21; P = 0.01), whereas the hazard of pregnancy was decreased in multiparous cows (HR = 0.66; P < 0.001) as well as in thin cows (HR = 0.78; P < 0.01). An interaction between cyclicity and treatment was found, where cows on the TAI treatment were unaffected by cyclicity (HR = 1.00; P = 0.04), but cows that were not cycling by 50 DIM within the ACT treatment had a reduced hazard of pregnancy in  59   comparison with those which were cycling (HR = 0.74; P = 0.04). Inclusion of only cows that were compliant with the treatment did not impact the effect of treatment on hazard of pregnancy (P > 0.10), but a tendency for a farm by leg health interaction was found (P = 0.08). On Farm A, leg health had no impact on hazard of pregnancy (HR = 1.00; P = 0.08); however, on Farm B cows with poor leg health had a decreased hazard of pregnancy (HR = 0.77; P = 0.08). AI category was found to have an impact on hazard of pregnancy, where cows bred at the time of estrus for their first postpartum insemination had a greater hazard of pregnancy than those bred at the end of the timed AI protocol (HR = 1.19; P = 0.03). Survival curves of hazard of pregnancy by 300 DIM for treatment, compliant to treatment, AI category, and cyclicity by treatment and leg health by farm interactions are presented in Figure 2.2.  The mean and range duration and intensity of estrus was found to be 11.1 h ± 5.0 (2 to 24 h) and 73.6 ± 20.1 (35 to 100), respectively. Estrous expression was not affected by BCS, gait score or milk production; there was not enough variation in hock score within the subset of data to be included within the analysis. Cyclicity affected estrous expression where cows that were cycling by 50 DIM had estrus episodes with greater peak activity (67.0 vs. 74.1 index; P = 0.02) and a longer duration (9.0 vs. 11.2 h; P < 0.01). Primiparous cows were found to have estrus episodes with greater peak activity (P < 0.01) and longer duration (P < 0.01) than multiparous cows (Table 2.7). There was an effect of the timing of the estrus episode on the expression of estrus (P < 0.001; Table 2.7). Estrus episodes that occurred before enrollment in the experiment had lower peak activity and a shorter duration than episodes occurring at any other time. Expression of estrus occurring during the Ovsynch protocol tended to be lower than those after the second prostaglandin but the same as estrus  60   occurring after the first prostaglandin; estrus occurring after the first and second prostaglandins did not differ (Table 2.7).  Finally, when only including estrus events that were measured by the AAM and AI was performed (Farm A; n = 199), estrus events with longer duration were more likely to result in pregnancy, where the odds of pregnancy for a long estrus event was 2.2 times greater than the odds of short estrus events (P = 0.02); there was no impact of peak activity found on pregnancy outcomes in this study.  2.4 Discussion  Incorporation of the detection of estrus during a presynch-ovsynch program for first AI in lactating dairy cows is a practice commonly used on North American dairy farms. In general, the challenge is to measure and compare success rates (detection of estrus and pregnancy per AI) of farm reproductive management programs that use different protocols, voluntary waiting periods, and have farm differences in milk production, thermal stress and general housing and management conditions. Previous studies have shown that “cherry picking” after the end of a PGF2α-based presynchronization protocol can be efficient and comparable (hazard ratios for days to conception) with timed AI only programs (Neves et al., 2012; Neves and LeBlanc, 2015; Dolecheck et al., 2016); however, this is not always the case when comparing only pregnancy success at first AI. The increase in AAM available on the market has created more options for producers to potentially increase estrus detection rates and modify current reproduction programs. Incorporation of these new technologies  61   lead to questions about factors that may impact their effectiveness. This study showed that using AAM for the detection of estrus within a presynch-ovsynch program resulted in similar pregnancy per AI compared with a reproduction program that was strictly based on timed AI. This and other studies(Neves et al., 2012; Valenza et al., 2012; Fricke et al., 2014b; Dolecheck et al., 2016) also found a strong effect of farm, suggesting that the use of AAM are likely more prone to individual variations, particularly when compared with more established timed AI protocols. In addition, this study showed that lameness had a negative impact on pregnancy success on the farm with a greater prevalence of leg health injuries and the repercussions of this are seen in increased days until conception. Leg health and poor BCS had no impact on peak intensity and duration as measured by AAM.  Within this study,  cows bred at timed AI were more likely to become pregnant than those bred at the time of estrus; however, cows bred at estrus for their first postpartum AI were found to have shorter days to conception than those bred at timed AI. These results agree with some that also reported an increase in pregnancy per AI when insemination were carried out at timed AI when in comparison with those at estrus (Cerri et al., 2004; Gümen et al., 2012; Fricke et al., 2014b; Stevenson et al., 2014). These results are in contrast to others that found no difference at first postpartum AI(Gümen and Seguin, 2003; Dalton et al., 2005; Chebel and Santos, 2010) or subsequent AI (Giordano et al., 2015; Dolecheck et al., 2016). The experimental design of this study and other similarly designed studies make it difficult to conclude the major cause for this difference (or lack thereof) in fertility between the two categories of inseminations considering cows detected in estrus were bred approximately 10 d earlier than cows bred in the TAI treatment. Previous studies have  62   reported that fewer DIM at first breeding can have negative effects on conception rates  (Tenhagen et al., 2003; Stevenson and Phatak, 2005), but no studies have compared the two categories at similar DIM. There were no longer-term impacts of treatment on average days open for the two groups. This is consistent with other studies (Neves et al., 2012; Fricke et al., 2014b; Dolecheck et al., 2016), which reported no difference in days to conception between ACT and TAI reproductive programs. In contrast, Neves et al. (2012) found a reduction in days from calving to pregnancy when they only included cows that were compliant to their respective programs. In addition, some studies have even found reproductive programs based on inseminations at estrus to have shorter days open even if they have reduced pregnancy per AI at first AI when comparing inseminations occurring at estrus vs. timed AI (Stevenson et al., 2014). Duration of estrus found in the present study was similar to those reported in Madureira et al. (2015) as well as others using AAM (Roelofs et al., 2005a; Aungier et al., 2012), but shorter than reported by Valenza et al. (2012). Furthermore, research reporting the duration of estrus using visual observations also found similar values (13.4 h, Roelofs et al., 2004; and 11.8 h, Roelofs et al., 2005). Consistent with the present study, Madureira et al. (2015) reported that pregnancy per AI increases with increased expression of estrus as measured using two different AAM (neck and leg mounted). Although we did not detect an effect of estrus intensity on pregnancy per AI as was found in Madureira et al. (2015), we did observe an effect of estrus duration on fertility. The lack of an impact by intensity may be due in part to the small numbers of cows that had both estrous expression data and were inseminated. Estrous expression was found to be different when DIM of the estrus event was  63   considered. Estrus episodes that occurred before the presynchronization had the lowest estrous expression and events after the second PGF2α had the highest, but whether this is an effect of DIM or an effect of hormone intervention cannot be concluded from this study. It was hypothesized that cows with poor BCS would have limited expression of estrus due to negative energy balance (Butler, 2003) and delays in cyclicity (Chebel and Santos, 2010). Another study from our group also found significant effects of BCS on estrous expression (Madureira et al., 2015a). In the present study we found that BCS did not affect detection or expression of estrus after a synchronization protocol but it did impact the likelihood of successful pregnancy on both farms.  Lameness was not found to have an effect on the expression of estrus as measured by the AAM. Previous research has demonstrated a reduction in daily steps (O’Callaghan et al., 2003; López-Gatius et al., 2005b) and steps at the time of estrus (Walker et al., 2008) in lame cows. Alternatively, Chapinal et al. (2010) found no change in daily steps between lame and non-lame cows, but did find that lame cows walked slower and spent more time lying down per day. Other studies using leg-mounted accelerometers and pedometers have found changes in walking activity caused by lameness (Alsaaod et al., 2012; de Mol et al., 2013; Thorup et al., 2015); however, there has been little research measuring activity using a neck-mounted accelerometer at the time of estrus (in contrast to daily activity), as performed in the current study.  Previous research on the detection of estrus after estrus synchronization protocols have found anywhere from 66 to 71% of the cows in estrus using AAM (Valenza et al., 2012; Fricke et al., 2014b). On the other hand, in a project consisting of 7 commercial farms, Chebel  64   et al. (2010) detected 48.7% of cows in estrus after estrus synchronization using visual observation and tail chalk with a range of 26.7 to 59.8% between farms. Results from the present study showed a discrepancy in the detection of estrus between the two farms (65.0 vs 34.4%), demonstrating that not only parity (Fricke et al., 2014b), cyclicity (Chebel and Santos, 2010), and BCS (DeJarnette et al., 2001) influence estrus detection rates, but farm variation is a key player on the efficacy of these programs.   This study observed 20.2% of cows to be anovular by 50 DIM and that cyclicity had negative impacts on pregnancy outcomes, days to conception, the detection of estrus after the presynchronization protocol, as well as the duration and peak intensity of estrus. Cyclicity was also found to have an interaction with treatment for pregnancy per AI as well as days to conception, where cows that were anovulatory and enrolled in the ACT treatment were the most compromised in both reproductive measures. This prevalence of anovular cows is similar to those found in the United States (average 23% from 50 to 65 DIM; Bamber et al., 2009). The detrimental effects of prolonged anovulation postpartum have been previously reported in the form of lower conception rates, increased days to conception (Chebel and Santos, 2010), increased pregnancy loss (Sterry et al., 2006; Silva et al., 2007) and reduced conception rates at the first postpartum AI (Chebel and Santos, 2010). Yániz et al. (2006) described that cows whom had more estrus events from parturition until 50 DIM had greater pregnancy success by 90 DIM.  Results from this study suggest that parity is also a risk factor for poor fertility as it increases days to conception and reduces pregnancy per AI and estrous expression. Greater parity has previously been reported to impact conception rate (Tenhagen et al., 2004b;  65   Fricke et al., 2014a; Madureira et al., 2015a) and increase days to conception (Fricke et al., 2014b). Estrous expression is impacted by parity as well, shown in this study by a decrease in intensity and duration of estrus, and in other studies where a change in walking activity at estrus has been shown to decrease as lactation number increases (Roelofs et al., 2005a; Yániz et al., 2006; Madureira et al., 2015a). This work is contrary to others that reported parity had no effect on intensity and/or duration of behaviours at estrus (Veerkamp et al., 2000; Løvendahl and Chagunda, 2010; Valenza et al., 2012). Using similar treatments as this study, Fricke et al. (2014b) found a parity by treatment interaction, where multiparous and primiparous cows had similar pregnancy per AI in their equivalent ACT treatment, whereas multiparous had reduced pregnancy per AI in their equivalent TAI treatment. In the present study, we also found an interaction between parity and treatment, but contrary to Fricke et al. (2014b) we found that primiparous and multiparous cows had similar pregnancy per AI for the TAI treatment, but primiparous had greater odds of pregnancy than multiparous cows when enrolled in the ACT treatment; these results are similar to those found by Stevenson et al. (2014).   The present study, as well as others evaluating the efficacy of AAM, consistently found an interaction between treatment and farm regarding the detection of estrus and pregnancy rates. This adds to the discussion about ideal reproductive programs and strengthens the need for consideration of the specific strengths of each farm (e.g.: proper and consistent use of AAM, complying with injection schedules, etc.) during on-farm decision-making regarding reproductive management (Tenhagen et al., 2004a; Stevenson and Phatak, 2005). In an economic model by Giordano et al. (2012) comparing 100% timed  66   AI reproductive programs with programs that incorporate the use of detection of estrus, the importance of farm specific reproductive management programs was recently portrayed. The study displayed that farms with poor pregnancy per AI when inseminating at detected estrus benefited economically by completing timed AI protocols. However, farms that can obtain 30% to 35% conception rates from inseminations at the time of estrus are always more profitable by submitting their cows to AI at the time of estrus instead of completing the full timed AI protocol. In a different economic assessment, researchers assessed reproductive performance and profit using a simulation model of different reproductive programs composed of either detection for estrus, timed AI or a mix of both methods (Galvão et al., 2013). They concluded that both profit and reproductive performance was maximized when farms combined both the detection of estrus and timed AI protocols into their reproductive management; however, if herds can achieve high service rates with either method they may be able to achieve greater profits by focusing on just one.   2.5 Conclusions  This study demonstrated that using AAM for detection of estrus within a presynch-ovsynch program resulted in similar pregnancy per AI and days open compared with a reproduction program that was strictly based on timed AI for first AI. It is important to note that there was a significant impact of farm, suggesting that the use of AAM are probably more prone to individual farm variations, particularly when compared with more established timed AI protocols. In addition, this study demonstrated that lameness was indeed a problem on the farm with the greater prevalence of leg health injuries where it increased the days to  67   conception and decreased the odds of pregnancy, but both leg health and poor BCS had no impact on peak intensity and duration as measured by AAM.        68   Table 2.1: Frequency distributions of explanatory variables for the entire study and stratified by farm.  Factor Entire Study Farm A Farm B Farm A vs. Farm B n1  (%) n  (%) n (%)  (P-value) Cyclicity2  878  443  435  < 0.001 Cycling 701 79.8 389 87.8 312 71.7  Anovular 177 20.2 54 12.2 123 28.3  Leg health3 909  465  444  < 0.001 Adequate 393 43.2 299 76.1 94 21.2  Poor 516 56.8 166 35.7 350 78.8  BCS4 910  466  444  < 0.001 Thin 510 56.0 293 62.9 217 48.9  Moderate 400 44.0 173 37.1 227 51.1  Parity 918  466  452  NS Primiparous 312 34.0 148 31.8 164 36.28  Multiparous 606 66.0 318 68.2 288 63.7  Milk production5 918  466  452  < 0.001 High 459 50.0 199 42.7 260 57.5  Low 459 50.0 267 57.3 192 42.5   Timing of estrus6 -  666  -    Early - - 177 26.6 - -   PGF 1 - - 174 26.1 - -   PGF2 - - 238 35.7 - -   OVS - - 77 11.6 - -   1 Does not always total to 918 cows per variable due to missing observations. 2 Cycling was defined as the presence of at least one corpus luteum present at one of two ultrasound exams performed at the time of the presynchronization. 3 Leg health was defined as adequate if the cow had a hock and gait score both ≤ 2, cows with either a hock score, gait score or both > 2 were considered as having poor leg health.   69   4 BCS ≤ 2.75 was defined as thin, while > 2.75 was defined as moderate.  5 Milk production was measured as the 305-d mature-equivalent yield and divided into two groups using the median 6 Data only available for Farm A.   70   Table 2.2: Pregnancy per AI (P/AI) and DIM (mean ± SE) for treatments (ACT vs. TAI), cows complaint with the treatment1 (ACT vs. TAI), and AI categories (estrus vs. timed AI) for the entire study and stratified by farm. Factor n2 P/AI (%)3 P-value DIM (d) SE P-value Treatment       Entire study 848  0.39   < 0.01 ACT 439 30.8  68.6 0.4  TAI 409 33.5  74.4 0.4  Compliant to treatment1 779  0.13   < 0.01 ACT 439 30.8  68.6 0.4  TAI 340 35.9  75.6 0.4  Farm A 436  0.12   < 0.01 ACT 233 30.5  69.3 0.5  TAI 203 37.4  79.4 0.5  Farm B 412  0.75   NS ACT 206 31.1  67.7 0.5  TAI 206 29.6  69.0 0.5  AI Category       Entire study 836  0.11   < 0.01 Estrus 286 28.3  63.2 0.3  Timed AI 550 33.8  75.7 0.2  Farm A 426  0.26   < 0.01 Estrus 171 30.4  65.4 0.4  Timed AI 255 35.7  80.1 0.4   71   Farm B 410  0.17   < 0.01 Estrus 115 25.2  61.1 0.5  Timed AI 295 32.2  71.4 0.3  1 Cows compliant to treatment excludes any cows that were inseminated outside of the proposed treatment.  2 Does not total 918 due to missing pregnancy diagnoses. 3 Calculated using frequency tables.   72   Table 2.3: Pregnancy per AI (P/AI) and odds ratios (OR) with 95% confidence intervals (95% CI) for pregnancy outcomes for the entire study and for cows that were inseminated correctly according to their assigned treatment (compliant to treatment). Factor  Entire Study  Compliant to Treatment P/AI (%; n/n1) OR 95% CI P-value P/AI (%; n/n) OR 95% CI P-value  Treatment 848   NS 779   0.09 ACT 30.7 (135/439)    30.7 (135/439) ref.   TAI 33.5 (137/409)    35.9 (122/340) 1.13 0.96-1.80  Parity 848   < 0.01 779   < 0.01 Primiparous 41.4 (123/297) 1.60 1.15-2.23  43.5 (117/269) 1.75 1.24-2.46  Multiparous 27.0 (149/551) ref.   27.5 (140/510) ref.   Cyclicity2 819   0.03 750   0.01 Cycling 34.1 (223/655) 1.56 1.04-2.34  35.4 (214/605) 1.77 1.14-2.74  Anovular 23.8  (39/164) ref.   22.8  (33/145) ref.   BCS3 843   < 0.01 774   < 0.01 Low 25.9 (118/456) ref.   26.8 (113/422) ref.   Moderate 39.5 (153/387) 1.63 1.20-2.23  40.6 (143/352) 1.58 1.15-2.19  Milk production4 848   0.03 779   0.04 Low 35.7 (142/398) 1.41 1.04-1.90  36.7 (135/368) 1.40 1.02-1.92   73    High 28.9 (130/450) ref.   29.7 (122/411) ref.   Leg Health5 843   0.10 774   0.09 Poor 27.7 (131/473) ref.   28.2 (122/432) ref.   Adequate 37.8 (140/370) 1.30 0.94-1.78  39.2 (134/342) 1.33 0.95-1.85   Does not total 918 due to missing pregnancy diagnoses. 2 Cycling was defined as the presence of at least one corpus luteum present at one of two ultrasound exams performed at 0 the time of the presynchronization. 1 3 BCS ≤ 2.75 was defined as thin, while > 2.75 was defined as moderate.  2 4 Milk production was measured as the 305-d mature-equivalent yield and divided into two groups using the median 3 5 Leg health was defined as adequate if the cow had a hock and gait score both ≤ 2, cows with either a hock score, gait 4 score or both > 2 were considered as having poor leg health. 5  6   7  74   Table 2.4: Pregnancy per AI (P/AI) and odds ratios (OR) with 95% confidence intervals (95% CI) for pregnancy outcome interactions for the entire study and for cows that were inseminated correctly according to their assigned treatment (compliant to treatment) using a multivariable logistic regression. Factor Entire Study Compliant to Treatment P/AI  (%; n/n) OR 95% CI P-value P/AI  (%; n/n) OR 95% CI P-value Leg health1*Parity    0.10    0.04 Multiparous:         Poor 26.5 (97/366) ref.   27.2 (91/335) ref.   Adequate 28.2 (51/181) 1.05 0.69-1.59  28.1 (48/171) 1.00 0.65-1.53  Primiparous:         Poor 31.8 (34/107) ref.   32.0  (31/97) ref.   Adequate 47.1 (89/189) 1.83 1.09-3.08  50.3 (86/171) 2.09 1.20-3.61  Treatment*Cyclicity2    0.07    NS  ACT:         Cyclic 33.9 (115/339) 2.46 1.24-4.86  33.9 (115/339)    Anovular 16.4  (12/73) ref.   16.4  (12/73)    TAI:          75   Cyclic 34.2 (108/316) 1.12 0.67-1.87  37.2 (99/266)    Anovular 29.7  (27/91) ref.   29.2  (21/72)    Treatment*Parity    0.02    0.06 ACT:         Primiparous 45.1 (73/162) 2.36 1.49-3.73  45.1 (73/162) 2.34 1.49-3.69  Multiparous 22.4 (62/277) ref.   22.4 (62/277) ref.   TAI:         Primiparous 37.0 (50/135) 1.09 0.69-1.73  41.1 (44/107) 1.22 0.74-2.02  Multiparous 31.8 (87/274) ref.   33.5 (78/233) ref.   1 Leg health was defined as adequate if the cow had a hock and gait score both ≤ 2, cows with either a hock score, 8 gait score or both > 2 were considered as having poor leg health. 9 2 Cycling was defined as the presence of at least one corpus luteum present at one of two ultrasound exams 10 performed at the time of the presynchronization.  11  12   13  76   Table 2.5: Pregnancy per AI (P/AI) and odds ratios (OR) with 95% confidence intervals (95% CI) for pregnancy outcomes stratified by farm.  Factor Farm A Farm B P/AI (%; n/n) OR 95% CI P-value P/AI (%; n/n) OR 95% CI P-value Treatment    NS    NS ACT 30.4  (71/233)    31.1 (64/206)    TAI 37.4  (76/203)    29.6 (61/206)    Parity    < 0.01    NS Primiparous 45.8  (65/142) 1.96 1.26-3.06  37.4 (58/155)    Multiparous 27.9  (82/294) ref.   26.1 (67/257)    Cyclicity1     0.02    NS Cycling 35.9 (131/365) 2.64 1.19-5.86  31.7 (92/290)    Anovular 16.3  (8/49)    27.0 (31.115)    BCS2    0.03    < 0.01  Low 28.6  (76/266) ref.   22.1 (42/190) ref.    77    Moderate 41.8  (71/170) 1.59 1.03-2.43  37.8 (82/217) 1.91 1.21-3.01  Milk production3    NS    0.03 Low 34.3  (82/239)    37.7 (60/159) ref.   High 33.0  (65/197)    25.7 (65/253) 1.65 1.06-2.57  Leg health4    NS    0.06 Poor 27.8  (42/151)    27.6 (89/322)  ref.  Adequate 36.8 (105/285)    41.2 (35/85) 1.66 0.99-2.79  1 Cycling was defined as the presence of at least one corpus luteum present at one of two ultrasound exams 14 performed at the time of the presynchronization. 15 2 BCS ≤ 2.75 was defined as thin, while > 2.75 was defined as moderate.  16 3 Milk production was measured as the 305-d mature-equivalent yield and divided into two groups using the 17 median.  18 4 Leg health was defined as adequate if the cow had a hock and gait score both ≤ 2, cows with either a hock score, gait score or both > 2 19 were considered as having poor leg health. 20  21  78   Table 2.6: Factors impacting the proportion and odds ratio (OR) with 95% confidence intervals (95% CI) of cows detected in estrus after a presynchronization protocol using an automated activity monitor. Factor Estrus Detection  (%; n/n) P-value OR 95% CI P-value Farm 444 < 0.001   < 0.001 Farm A 65.1 (151/232)  3.47 2.29-5.27  Farm B 34.4 (73/212)  ref.   Cyclcity1  420 < 0.01   0.08 Cycling 52.2 (181/347)  1.64 0.94-2.87  Anovular 34.3 (25/73)  ref.   Parity 444 0.01   < 0.01 Primiparous 58.2 (92/158)  1.82 1.18-2.81  Multiparous 46.2 (132/286)  ref.   BCS2 440 NS   NS  Low 49.6 (122/246)      Moderate 51.6 (100/194)     Leg Health3 440 < 0.001   NS Poor 40.8 (98/240)     Adequate 62.0 (124/200)     1 Cycling was defined as the presence of at least one corpus luteum present at one of two ultrasound exams performed at the time of the presynchronization. 2 BCS ≤ 2.75 was defined as thin, while > 2.75 was defined as moderate. 3 Leg health was defined as adequate if the cow had a hock and gait score both ≤ 2, cows with either a hock score, gait score or both > 2 were considered as having poor leg health.   79   Table 2.7: Factors impacting estrous expression parameters: peak activity1 and duration2 (LS means ± SE) as measured by an automated activity monitor. Factor Peak Activity (Index) SE P-value Duration (h) SE P-value Cyclicity3   0.07   0.02 Cycling 74.2 0.99  11.3 0.25  Anovular 68.6 2.90  9.6 0.72  Parity   < 0.01   < 0.01 Primiparous 74.3 1.93  11.1 0.48  Multiparous 68.5 1.63  9.8 0.42  BCS4   NS   NS  Low 71.4 1.71  10.5 0.42   Moderate 71.5 1.87  10.3 0.47  Timing of estrus5   < 0.01   < 0.01 Early 65.4b 2.00  9.0a 0.50  PGF 1 73.3bc 1.97  10.7b 0.49  PGF2 75.6c 1.90  11.6c 0.47  OVS 71.4b 2.49  10.4b 0.62  1 Peak activity was defined as the highest index measured by the automated activity monitor.  2 Duration was measured as the time (h) in which the index spent above the activity threshold.  3 Cycling was defined as the presence of at least one corpus luteum present at one of two ultrasound exams performed at the time of the presynchronization. 4 BCS ≤ 2.75 was defined as thin, while > 2.75 was defined as moderate.  5 Timing of estrus was classified as, Early: estrus episodes occurring before the first PGF2α of the presynchronization protocol, PGF1: after the first PGF2α and before the second PGF2α of the presynchronization protocol, PGF2: after the second PGF of the presynchronization protocol but before the start of the Ovsynch protocol and OVS: between the first injection of GnRH until timed AI of the Ovsynch protocol.   80   Figure 2.1: Schematic figure of the reproductive treatments used. Cows were assigned randomly to two treatments after a presynchronization protocol. Cows on the ACT treatment were bred by detection of estrus, and the remaining unbred cows were enrolled in an Ovsynch 56 protocol and bred by timed AI. Cows on the TAI treatment were all enrolled in an Ovsynch 56 protocol and bred by timed AI.     Timed AI Presynchronization Ovsynch Bred by detection of estrus Not bred by detection of estrus Ovsynch If not bred by activity Timed AI Activity Treatment  (ACT) Timed AI Treatment  (TAI) 47 ± 7 DIM  83 ± 7 DIM  83 ± 7 DIM  + 14 d + 12 d + 12 d + 10 d + 10 d  81   Figure 2.2: Survival curves of DIM to conception until 300 DIM for: A) treatment (P > 0.05), B) cows compliant to both treatments (P > 0.05), C) AI category (insemination at estrus or timed AI) (P = 0.03), D) treatment by cyclicity interaction (P = 0.04), and E) farm by leg health interaction (P = 0.08). A)  B)   0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 50 100 150 200 250 300 Proportion of non-pregnant cows Days in milk (d) ACT TAI 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 50 100 150 200 250 300 Proportion of non-pregnant cows Days in milk (d) ACT TAI  82   C)  D)      0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 50 100 150 200 250 300 Proportion of non-pregnant cows Days in milk (d) Estrus Timed AI 00.10.20.30.40.50.60.70.80.910 50 100 150 200 250 300Proportion of non-pregnant cowsDays in milk (d)ACT - AnovularTAI - AnovularACT - CyclicTAI - Cyclic 83   E)     0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 50 100 150 200 250 300 Proportion of non-pregnant cows Days in milk (d) Farm A - Poor Leg Health Farm A - Adequate Leg Health Farm B - Poor Leg Health Farm B - Adequate Leg Health  84   Chapter 3: Rumen-reticular temperature during estrus and ovulation in dairy cows: effects of estrous expression2  3.1 Introduction  New technologies have been developed to improve dairy cattle health and reproductive efficiency on dairy farms. The invention of hormonal protocols such as the Ovsynch have greatly improved reproductive efficiency in lactating dairy cows by increasing insemination rates, but, a growing body of research indicates that insemination programs based on the detection of estrus are equally effective (Fricke et al., 2014b; Neves and LeBlanc, 2015; Denis-Robichaud et al., 2018b). The monitoring of core body temperature to assess febrile states has been widely used in human and veterinary sciences. The recent development of rumen-reticular thermometer technologies that monitor core body temperature provide continuous monitoring of core body temperature in real-time. The ability to automatically measure continuous body temperature expands the applicability for health monitoring, and has the potential to also be used as a predictor of ovulation. Vaginal temperature increases near the onset of estrus and decreases near ovulation (Rajamahendran et al., 1989; Mosher et al., 1990; Talukder et al., 2014).  The recent surge in interest and use of automated technologies for estrus detection requires that these technologies be validated in terms of their performance.                                                               2 A version of this chapter has been submitted for publication: Burnett, T.A., L.B. Polsky, M. Kaur, and R.L.A. Cerri. Rumen-reticular temperature during estrus and ovulation in dairy cows: Effects of estrous expression.  85   Methods of measuring ovulation are limited by the need for either blood or milk hormone analyses or ultrasonography by way of rectal palpation; neither procedure is able to detect the precise time of ovulation. The measurement of the LH surge, which precedes ovulation by approximately 26 hr (Bloch et al., 2006) with 0.5 h deviations (Bloch et al., 2006), is the key event preceding ovulation. However, to be able to measure the LH surge, blood samples need to be taken at short time intervals, thus limiting its practical use on farms. Although the timing of ovulation can be detected using ultrasonography, this methodology also requires a substantial time investment (e.g.: palpating once every 3 hr). In contrast, automated activity monitors (AAM) can accurately detect behavioural changes that are associated with estrus, such as increased number of steps (Madureira et al., 2015a), and reduced rumination (Reith and Hoy, 2012), changes in feed intake (Halli et al., 2015) and lying time (Silper et al., 2017). However, variation in the time to ovulation relative to the onset of estrus remains, a critical factor for a successful pregnancy.  The goals of this study were to 1) determine if changes in rumen-reticular temperature at estrus or ovulation exist and, 2) determine if these changes are impacted by the intensity of estrous expression. Our hypothesis was that increased physical activity during estrus would result in an increase rumen-reticular temperature, and the increase would be tempered by the proximity of ovulation, and that these variations in temperature would be captured by the rumen-reticular temperature logger.     86   3.2 Materials and methods  This experiment was conducted from June until October in 2015 at The University of British Columbia’s Dairy Education and Research Centre, Agassiz, BC. All procedures were approved by the University of British Columbia’s Animal Care Committee (protocols # A15-0089 and #A10-0290).  3.2.1 Animals and housing  A total of 225 episodes of estrus arising from 102 lactating Holstein cows, each having on average 2.2 ± 1.3 (mean ± SD) estrus episodes were used in this study. The cows used in the present study were a subset of animals initially described in a companion study (Burnett et al., 2018).  To be included in the present study cows had to be  30 DIM, had an estrus alert on the AAM, and be equipped with a rumen-reticular temperature bolus.   Cows had a mean (± SD) DIM of 98 ± 57, parity of 2.4 ± 1.6, BCS of 2.78 ± 0.20, gait score (GS) of 2.3 ± 0.70 and a 305-d mature-equivalent yield of 12,600 ± 1,700 kg of milk at each estrus event. Animals were housed in a naturally ventilated wood-framed barn with a free-stall design, equipped with deep sand-bedded stalls and milked twice daily at 0500 h and 1500 h in a parallel milking parlour. Fresh TMR was delivered twice daily at approximately 0700 h and 1600 h. The TMR was formulated following the Nutrient Requirements of Dairy Cattle (NRC) guidelines (NRC, 2001) to meet or exceed the requirements of a 620 kg Holstein cow producing 40 kg/d of 3.5% fat corrected milk; all animals had ad libitum access to both TMR and water.   87   All cows had their body condition and gait scored at the time of enrolment into the study. Body condition was scored on a 5-point scale from thin (1) to obese (5) as outlined by Edmonson et al. (1989). Cows were later classified as thin (< 2.75), average (= 2.75) or moderate (> 2.75). Gait score (GS) was determined on a 5-point scale from normal (1) to severely lame (5) as outlined by Flower and Weary (2006). Animals were later classified as sound (≤ 2) and lame (> 2). Health and production information was collected by the dairy herd personnel with the assistance of the herd veterinarian, and confirmed and recorded by the project leader using the on-farm Dairy Comp 305 software (Valley Agricultural Software, Tulare, CA).   3.2.2 Study design  In this observational cohort study animals were continuously monitored by a collar-mounted automated activity monitor (Heatime®, H-Tags, SCR Engineers, Netanya, Israel) that was fitted on the upper left side of the neck beginning 10 d after parturition. Animals were enrolled into the study when the monitor identified them as having crossed the alert threshold, set at 35 index; an index activity of 35 equates to approximately a 6-SD change in activity compared with a baseline set by the AAM system. We used peak activity (maximum activity during an estrus episode) and duration (amount of time the animal spent with activity above the threshold) to describe the expression of estrus (see also Madureira et al. (2015)).  88   Weighted rumen-reticular thermometer boluses (TempTrack®, DVM Systems, Greeley, CO) were administered orally into the digestive tract of the cows using a specialized gun at 35 ± 7 DIM. Associated software allowed for hourly temperature monitoring. The AAM was checked twice daily during milking times for cows in estrus. Animals classified as in estrus were then monitored for ovulation using ultrasonography immediately following milking for a maximum of 6 examinations (approximately 60 h after the first ultrasound). Environmental temperature and humidity were recorded hourly using an onsite weather station (Agriculture and Agri-Food Canada, Agassiz, BC). The Temperature Humidity Index (THI) was calculated following Mader et al. (2006): THI = (0.8 * Temperature (oC)) + ((Humidity/100)*(Temperature -14.3)) + 46.4. Maximum THI was determined for the morning (0000 to 0759 h), afternoon (0800 to 1559 h) and night (1600 to 2359 h) periods; the THI at the time of onset of estrus was determined based on these periods.    3.2.3 Determination of estrus, ovulation and ovulation times  As mentioned, cows had their ovaries examined by ultrasound (Ibex Pro; E.I. Medical Imaging, Loveland, CO) using a 7.5 MHz linear-array rectal probe at enrolment, and then twice daily, until a maximum of 6 per rectum examinations. The presence and diameter of the 3 largest follicles and corpus lutea were measured and recorded. Cows were classified as in estrus if they had at least one dominant pre-ovulatory follicle greater than 15 mm and an absence of a large corpus luteum (>20 mm) at the time of the estrus alert. Ovulation was determined by the disappearance of the dominant pre-ovulatory follicle; the time of ovulation was determined as the median time between the two ultrasound examinations  89   where the pre-ovulatory follicle disappeared. Ovulation intervals were calculated as the time from when an alert on the AAM occurred until the time of ovulation.  All animals with estrus alerts had their ovaries scanned using ultrasonography 7 days after their last palpation for the presence of a new corpus luteum which was used for confirmation of estrus or false alerts. A new corpus luteum was used to confirm ovulation for true estrus alerts followed by ovulation. For estrus alerts without ovulation confirmation was obtained using a maximum of 6 ultrasound examinations; the presence of a new corpus luteum was used to determine late ovulations and the absence of a new corpus luteum was indicative of failed ovulations. For false alerts, the absence of a new corpus luteum was used to confirm that the estrus alert was correctly classified as false. For estrus events with late or failed ovulations and false alerts of estrus, temperature change data were removed from the analysis.  3.2.4 Rumen-reticular temperature data  Changes in rumen-reticular temperature at estrus were determined for the entire duration of estrus, as measured by the AAM. To determine changes in rumen-reticular temperature around ovulation, temperature data were collected from the period between the ultrasound examination that confirmed ovulation and the preceding exam. The maximum temperature change (MTC; i.e.: positive amplitude), minimum temperature change (MiTC; i.e. negative amplitude) and the positive area under the curve relative to baseline (AUC) were calculated at estrus (MTCE/MiTCE/AUCE) and around ovulation  90   (MTCO/MiTCO/AUCO). Measures of AUC were used as a method of measuring the intensity and length of time each animal spent above their respective baselines during estrus and ovulation. The AUC was calculated using the trapezoidal rule, where only positive areas were considered. To determine if temperature changes around estrus could be potentially used within an alerting system for ovulation timing, we calculated the magnitude of change in temperature relative to a rolling baseline value using the following formula: (Temperature - Baseline Temperature)/ Baseline Standard Deviation. The baseline values and standard deviations were calculated hourly using the previous 5 d of hourly temperature readings after they had been corrected for bouts of water intake using a proprietary manufacture’s algorithm (TempTrack®, DVM Systems, Greeley, CO). Data were cleaned to only include 12 hours before the AAM alert until 12 hours after ovulation. Ovulation has previously been associated with a drop in temperature (Miura et al., 2017) thus alerts were then categorized as estrus or ovulation alerts based on the direction of change, where estrus alerts were of positive change and ovulation of negative change. An alert was considered as two temperature readings in a row that exceeded the following thresholds: 0.5 STD, 1 STD, 1.5 STD or 2 STD.  3.2.5 Statistical analyses  All statistical analyses were carried out with SAS Studio (version 5.1; SAS Inst. Inc., Cary, NC) with estrus episode as the experimental unit. The data were examined for missing  91   data points and events missing greater than 25% of the hourly data points, either at estrus or around ovulation, were excluded from the analysis. Estrus events that were confirmed as false alerts were also excluded from the analysis. Before analyses, data were checked for normality using the UNIVARIATE procedure and probability distribution plots. The AUC and MTC were used as dependent variables and were tested for the effects of sampling time (at estrus vs. around ovulation), estrous expression, THI at the time of onset of estrus, parity, BCS, GS, and stage of lactation using ANOVA with sampling time as a repeated measure and estrus event nested in cow as a random effect using the MIXED procedure. Due to collinearity, neither peak activity and duration nor AUC and MTC were ever used within the same model. Peak activity was categorized using the median of 80 index into high and low intensity of events. Duration of estrus was also categorized by the median, 12 hours, into short and long estrus events. THI at the onset of estrus was categorized as high and low using a THI index of 72. Cows with more than one lactation were classified as multiparous and those with only one lactation were considered primiparous.   3.3 Results  3.3.1 Rumen-reticular temperature  All measures of temperature change (AUC, MTC, and MiTC) were affected by sampling time; where AUC and MTC were greater at estrus than around ovulation (2.4 ± 0.2 and 1.7 ± 0.20C2 for AUCE and AUCO, respectively; P < 0.01; 0.55 ± 0.03 and 0.32 ± 0.040C for MTCE and  92   MTCO, respectively; P < 0.001) and MiTC was more negative at ovulation than at estrus (-0.27 ± 0.050C and -0.58 ± 0.050C  for MiTCE and MiTCO, respectively; P < 0.001). However, there was a three-way interaction present between sampling time, estrous expression and THI at the time of estrus onset for AUC (Figure 3.1), MTC (Figure 3.2) and MiTC (Figure 3.3); this was consistent whether estrous expression was measured using peak activity or duration of estrus, with the exception of the three-way interaction of MiTC, peak activity, and THI which was not significant. Cows with greater estrous expression had greater AUC and MTC at the time of estrus than at ovulation, but cows with lesser estrous expression did not, irrespective of THI. It is important to note that although cows with less estrous expression had equal AUC and MTC at both estrus and ovulation, they failed to return to basal levels, as neither AUC nor MTC had returned to zero. Cows with greater estrous expression had greater AUC and MTC relative to those with lesser expression at the time of estrus, but the magnitude of these changes was reliant on THI. On days with higher THI, cows with greater estrous expression had larger changes of both AUC and MTC than those with lesser estrous expression; on days with lower THI the effect of estrous expression on rumen-reticular temperature change was dampened. Around ovulation, there was generally no differences in either AUC nor MTC relative to estrous expression or THI. The associations of sampling time, estrous expression, and THI at the onset of estrus with AUC, MTC and MiTCH have been depicted in Figures 3.1, 3.2 and 3.3, respectively. BCS was found to have an interaction with sampling time for MTC and MiTCH, where MTCO and MiTCO were consistent across BCS categories but MTCE and MiTCE varied by BCS (MTCO: 0.29 ± 0.06, 0.36 ± 0.04, and 0.31 ± 0.05 CO; MTCE: 0.67 ± 0.06, 0.53 ± 0.04, and 0.45  93   ± 0.05 OC; P = 0.03; MiTCO: -0.73 ± 0.09, -0.55 ± 0.05, and -0.44 ± 0.08 CO; MiTCE: -0.29 ± 0.08, -0.28 ± 0.05, and -0.19± 0.07 OC for thin, average and moderately conditioned animals; P = 0.05; Figure 3.4); BCS had no impact on AUC at either sampling times. Parity, stage of lactation, GS, and milk production all failed to impact AUC, MTC or MiTCH at estrus or around ovulation.  3.3.2 Temperature alerts  Temperature alerts for estrus were found for 87.0, 76.8, 56.8, and 35.3% of estrus events when using 0.5 STD, 1.0 STD, 1.5 STD and 2.0 STD from baseline as thresholds and temperature alerts for ovulation were found for 85.8, 65.2, 33.7, and 18.9% of events, respectively. Although more temperature alerts were found using the 0.5 STD threshold, alerts varied significantly. Even with the highest threshold for alert (2 STD), the precision of the estrus and ovulation alerts were not more precise (smaller standard deviations from the mean), but only a small proportion of events were alerted (Table 3.1). In general, temperature alerts for estrus occurred closer to AAM alerts but before the time of ovulation. Temperature alerts for ovulation occurred after AAM alerts and before ovulation. Temperature alerts for estrus varied less around the median in comparison with alerts for ovulation (Figure 3.5). The mean time (± SD) intervals between AAM alerts, alerts for estrus and ovulation based on changes in temperature, and timing of ovulation have been summarized in Table 3.1.    94   3.4 Discussion  This is the first study to report the impacts of intensity and duration of estrous expression on changes in body temperature associated with estrus and ovulation. The outcomes of this manuscript are supported by results of Suthar et al. (2011) and Wrenn et al. (1958) who found higher vaginal temperatures during estrus and lower vaginal temperature around ovulation in lactating dairy cows. Interestingly, Suthar et al. (2011) carried out their study in a tie-stall barn, suggesting that changes in temperature around estrus occur even in the absence of physical exertion. Previous research using vaginal temperature has accounted much of the increase in temperature at estrus to increased blood flow to the vagina caused by increased concentrations of estradiol (Abrams et al., 1973); however, as the current study used rumen-reticular temperature, increased thermal conductance in the vagina is likely not the only factor impacting body temperature at the time of estrus. In agreement, Cooper-Prado et al. (2011) and Randi et al., (2018) also reported rumen-reticular and external auditory canal temperature to increase at the onset of estrus, respectively, further suggesting that thermal conductance is not the only reason for changes in temperature at estrus. Suthar et al. (2011) also suggests that changes in temperature may not only be from physical activity but may be associated with factors that regulate gonadotropin-releasing hormone having an impact on central thermoregulation. Fisher et al. (2008) demonstrated that the LH surge was associated with an increase in vaginal temperature, suggesting that LH may have thermogenic properties. However, in the current study, we did not find rumen-reticular temperature to be increased at estrus in cows with low estrous expression, suggesting that physical movement may be more important for  95   changes in rumen-reticular temperature than for other measures of body temperature.   Although not tested in this study, previous reports arising from our laboratory using similar AAM have demonstrated that greater estrous expression is associated with higher fertility (Madureira et al., 2015a; Burnett et al., 2017a; Polsky et al., 2017) and lower anovulation rates (Burnett et al., 2018; Madureira et al., 2019). Greater estrous expression in the current study was found to generate higher amplitude changes in temperature at estrus but have generally unchanged temperature changes around ovulation, suggesting that greater increases in temperature at estrus may be associated with positive impacts on fertility as well. Further research is warranted to establish if changes in temperature related to estrus and ovulation are associated with fertility. From the current study, we are not able to decipher if the heightened temperature response in animals with higher intensity and longer duration of estrous expression is solely due to increased movement or if there are other underlying factors. Increased rumen-reticular temperature at estrus observed in the current study supports the results from previous research using vaginal temperature where temperature surges were associated with estrus and thus a predictor of ovulation in dairy cows. Fisher et al. (2008) reported a high correlation between the time of LH surge and peak vaginal temperature in non-lactating dairy cows. The authors concluded that an increase in vaginal temperature occurred within 6 h of the LH surge. Using vaginal thermometers in heifers, Mosher et al. (1990) reported that the interval between onset of temperature increase until ovulation was 21.1 h, whereas the interval from LH peak to the time of ovulation was 21.4 h. These results suggested that the onset of an increase in vaginal temperature may be a good  96   proxy for the LH surge and thus ovulation. Additionally, Clapper et al. (1990) reported less variation in the interval from the LH surge to the onset of vaginal temperature spike compared with the interval from peak estradiol to the onset of vaginal temperature peak. In the present study, rumen-reticular temperature did increase at estrus, but no agreements between temperature alerts, using standard deviations from baseline, and ovulation times were found. Although 88 – 19% of the estrus events captured by the AAM also displayed a temperature alert, depending on the threshold used, the time range between the temperature alerts and ovulation was too wide to be used practically for breeding decisions or the detection of ovulation for research.  Similar to the current study, Talukder et al. (2014) were able to obtain high sensitivity for estrus by measuring vulvar temperature using infrared technology, but unable to obtain acceptable values for specificity and the range in time from the estrus alert to ovulation was quite large (16 – 60 h), where 73% of ovulations occurred 24 to 47 h after the alert. Contrarily, Kyle et al. (1998) reported, by using a threshold of a change in vaginal temperature from baseline of 0.4 0C for a minimum of 3 h, to have obtained a detection sensitivity and a positive predictive value of 89.4 and 85.7%, respectively, which was higher sensitivity than using visual observations. Additionally, a recent study using changes in skin surface temperature demonstrated that they were able to increase their specificity and positive predictive values by including a necessary decrease in temperature within the prior 72 h (Miura et al., 2017). Although changes in temperature has potential for being an indicator of estrus, more research is necessary to decipher more accurate ways of automatically monitoring ovulation times in dairy cows. Future studies should include  97   secondary parameters, such as activity, lying, rumination and feeding time, which could be used in unison with rumen-reticular temperature to make predictions more accurate. During days with greater THI, changes in rumen-reticular temperature at estrus were found to be greater than on days with lower THI. The impacts of estrous expression on these temperature changes were less drastic under cooler and less humid conditions. Contrary to this study, Cooper-Prado et al. (2011) did not find an impact of ambient temperature on changes in rumen-reticular temperature at the time of estrus. Similarly, Sakatani et al. (2016) did not find an impact of season on changes in vaginal temperature at estrus. However, they also did not find an impact of season on estrous expression and within that study, non-lactating Japanese Black cows were used, thus it is possible that there are breed differences in terms of coping with changes in environmental temperatures. Body temperature is a balance between accumulated heat, either through physiological processes (e.g.: milk production, physical exercise) or the environment, and the ability to dissipate heat (West, 2003). Under lower THI conditions, cows with high estrous expression had a smaller increase in rumen-reticular temperature at estrus than those in higher THI conditions. This may be attributed to animals being better able to dissipate heat in colder and less humid environments (West, 2003).  Cows with low BCS had higher changes in rumen-reticular temperature at estrus than average and moderately conditioned animals. This is contrary to the literature on estrous expression, which generally states that thin animals have lower estrous expression (Aungier et al., 2012; Madureira et al., 2015a; Polsky et al., 2017). Additionally, no association of body temperature with the BCS of healthy cows within the first 10 DIM (Suthar et al., 2012) or  98   non-lactating cows (Scanavez et al., 2018) were found in previous studies. However, the current study was measuring the change in body temperature from basal levels, and may explain why we found different results than other researchers who were reporting factors impacting basal temperature. Future research is necessary to determine if there may be certain factors that either increase the amount of heat accumulated or decrease the ability of thin cows to dissipate heat.   3.5 Conclusions  This study demonstrated that changes in rumen-reticular temperature were higher at the time of estrus than around the time of ovulation; however, changes in temperature at estrus was largely explained by the intensity of estrous expression. Changes in rumen-reticular temperature at estrus are impacted by estrous expression, where estrus events characterized as having greater estrous expression had higher changes in temperature than those with low estrous expression. The magnitude of these temperature changes was impacted by the environmental temperature and humidity on the day of the event. Although differences in rumen-reticular temperature can be found at estrus, the resulting variation prevents us from recommending this technique as an early alert system for estrus or ovulation at this time. Future research should focus on how rumen-reticular temperature can be used more efficiently and accurately in determining the timing of ovulation, as well as a more in depth analyses using temperature in parallel with other automated sensors.  99   Table 3.1: Descriptive statistics of time intervals between the time of ovulation, automated activity monitor (AAM) alerts and temperature alerts for estrus an ovulation based on standard deviation changes in rumen-reticular temperature in dairy cows. On average, temperature alerts came after AAM alerts for estrus, but before ovulation.   Interval Alerted1  (%; n/n) Mean Time (h) SD Minimum Maximum AAM Alert – Temperature Estrus Alert      0.5 STD 87.9  (167/190) -4.2 9.0 -12.0 45.3 1 STD 76.8  (146/190) 0.8 11.5 -12.0 69.5 1.5 STD 56.8 (108/190) 2.0 9.5 -12.0 29.2 2 STD 35.3 (67/190) 4.0 10.0 -11.9 33.6 Temperature Estrus Alert – Ovulation      0.5 STD - -29.3 12.8 -98.3 1.4 1 STD - -24.1 13.4 -98.3 4.9 1.5 STD - -22.5 14.1 -98.3 6.5 2 STD - -19.8 15.2 -98.3 6.8 AAM Alert – Temperature Ovulation Alert      0.5 STD 85.8 (163/190) 2.7 14.0 -12.0 39.1 1 STD 65.2 (124/190) 9.7 16.9 -12.0 74.2  100   1.5 STD 33.7 (64/190) 10.4 16.9 -11.8 74.2 2 STD 18.9 (36/190) 13.0 15.1 -10.7 35.5 Temperature Ovulation Alert – Ovulation      0.5 STD - -23.1 18.1 -89.6 11.4 1 STD - -16.1 18.9 -87.6 11.6 1.5 STD - -17.1 20.9 -87.6 10.9 2 STD - -13.4 17.3 -50.2 11.6 1 Proportion of AAM estrus alerts which were also alerted based on changes in rumen-reticular temperature.      101   Figure 3.1: Area under the curve of rumen-reticular temperature relative to baseline at the time of estrus and ovulation for estrus events with high and low peak activity (panel A; P < 0.001) and long and short duration of estrus (panel B; P < 0.001). Superscripts of letters a-d denote significant differences (P < 0.05), while letters x-y denote tendencies (0.05  P < 0.10). High peak activity: estrous expression greater than the median of 80 index on the AAM. Low peak activity: estrous expression less than the median. Long duration: estrous expression lasting longer than the median of 12 hr. Short duration: estrous expression lasting less than the median. High THI: THI at the onset of estrus greater than 72. Low THI: THI at the onset of estrus less than 72. A)  B)    102   Figure 3.2: Maximum change in rumen-reticular temperature at the time of estrus and ovulation for estrus events with high and low peak activity (panel A; P < 0.01) and long and short duration of estrus (panel B; P < 0.01). Superscripts of letters a-d denote significant differences (P < 0.05), while letters x-y denote tendencies (0.05  P < 0.10). High peak activity: estrous expression greater than the median of 80 index on the AAM. Low peak activity: estrous expression less than the median. Long duration: estrous expression lasting longer than the median of 12 hr. Short duration: estrous expression lasting less than the median. High THI: THI at the onset of estrus greater than 72. Low THI: THI at the onset of estrus less than 72. A)  B)    103   Figure 3.3: Maximum negative change in rumen-reticular temperature at the time of estrus and ovulation for estrus events with high and low peak activity (panel A; P < 0.01) and long and short duration of estrus (panel B; P < 0.01). Superscripts of letters a-d denote significant differences (P < 0.05), while letters w-z denote tendencies (0.05  P < 0.10). High peak activity: estrous expression greater than the median of 80 index on the AAM. Low peak activity: estrous expression less than the median. Long duration: estrous expression lasting longer than the median of 12 hr. Short duration: estrous expression lasting less than the median. High THI: THI at the onset of estrus greater than 72. Low THI: THI at the onset of estrus less than 72. A)  B)    104   Figure 3.4: Maximum postive (Panel A) and negative (Panel B) change in rumen-reticular temperature of at the time of estrus and ovulation for estrus events with varying body condition score. Superscripts of letters a-d denote significant differences (P < 0.05), while letters x-y denote tendencies (0.05  P < 0.10). Thin: BCS <2.75, Average: BCS = 2.75, Moderate: BCS > 2.75. A)  B)  105   Figure 3.5: Boxplots demonstrating the distribution of time intervals between automated activity monitor (AAM) alerts, alerts for estrus and ovulation based off of rumen-reticular temperature change, and the timing of ovulation.     Temperature Estrus Alerts temperature Temperature Ovulation Alerts  Time interval (hr) Legend: AAM alert – Estrus alert (Temp.)   0.5 STD      1.0 STD  1.5 STD      2.0 STD Estrus alert (Temp.) – Ovulation  0.5 STD      1.0 STD  1.5 STD      2.0 STD AAM alert – Ovulation alert (Temp.)  0.5 STD      1.0 STD  1.5 STD      2.0 STD Ovulation alert (Temp.) – Ovulation  0.5 STD      1.0 STD  1.5 STD      2.0 STD  106   Chapter 4: Effect of estrous expression on timing and failure of ovulation of Holstein dairy cows using automated activity monitors3 4.1 Introduction   Timing of AI relative to ovulation is important for successful fertilization. Inseminations occurring too long after the onset of estrus will not permit sufficient time for capacitation of the spermatozoa resulting in the fertilization of poor quality oocytes, while insemination too early will result in the death of many spermatozoa before the oocyte arrives at the ampulla for fertilization (Saacke et al., 2000; Dalton et al., 2001). Previous reports have discussed the optimal timing of AI relative to estrous behaviors. The most optimal interval for AI after the onset of visually detected estrus was 7 to 12 h (Hall et al., 1959). Another study reported the greatest conception rates for inseminations occurring 4 to 12 h after the onset of estrus using a pressure-sensing system (Dransfield et al., 1998).  Using an automated activity monitor (AAM), Stevenson et al. (2014) reported a difference between primiparous and multiparous cows, where the optimal time for AI relative to onset of estrus was 13 to 16 h post-alert whereas primiparous cows had relatively stable conception risks when being inseminated any time between 0 to 16 h post-alert. Using visual detection, Van Eerdenburg et al. (2002) reported that cows that ovulated                                                              3 A version of this chapter has been accepted for publication: Burnett, T.A., L.B. Polsky, M. Kaur, and R.L.A. Cerri. 2018. Effect of estrous expression on timing and failure of ovulation of Holstein dairy cows using automated activity monitors. J. Dairy Sci. 101:11310-11320.  107   between 0 and 24 h post-AI had greater conception rates and also had more estrous expression than those that ovulated before insemination or later than 24 h.  High-producing cows have reduced estrous expression (Lopez et al., 2004; Rivera et al., 2010), which can be further compromised by current management practices in which cows are kept in free-stall barns with concrete flooring and little or no access to dirt lots (Stevenson, 2001). Decreased estrous expression has not only made the detection of cows for AI more difficult, but has also been seen to impact the fertility of the cow. Dransfield et al. (1998) reported that cows that stood to be mounted more times per estrus had greater conception rates. Performing standing estrus (Bijker et al., 2015), having increased intensity of estrus (Madureira et al., 2015a) or walking activity (López-Gatius et al., 2005b) have all individually been shown to better pregnancy success in dairy cows.  Other common factors impacting fertility are issues with physical health such as lameness and BCS. Reduced BCS is associated with reduced conception rates at first AI postpartum, more days open, and postponed resumption of cyclicity resulting in a longer interval from parturition to first estrus and first service (Roche et al., 2009). Carry-over effects of reduced BCS during the transition period have been demonstrated to be detrimental even after the voluntary waiting period, where the likelihood of pregnancy has been found to be reduced with each 0.5-unit decrease of BCS (Buckley et al., 2003; Roche et al., 2007). Lameness also is associated with less pregnancy per AI (Burnett et al., 2017a), and increased number of inseminations per conception (Hernandez et al., 2001; Melendez et al., 2003), days to conception (Burnett et al., 2017a) and prevalence of ovarian cysts (Melendez et al., 2003). It is not well understood how expression of estrus, BCS, and lameness affect  108   fertility. One potential cause may be from disruptions of the events that lead to ovulation. In the present study, we aimed to determine if the reductions in fertility that are found with lesser estrous expression, as well as with lameness and BCS, are associated with the interval from activity alert to ovulation using AAMs.  The main goal of this project was to determine the effect of estrous expression, as measured by two different AAMs, on ovulation timing and ovulation failure. We hypothesized that decreased estrous expression and compromised physical health (i.e., poor BCS and lameness) would increase the proportion of cows with ovulation failure and those ovulating outside the ideal timing relative to the onset of estrus and consequently insemination.  4.2 Materials and methods   This experiment was conducted at the University of British Columbia’s Dairy Education and Research Centre, Agassiz, BC. All procedures were approved by the Animal Care Committee of the University of British Columbia. Animals used in this experiment were cared for as outlined by the guidelines provided by the Canadian Council for Animal (2009).  4.2.1 Animals and housing   A total of 850 episodes of estrus were used from 293 high-producing Holstein dairy cows in this study; an average of (mean ± SD) 2.5 ± 1.7 episodes were used per cow. Cows had a mean DIM of 103 ± 56, lactation of 2.7 ± 1.7, BCS of 2.80 ± 0.20, gait score (GS) of 2.2 ±  109   0.8, hock score (HS) of 1.3 ± 0.5, and a 305-d mature-equivalent yield of 12,500 ± 1,900 kg of milk. Cows were housed in a naturally ventilated wood-framed barn with a free-stall design, equipped with deep sand-bedded stalls. Cows were milked twice daily at 0500 and 1500 h through a conventional milking parlour. Fresh TMR was delivered twice daily at approximately 0700 and 1600 h. The TMR was formulated following the Nutrient Requirements of Dairy Cattle (NRC) guidelines (NRC, 2001) to meet or exceed the requirements of a 620 kg Holstein cow producing 40 kg/d of 3.5% fat corrected milk. All cows had ad libitum access to both feed and water.  4.2.2 Study design   This was an observational cohort study. Cows were enrolled onto the study when they crossed the alert threshold, set by the company recommendations for the collar-mounted AAM (AAMC). At enrolment, cows were body condition, gait and hock scored, and scanned by transrectal ultrasonography to determine ovarian structures using an ultrasound (Ibex Pro; E.I. Medical Imaging, Loveland, CO) with a 7.5 MHz linear-array probe. Inclusion criteria for the present study was that cows had to be greater than or equal to 30 DIM and have an estrus alert on the AAMC. The AAMs were checked twice daily during milking times for cows in estrus. Cows classified as in estrus were then monitored for ovulation using ultrasonography after each milking (2 x per d) at approximately 0800 and 1700 h for a maximum of 6 times (approximately 60 h after the first ultrasound).  110   Each cow was equipped with an AAMC (Heatime®, H-Tags, SCR Engineers, Netanya, Israel) before 10 d post calving and a leg-mounted AAM (AAML: Boumatic Heat-seeker-TX®, Boumatic Dairy Equipment Co., Madison, WI) on the day of calving of their first lactation.  4.2.3 Determination of estrus, ovulation and ovulation times   Cows were classified to be in estrus if they had at least one dominant pre-ovulatory follicle greater than 15 mm and an absence of a large corpus luteum (> 20 mm) at the time of the estrus alert. Ovulation was determined by the disappearance of the dominant pre-ovulatory follicle; the time of ovulation was determined as the median time between the ultrasound where the pre-ovulatory follicle disappeared and the ultrasound preceding its disappearance. For estrus events where ovulation occurred after the maximum of 6 ultrasounds, ovulation time was calculated as the time of the last ultrasound plus 7.75 h (the average median hours between all ultrasounds). Intervals from activity alert to ovulation were calculated as the time from when an alert on the respective AAM occurred until the time of ovulation. Intervals to ovulation were classified into 12 h periods as: short (< 20 h), medium (20-31 h) or long (≥ 32 h); 12 h blocks were used because assessment for ovulation was only carried out twice daily. All cows with estrus alerts, including false alerts, were scanned using an ultrasound 7 d after the alert for the presence of a new corpus luteum, which was used for confirmation of estrus and ovulation classifications. For true estrus alerts where ovulation was detected, a new corpus luteum was used to confirm ovulation. For true estrus alerts where ovulation was not detected by the maximum of 6 ovarian ultrasound checks, the presence of a new  111   corpus luteum was used to determine late ovulations and the absence of a new corpus luteum was indicative of failed ovulation. For false estrus alerts, the absence of a new corpus luteum was used to confirm that the estrus alert was correctly classified as false.   4.2.4 Expression of estrus  Physical activity was continuously monitored using both AAMs. The threshold was set at an activity index of 35 for the AAMC and at a relative increase in activity of 180% for the AAML; an index activity of 35 equates approximately to a 6-standard deviation increase in activity in relation to baseline within the AAMC system. Two traits, previously used by Madureira et al. (2015), were used to describe the expression of estrus: 1) peak activity and 2) duration. Peak activity was defined as the maximum activity (in index or percentage relative increase for collar- and leg-mounted monitors, respectively) during an episode of estrus. Duration of estrus was defined as the time the activity of the cow exceeded threshold values set by the AAM software. The term estrous expression will be used to encompass both traits unless specified. For the AAML, only peak activity could be calculated, although both peak activity and duration were calculated for the AAMC. Peak activity and duration of estrus were categorized into groups above and below their respective medians (Peak activity AAMC: High ≥ 80 index, Low < 80 index; Peak activity AAML: High ≥ 331 relative increase (%), Low < 331 relative increase (%); Duration AAMC: Long ≥ 12 hr, Short < 12 hr).   112   4.2.5 Health scoring   All cows were body condition, gait and hock scored at the time of enrollment onto the study. Body condition was scored on a 5-point scale from thin (1) to obese (5) as outlined by Edmonson et al. (1989). Cows were later categorized as Thin (< 2.50), Average (= 2.75) or Moderate (> 3.00). Lameness was scored on a 5-point scale from sound (1) to severely lame (5) as outlined by Flower and Weary (2009). Cows were later classified as Sound (≤ 2) and Lame (> 3). Hock lesions were scored on a 4-point scale modified from the Hock Assessment Chart for Cattle developed by Cornell Cooperative Extension (http://hdl.handle.net/1813/36913) as follows: (1) no swelling and without balding, (2) minor swelling without balding, (3) minor to moderate swelling with bald area and (4) severe swelling with bald area, with or without broken or scabbed skin. As the incidence of hock lesions of ≥ 3 was only 1.8%, there was not enough variation, and the variable was removed from the study. Milk weights were recorded automatically twice daily at milking through a conventional milking parlor (Boumatic Dairy Equipment Co., Madison, WI) and the data were automatically transferred and stored on Dairy Comp 305 software (Valley Agricultural Software, Tulare, CA). Health and production information was collected by the dairy herd personnel with the assistance of the herd veterinarian, and confirmed and recorded by the project leader using the on-farm Dairy Comp 305 software.   113   4.2.6 Ultrasonography and pregnancy diagnosis  Ovaries of cows in estrus were examined by ultrasound (Ibex Pro; E.I. Medical Imaging, Loveland, CO) using a 7.5 MHz linear-array rectal probe at enrolment, and twice daily, until a maximum of 6 palpations. Presence and diameter of the 3 largest follicles and corpora lutea were measured and recorded. Cows were inseminated at estrus using the AM/PM rule in relation to when the estrus alert was triggered, the mean ( SD) time from alert to breeding was 11.0  5.3 and 10.1  6.5 h for the AAMC and AAML, respectively. The voluntary waiting period was 60 DIM and the cows had a mean number of inseminations of 2.6 ± 1.7, with a range from first to ninth service.  All cows were examined by way of ultrasound for pregnancy at 36 ± 7 d post-AI for the detection of an embryonic vesicle with a viable embryo (presence of heartbeat). Pregnancy diagnosis was carried out by the herd veterinarians.  4.2.7 Statistical analyses  All statistical analyses in this study were carried out with SAS Studio (version 3.4; SAS Inst. Inc., Cary, NC) with episode of estrus as the experimental unit. Prior to all analyses data were checked for normality using the UNIVARIATE procedure and probability distribution plots.  Due to collinearity, AAMC peak activity, AAMC duration and AAML peak activity were not used together in the same model as independent variables.   114   Peak activity and duration of estrus were used as continuous dependent variables and tested for the fixed effects of parity, BCS, GS, stage of lactation (≤ 60, 61-119, and ≥ 120 DIM), interval to ovulation (< 20, 20-31, and ≥ 31 hr), and false alerts using ANOVA with cow as a random effect using the MIXED procedure. Pregnancy per AI and ovulation failure were binomial dependent variables assessed using the GLIMMIX procedure with cow as the random effect, estrus event as the experimental unit and parity, BCS, GS, stage of lactation (≤ 60, 61-119, and ≥ 120 DIM), and estrous expression (AAMC peak activity, AAMC duration or AAML peak activity) as fixed effects. Peak activity and duration were categorized by the median for both AAMC (Peak activity: high and low; Duration: long and short) and AAML (Peak activity: high and low). The LOGISTIC procedure, with backwards elimination using Wald’s statistic criterion when P < 0.15, was used to determine which independent variables (same as above) impacted the occurrence of long and short intervals from activity alert to ovulation (using the median of 20 hr) and false alerts. Results from the MIXED and GLIMMIX procedures are presented as LSMEANS, while those from LOGISTIC are presented as odds ratios. Values were considered a tendency when having a P-value of  0.10 and as statistically significant when  0.05. Time of day of estrus alert was used as a covariate as previous research has reported it to impact the expression of estrus (Stevenson et al., 2014).      115   4.3 Results  4.3.1 Intervals from activity alert to ovulation  The average time from AAMC alert to ovulation was (mean ± SD) 25.8 ± 10.2 h, and the average time from AAML alert to ovulation was 24.7 ± 9.3 h. Descriptive statistics of AAMC intervals to ovulation, AAML intervals to ovulation, and AI to ovulation intervals are summarized in Table 4.1. The average peak activity (AAMC and AAML) and duration (AAMC) found in this study were 72.4 ± 20.7 index, 380 ± 146 relative increase (%), and 11.1 ± 5.4 h, respectively. Changes in estrous expression were associated with differences in intervals to ovulation. Cows with a short interval to ovulation exhibited less estrous expression than cows with medium and long length intervals to ovulation using the AAMC (P < 0.001), while when using the AAML, cows with short intervals to ovulation performed less estrous expression than cows with medium but the same as those with long intervals to ovulation (P = 0.02). Data on estrous expression and intervals from activity alert to ovulation have been summarized in Table 4.2. When using the AAMC, the odds of events of estrus with less peak activity or shorter duration having a short interval to ovulation (below the median of 20 h) were 2.6 and 4.7 times greater than the odds of an estrus event with greater activity (P <0.001) or longer duration (P <0.001), respectively. When using the AAML, similar results were found, where the odds of an estrus event with lesser activity having a short interval to ovulation was 1.9 times greater than the odds of estrus events with increased peak activity  116   (P < 0.01). The frequency of estrus events with low and high estrous expression in each of the ovulation interval categories has been summarized in Figure 4.1.   When using the AAMC, estrus events occurring in early lactation (30 - 60 DIM) had 1.7 and 2.3 greater odds of having an interval to ovulation less than the median (< 20 hr) than those in mid (61-119) or late lactation (≥ 120 DIM), respectively (P = 0.01). When using the AAML, the odds of estrus events early in lactation having an interval to ovulation of < 20 h was 2.3 times greater than the odds of estrus events in late lactation (P < 0.01).  In addition, the odds of a lame cow to have an interval to ovulation ≥ 20 hr were 1.9 times greater than the odds of a sound cow (P = 0.01). No effects of GS were found on intervals to ovulation when using the AAML (P = 0.12). No effects of BCS, parity or the size of the pre-ovulatory follicle were found on intervals to ovulation for either monitor.  4.3.2 Ovulation failure and false alerts  Considering all the estrus alerts that appeared on the AAMC, 81.7% of these alerts resulted in ovulation. When excluding false alerts, 93.3% of the estrus alerts resulted in ovulation, equating to 6.7% ovulation failure. Ovulation failure was affected by estrous expression and parity but not by stage of lactation, BCS, GS, milk production or the size of the pre-ovulatory follicle. Events of estrus with less estrous expression had increased ovulation failure, where 1.9 ± 1.4% and 9.5 ± 1.7% and 8.9 ± 1.7% and 2.1 ± 1.4% ovulation failure was found for high and low peak activity and short and long duration episodes of estrus using the AMMC, respectively (P < 0.001). Using the AAML, the same relationship was found where  117   events of estrus with less estrous expression experienced more ovulation failure (P = 0.03; Table 4.3). Additionally, parity was found to impact ovulation failure, where multiparous cows had increased ovulation failure when compared with primiparous cows (P = 0.05). Ovulation failure results have been summarized in Table 4.3. The positive predictive value (the number of estrus alerts that correctly identified a cow in estrus divided by the total alerts on the monitor) for the AAMC was 87.8%. As the enrollment criteria of this experiment was based on estrus alerts from the AAMC, and not by the AAML, we could not calculate a positive predictive value for the AAML. Alerts on the AAMC that were false were found to have reduced estrous expression in comparison with true positive alerts. The duration of false alert events was much shorter than true positive estrus events (4.9 ± 0.6 vs. 11.8 ± 0.3; P < 0.001) and had a much smaller peak of activity (48.4 ± 2.1 vs. 75.7 ± 0.9; P < 0.001).  The odds of estrus events that had a short duration or a smaller peak activity to be a false alert were 12.5 and 17.6 times greater, respectively, than those with a long duration or greater peak activity (P < 0.001). Additionally, stage of lactation, parity and BCS also affected the odds of false alerts, but no impact of GS was found. Cows in early lactation had the smallest odds of showing false alerts, as both those in mid and late lactation had greater odds of showing false alerts (P = 0.04). Primiparous cows had greater odds than multiparous cows (P = 0.04) to show false alerts and a tendency for BCS to impact false alerts was found. Thin cows had the smallest odds of showing false alerts in relation to average and moderately body conditioned cows (P = 0.06). Odds ratios for all potential impactors of false alerts are summarized in Table 4.3.     118   4.3.3 Estrous expression and fertility  Estrous expression when using the AAMC, was found to be affected by parity, BCS and stage of lactation. Primiparous cows had greater peak activity (77.5 ± 1.5 vs. 73.4 ± 0.9 index; P = 0.02) and longer duration (12.7 ± 0.4 vs. 11.5 ± 0.3 h; P = 0.03) estrus events than multiparous cows. There was a tendency for thin cows to have less activity than both average and moderately body conditioned cows (72.3 ± 2.0 vs. 77.0 ± 1.1 vs. 77.0 ± 1.4 index; P = 0.08); no effects of BCS were found on the duration of estrus. Events of estrus were found to have a shorter duration earlier in lactation than during mid and late lactation (11.3 ± 0.4 vs. 12.6 ± 0.4 vs. 12.4 ± 0.4; P = 0.02); no effect of stage of lactation was found on peak activity. No effects of GS, milk production or the size of the pre-ovulatory follicle were found on the expression of estrus using the AAMC; estrous expression using the AAML was not impacted by any of the used variables. Expression of estrus also influenced fertility associated with each episode of estrus. Estrus events with greater peak activity on the AAMC and AAML were found to have greater pregnancy per AI (P/AI) than those with less activity (AAMC: 42.3 ± 0.4 vs. 31.7 ± 0.4%; P = 0.02; AAML: 43.1± 0.4 vs. 36.3 ± 0.4%; P = 0.10). Furthermore, estrus events with long duration on the AAMC were also found to have greater P/AI than those with short duration (42.4 ± 0.4 vs. 32.6 ± 0.4%; P = 0.03). When failed ovulations were excluded from the analysis, as events with less estrous expression are shown to have more ovulation failure than those with greater, estrous expression was still found to have a large impact on P/AI when using the AAMC but not for the AAML (Table 4.4). Parity and GS were also found to  119   affect P/AI, whereas stage of lactation, BCS, milk production and the size of the pre-ovulatory follicle were not. Factors impacting pregnancy per AI have been summarized in Table 4.4.  4.4 Discussion  Factors that influence differences in timing of ovulation relative to the onset of estrus may be useful for making more informed on-farm breeding decisions. This study demonstrates that intervals from activity alerts, using AAMs, to ovulation were impacted by estrous expression, stage of lactation, parity, and lameness. Furthermore, we were also able to demonstrate that ovulation failure is influenced by estrous expression, and reiterate the importance of estrous expression on fertility, particularly as this information can become readily available with the use of AAMs.  The intervals to ovulation of both AAMs found in this study are similar to Valenza et al. (2012; 28.7 h); using a similar neck-mounted AAM, as well as other studies using pedometers (29.3 h; Roelofs et al., 2005, 2006), pressure-sensing systems (27.6 h; Walker et al., 1996) and visual detection (30.6 h; Roelofs et al., 2005b). Stevenson et al. (2014) also reported similar intervals to ovulation using a pressure-sensing system (26.4 h) and a collar-mounted AAM (25.7 h). Suthar et al. (2011), on the other hand, reported a shorter interval to ovulation (20.6 h) using visual observation on tie-stall-housed cows.  Generally, we found that ovulation was hastened in episodes of estrus with less estrous expression on AAMs and those that occurred early in lactation but was delayed in cows that were considered lame. Previous research using AAMs (Hockey et al., 2010; Valenza et al., 2012; Stevenson et al., 2014), visual estrus detection (Walker et al., 1996) and  120   pressure-sensing devices (Stevenson et al., 2014) also have reported that the interval from the onset of estrus to ovulation increased with longer duration episodes of estrus, however, this relationship has not been often related to the intensity of estrous expression. It has been suggested that differences in intervals from onset of activity to ovulation may be due to the timing of the LH surge relative to the onset of estrous behaviours (Bloch et al., 2006). Bloch et al. (2006) reported that cows with extended estrus to ovulation intervals also had lesser concentrations of estradiol at the beginning of estrus, even though there was no difference in pre-ovulatory follicle diameter, and suggest that these lesser concentrations of estradiol may be why the LH surge is delayed. Walker et al. (1996) suggested that estrus events with longer duration and thus longer intervals to ovulation may occur because circulating estradiol concentrations are adequate to cross the threshold needed for estrous expression but insufficient for ovulation. In addition, Kaim et al. (2003) noted that within a subset of cows that had severely delayed ovulation (> 50 h after onset of estrus) cows had significantly smaller LH surges in comparison with cows that ovulated within a more normal range. Although late and small magnitude LH surges have been associated with drastically delayed ovulation (Kaim et al., 2003) and in the case of lame cows (Dobson et al., 2007), the impact of estrous expression on the LH surge has not been fully elucidated. The occurrence of ovulation during a more ideal timing relative to insemination may be why episodes of estrus with greater estrous expression have been reported as being more fertile, both in this and previous studies (Silper et al., 2017; Burnett et al., 2018; Madureira et al., 2019). Moreover, a recent study by Madureira et al. (2015) found a poor association between estradiol concentration measured at the time of alert and the intensity of estrous  121   expression. Hockey et al. (2010) and Saumande and Humblot (2005) noted a correlation with the size of the ovulatory follicle and the interval from onset of activity to ovulation, however, both Van Eerdenburg et al. (2002) and Madureira et al. (2015) reported no influence of pre-ovulatory follicle size on estrous expression. Similarly, Bloch et al. (2006) reported no differences in pre-ovulatory follicle size among estrus to ovulation interval lengths. In the present study, estradiol was not measured, but no effect of the pre-ovulatory follicle diameter on the interval from onset of estrus to ovulation or on estrous expression was found. Further research is needed to determine which physiological events can account for the changes in the interval from activity alert to ovulation associated with estrous expression. Estrous expression also had a large impact on ovulation failure and fertility regardless of which AAM was used. Although cows with greater estrous expression had significantly less ovulation failure (3.5 vs. 11.5%), it is probably not the only explanation for the differences seen in P/AI between cows that express a greater or lesser intensity of estrous behaviors in this study. When removing cows that did not ovulate, the same pattern of increased P/AI at events of estrus with greater estrous expression was maintained overall, suggesting that there must be other processes (e.g., uterine environment), that are at play to explain changes in fertility observed herein.   Similar to the current study, Valenza et al. (2012) reported 95% of cows that were both in estrus and indicated by an alert on an AAM successfully ovulated. In another study that used progesterone to discriminate between true and false estrus alerts on an AAM, they reported that 91.5% of cows that had an alert with low progesterone (< 1 ng/mL)  122   successfully ovulated, and within that 85.5% ovulated < 36 h after the alert (Stevenson et al., 2014). Others have reported similar ovulation failure rates (6.5%) as well as an effect of season (12.4% warm season; 3.4% cool season; López-Gatius et al., 2005a), although the current study did not observe a similar seasonal effect. Ovulation failure is generally thought to result from a lack of responsiveness of the hypothalamus to increased circulating concentrations of estradiol and thus failure to trigger the GnRH surge and ultimately the LH surge (Wiltbank et al., 2002). Factors impacting ovulation failure are mainly those associated with infection and stress. For example, heat stress (López-Gatius et al., 2005a), uterine infection (Sheldon et al., 2006), and ACTH (Ribadu et al., 1999, 2000) or endotoxin treatments (Peter et al., 1989) are all stress factors associated with ovulation failure. Although not all anovulation is caused by cystic ovarian follicles (Wiltbank et al., 2002; Vanholder et al., 2006), additional risk factors of cystic ovarian follicles have been reported as milk production (Laporte et al., 1994; Heuer et al., 1999), parity (Laporte et al., 1994), and stage of lactation (Vanholder et al., 2006). Moreover, cystic cows have been reported as having a lack of standing estrus, or frequent bouts of estrual behaviours (Wiltbank et al., 2002) but there are very few reports on this subject. To the best of our knowledge, this is the first study to report the impacts of estrous expression (i.e. intensity and duration) using an AAM on ovulation failure.  Stage of lactation was found to impact estrous expression and the interval from activity alert to ovulation, but not ovulation failure or fertility. Impacts of stage of lactation on estrous expression could be from many different factors, including: the number of estrous cycles that have occurred since parturition (Lopez et al., 2004; Aungier et al., 2012), the  123   amount of milk being produced (Lopez et al., 2004) or the presence of negative energy balance (Butler, 2000). Impacts of BCS on estrous expression may be symptomatic of negative energy balance. Previous research has demonstrated the harmful impacts of negative energy balance and poor BCS on fertility (Butler, 2000; Moreira et al., 2000) and estrous expression (Madureira et al., 2015a). In contrast, in the present study, we are unable to decipher which of these factors related with stage of lactation are most influential on estrous expression. Previous research also supports that stage in lactation may not have a key impact on ovulation failure (López-Gatius et al., 2005a). Conversely, Demetrio et al. (2007) found cows with increased DIM had a greater chance of ovulation failure. Negative energy balance postpartum has been shown to decrease pulsatile LH secretion and IGF-1 concentrations (Diskin et al., 2003) which are important for follicular growth as well as ovulation; however, the lack of impact of stage of lactation on ovulation failure also may have resulted from the exclusion of estrus events that occurred before 30 DIM.   Parity was also found to be an important factor that impacted ovulation failure, estrous expression and fertility, but not intervals to ovulation. Other research has also reported primiparous cows to have longer duration episodes of estrus and to take more total steps per estrus using pedometers (Roelofs et al., 2005a), and express more behaviours at estrus when using visual observation (Roelofs et al., 2005b; a). In contrast, Van Eerdenburg et al. (2002) found no effects of parity on behavioural scoring, nor did Valenza et al. (2012) and Stevenson et al. (2014) find differences in estrus duration using AAMs.  Similar to the current study, Roelofs et al. (2005b; a) and Walker et al. (1996) found no difference in interval from the beginning of estrus to ovulation between multiparous and primiparous  124   cows using pedometers and the visual detection of estrus, whereas Stevenson et al. (2014) found primiparous cows to have shorter ovulation intervals from onset of activity. In contrast, Roelofs et al. (2005b; a) did find that primiparous cows had a shorter interval from the end of estrus to ovulation when using pedometers (Roelofs et al., 2005a). The difference in the interval from the end of estrus to ovulation was speculated to be the result of greater duration of estrus in primiparous cows.  The current study was not designed to report how accurate the AAMs were able to alert estrus events, but was able to determine which characteristics predispose cows to false alerts. False alerts impact reproductive performance through inappropriate use of labour and costs associated with inseminations of cows that are not in estrus, as well as the potential of harming a pre-existing pregnancy that may be present but not yet diagnosed. In addition, false alerts can be damaging as they can cause producers to lose trust in the technology or become frustrated, leading to decreased use of the system. Estrous expression was found to be important for deciphering false alerts, as estrus events with less estrous expression were more likely to be false than events having greater estrous expression. The current study found less estrous expression, greater BCS, later days in milk and being primiparous to be risk factors that increased the odds of showing more false alerts than their counterparts. A previous study demonstrated that false estrus alerts had less actual steps when using a pedometer in comparison with correctly alerted estrus events (477  322 steps vs. 841  259 steps); however, they did not find the same relationship when looking at mean and standard deviation of steps at estrus alert (Roelofs et al., 2005a). Similarly, a previous study reported that increased time postpartum increased the odds of a cow to demonstrate a false- 125   positive alert (Holman et al., 2011a); that study found no impact of BCS, parity GS, or milk yield on false alerts using different automated technologies as well as visual detection. Results from this study may suggest that cows that are in a healthier state (i.e.: greater BCS, later in lactation) may engage in certain activities that may be more likely to cause them to be falsely identified as in estrus.   Interestingly, the factors reported by (Aungier et al., 2012) to impact the ability of the AAM to correctly identify estrus are similar to those that we found to impact false alerts. Aungier et al. (2012) reported estrous expression to have a significant impact on alerts, where the odds of a correct alert increased by 20% and 91% with every 1-unit increase in peak activity and 2-h increase duration, respectively. In addition, (Aungier et al., 2012) reported impacts of BCS, milk yield and the quantity of postpartum pre-ovulatory follicular phases on the likelihood of positive alerts, but contrary to the current study did not find impacts of parity on alerts. They reported that estrus events occurring at second or subsequent pre-ovulatory cycles, rather than the first postpartum, had increased odds of 8.2 for being correctly identified by the monitor and that the odds of a pre-ovulatory phase being identified by the AAM improved by 1.4 for every increase in 0.25 BCS unit. (Aungier et al., 2012) also found that cows producing less milk at the time of estrus were more likely to be identified.  In the current study we did not find an impact of milk yield on the occurrence of false alerts. Neither this study or (Aungier et al., 2012) reported any impacts of lameness on the occurrence of false or positive alerts.     126   4.5 Conclusions  This study demonstrated that the expression of estrus, measured by two different AAM systems, is important, not only for estrus detection, but also for its impacts on the interval from activity alert to ovulation, ovulation failure and fertility. Timing of ovulation is a central factor that impacts on-farm breeding decisions, thus understanding the factors associated with it can yield important information to refine a more efficient use of AAMs. In addition to estrous expression, stage of lactation and lameness were also found to influence the timing of ovulation relative to an estrus alert. Furthermore, we demonstrated that estrous expression influenced ovulation failure, although the occurrence of ovulation failure cannot solely explain the difference in fertility between estrus events with greater and lesser estrous expression.    127   Table 4.1: Descriptive statistics of ovulation timing variables.  Factor n Mean Interval Time (h) SD Minimum Maximum AI - Ovulation 426 15.0 8.8 -8.0 62.7 AAMC      AAM Alert – Ovulation 627 25.8 10.2 -2.5 87.1 AAM Alert – AI 498 11.0 5.2 -19.3 23.5 AAML      AAM Alert – Ovulation 499 24.7 9.2 -0.2 89.8 AAM Alert – AI 384 10.1 6.5 -21.7 27.8            128   Table 4.2: Associations of intervals from activity alert to ovulation and estrous expression traits measured using two automated activity monitors.  Interval to ovulation (h)   < 20 20-31 ≥ 32  Factor Mean SE Mean SE Mean SE P -value Peak activity AAMC (index)1 69.9a 1.6 80.7b 1.0 78.4b 1.9 <0.001 Peak activity AAML (%-relative increase)1 361.9a 13.4 405.5b 8.9 388.8ab 18.1 0.02 Duration of estrus AAMC (h)2 10.1a 0.4 12.9b 0.3 13.3b 0.5 <0.001 1 Peak activity was defined as the maximum activity during an episode of estrus.  2 Duration of estrus was defined as the time the activity of the cow exceeded threshold values set by the AAM software. a–b Values with different superscripts in the same row are significantly different (P < 0.05).        129   Table 4.3: Factors impacting ovulation failure and false alerts using automated activity monitoring systems. Factor Classification Ovulation Failure  (%) SE P-value False Alerts (Odds Ratio) CI P-value  Peak activity AAMC (index)1 High (≥ 80)  1.9 1.4 < 0.001 ref.  < 0.001  Low (< 80) 9.5 1.7 17.6 8.3 – 37.3 Peak activity AAML (%-relative increase)1 High (≥ 331)  2.3 1.4 0.03 - - N/A2  Low (< 331) 6.2 1.5 - - Duration of estrus AAMC (hr)1 Long (≥ 12)  2.1 1.4 < 0.001 ref.  < 0.001  Short (< 12) 8.9 1.7 12.5 6.3 – 24.7 Parity Multiparous  8.0 1.2 0.05 ref.  0.05  Primiparous  3.2 2.1 1.8 1.0 – 3.2 Body Condition Thin (< 2.75)  6.2 2.8 NS ref.  0.06  130    Average (2.75)  5.3 1.5 3.0 1.1 – 8.0  Moderate (> 2.75)  6.1 2.8 3.5 1.3 – 10.0 Stage of lactation Early (30 – 60 DIM)  5.5 1.8 NS ref.  < 0.01  Mid (61 – 119) 6.7 1.6 3.4 1.7 – 7.0  Late (≥ 120 DIM) 3.5 1.9 2.9 1.3 – 6.1 Milk production 1000 kg of 305d mature equivalent yield  -0.005 0.005 NS 1.00 1.0 – 1.0  NS Gait score Sound (≤ 2)  6.1 1.4 NS ref.  NS  Lame (> 2) 4.2 2.0 0.84 0.5 – 1.5 Pre-ovulatory follicle Follicle diameter (mm)  -0.001 0.002 NS - - N/A2 1 Traits describing the expression of estrus were categorized above and below their respective medians. 2 Factors with ‘N/A’ as a P-value are those which were not included within the analysis for that specific outcome variable.      131   Table 4.4: Factors impacting pregnancy per AI including all true events and when excluding non-ovulated events. Factor Classification All Events  Only Ovulated Events n Pregnancy per AI (%) SE P-value n Pregnancy per AI (%) SE P-value Peak activity AAMC (index) High (≥ 80) 485 42.3 0.4 0.02 451 45.6 0.3 0.04  Low (< 80) 31.7 0.4 35.7 0.5 Peak activity AAML (%-relative increase) High (≥ 331) 497 43.2 0.4 0.10 430 43.3 0.4 0.27  Low (< 331) 36.4 0.4 38.3 0.4 Duration of estrus AAMC (hr) Long (≥ 12) 490 42.7 0.4 0.02 450 45.5 0.4 0.07  Short (< 12) 32.1 0.4 37.1 0.4 Gait score Sound (≤ 2) 485 41.7 0.3 0.04 451 46.0 0.3 0.03  Lame (> 2) 32.3 0.5 35.2 0.5 Parity Multiparous 485 28.7 0.3 < 0.01 451 32.2 0.3 < 0.01  Primiparous 45.3 0.6 49.0 0.5 1 AAMC = collar-mounted automated activity monitor; AAML = leg-mounted automated activity monitor.  132   Figure 4.1: Frequency of estrus events for estrus alert to ovulation intervals relative to peak activity of estrus on the AAMC (panel A; P < 0.001) and the AAML (panel B; P < 0.01), and duration of estrus on the AAMC (panel C; P < 0.001). Peak activity was defined as the maximum activity during an episode of estrus. Duration of estrus was defined as the time the activity of the cow exceeded threshold values set by the AAM software. Peak activity and duration of estrus were classified above and below their respective medians relative to the AAM (80 index, 331 %-relative increase and 12 h for AAMC peak activity, AAML peak activity and AAMC duration, respectively). A)  B)       010203040506070<20 20-31 ≥32Frequency of Events (%)AAMC Interval to Ovulation (h)High PeakLow Peak010203040506070<20 20-31 ≥32Frequency of Events (%)AAML Interval to Ovulation (h) High PeakLow Peak 133   C)     010203040506070<20 20-31 ≥32Frequency of Events (%)AAMC Interal to Ovulation (h)LongShort 134   Chapter 5: The impact of GnRH administration at the time of artificial insemination on conception risk and its association with estrous expression4 5.1 Introduction  Achieving high reproductive performance is a primary goal on commercial dairy farms. Different reproductive management practices are available to producers, but automated activity monitors (AAM) are being adopted as a sustainable alternative or complement to visual heat detection and systematic timed artificial insemination. In a recent Canadian survey half of the farmers that responded reported using AAM to aid in estrus detection in at least a portion of their herd (Denis-Robichaud et al., 2016).  Despite the growing interest in using this technology, challenges remain regarding interpretation of the data to optimize breeding decisions.  The occurrence of estrus has been shown to improve fertility of timed AI protocols by increasing conception rates (Galvão et al., 2004; Pereira et al., 2014) and reducing pregnancy losses (Galvão et al., 2004; Pereira et al., 2014, 2016). Similarly, benefits of estrous expression in recipient cows of embryo transfer have been noted (Jinks et al., 2013; Pereira et al., 2016). Not only has the occurrence of estrus been shown to impact fertility, but the intensity of estrous expression has also been associated with fertility when timed AI                                                              4 A version of this chapter has been submitted for publication: Burnett, T.A., A.M.L. Madureira, J.W. Bauer and R.L.A. Cerri. The impact of GnRH administration at the time of artificial insemination on conception risk and its association with estrous expression.  135   protocols are implemented (Madureira et al., 2019) and at spontaneously occurring estrus events (Madureira et al., 2015a). Using visual detection, the number of times a cow stands to be mounted has been associated with increased conception rates (Gwazdauskas et al., 1981; Reimers et al., 1985). Recently, quantification of different behavioural changes associated with estrus, such as walking activity (López-Gatius et al., 2005b; Madureira et al., 2015a), lying behaviour (Silper et al., 2017) and activity using neck movement (Madureira et al., 2015a; Burnett et al., 2017a, 2018) measured using AAM have been associated with fertility. In addition to conception, ovulation timing and failure has been shown to differ with estrous expression. Cows with lesser estrus intensity (Burnett et al., 2018) and shorter duration of estrus (Hockey et al., 2010; Valenza et al., 2012; Stevenson et al., 2014) have been associated with shorter intervals from the time of AAM estrus alert to ovulation. Moreover, less estrous expression has also been associated with increased ovulation failure both at spontaneous estrus events (Burnett et al., 2018) and at timed AI (Silper et al., 2017; Madureira et al., 2019). The occurrence of estrus at the end of timed AI protocols using tail chalk have also reported to increase ovulation rates (Galvão et al., 2004; Pereira et al., 2014).  Not only is ovulation modulated by estrous expression, but recent evidence suggests that the function of the corpus luteum is modified by estrous expression. The occurrence of estrus after the implementation of timed AI protocols (Pereira et al., 2014, 2016) and embryo transfer (Pereira et al., 2016) have been found to result in greater circulating progesterone concentrations 7 d post-estrus. Madureira et al. (2019) also showed that animals with a greater change in steps on the day of timed AI compared with baseline had greater concentrations of circulating progesterone 7 d-post AI. Increased progesterone 7 d-post AI  136   (Stronge et al., 2005; Pereira et al., 2014, 2016) or a  quicker rise in progesterone post-ovulation (Larson et al., 1997) have both been associated with greater conception rates. Previously, gonadotropin-releasing hormone (GnRH) has been shown to indirectly cause ovulation through the action of luteinizing hormone (LH) (Thatcher et al., 1993). This has led to the suggestion that an increase in the percentage of large luteal cells and production of progesterone by the corpus luteum, facilitating a faster rise in progesterone post-ovulation (Mee et al., 1993).  Administration of GnRH at the time of AI has been studied for its effects on pregnancy per AI, but findings have been mixed for both first service and repeat breeders (first service reviewed by Mee et al. (1990) and repeat breeders reviewed by Stevenson et al. (1990)). However, no research has been carried out to determine the effects of different intensities of estrous expression with and without GnRH administration on fertility.  Upon reviewing the available literature we suggest that cows with lesser estrous expression may have compromised fertility due to improper ovulation times, ovulation failure, or poorer functioning corpus lutea post-ovulation. Thus, the objective of this study was to determine if the fertility of spontaneous estrus events with lesser estrous expression increases when cows are stimulated to ovulate when administrated GnRH at the moment of AI. Our hypothesis was that the administration of GnRH at the moment of AI of cows would decrease ovulation failure and increase fertility in cows showing lower intensity of estrous expression, but not in cows showing greater intensity of estrous expression.   137   5.2 Materials and methods  This experiment was conducted from February 2017 until October in 2018 at the University of British Columbia’s Dairy Education and Research Centre, in Agassiz, BC, Canada as well as two local commercial farms. All procedures were approved by the Animal Care Committee of the University of British Columbia (protocol # A18-0039).   5.2.1 Animals and housing  This study took place on three commercial dairy farms located in the lower Fraser Valley region of British Columbia. A total of 2,007 episodes of estrus (Farm A: 964; Farm B: 383; Farm C: 671) were used from 852 lactating Holstein cows (Farm A: 330; Farm B: 204; Farm C: 318); an average of (mean ± SD) 2.4 ± 1.6 episodes per cow. On two farms, animals were housed in wood-framed barns with a free-stall design, equipped with deep sand-bedded stalls, fans and a manure scraper. On the third farm, stalls were bedded with sawdust and manure management was carried out using a flush system. On two farms, cows were milked twice daily at 0500 h and 1500 h within conventional milking parlours, while the other used automated milking robots (Monobox; GEA Farm Technologies, Germany). Fresh TMR was delivered twice daily on all farms at approximately 0700 h and 1600 h. The TMR on each farm was formulated following the Nutrient Requirements of Dairy Cattle (NRC) guidelines (NRC, 2001) to meet or exceed the requirements of a 620 kg Holstein cow  138   producing 40 kg/d of 3.5% fat corrected milk; all cows on all farms had ad libitum access to both TMR and water. All cows had their body condition (BCS) and gait (GS) scored within 3 d of their estrus alert. Body condition and gait were scored using 5-point scales from thin (1) to obese (5) and sound (1) to severely lame (5) as outlined by Edmonson et al. (1989) and Flower and Weary (2006), respectively. Cows were later classified as thin (< 2.75), average (= 2.75) or moderate (> 2.75) and sound (≤ 2) and lame (> 2). Health and production information was collected by the dairy herd personnel with the assistance of the herd veterinarian, and confirmed and recorded by the project leader from the on-farm Dairy Comp 305 software (Valley Agricultural Software, Tulare, CA). The mean DIM, parity, BCS, GS and 305-d mature-equivalent yield are summarized by farm (see Table 5.1).  5.2.2 Study design  In this randomized controlled trial, animals were continuously monitored using leg-mounted pedometers (Farm A: Afi PedoPlus ®, Afikim, Isreal; Farm B and C: Rescounter II ®, GEA Farm Technologies, Germany), fitted immediately following calving. Animals were enrolled onto the study when the monitor identified them as having crossed the alert threshold, as set by the manufacturer of each AAM system. The AAMs were checked once (Farm C) or twice (Farms A and B) daily during milking times for cows in estrus, and cows were inseminated either in the morning (Farm C) or using the AM/PM rule (Farms A and B). At the moment of AI, cows were randomly assigned to receive an intramuscular injection of  139   100 μg of gonadorelin (GnRH; Factrel, Zoetis, Kirkland, QC, Canada; 2 ml). Peak activity (see Madureira et al. (2015)) was used to describe the intensity of the expression of estrus using the data from the AAMs. Peak activity was defined as the maximum activity during an estrus episode.  A subset of cows on Farm A were assessed for the occurrence of ovulation at 24 h (n = 428), 48 hr (n = 822) and 7 d (n = 957) post-alert. On two farms (Farms A and B) cows had blood samples collected at estrus (n = 430) and 7 d post-alert (n = 407) for the analysis of progesterone.   5.2.3 Determination of estrus, ovulation and pregnancy  Cows had their ovaries examined by per rectum ultrasound (Ibex Pro; E.I. Medical Imaging, Loveland, CO) using a 7.5 MHz linear-array rectal at enrolment, 24 h, 48 h and 7 d post-alert (Farm A only). The presence and diameter of the 3 largest follicles and corpus lutea were measured and recorded at each examination. Exams at alerts were used to determine if the cow was truly in estrus, exams at 24 h and 48 h were used to determine general timing of ovulations, and d 7 examinations were used to confirm ovulations. Exams for ovulation were performed at the milking closest to the desired time point, such that 24 h exams occurred 23.5 ± 4.5 h post-alert and 48 h exams 50.1 ± 5.2 post-alert. Cows were classified as in estrus if they had at least one dominant pre-ovulatory follicle greater than 15 mm and an absence of corpus luteum or a corpus luteum that was less than 25 mm at the time of the estrus alert. Ovulation was determined by the disappearance of the dominant  140   pre-ovulatory follicle. A new corpus luteum at d 7 was used to confirm ovulation for true estrus alerts where ovulation was detected. For true estrus alerts where ovulation was not detected by 48 h post-alert the absence of a new corpus luteum at d 7 was indicative of failed ovulations. For false alerts, the absence of a new corpus luteum at d 7 was used to confirm that the estrus alert was correctly classified as false. At the event that both ovaries had only small structures at alert, but a new corpus luteum was formed by 7 d, cows were assumed to have ovulated before the first palpation.  On all farms, pregnancy diagnosis was performed at 36 ± 7 days post AI for the detection of an embryonic vesicle with a viable embryo (presence of a heartbeat). Diagnosis was carried out by the regular herd veterinarian of each farm.    5.2.4 Blood collection and progesterone analysis  A subset of cows from Farms A and B had their blood collected on the day of the estrus alert and at 7 d post-alert for the analysis of progesterone. Due to Farm B only being visited twice weekly, blood samples occurred within 3 d of estrus and at 9 ± 2 DIM; blood samples from Farm A occurred on the exact DIM as specified. For the analyses of progesterone at estrus, only samples collected within 2 d of the alert were used.  Blood samples (7 ml) were collected via venipuncture from the median coccygeal vein or artery using K2EDTA coated tubes (Vacutainer system, Becton Dickinson, Rutherford, NJ). All samples were stored at 4°C until centrifugation. Samples were  141   centrifuged on the same day of collection at 2,700 × g for 15 min and plasma was pipetted into aliquots and subsequently stored at -17°C until analysis.  Progesterone was analysed using a commercially available ELISA kit that is labelled for its use in bovine plasma and serum (Ovucheck Plasma; Biovet, St-Hyacinthe, Quebec). This is a monoclonal antibody kit which measures the optical density of standards and plasma samples and is read in a microplate absorbance reader set at a 405 nm wavelength. Mean intra- and inter-assay CV were 7.8 and 12.4 %, respectively.  5.2.5 Statistical analyses  All statistical analyses in this study was carried out with SAS Studio (version 5.1; SAS Inst. Inc., Cary, NC) with estrus episode as the experimental unit. Prior to all analyses data were checked for normality using the UNIVARIATE procedure and probability distribution plots. Estrus events that were confirmed as false alerts were excluded from the analysis. Pregnancy per AI, ovulation rates at 24 h, 48 h and 7 d post-alert, and progesterone concentration at 7 d post-AI were used as dependent variables and were tested for the effects of GnRH treatment, estrous expression, parity, BCS, GS, stage of lactation, insemination number and milk production using ANOVA with estrus event nested in cow as a random effect using the MIXED procedure. Peak activity was categorized using the median of each farm. Cows with more than one lactation were classified as multiparous cows and those with only one lactation were considered primiparous. Estrus events that occurred before 60 DIM  142   were classified as early, greater than 60 DIM but  120 DIM were classified as mid-lactation, and any events that occurred after 120 DIM were classified as late lactation.  Ovulation rate was calculated as the hazard risk of ovulation by 7 d post-estrus using the PHREG procedure. Cox proportional hazard regression models included days in which the cow had not yet ovulated as the outcome variable and treatment, estrous expression, parity, BCS, and GS were used as explanatory variable. Only variable with P  0.15 Observation were right-censored at 7 d if ovulation had not been previously confirmed. Survival curves were drawn from the proportion of cows that had not ovulated at each time point given by the LIFETEST procedure.   5.3 Results 5.3.1 Pregnancy per AI and ovulation  Treatment of GnRH at the time of AI tended to increase pregnancy per AI (42.8  1.7 vs. 36.8  1.8 %; P = 0.08). In addition, we noted an interaction between treatment and estrous expression on pregnancy per AI. Within the control group, cows with higher estrous expression had greater pregnancy per AI than those with lesser expression; whereas, GnRH administration increased pregnancy per AI for cows with low estrous expression but not those with high expression (Figure 5.1; P < 0.001).  Although there were differences in herd characteristics (Table 5.1), there was no impact of farm on the interaction between estrous expression and treatment. There was a tendency for an interaction between farm and  143   treatment, where there was no impact of treatment alone on pregnancy per AI on Farm C (Farm A – GnRH: 43.4  2.6 Control: 39.5  2.8; Farm B – GnRH: 48.4  3.4 Control: 34.8  3.8; Farm C – GnRH: 36.6  2.8 Control: 36.2  2.7 % pregnancy per AI; P = 0.06). An interaction between insemination number and treatment was found. In general, as insemination number increased, pregnancy per AI decreased. However, at lower insemination numbers postpartum, GnRH administration appeared to provide greater benefits than at greater insemination numbers postpartum (Figure 5.2; P < 0.01). Overall, there was no effect of GnRH administration, estrous expression, or an interaction, on the rate in which cows ovulated (Figure 5.3; P = 0.17). However, when considering the proportion of cows that ovulated by 24 h, 48 h and 7 d post-estrus, interactions between GnRH administration and estrous expression were found. By 24 h post-alert, cows with greater estrous expression had lower ovulation rates than those with lesser estrous expression within control cows, but in the GnRH cows there were no differences in the proportion of cows that ovulated between cows with high and low estrous expression (Figure 5.4; P < 0.001). By 48 h post-estrus, cows with lesser estrous expression receiving GnRH had lower ovulation rates than all other groups (Figure 5.4; P = 0.04.). Finally, at 7 d post-alert, ovulation did not differ by treatment for cows with low estrous expression, but the administration of GnRH increased ovulation rates for cows with high estrous expression (Figure 5.4; P = 0.03). Parity, gait score, BCS, stage of lactation nor 305-d lactation yield were associated with differences in the proportion of cows that ovulated by 24 h, 48 h or 7 d post-alert.   144   5.3.2 Progesterone  Cows with greater estrous expression had greater progesterone concentrations 7 d post-alert (5.0  0.1 vs. 4.6  0.1 ng/mL; P = 0.04), but no difference was found at estrus. Treatment with GnRH did not impact progesterone concentration at 7 d post-estrus, nor was there an interaction found between GnRH treatment and estrous expression. There was an interaction of treatment with progesterone concentration at alert on pregnancy per AI, where increasing progesterone concentrations at AI were associated with lower pregnancy per AI, but this relationship was dampened for cows administered GnRH at the moment of AI (Figure 5.5; P = 0.05); this was consistent by estrous expression. Ovulation rate was not impacted by progesterone concentration at the time of alert.  5.4 Discussion  This study demonstrated that the outcome of administrating GnRH at the moment of AI at spontaneous estrus events may be reliant on the intensity of estrous expression. Cows with lesser estrous expression benefitted from GnRH administration while cows with greater estrous expression did not. To the best of our knowledge, the impact of this interaction on pregnancy per AI has not previously been reported in dairy cattle. A tendency for an overall impact of GnRH treatment (irrespective of estrous expression) on pregnancy per AI was found in previous research (Stevenson et al., 1990; López-Gatius et al., 2006). In the present study, this association varied by farm, similar to a study by Anderson and Malmo  145   (1985), where 1 out of the 3 farms failed to show an overall impact of GnRH. Interestingly, the interaction of estrous expression and GnRH administration on pregnancy per AI was consistent on all 3 participating farms even though all farm characteristics were not consistent, suggesting that administering GnRH to cows with suboptimal estrous expression is a practice that can be adopted successfully under varying conditions.  A similar study, using beef cattle and tail chalk, found similar results, where cows with lesser estrous expression at the moment of timed AI had greater pregnancy per AI when receiving GnRH at AI in comparison with cows not receiving GnRH (Rodrigues et al., 2019). In concordance with our study, a meta-analysis of 40 trials, including over 19,000 cows, found an overall positive impact of GnRH application at the moment of AI (Morgan and Lean, 1993). These authors argued that the variation noted within many of the included studies was due in large part to small samples sizes but we suggest that some of the variability may be due to differences in estrous expression. Our original hypothesis that GnRH administration would increase fertility of cows with low estrous expression through increased ovulation and corpus luteum function was not supported by our findings. Although there were differences in the proportion of cows that had ovulated by 24 h, 48 h and 7 d post-estrus, the overall rate of ovulation did not differ by estrous expression, GnRH administration or their interaction. Although progesterone concentrations were increased 7 d post-estrus for cows with greater estrous expression, this was not impacted by GnRH administration.  The impact of GnRH on ovulation rates of cystic cows (Bierschwal et al., 1975; Ijaz et al., 1987; Nanda et al., 1988), during synchronization protocols (Bello et al., 2006; Galvão  146   and Santos, 2010) and as a treatment to induce cyclicity in the early postpartum (Gümen and Seguin, 2003; Bittar et al., 2014) have been thoroughly researched. However, few studies have reported the impacts of GnRH at AI on the ovulation rates of spontaneous estrus events. Similar to the current study, but using an estradiol-based timed AI protocol, Rodrigues et al. (2019) reported that the proportion of cows which had ovulated by 7 d post-AI was unchanged by GnRH administration at AI in cows with lesser estrous expression. However, in contrast with the current study, they found that cows with greater estrous expression and without the use of GnRH had greater ovulation rates than cows with lesser estrous expression irrespective of GnRH treatment. Additionally, the administration of buserelin, a GnRH analogue, at the moment of AI has also been reported to not impact the interval from the onset of estrus to ovulation (Ryan et al., 1994). The administration of GnRH is a direct link to ovulation due to its effects on LH secretion and pulse frequency (Bloch et al., 2006). Administration of GnRH at the onset of estrus increased the intensity (both the amplitude and area under the curve) of the preovulatory LH surge, and shortened the interval from the onset of estrus to the LH surge and subsequently ovulation (Kaim et al., 2003). In fact, in that study, the administration of GnRH at estrus caused a dramatic reduction in the proportion of cows that ovulated more than 30 h after the onset of estrus, thus shortening the ovulation interval and reducing ovulation failure. Lee et al. (1985) reported that cows receiving GnRH at first postpartum AI, instead of at estrus, had increased concentrations of LH in the hours following administration, in addition to the naturally occurring LH surge occurring around the onset of estrus; unfortunately, these authors failed to monitor ovulation rates.  147   Circulating concentrations of progesterone and estradiol have been shown to impact the intensity and timing of a GnRH-induced LH surge (Stevenson and Pulley, 2016). Cows with increased concentrations of circulating progesterone or decreased concentrations of estradiol at the moment of GnRH injection had reduced LH surges and ovulation rates (Stevenson and Pulley, 2016). Additionally, the time from GnRH injection to the peak of the LH surge was increased in cows with low progesterone in comparison with those with high progesterone (Stevenson and Pulley, 2016). Since cows with greater estrous expression have been suggested to have lower progesterone at the time of AI, these results may be similar to those demonstrated by Burnett et al. (2018),  where cows with low estrous expression had lower ovulation rates, but those that did ovulate had shorter ovulation intervals from the onset of estrus. These associations may help explain why GnRH administration modulated the proportion of cows which had ovulated by 24 h, 48 h and 7 d post-estrus differently in cows with high and low estrous expression. Future research is warranted to determine the relationship between estrous expression, ovulation and GnRH administration.  Results from the current study suggest that GnRH administration may have a greater impact on cows that are being inseminated while concentrations of progesterone are greater than optimal. Previous research has reported a lower occurrence of estrous behaviour (Pereira et al., 2016) and ovulation (Madureira et al., 2019) when progesterone concentrations are high at timed AI.  Research using superovulated heifers has demonstrated that by inducing a larger LH surge through GnRH administration resulted in faster maturation of the preovulatory oocyte (Laurinčík et al., 1991) and a greater percentage of fertilized embryos 8 d after AI (Wubishet et al., 1986). This may suggest that the intensity of  148   the LH surge may not only be important for ovulation but also fertility. Little research has investigated the impact of GnRH administration at the moment of AI on pregnancy outcomes in cows with higher progesterone at spontaneous estrus events. However, there is some evidence demonstrating that cows with higher progesterone concentrations have lower GnRH-induced LH surges and ovulatory responses (Colazo et al., 2008; Dias et al., 2010). Stevenson and Pulley (2016) demonstrated similar results when measuring the intensity and timing of GnRH-induced LH surges relative to GnRH administration within synchronization protocols. Additionally, there is research to suggest that increased estradiol concentrations can increase the responsiveness of the pituitary to GnRH (Nett et al., 2002; Dias et al., 2010; Stevenson and Pulley, 2016). In fact, Dias et al. (2010) demonstrated that treatment of estradiol before GnRH administration promoted pituitary responsiveness even in heifers with elevated progesterone concentrations. As progesterone and estradiol are inversed at the time of estrus (low progesterone, high estradiol), there is a possibility that estradiol may also play a role in the interaction that we see between progesterone concentrations at AI and GnRH treatment.  Varying outcomes have been reported on the impact of GnRH administration on progesterone concentrations in the days following estrus, including reports that progesterone concentrations increase (Mee et al., 1993), decrease (Lucy and Stevenson, 1986; Ryan et al., 1994) or remain the same (Lewis et al., 1990; Mee et al., 1990). Similar to the current study, Rodrigues et al. (2019) reported cows with greater estrous expression to have increased progesterone concentrations 7 d post-AI and that GnRH administration to cows with lesser estrous expression did not impact progesterone concentrations.  149   Furthermore, previous studies have reported no changes in progesterone in the week following AI (Lee et al., 1985) or between 8 and 14 d post-AI (Mee et al., 1990) when GnRH was given at the time of AI; however, the latter study did find that progesterone concentrations tended to decrease when GnRH was administered at the onset of estrus. In contrast, Kaim et al. (2003) reported GnRH administration at the onset of estrus did not increase progesterone concentrations post-ovulation, nor did they report changes in corpus luteum size. Mee et al. (1993) conducted 4 experiments where, in general, concentrations of progesterone were found to have a steeper rise and have greater concentrations in cows treated with GnRH 12 h post-estrus; however, the exact day relative to estrus in which progesterone was found to have greater concentrations was not consistent and varied from 4 to 8 d post-estrus. A potential limitation of our study is that the sampling schedule was not adequate to accurately depict differences in progesterone concentrations. In addition, because we were interested in the ability of the corpus luteum to produce progesterone, cows which did not ovulate did not have progesterone concentrations measured 7 d post-estrus; possibly explaining the lack of difference in progesterone concentrations due to GnRH administration.  Others have reported that the administration of GnRH or GnRH analogues are more beneficial on repeat breeders in comparison with first AI postpartum (see meta-analysis by Morgan and Lean, 1993). Similar to our findings, Ryan et al. (1991) reported that the administration of a GnRH analogue was most beneficial at 40-59 DIM in comparison with to other inseminations time points. In Australian herds managed with seasonal calving, GnRH at AI was reported to increase conception risk, but only in cows bred at less than 40 DIM  150   (Shephard et al., 2014);  considerably earlier than the voluntary wait period used in our study. In contrast, Lewis et al. (1990) and Anderson and Malmo (1985) found no impacts of GnRH administration on pregnancy per AI at either first, second or third service. The majority of work discussing the impact of GnRH administration at different service numbers fails to directly compare the interaction of service number with GnRH treatment. Instead most focus on magnitude changes in conception at each service or only use one population of service number (e.g.: only repeat breeders) and then compare their results to previously published works. This difference may in part help explain why the current study found results differing from many previous work reported to date.   5.5 Conclusions  This study demonstrates that administration of GnRH at the time of AI is beneficial for cows with lesser estrous expression, and in cows with greater progesterone concentrations at the time of estrus. Administration of GnRH altered the proportion of cows that ovulated by 24 h and 48 h, but by 7 d post-AI only cows with greater estrous expression benefitted from receiving GnRH; these changes in ovulation were not correlated to an increase in fertility. Progesterone concentrations were not affected by GnRH administration, but GnRH did seem to provide some benefits in terms of fertility in cows with greater concentrations of progesterone at the moment of spontaneous estrus. Beneficial impacts of GnRH for cows with lesser estrous expression was consistent within farm, suggesting that administering GnRH to cows with suboptimal estrous expression is a practice that can be  151   adopted successfully under varying conditions. Future research is warranted to determine the mechanism responsible for this relationship between estrous expression and GnRH administration.           152   Table 5.1: Cow characteristics (mean ± SD)  for all estrus events that were inseminated within the study, separated by farm.      Estrus Characteristic Farm A Farm B Farm C P - value Overall DIM 111 ± 63 117 ± 50 131 ± 73 < 0.001 121 ± 65 Insemination number 2.7 ± 1.9 2.2 ± 1.4 3.0 ± 2.5 < 0.001 2.7 ± 2.0 Lactation 2.7 ± 1.6 1.8 ± 1.1 2.4 ± 1.4 <0.001 2.4 ± 1.5 BCS (% thin) 15.7 15.3 11.3 0.11 14.8 Gaits core (% lame) 44.2 62.5 65.1 <0.001 54.0 Average 305-d mature equivalent milk yield (kg of milk) 12,950 ± 1,950 12,900 ± 2,200 11,700 ± 2,000 <0.001 12,550 ± 2,100  153   Figure 5.1: The impact of GnRH administration at the time of artificial insemination on pregnancy per AI relative to the intensity of estrous expression. Superscripts of letters a-b denote differences (P < 0.05). High = estrous expression greater than the median of each farm. Low = estrous expression lesser than the median.            05101520253035404550High LowPregnancy	per	AI	(%)Estrous	ExpressionGnRH Controlaaab 154   Figure 5.2: The interaction, with 95% confidence intervals, of the impact of GnRH administration at varying insemination numbers on pregnancy per AI (P < 0.01).       Legend: Treatment GnRH Control Pregnancy per AI (%) Insemination number postpartum 40 20 0 2.5 5 7.5 10 12.5  155   Figure 5.3: Survival curve analysis for ovulation for the interaction between GnRH administration at the moment of AI and estrous expression as measured using automated activity monitors (P = 0.17). High = estrous expression greater than the median of each farm. Low = estrous expression lesser than the median.     00.10.20.30.40.50.60.70.80.910 1 2 3 4 5 6 7 8Proportion of non-ovulated cowsDays from estrusControl - HighControl - LowGnRH - HighGnRH - Low 156   Figure 5.4: The impact of GnRH administration at the time of artificial insemination on ovulation rates by 24 h (panel A), 48 h (panel B) and 7 d (panel C) post-alert relative to the intensity of estrous expression as measured using an automated activity monitor. Superscripts of letters a-d denote significant differences (P < 0.05), letters x-z denote tendency (0.05 < P < 0.10). High = estrous expression greater than the median of each farm. Low = estrous expression lesser than the median. A)  B)  05101520253035404550High LowOvulation	rate	by	24	h	(%)Estrous	ExpressionGnRH Controlabac50556065707580859095100High LowOvulation	rate	by	48	h	(%)Estrous	ExpressionGnRH Controlaaxbya 157   C)    50556065707580859095100High LowOvulation	rate	by	7	d	(%)Estrous	ExpressionGnRH Controlab b b 158   Figure 5.5: The interaction, with 95% confidence intervals, of the impact of GnRH administration at AI on pregnancy per AI dependant on the concentration of progesterone at estrus (P = 0.05).            Legend: Treatment GnRH Control Pregnancy per AI (%) Progesterone concentration at estrus (ng/mL) 80 60 40 20 0 0 0.5 1.0 1.5 2.0 2.5  159   Chapter 6: General Discussion 6.1 Thesis findings  The overall objectives of this thesis were to contribute to the current knowledge on automated technologies for the detection of estrus and ovulation by demonstrating their applicability within reproductive management, investigating automated estrus alerts, and to explore the relationship of estrous expression intensity with fertility and ovulation in dairy cattle. In Chapter 1, the physiological basis of the estrous cycle was reviewed; different behavioural changes that occur at estrus and the current state of automated activity monitors for the detection of estrus and ovulation were described. Chapter 1 concluded with a discussion of factors that may impact the intensity of estrous expression, and subsequently how estrous expression may be related to fertility.  At the time, it was generally accepted that AAM were able to detect estrus events at comparable rates to that of visual estrus detection (Dolecheck et al., 2015; Roelofs and van Erp-van der Kooij, 2015), but there was a lack of knowledge on the incorporation of AAM into reproductive management. In Chapter 2, it was demonstrated that using AAM for the detection of estrus within a timed AI protocol for the first AI postpartum resulted in similar overall reproductive efficiency when compared with a treatment using solely timed AI.  Thus, the use of AAM within reproductive management has the potential to reduce the amount of hormonal interventions per insemination per cow without reducing reproductive efficiency. The findings described in Chapter 2 insights into the association between estrous expression and fertility; cows with more intense estrous expression, as measured by the AAM, had  160   greater pregnancy per AI compared to cows with lesser estrous expression. Estrus intensity was reduced in early lactation, in multiparous cows and in cows that had not returned to cyclicity by the end of the voluntary waiting period. Interestingly, although a lower proportion of cows with poor leg health were detected in estrus during the experimental period, the intensity of estrous expression on the AAM was unchanged between cows with and without lameness and/or hock lesions. In conclusion, Chapter 2 showed that AAM can be effective in reproductive management strategies commonly used on dairy farms, and confirms that cows with greater estrous expression have greater fertility. This body of work also added information about how estrous expression can vary by different cow specific factors.  Chapter 3 focussed on the use of rumen-reticular temperature changes as a novel method for the detection of estrus and ovulation. Methods for monitoring ovulation are quite limited and previous research has reported increases in vaginal temperature to be quite closely related to ovulation (Bloch et al., 2006), but vaginal temperature is difficult to continuously measure over long periods of time. Thus, the objectives were to determine if rumen-reticular temperature boluses could be used as a novel measure to detect estrus and ovulation, and determine if temperature changes around these times were impacted by differences in estrous expression and ambient temperature. Changes in rumen-reticular temperature around the time of estrus were noted, but these changes are most prominent for cows with greater intensity of estrous expression and on days with a greater temperature and humidity index. Around the time of ovulation, cows still had increased rumen-reticular temperature relative to basal levels, but they were less than around estrus and there were  161   no differences depending on the level of estrous expression or environmental temperature. Unfortunately we were unable to provide evidence that changes in rumen-reticular temperature can be practically used as an estrus alert on dairy farms.  The remainder of the thesis focussed on understanding the relationship between estrous expression and fertility. In Chapter 4, the impact of estrous expression on ovulation timing and failure was investigated. As mentioned above, cows with greater estrous expression have increased fertility; it was hypothesized that these changes in fertility may be due to differences in the timing or failure of ovulation. Within this study, cows with decreased estrous expression again were found to have decreased fertility and were also found to have shorter intervals from the time of AAM estrus alert to ovulation, but also to have more ovulation failure; this was consistent across both AAM systems used in the study. Ovulation intervals were also impacted by stage of lactation, and lameness but not by BCS, parity, or size of the pre-ovulatory follicle. Chapter 4 also described differences in false alerts in comparison to true positive estrus alerts, where false alerts were found to have lower intensity and shorter duration than true alerts; an aspect that could potentially be very useful when making breeding decisions based on AAM alerts alone. Additionally, it was found that early lactation, multiparous cows, and thin cows (BCS < 2.75) were less likely to show false alerts on the AAM – suggesting there are groups of animals which perform certain behaviours that are more likely to be erroneously alerted as estrus on the AAM. Overall, Chapter 4 demonstrated that there are differences in ovulation timing and failure depending on the intensity of estrous expression, suggesting that it may be one of the underlying mechanisms behind the association of estrous expression and fertility.   162   Expanding on the results of Chapter 4, Chapter 5 investigated if the decreased fertility of cows with reduced estrous expression could be mitigated by stimulating ovulation by administrating GnRH at the moment of AI.  It was hypothesized that GnRH administration at AI would decrease ovulation failure and increase progesterone concentrations in the week following AI and consequently increase the pregnancy success of cows with reduced estrous expression. Administration of GnRH was found to increase pregnancy per AI of cows with reduced estrous expression while cows with greater estrous expression remained unchanged due to treatment. Although the overall impact of GnRH treatment varied by farm, the interaction between estrous expression and GnRH treatment on pregnancy success was consistent – suggesting administration of GnRH to cows with reduced estrous expression is a practice that would be applicable across many different settings.  However, ovulation rates and progesterone concentrations post-AI were not found to be the driving factors behind the observed changes in fertility. Although ovulation rates 24 and 48 h after estrus were modulated due to GnRH administration, by 7 d post-estrus there was no difference in the proportion of cows that had ovulated when comparing to the control group of cows with reduced estrous expression. Similarly, although cows with greater estrous expression were found to have higher progesterone concentrations 7 d post-estrus, there was no impact of GnRH treatment on progesterone concentrations. However, the administration of GnRH benefited cows that had greater concentrations of progesterone at the time of AI. Overall, Chapter 5 demonstrated that the fertility of cows with reduced estrous expression can be increased by administrating GnRH at the moment of AI, but it does not seem to be associated with ovulation rates or greater concentrations of progesterone post-ovulation.   163   6.2 Implications, limitations and future directions    Collectively these chapters provide evidence that AAM can be incorporated within the dairy industry as they are successfully able to detect cows in estrus, but also provide information that can aid producers in making evidence-based breeding decisions. For example, farmers could use the intensity of estrous expression measured by the AAM to make decisions on their use of greater risk or higher priced options such as semen quality, semen type (e.g.: sexed vs. conventional), and embryo transfer based on the changes in pregnancy risk associated with more intense estrous expression.  Unfortunately to date there has only been a dearth of research describing the economic impacts behind omitting inseminations. There is merit to understanding the economic repercussions of guaranteeing a cow remains open (not pregnant) until the next estrus, in comparison with the reduced risk of pregnancy and potential economic losses due to a failed attempt of a more expensive reproductive technology. One potential strategy could be to inseminate cows with reduced estrous expression with poorer quality semen, such as an unproven sire, as a way of reducing semen costs.  But again, no research has investigated the long-term economic impacts of introducing poorer quality genetics into the herd in comparison with the saved expense from using a cheaper semen.   This thesis also demonstrated two methods of hormone intervention reduction in comparison with using only timed AI protocols. Firstly, my work demonstrated that completion of a Presynch-Ovsynch protocol is not necessary for cows that are detected in estrus within the timeframe of the protocol. A standard Presynch-Ovsynch protocol involves  164   5 injections, but by implementing an effective estrus detection strategy producers can reduce up to 4 of those injections depending on when the cow exhibits estrus. As discussed in Chapter 1, there is increasing reluctance of consumers to purchase products that are produced with the use of hormones; a phenomenon that has resulted in decreased usage of synchronization protocols in the EU (Chanvallon et al., 2014) but also a corresponding focus by the dairy industry in rationalizing the use of hormone interventions in dairy reproductive management (Saint-Dizier and Chastant-Maillard, 2012; Higgins et al., 2013). Results from Chapter 2 found that, depending on the farm, anywhere from 35 to 65% of cows were detected in estrus between the end of the pre-synchronization and the beginning of the Ovsynch protocol, making these cows ideal candidates for reduced hormone injections. Secondly, Chapter 5 demonstrated that the addition of a single injection of GnRH at the moment of AI can increase the conception risk of cows with reduced estrous expression from 33 to 41%. Hypothetically, if a producer only administered GnRH to cows with reduced estrous expression and kept cows with greater estrous expression untreated, the overall conception risk would increase to approximately 40%. This would be a conception risk arguably equal or greater than that expected with the use of the standard Ovsycnh alone (Pursley et al., 1997b; a, 1998) or from pre-synchronization followed by Ovsynch (Cordoba and Fricke, 2001; Moreira et al., 2001; Portaluppi and Stevenson, 2005) which require anywhere from 3 to 6 injections per cow.  Inseminating cows using timed AI also has the potentially of dramatically reducing labour costs associated with the detection of estrus, allowing for reallocation of resources and labour given that injection and AI schedules are fixed. However, some have cautioned  165   producers from believing that all efforts focused on estrus detection can be removed unless all cows are subject to strict resynchronization protocols (Jordan et al., 2002) – a management practice that will likely continue to come under criticism as public awareness of this practice increases (Pieper et al., 2016). Using a dynamic model, Galvão et al. (2013) discussed the economic returns of using estrus detection alone at two estrus detection rates (40 and 60%) and two accuracy levels (85 and 95%), timed artificial insemination alone with two compliance rates (85 and 95%), and then a combination of both. From an economic perspective and also in terms of reproductive performance, the combination of estrus detection and timed AI was generally found to be more profitable than either system alone. However, the authors also noted that if a farm is able to achieve either high compliance of injections (95%) or high accuracy (95%) and estrus detection (60%) there may not be any additional cost savings to doing both.  Although there was no difference between reproductive programs using AAM within a timed AI protocol when compared with 100% timed AI in terms of conception risk and days from calving to conception in Chapter 2, the conception risk of animals bred at the end of the timed AI protocol was greater than those that were inseminated at the moment of estrus alert; similar results have also been reported by others (Neves et al., 2012; Fricke et al., 2014a).  The interpretation of these results is difficult due to the difference in DIM of the two groups. Cows bred at the end of the timed AI protocol automatically had greater DIM than those bred at estrus. Thus, it is impossible to determine if the improved conception of cows bred at the end of the timed AI protocol is due to the protocol itself or due to having greater DIM. Delaying first AI has been associated with increased conception rates both in  166   vivo (Tenhagen et al., 2003; Stevenson and Phatak, 2005) and those carried out using machine learning algorithms (Caraviello et al., 2006b; Schefers et al., 2010). In this thesis (Chapters 2 and 4), it was also determined that cows have reduced estrous expression early postpartum. Together this may suggest that a strategy to increase conception rates of spontaneous estrus events could be to allow more time before the first postpartum AI; however, drastic delays in conception are not ideal either as there may be economic repercussions (Weller and Folman, 1990)  Different health characteristics such as age, lameness, and cyclicity can impact the detection of estrus (Chapter 2) and the intensity of estrus (Chapter 2 and 4). Thus, there is potential in creating new reproductive programs based on such factors to discriminate which cows are more at risk for reduced estrous expression and less likely to demonstrate estrous behaviours. In this case, these types of at risk cows could be placed on a management regime with the use of hormonal intervention to ensure submission rates; whereas, the healthier cows could be managed without such interventions. This type of strategy could move reproductive management into an individualized management and away from blanket treatments; thus, using hormonal protocols more strategically. Another potential option could be determining if at risk cows would benefit from changes in the thresholds for alert. However, reducing the estrus alert threshold, even for a small population of cows, will increase the number of false alerts and those that are detected will have very poor estrous expression and thus poor pregnancy outcomes. Clearly, determining why these cows are exhibiting such low rates of estrus and estrous expression must be a priority.  167   Risk factors identified in this thesis that are associated with reduced estrus detection include cyclicity, parity and leg health.  Interestingly, non-cycling and multiparous cows were indeed found to have reduced estrous expression, but no association of poor leg health was found. Additionally, thin cows and estrus events that occurred early in lactation were associated with reduced estrous expression throughout this thesis. A common theme across these factors is health, and many of them can be linked to the transition period. Previous research has shown that conception risk can be increased if days to first service is postponed, and has been hypothesized to be due to cows having more time to recover after parturition (Tenhagen et al., 2003). Other studies have associated uterine disease with decreased luteal function (Sheldon et al., 2000; Williams et al., 2007; Strüve et al., 2013), delayed return to cyclicity (Földi et al., 2006) and low reproductive performance (Bonnett et al., 1993; Sheldon et al., 2000); similar decreases in fertility have also been found with other transition diseases such as subclinical ketosis, retained placenta, and lameness (Fourichon et al., 2000). Additionally, Rutherford et al. (2016) reported subclinical ketosis to decrease both the intensity and duration of estrous expression at both the first observed estrus as well as the estrus that led to the first AI postpartum. These diseases may account for differences in estrous detection and expression found early postpartum and in cows that are compromised. Currently there is a lack of research directly linking postpartum disease to estrous expression; future research should investigate if better postpartum care could be used as a prevention for poor detection and expression of estrus.  In Chapter 3, the use of rumen-reticular temperature failed to accurately identify estrus alerts. However, there were changes in temperature found at the time of estrus,  168   suggesting there is still potential for this metric to be used in unison with other measures to make a more accurate alert. Previous studies have reported success from combining information from more than one behaviour using AAMs (Firk et al., 2002; Holman et al., 2011b; Michaelis et al., 2014). As previously discussed, these improvements in accuracy may be due to the effectiveness of one technology over the other, but more likely due to differences in the pattern of behaviours exhibited by different cows. Temperature patterns may be a useful measure for producers when incorporated into automated systems; particularly, as temperature is important for detecting febrile states in sick cows.  This thesis contributes to the body of knowledge supporting the association between estrous expression and fertility, but the mechanism behind this relationship remains unknown. Although ovulation was predicted to be a major factor accounting for fertility difference, we were unable to confirm this in the work described in Chapters 4 and 5. Although variations in ovulation failure were found between cows with varying intensities of estrous expression in Chapter 4, removal of non-ovulated cows from the analysis did not change the relationship between estrous expression and fertility. If ovulation was a driving factor, it would be expected that removal of non-ovulated cows would remove the relationship between estrous expression and fertility. Furthermore, in Chapter 5, fertility improved in cows with reduced estrous expression without modulation of overall ovulation rates, again suggesting that ovulation is not the primary driver in this relationship. One limitation of our work with estrous expression is that it is unknown how differences in estrous expression are related across farms. The work presented in Chapters 2 and 5 as well as in other studies (Neves et al., 2012; Fricke et al., 2014b; Denis-Robichaud  169   et al., 2018b) provides evidence that there are stark farm differences in how cows react to reproductive management strategies. As reviewed in Chapter 1, there are many environmental and physical factors, such as flooring type (Vailes and Britt, 1990), heat stress (López-Gatius et al., 2005b), body condition (Pennington et al., 1986; Madureira et al., 2015a), and stage of lactation (Peralta et al., 2005), that can impact estrous expression, thus we would expect estrous expression intensity to vary by farm. Unfortunately, the work described in this thesis was under powered to assess farm differences and thus we were unable to investigate the variability in estrous expression across farms and how this variability is related to fertility.    6.3 General conclusions   Automated activity monitors are useful technologies for the dairy industry, becoming effective aids for reproductive management. The use of these types of technologies reduces dairy producers’ reliance on hormone protocols and gives them practical options for the incorporation of greater rates of spontaneous estrus detection into reproductive management. Estrous expression can be used to predict the fertility and ovulation failure of dairy cows.  Although these differences in fertility were not found to be fully associated with ovulation failure, their effects were found to be mitigated by the use of GnRH, which has the potential to improve conception rates. Further research is merited to decipher the interrelationship between estrous expression, ovulation and fertility.     170   References Abrams, R.M., W.W. Thatcher, F.W. Bazer, and C.J. Wilcox. 1973. Effect of estradiol-17β on vaginal thermal conductance in cattle. J. Dairy Sci. 56:1058–1062. doi:10.3168/jds.S0022-0302(73)85305-6. Adams, A.E., F.J. Olea-Popelka, and I.N. Roman-Muniz. 2013. Using temperature-sensing reticular boluses to aid in the detection of production diseases in dairy cows. J. Dairy Sci. 96:1549–1555. doi:10.3168/jds.2012-5822. Allrich, R.D. 1994. Endocrine and neural control of estrus in dairy cows. J. Dairy Sci. 77:2738–2744. doi:10.3168/jds.S0022-0302(94)77216-7. Alsaaod, M., C. Römer, J. Kleinmanns, K. Hendriksen, S. Rose-Meierhöfer, L. Plümer, and W. Büscher. 2012. Electronic detection of lameness in dairy cows through measuring pedometric activity and lying behavior. Appl. Anim. Behav. Sci. 142:134–141. doi:10.1016/j.applanim.2012.10.001. Anderson, G.A., and J. Malmo. 1985. Pregnancy rate of cows given synthetic gonadotrophin-releasing hormone at the time of service. Aust. Vet. J. 62:222–224. doi:10.1111/j.1751-0813.1985.tb07317.x. Arck, P., P.J. Hansen, B. Mulac Jericevic, M.-P. Piccinni, and J. Szekeres-Bartho. 2007. Progesterone during pregnancy: Endocrine-immune cross talk in mammalian species and the role of stress. Am. J. Reprod. Immunol. 58:268–279. doi:10.1111/j.1600-0897.2007.00512.x. At-Taras, E.E., and S.L. Spahr. 2001. Detection and characterization of estrus in dairy cattle with an electronic heatmount detector and an electronic activity tag. J. Dairy Sci. 84:792–798. doi:10.3168/jds.S0022-0302(01)74535-3. Aungier, S.P.M., J.F. Roche, P. Duffy, S. Scully, and M.A. Crowe. 2015. The relationship between activity clusters detected by an automatic activity monitor and endocrine changes during the periestrous period in lactating dairy cows. J. Dairy Sci. 98:1666–1684. doi:10.3168/jds.2013-7405. Aungier, S.P.M., J.F. Roche, M. Sheehy, M.A. Crowe, A. Gümen, G.E. Lamming, L.J. Burton,  171   J.A.S. Lancaster, J.H. Williamson, and C.W. Holmes. 2012. Effects of management and health on the use of activity monitoring for estrus detection in dairy cows. J. Dairy Sci. 95:2452–66. doi:10.3168/jds.2011-4653. Bamber, R.L., G.E. Shook, M.C. Wiltbank, J.E.P. Santos, and P.M. Fricke. 2009. Genetic parameters for anovulation and pregnancy loss in dairy cattle. J. Dairy Sci. 92:5739–5753. doi:10.3168/jds.2009-2226. Barkema, H.W., M.A.G. von Keyserlingk, J.P. Kastelic, T.J.G.M. Lam, C. Luby, J.-P. Roy, S.J. LeBlanc, G.P. Keefe, and D.F. Kelton. 2015. Invited review: Changes in the dairy industry affecting dairy cattle health and welfare. J. Dairy Sci. 98:7426–7445. doi:10.3168/jds.2015-9377. Barkema, H.W., J.D. Westrik, K.A.S. van Keulen, Y.H. Schukken, and A. Brand. 1994. The effects of lameness on reproductive performance, milk production and culling in Dutch dairy farms. Prev. Vet. Med. 20:249–259. doi:10.1016/0167-5877(94)90058-2. Bartol, F.F., W.W. Thatcher, G.S. Lewis, E.L. Bliss, M. Drost, and F.W. Bazer. 1981. Effect of estradiol-17beta on PGF and total protein content in bovine uterine flushings and peripheral plasma concentration of 13, 14-dihydro-15-keto-PGF(2alpha). Theriogenology 15:345–358. Bauman, C.A., H.W. Barkema, J. Dubuc, G.P. Keefe, and D.F. Kelton. 2016. Identifying management and disease priorities of Canadian dairy industry stakeholders. J. Dairy Sci. 99:10194–10203. doi:10.3168/JDS.2016-11057. Beaudeau, F., A. Henken, C. Fourichon, K. Frankena, and H. Seegers. 1993. Associations between health disorders and culling of dairy cows: a review. Livest. Prod. Sci. 35:213–236. doi:10.1016/0301-6226(93)90094-X. Bell, M.J., E. Wall, G. Russell, D.J. Roberts, and G. Simm. 2010. Risk factors for culling in Holstein-Friesian dairy cows. Vet. Rec. 167:238–240. doi:10.1136/vr.c4267. Bello, N.M., J.P. Steibel, and J.R. Pursley. 2006. Optimizing ovulation to first GnRH improved outcomes to each hormonal injection of Ovsynch in lactating dairy cows. J. Dairy Sci. 89:3413–3424. doi:10.3168/JDS.S0022-0302(06)72378-5. Bernard, C., J.P. Valet, R. Béland, and R.D. Lambert. 1983. Prediction of bovine ovulation by  172   a rapid radioimmunoassay for plasma LH. J. Reprod. Fertil. 68:425–30. doi:10.1530/JRF.0.0680425. Bertilsson, J., B. Berglund, G. Ratnayake, K. Svennersten-Sjaunja, and H. Wiktorsson. 1997. Optimising lactation cycles for the high-yielding dairy cow. A European perspective. Livest. Prod. Sci. 50:5–13. doi:10.1016/S0301-6226(97)00068-7. Bewley, J.M., M.E. Einstein, M.W. Grott, and M.M. Schutz. 2008. Comparison of reticular and rectal core body temperatures in lactating dairy cows. J. Dairy Sci. 91:4661–4672. doi:10.3168/jds.2007-0835. Bierschwal, C.J., H.A. Garverick, C.E. Martin, R.S. Youngquist, T.C. Cantley, and M.D. Brown. 1975. Clinical response of dairy cows with ovarian cysts to GnRH. J. Anim. Sci. 41:1660–1665. doi:10.2527/jas1975.4161660x. Bijker, I., R.M. Christley, R.F. Smith, and H. Dobson. 2015. Effect of signs of oestrus, disease stressors and cow activity on pregnancy rate following artificial insemination. Vet. Rec. 176:411–416. doi:10.1136/vr.102776. Bisinotto, R.S., R.C. Chebel, and J.E.P. Santos. 2010. Follicular wave of the ovulatory follicle and not cyclic status influences fertility of dairy cows. J. Dairy Sci. 93:3578–3587. doi:10.3168/jds.2010-3047. Bisinotto, R.S., E.S. Ribeiro, and J.E.P. Santos. 2014. Synchronisation of ovulation for management of reproduction in dairy cows. Animal Suppl 1:151–159. doi:10.1017/S1751731114000858. Bittar, J.H.J., P.J. Pinedo, C.A. Risco, J.E.P. Santos, W.W. Thatcher, K.E. Hencken, S. Croyle, M. Gobikrushanth, C.C. Barbosa, A. Vieira-Neto, and K.N. Galvão. 2014. Inducing ovulation early postpartum influences uterine health and fertility in dairy cows. J. Dairy Sci. 97:3558–3569. doi:10.3168/JDS.2013-7533. Bloch, A., Y. Folman, M. Kaim, Z. Roth, R. Braw-Tal, and D. Wolfenson. 2006. Endocrine alterations associated with extended time interval between estrus and ovulation in high-yield dairy cows. J. Dairy Sci. 89:4694–4702. doi:10.3168/jds.S0022-0302(06)72520-6. Bonafos, L.D., K. Kot, and O.J. Ginther. 1995. Physical characteristics of the uterus during the  173   bovine estrous cycle and early pregnancy. Theriogenology 43:713–721. doi:10.1016/0093-691X(95)00014-Y. Bonnett, B.N., S.W. Martin, and A.H. Meek. 1993. Associations of clinical findings, bacteriological and histological results of endometrial biopsy with reproductive performance of postpartum dairy cows. Prev. Vet. Med. 15:205–220. doi:10.1016/0167-5877(93)90114-9. Borchers, M.R., and J.M. Bewley. 2015. An assessment of producer precision dairy farming technology use, prepurchase considerations, and usefulness. J. Dairy Sci. 98:4198–4205. doi:10.3168/JDS.2014-8963. Boyle, L.A., J.F. Mee, and P.J. Kiernan. 2007. The effect of rubber versus concrete passageways in cubicle housing on claw health and reproduction of pluriparous dairy cows. Appl. Anim. Behav. Sci. 106:1–12. doi:10.1016/J.APPLANIM.2006.07.011. Brickell, J.S., and D.C. Wathes. 2011. A descriptive study of the survival of Holstein-Friesian heifers through to third calving on English dairy farms. J. Dairy Sci. 94:1831–1838. doi:10.3168/jds.2010-3710. Bruno, R.G.S., A.M. Farias, J.A. Hernández-Rivera, A.E. Navarrette, D.E. Hawkins, and T.R. Bilby. 2013. Effect of gonadotropin-releasing hormone or prostaglandin F2α-based estrus synchronization programs for first or subsequent artificial insemination in lactating dairy cows. J. Dairy Sci. 96:1556–1567. doi:10.3168/jds.2012-5885. Brusveen, D.J., A.H. Souza, and M.C. Wiltbank. 2009. Effects of additional prostaglandin F2α and estradiol-17β during Ovsynch in lactating dairy cows. J. Dairy Sci. 92:1412–1422. doi:10.3168/jds.2008-1289. Buckley, F., K. O’Sullivan, J.F. Mee, R.D. Evans, and P. Dillon. 2003. Relationships among milk yield, body condition, cow weight, and reproduction in spring-calved Holstein-Friesians.. J. Dairy Sci. 86:2308–2319. doi:10.3168/jds.S0022-0302(03)73823-5. Buhi, W.C. 2002. Characterization and biological roles of oviduct-specific, oestrogen-dependent glycoprotein. Reproduction 123:355–362. Burnett, T.A., A.M.L. Madureira, B.F. Silper, A.C.C. Fernandes, and R.L.A. Cerri. 2017a. Integrating an automated activity monitor into an artificial insemination program and  174   the associated risk factors affecting reproductive performance of dairy cows. J. Dairy Sci. 100:5005–5018. doi:10.3168/jds.2016-12246. Burnett, T.A., L. Polsky, M. Kaur, and R.L.A. Cerri. 2018. Effect of estrous expression on timing and failure of ovulation of Holstein dairy cows using automated activity monitors. J. Dairy Sci. 101:11310–11320. doi:10.3168/JDS.2018-15151. Burnett, T.A., E. Vasseur, D. Pellerin, D.B. Haley, A.M. de Passillé, J. Rushen, and R.L.A. Cerri. 2017b. Factors of cow comfort associated with herd-level reproductive outcomes on Canadian dairy farms. J. Dairy Sci. 100:290. Butler, W.R. 2000. Nutritional interactions with reproductive performance in dairy cattle. Anim. Reprod. Sci. 60–61:449–457. Butler, W.R. 2003. Energy balance relationships with follicular development, ovulation and fertility in postpartum dairy cows. Livest. Prod. Sci. 83:211–218. doi:10.1016/S0301-6226(03)00112-X. Campbell, B.., R.J. Scaramuzzi, and R. Webb. 1995. Control of antral follicle development and selection in sheep and cattle. J. Reprod. Fertil. Suppl. 49:335–350. Canadian Council for Animal. 2009. The care and use of farm animals in research, teaching and testing. Ottawa, Ontario, Canada. Caraty, A., B. Delaleu, D. Chesneau, and C. Fabre-Nys. 2002. Sequential role of E2 and GnRH for the expression of estrous behavior in ewes.. Endocrinology 143:139–145. Caraviello, D.Z., K. a Weigel, P.M. Fricke, M.C. Wiltbank, M.J. Florent, N.B. Cook, K. V Nordlund, N.R. Zwald, and C.L. Rawson. 2006a. Survey of management practices on reproductive performance of dairy cattle on large US commercial farms. J. Dairy Sci. 89:4723–4735. doi:10.3168/jds.S0022-0302(06)72522-X. Caraviello, D.Z., K.A. Weigel, M. Craven, D. Gianola, N.B. Cook, K.V. Nordlund, P.M. Fricke, and M.C. Wiltbank. 2006b. Analysis of reproductive performance of lactating cows on large dairy farms using machine learning algorithms. J. Dairy Sci. 89:4703–4722. doi:10.3168/jds.S0022-0302(06)72521-8.  175   Caraviello, D.Z., K.A. Weigel, P.M. Fricke, M.C. Wiltbank, M.J. Florent, N.B. Cook, K.V. Nordlund, N.R. Zwald, and C.L. Rawson. 2006c. Survey of management practices on reproductive performance of dairy cattle on large US commercial farms. J. Dairy Sci. 89:4723–4735. doi:10.3168/JDS.S0022-0302(06)72522-X. Carrick, M.J., and J.N. Shelton. 1969. Oestrogen-progesterone relationships in the induction of oestrus in spayed heifers. J. Endocrinol. 45:99–109. doi:10.1677/JOE.0.0450099. Cartmill, J.A., S.Z. El-Zarkouny, B.A. Hensley, T.G. Rozell, J.F. Smith, and J.S. Stevenson. 2001. An alternative AI breeding protocol for dairy cows exposed to elevated ambient temperatures before or after calving or both. J. Dairy Sci. 84:799–806. doi:10.3168/JDS.S0022-0302(01)74536-5. Cavalieri, J., V. Eagles, M. Ryan, and K.L. Macmillan. 2003a. Comparison of four methods for detection of oestrus in dairy cows with resynchronised oestrous cycles. Aust. Vet. J. 81:422–425. Cavalieri, J., L.R. Flinker, G.A. Anderson, and K.L. Macmillan. 2003b. Characteristics of oestrus measured using visual observation and radiotelemetry. Anim. Reprod. Sci. 76:1–12. Cerri, R.L.A., R.C. Chebel, F. Rivera, C.D. Narciso, R.A. Oliveira, W.W. Thatcher, and J.E.P. Santos. 2011. Concentration of progesterone during the development of the ovulatory follicle: I. Ovarian and embryonic responses. J. Dairy Sci. 94:3342–3351. doi:10.3168/jds.2010-3734. Cerri, R.L.A., J.E.P. Santos, S.O. Juchem, K.N. Galvão, and R.C. Chebel. 2004. Timed artificial insemination with estradiol cypionate or insemination at estrus in high-producing dairy cows. J. Dairy Sci. 87:3704–3715. doi:10.3168/jds.S0022-0302(04)73509-2. Chanvallon, A., S. Coyral-Castel, J. Gatien, J.-M. Lamy, D. Ribaud, C. Allain, P. Clément, and P. Salvetti. 2014. Comparison of three devices for the automated detection of estrus in dairy cows. Theriogenology 82:734–741. doi:10.1016/j.theriogenology.2014.06.010. Chapinal, N., A.M. de Passillé, J. Rushen, and S. Wagner. 2010. Automated methods for detecting lameness and measuring analgesia in dairy cattle. J. Dairy Sci. 93:2007–2013. doi:10.3168/jds.2009-2803.  176   Charmandari, E., C. Tsigos, and G. Chrousos. 2005. Endocrinology of the stress response. Annu. Rev. Physiol. 67:259–284. doi:10.1146/annurev.physiol.67.040403.120816. Chebel, R.C., M.J. Al-Hassan, P.M. Fricke, J.E.P. Santos, J.R. Lima, C.A. Martel, J.S. Stevenson, R. Garcia, and R.L. Ax. 2010. Supplementation of progesterone via controlled internal drug release inserts during ovulation synchronization protocols in lactating dairy cows. J. Dairy Sci. 93:922–931. doi:10.3168/jds.2009-2301. Chebel, R.C., and J.E.P. Santos. 2010. Effect of inseminating cows in estrus following a presynchronization protocol on reproductive and lactation performances. J. Dairy Sci. 93:4632–4643. doi:10.3168/jds.2010-3179. Chenault, J.R., W.W. Thatcher, P.S. Kalra, R.M. Abrams, and C.J. Wilcox. 1975. Transitory changes in plasma progestins, estradiol, and luteinizing hormone approaching ovulation in the bovine. J. Dairy Sci. 58:709–717. doi:10.3168/jds.S0022-0302(75)84632-7. Clapper, J.A., J.S. Ottobre, A.C. Ottobre, and D.L. Zartman. 1990. Estrual rise in body temperature in the bovine I. Temporal relationships with serum patterns of reproductive hormones. Anim. Reprod. Sci. 23:89–98. doi:10.1016/0378-4320(90)90051-G. Clemente, M., J. de La Fuente, T. Fair, A. Al Naib, A. Gutierrez-Adan, J.F. Roche, D. Rizos, and P. Lonergan. 2009. Progesterone and conceptus elongation in cattle: a direct effect on the embryo or an indirect effect via the endometrium?. Reproduction 138:507–517. doi:10.1530/REP-09-0152. Coe, B.L., and R.D. Allrich. 1989. Relationship between endogenous estradiol-17b and estrous behavior in heifers.. J. Anim. Sci. 67:1546–1551. Colazo, M.G., J.P. Kastelic, H. Davis, M.D. Rutledge, M.F. Martinez, J.A. Small, and R.J. Mapletoft. 2008. Effects of plasma progesterone concentrations on LH release and ovulation in beef cattle given GnRH. Domest. Anim. Endocrinol. 34:109–117. doi:10.1016/J.DOMANIEND.2006.11.004. Cook, D.L., T.A. Winters, L.A. Horstman, and R.D. Allrich. 1986. Induction of estrus in ovariectomized cows and heifers: effects of estradiol benzoate and gonadotropin releasing hormone.. J. Anim. Sci. 63:546–550.  177   Cooper-Prado, M.J., N.M. Long, E.C. Wright, C.L. Goad, and R.P. Wettemann. 2011. Relationship of ruminal temperature with parturition and estrus of beef cows. J. Anim. Sci. 89:1020–1027. doi:10.2527/jas.2010-3434. Cordoba, M.C., and P.M. Fricke. 2001. Evaluation of two hormonal protocols for synchronization of ovulation and timed artificial insemination in dairy cows managed in grazing-based dairies. J. Dairy Sci. 84:2700–2708. doi:10.3168/jds.S0022-0302(01)74724-8. Crighton, D.B., and J.P. Foster. 1977. Luteinizing hormone release after two injections of synthetic luteinizing hormone releasing hormone in the ewe. J. Endocrinol. 72:59–67. doi:10.1677/JOE.0.0720059. Cunha, A.P., J.N. Guenther, M.J. Maroney, J.O. Giordano, A.B. Nascimento, S. Bas, H. Ayres, and M.C. Wiltbank. 2008. Effects of high vs. low progesterone concentrations during Ovsynch on double ovulation rate and pregnancies per AI in high producing dairy cows. J. Dairy Sci. Suppl 91:246. Dalton, J.C., R. Manzo, A. Ahmadzadeh, B. Shafii, W.J. Price, and J.M. DeJarnette. 2005. Short communication: Conception rates following detection of estrus and timed AI in dairy cows synchronized using GnRH and PGF2alpha. J. Dairy Sci. 88:4313–4316. doi:10.3168/jds.S0022-0302(05)73117-9. Dalton, J.C., S. Nadir, J.H. Bame, M. Noftsinger, R.L. Nebel, and R.G. Saacke. 2001. Effect of time of insemination on number of accessory sperm, fertilization rate, and embryo quality in non-lactating dairy cattle. J. Dairy Sci. 84:2413–2418. doi:10.3168/jds.S0022-0302(01)74690-5. Davidge, S.T., J.L. Wiebold, P.L. Senger, and J.K. Hillers. 1987. Influence of varying levels of blood progesterone upon estrous behavior in cattle. J. Anim. Sci. 64:126–132. doi:10.2527/jas1987.641126x. DeJarnette, J.M., R.R. Salverson, and C.E. Marshall. 2001. Incidence of premature estrus in lactating dairy cows and conception rates to standing estrus or fixed-time inseminations after synchronization using GnRH and PGF2α. Anim. Reprod. Sci. 67:27–35. doi:10.1016/S0378-4320(01)00107-5. Demetrio, D.G.B., R.M. Santos, C.G.B. Demetrio, and J.L.M. Vasconcelos. 2007. Factors affecting conception rates following artificial insemination or embryo transfer in  178   lactating Holstein cows. J. Dairy Sci. 90:5073–5082. doi:10.3168/jds.2007-0223. Denis-Robichaud, J., R.L.. Cerri, A. Jones-Bitton, and S.J. LeBlanc. 2018a. Dairy producers’ attitudes toward reproductive management and performance on Canadian dairy farms. J. Dairy Sci. 101:850–860. doi:10.3168/jds.2016-12416. Denis-Robichaud, J., R.L.A. Cerri, A. Jones-Bitton, and S.J. LeBlanc. 2016. Survey of reproduction management on Canadian dairy farms. J. Dairy Sci. 99:9339–9351. doi:10.3168/jds.2016-11445. Denis-Robichaud, J., R.L.A. Cerri, A. Jones-Bitton, and S.J. LeBlanc. 2018b. Performance of automated activity monitoring systems used in combination with timed artificial insemination compared to timed artificial insemination only in early lactation in dairy cows. J. Dairy Sci. 101:624–636. doi:10.3168/jds.2016-12256. Denis-Robichaud, J., S. LeBlanc, A. Jones-Bitton, B. Silper, and R.L. Aoki Cerri. 2018c. Pilot study to evaluate the association between the length of the luteal phase and estrous activity detected by automated activity monitoring in dairy cows. Front. Vet. Sci. 5:1–10. doi:10.3389/FVETS.2018.00210. Dias, F.C.F., M.G. Colazo, J.P. Kastelic, R.J. Mapletoft, G.P. Adams, and J. Singh. 2010. Progesterone concentration, estradiol pretreatment, and dose of gonadotropin-releasing hormone affect gonadotropin-releasing hormone-mediated luteinizing hormone release in beef heifers. Domest. Anim. Endocrinol. 39:155–162. doi:10.1016/J.DOMANIEND.2010.04.002. Dieleman, S.J., M.M. Bevers, J. Poortman, and H.T. van Tol. 1983a. Steroid and pituitary hormone concentrations in the fluid of preovulatory bovine follicles relative to the peak of LH in the peripheral blood. J. Reprod. Fertil. 69:641–649. doi:10.1530/JRF.0.0690641. Dieleman, S.J., M.M. Bevers, H.T.M. Van Tol, and A.H. Willemse. 1986. Peripheral plasma concentrations of oestradiol, progesterone, cortisol, LH and prolactin during the oestrous cycle in the cow, with emphasis on the peri-oestrous period. Anim. Reprod. Sci. 10:275–292. doi:10.1016/0378-4320(86)90003-5. Dieleman, S.J., T.A. Kruip, P. Fontijne, W.H. de Jong, and G.C. van der Weyden. 1983b. Changes in oestradiol, progesterone and testosterone concentrations in follicular fluid and in the micromorphology of preovulatory bovine follicles relative to the peak of  179   luteinizing hormone. J. Endocrinol. 97:31–42. doi:10.1677/JOE.0.0970031. Diskin, M.G., D.R. Mackey, J.F. Roche, and J.M. Sreenan. 2003. Effects of nutrition and metabolic status on circulating hormones and ovarian follicle development in cattle. Anim. Reprod. Sci. 78:345–370. Diskin, M.G., and J.M. Sreenan. 2000a. Expression and detection of oestrus in cattle. Reprod. Nutr. Dev. 40:481–491. doi:10.1051/rnd:2000112. Diskin, M.G., and J.M. Sreenan. 2000b. Expression and detection of oestrus in cattle. Reprod. Nutr. Dev. 40:481–91. Dobson, H., R.F. Smith, M.D. Royal, C.H. Knight, and I.M. Sheldon. 2007. The high-producing dairy cow and its reproductive performance. Reprod. Domest. Anim. 42 Suppl 2:17–23. doi:10.1111/j.1439-0531.2007.00906.x. Dobson, H., S.L. Walker, M.J. Morris, J.E. Routly, and R.F. Smith. 2008. Why is it getting more difficult to successfully artificially inseminate dairy cows?. Animal 2:1104–1111. doi:10.1017/S175173110800236X. Dobson, H., J. Williams, J.E. Routly, D.N. Jones, J. Cameron, A. Holman-Coates, and R.F. Smith. 2018. Short communication: Chronology of different sexual behaviors and motion activity during estrus in dairy cows.. J. Dairy Sci. 101:8291–8295. doi:10.3168/jds.2017-14341. Dolecheck, K.A., W.J. Silvia, G. Heersche, Y.M. Chang, D.L. Ray, A.E. Stone, B.A. Wadsworth, and J.M. Bewley. 2015. Behavioral and physiological changes around estrus events identified using multiple automated monitoring technologies. J. Dairy Sci. 98:8723–8731. doi:10.3168/jds.2015-9645. Dolecheck, K.A., W.J. Silvia, G. Heersche, C.L. Wood, K.J. McQuerry, and J.M. Bewley. 2016. A comparison of timed artificial insemination and automated activity monitoring with hormone intervention in 3 commercial dairy herds. J. Dairy Sci. 99:1506–1514. doi:10.3168/jds.2015-9914. Dransfield, M.B.G., R.L. Nebel, R.E. Pearson, and L.D. Warnick. 1998. Timing of insemination for dairy cows identified in estrus by a radiotelemetric estrus detection system. J. Dairy Sci. 81:1874–1882. doi:10.3168/jds.S0022-0302(98)75758-3.  180   Edmonson, A.J., I.J. Lean, L.D. Weaver, T. Farver, and G. Webster. 1989. A body condition scoring chart for Holstein dairy cows. J. Dairy Sci. 72:68–78. doi:10.3168/jds.S0022-0302(89)79081-0. van Eerdenburg, F.J.C.M. 2008. Possible causes for diminished expression of oestrous behaviour. Vet. Q. 30:79–100. van Eerdenburg, F.J.C.M., I.A.J.J.M. Daemen, E.M. van der Beek, and F.W. van Leeuwen. 2000. Changes in estrogen-α receptor immunoreactivity during the estrous cycle in lactating dairy cattle. Brain Res. 880:219–223. doi:10.1016/S0006-8993(00)02870-5. Van Eerdenburg, F.J.C.M., D. Karthaus, M.A.M. Taverne, I. Mercis, and O. Szenci. 2002. The relationship between estrous behavioral score and time of ovulation in dairy cattle. J. Dairy Sci. 85:1150–1156. doi:10.3168/jds.S0022-0302(02)74177-5. Van Eerdenburg, F.J.C.M., H.S.H. Loeffler, and J.H. van Vliet. 1996. Detection of oestrus in dairy cows: A new approach to an old problem. Vet. Q. 18:52–54. doi:10.1080/01652176.1996.9694615. Elvinger, F., R.P. Natzke, P.J. Hansen, F. Baccari, C.J. Wilcox, C.J. Peel, and W.D. Steinhour. 1992. Interactions of heat stress and bovine somatotropin affecting physiology and immunology of lactating cows. J. Dairy Sci. 75:449–462. doi:10.3168/jds.S0022-0302(92)77781-9. Esslemont, R.J., and M.J. Bryant. 1976. Oestrous behaviour in a herd of dairy cows. Vet. Rec. 99:472–475. doi:10.1136/VR.99.24.472. Fabre-Nys, C., G.B. Martin, and G. Venier. 1993. Analysis of the hormonal control of female sexual behavior and the preovulatory LH surge in the ewe: Roles of quantity of estradiol and duration of its presence. Horm. Behav. 27:108–121. Ferguson, J.D., and A. Skidmore. 2013. Reproductive performance in a select sample of dairy herds. J. Dairy Sci. 96:1269–1289. doi:10.3168/jds.2012-5805. Firk, R., E. Stamer, W. Junge, and J. Krieter. 2002. Automation of oestrus detection in dairy cows: a review. Livest. Prod. Sci. 75:219–232. doi:10.1016/S0301-6226(01)00323-2.  181   Fisher, A.D., R. Morton, J.M.A. Dempsey, J.M. Henshall, and J.R. Hill. 2008. Evaluation of a new approach for the estimation of the time of the LH surge in dairy cows using vaginal temperature and electrodeless conductivity measurements. Theriogenology 70:1065–1074. doi:10.1016/j.theriogenology.2008.06.023. Flower, F.C., and D.M. Weary. 2006. Effect of hoof pathologies on subjective assessments of dairy cow gait. J. Dairy Sci. 89:139–146. doi:10.3168/jds.S0022-0302(06)72077-X. Flower, F.C., and D.M. Weary. 2009. Gait assessment in dairy cattle. Animal 3:87–95. doi:10.1017/S1751731108003194. Földi, J., M. Kulcsár, A. Pécsi, B. Huyghe, C. de Sa, J.A.C.M. Lohuis, P. Cox, and G. Huszenicza. 2006. Bacterial complications of postpartum uterine involution in cattle. Anim. Reprod. Sci. 96:265–281. doi:10.1016/J.ANIREPROSCI.2006.08.006. Foote, R.H. 1975. Estrus detection and estrus detection aids. J. Dairy Sci. 58:248–256. doi:10.3168/JDS.S0022-0302(75)84555-3. Forde, N., M.E. Beltman, G.B. Duffy, P. Duffy, J.P. Mehta, P. O’Gaora, J.F. Roche, P. Lonergan, and M.A. Crowe. 2011a. Changes in the endometrial transcriptome during the bovine estrous cycle: Effect of low circulating progesterone and consequences for conceptus elongation. Biol. Reprod. 84:266–278. doi:10.1095/biolreprod.110.085910. Forde, N., M.E. Beltman, P. Lonergan, M. Diskin, J.F. Roche, and M.A. Crowe. 2011b. Oestrous cycles in Bos taurus cattle. Anim. Reprod. Sci. 124:163–169. doi:10.1016/j.anireprosci.2010.08.025. Fourichon, C., H. Seegers, and X. Malher. 2000. Effect of disease on reproduction in the dairy cow: a meta-analysis. Theriogenology 53:1729–1759. doi:10.1016/S0093-691X(00)00311-3. Frewer, L.J., I.A. van der Lans, A.R.H. Fischer, M.J. Reinders, D. Menozzi, X. Zhang, I. van den Berg, and K.L. Zimmermann. 2013. Public perceptions of agri-food applications of genetic modification – A systematic review and meta-analysis. Trends Food Sci. Technol. 30:142–152. doi:10.1016/J.TIFS.2013.01.003. Fricke, P., P. Carvalho, J. Giordano, A. Valenza, G. Lopes, and M. Amundson. 2014a. Expression and detection of estrus in dairy cows: the role of new technologies. Animal  182   8:134–143. doi:10.1017/s1751731114000299. Fricke, P., J. Giordano, A. Valenza, G. Lopes, M. Amundson, and P. Carvalho. 2014b. Reproductive performance of lactating dairy cows managed for first service using timed artificial insemination with or without detection of estrus using an activity-monitoring system. J. Dairy Sci. 97:2771–2781. doi:10.3168/jds.2013-7366. Galvão, K.N., P. Federico, A. De Vries, and G.M. Schuenemann. 2013. Economic comparison of reproductive programs for dairy herds using estrus detection, timed artificial insemination, or a combination. J. Dairy Sci. 96:2681–2693. doi:10.3168/jds.2012-5982. Galvão, K.N., and J.E.P. Santos. 2010. Factors affecting synchronization and conception rate after the Ovsynch protocol in lactating Holstein cows. Reprod. Domest. Anim. 45:439–446. doi:10.1111/j.1439-0531.2008.01220.x. Galvão, K.N., J.E.P. Santos, S.O. Juchem, R.L.A. Cerri, A.C. Coscioni, and M. Villaseñor. 2004. Effect of addition of a progesterone intravaginal insert to a timed insemination protocol using estradiol cypionate on ovulation rate, pregnancy rate, and late embryonic loss in lactating dairy cows. J. Anim. Sci. 82:3508–3517. doi:10.2527/2004.82123508x. Gazal, O.S., L.S. Leshin, R.L. Stanko, M.G. Thomas, D.H. Keisler, L.L. Anderson, and G.L. Williams. 1998. Gonadotropin-releasing hormone secretion into third-ventricle cerebrospinal fluid of cattle: Correspondence with the tonic and surge release of luteinizing hormone and its tonic inhibition by suckling and neuropeptide Y. Biol. Reprod. 59:676–683. doi:10.1095/biolreprod59.3.676. Ghanem, M., T. Nakao, K. Nakatani, M. Akita, and T. Suzuki. 2006. Milk progesterone profile at and after artificial Insemination in repeat-breeding cows: effects on conception rate and embryonic death. Reprod. Domest. Anim. 41:180–183. doi:10.1111/j.1439-0531.2006.00667.x. Giordano, J.O., A.S. Kalantari, P.M. Fricke, M.C. Wiltbank, and V.E. Cabrera. 2012. A daily herd Markov-chain model to study the reproductive and economic impact of reproductive programs combining timed artificial insemination and estrus detection. J. Dairy Sci. 95:5442–5460. doi:10.3168/jds.2011-4972. Giordano, J.O., M.L. Stangaferro, R. Wijma, W.C. Chandler, and R.D. Watters. 2015.  183   Reproductive performance of dairy cows managed with a program aimed at increasing insemination of cows in estrus based on increased physical activity and fertility of timed artificial inseminations. J. Dairy Sci. 98:2488–2501. doi:10.3168/jds.2014-8961. Gómez, F., H. de Boer, and F.J.C.M. van Eerdenburg. 2003. Relationship between mild lameness and expression of oestrus in dairy cattle. Vet. Rec. 152:403–404. doi:10.1136/VR.152.13.403. Groenendaal, H., D.T. Galligan, and H.A. Mulder. 2004. An economic spreadsheet model to determine optimal breeding and replacement decisions for dairy cattle. J. Dairy Sci. 87:2146–2157. doi:10.3168/jds.S0022-0302(04)70034-X. Gümen, A., A. Keskin, G. Yilmazbas-Mecitoglu, E. Karakaya, A. Alkan, H. Okut, and M.C. Wiltbank. 2012. Effect of presynchronization strategy before Ovsynch on fertility at first service in lactating dairy cows. Theriogenology 78:1830–1838. doi:10.1016/j.theriogenology.2012.07.021. Gümen, A., and B. Seguin. 2003. Ovulation rate after GnRH or PGF2α administration in early postpartum dairy cows. Theriogenology 60:341–348. doi:10.1016/S0093-691X(02)01372-9. Gwazdauskas, F.C., J.A. Lineweaver, and W.E. Vinson. 1981. Rates of conception by artificial insemination of dairy cattle. J. Dairy Sci. 64:358–362. doi:10.3168/jds.S0022-0302(81)82575-1. Gwazdauskas, F.C., R.L. Nebel, D.J. Sprecher, W.D. Whittier, and M.L. McGilliard. 1990. Effectiveness of rump-mounted devices and androgenized females for detection of estrus in dairy cattle. J. Dairy Sci. 73:2965–2970. doi:10.3168/jds.S0022-0302(90)78985-0. Hadley, G.L., C.A. Wolf, and S.B. Harsh. 2006. Dairy cattle culling patterns, explanations, and implications. J. Dairy Sci. 89:2286–2296. doi:10.3168/jds.S0022-0302(06)72300-1. Hall, J.G., C. Branton, and E.J. Stone. 1959. Estrus, estrous cycles, ovulation time, time of service, and fertility of dairy cattle in Louisiana. J. Dairy Sci. 42:1086–1094. doi:10.3168/JDS.S0022-0302(59)90693-9. Halli, K., C. Koch, F.-J. Romberg, and S. Hoy. 2015. Investigations on automatically measured  184   feed intake amount in dairy cows during the oestrus period. Arch. Anim. Breed. 58:93–98. Hansel, W., and E.M. Convey. 1983. Physiology of the estrous cycle. J. Anim. Sci. 57:404–424. doi:10.2527/animalsci1983.57supplement_2404x. Hansen, P.J., and C.F. Aréchiga. 1999. Strategies for managing reproduction in the heat-stressed dairy cow. J. Anim. Sci. 77:36–50. Hardy, D.F., and J.F. DeBold. 1971. The relationship between levels of exogenous hormones and the display of lordosis by the female rat. Horm. Behav. 2:287–297. Hein, K.G., and R.D. Allrich. 1992. Influence of exogenous adrenocorticotropic hormone on estrous behavior in cattle. J. Anim. Sci. 70:243–7. Herlihy, M.M., M.A. Crowe, D.P. Berry, M.G. Diskin, and S.T. Butler. 2013. Factors associated with fertility outcomes in cows treated with protocols to synchronize estrus and ovulation in seasonal-calving, pasture-based dairy production systems. J. Dairy Sci. 96:1485–1498. doi:10.3168/JDS.2011-5250. Hernandez, J., J.K. Shearer, and D.W. Webb. 2001. Effect of lameness on the calving-to-conception interval in dairy cows. J. Am. Vet. Med. Assoc. 218:1611–1614. doi:10.2460/javma.2001.218.1611. Heuer, C., Y.H. Schukken, and P. Dobbelaar. 1999. Postpartum body condition score and results from the first test day milk as predictors of disease, fertility, yield, and culling in commercial dairy herds. J. Dairy Sci. 82:295–304. doi:10.3168/jds.S0022-0302(99)75236-7. Higgins, H.M., E. Ferguson, R.F. Smith, and M.J. Green. 2013. Using hormones to manage dairy cow fertility: The clinical and ethical beliefs of veterinary practitioners. PLoS One 8:e62993. doi:10.1371/journal.pone.0062993. Hillegass, J., F.S. Lima, M.F. Sá Filho, and J.E.P. Santos. 2008. Effect of time of artificial insemination and supplemental estradiol on reproduction of lactating dairy cows. J. Dairy Sci. 91:4226–4237. doi:10.3168/JDS.2007-0901.  185   Hockey, C., J. Morton, S. Norman, and M. McGowan. 2010. Evaluation of a neck mounted 2-hourly activity meter system for detecting cows about to ovulate in two paddock-based Australian dairy herds. Reprod. Domest. Anim. 45:e107-17. doi:10.1111/j.1439-0531.2009.01531.x. Holman, A., J. Thompson, J.E. Routly, J. Cameron, D.N. Jones, D. Grove-White, R.F. Smith, and H. Dobson. 2011a. Comparison of oestrus detection methods in dairy cattle. Vet. Rec. 169:47–52. doi:10.1136/vr.d2344. Holman, A., J. Thompson, J.E. Routly, J. Cameron, D.N. Jones, R.F. Smith, and H. Dobson. 2011b. Comparison of oestrus detection methods in dairy cattle. Vet. Rec. 169:47–52. doi:10.1136/vr.d2344. Hostiou, N., J. Fagon, S. Chauvat, A. Turlot, F. Kling, X. Boivin, and C. Allain. 2017. Impact of precision livestock farming on work and human-animal interactions on dairy farms. A review. Biosci. Biotechnol. Biochem. 21:1–8. Hunter, R.H.F. 2005. The Fallopian tubes in domestic mammals: how vital is their physiological activity?. Reprod. Nutr. Dev. 45:281–290. doi:10.1051/rnd:2005020. Hurnik, J.F., G.J. King, and H.A. Robertson. 1975. Estrous and related behaviour in postpartum Holstein cows. Appl. Anim. Ethol. 2:55–68. doi:10.1016/0304-3762(75)90065-6. Ijaz, A., M.L. Fahning, and R. Zemjanis. 1987. Treatment and control of cystic ovarian disease in dairy cattle: A review. Br. Vet. J. 143:226–237. doi:10.1016/0007-1935(87)90085-6. Inchaisri, C., R. Jorritsma, P.L.A.M. Vos, G.C. van der Weijden, and H. Hogeveen. 2010. Economic consequences of reproductive performance in dairy cattle. Theriogenology 74:835–846. doi:10.1016/j.theriogenology.2010.04.008. Ito, K., D.M. Weary, and M.A.G. von Keyserlingk. 2009. Lying behavior: Assessing within- and between-herd variation in free-stall-housed dairy cows. J. Dairy Sci. 92:4412–4420. doi:10.3168/jds.2009-2235. Jennes, L., J.A. Eyigor, J.A. Janovick, and P.M. Conn. 1997. Brain gonadotropin releasing hormone receptors: Localization and regulation.. Recent Prog. Horm. Res. 52:475–491.  186   Jinks, E.M., M.F. Smith, J.A. Atkins, K.G. Pohler, G.A. Perry, M.D. MacNeil, A.J. Roberts, R.C. Waterman, L.J. Alexander, and T.W. Geary. 2013. Preovulatory estradiol and the establishment and maintenance of pregnancy in suckled beef cows. J. Anim. Sci. 91:1176–1185. doi:10.2527/jas.2012-5611. Jónsson, R., M. Blanke, N.K. Poulsen, F. Caponetti, and S. Højsgaard. 2011. Oestrus detection in dairy cows from activity and lying data using on-line individual models. Comput. Electron. Agric. 76:6–15. doi:10.1016/J.COMPAG.2010.12.014. Jordan, E.R., M.J. Schouten, J.W. Quast, A.P. Belschner, and M.A. Tomaszewski. 2002. Comparison of two timed artificial insemination (TAI) protocols for management of first insemination postpartum. J. Dairy Sci. 85:1002–1008. doi:10.3168/jds.S0022-0302(02)74160-X. Jungblut, P.W., J. Gaues, A. Hughes, E. Kallweit, W. Sierralta, P. Szendro, and R.K. Wagner. 1976. Activation of transcription-regulating proteins by steroids. J. Steroid Biochem. 7:1109–1116. doi:10.1016/0022-4731(76)90041-8. Kaim, M., A. Bloch, D. Wolfenson, R. Braw-Tal, M. Rosenberg, H. Voet, and Y. Folman. 2003. Effects of GnRH administered to cows at the onset of estrus on timing of ovulation, endocrine responses, and conception. J. Dairy Sci. 86:2012–2021. doi:10.3168/JDS.S0022-0302(03)73790-4. Kamphuis, C., B. DelaRue, C.R. Burke, and J. Jago. 2012. Field evaluation of 2 collar-mounted activity meters for detecting cows in estrus on a large pasture-grazed dairy farm. J. Dairy Sci. 95:3045–3056. doi:10.3168/jds.2011-4934. Kerbrat, S., and C. Disenhaus. 2004. A proposition for an updated behavioural characterisation of the oestrus period in dairy cows. Appl. Anim. Behav. Sci. 87:223–238. doi:10.1016/j.applanim.2003.12.001. Kesner, J.S., V. Padmanabhan, and E.M. Convey. 1982. Estradiol induces and progesterone inhibits the preovulatory surges of luteinizing hormone and follicle-stimulating hormone in heifers. Biol. Reprod. 26:571–578. doi:10.1095/biolreprod26.4.571. Kou, H., Y. Zhao, K. Ren, X. Chen, Y. Lu, and D. Wang. 2017. Automated measurement of cattle surface temperature and its correlation with rectal temperature. PLoS One 12:e0175377. doi:10.1371/journal.pone.0175377.  187   Krpálková, L., V.E. Cabrera, J. Kvapilík, J. Burdych, and P. Crump. 2014. Associations between age at first calving, rearing average daily weight gain, herd milk yield and dairy herd production, reproduction, and profitability. J. Dairy Sci. 97:6573–6582. doi:10.3168/JDS.2013-7497. Kyle, B.L., A.D. Kennedy, and J.A. Small. 1998. Measurement of vaginal temperature by radiotelemetry for the prediction of estrus in beef cows. Theriogenology 49:1437–49. Kyle, S.D., C.J. Callahan, and R.D. Allrich. 1992. Effect of progesterone on the expression of estrus at the first postpartum ovulation in dairy cattle. J. Dairy Sci. 75:1456–1460. doi:10.3168/jds.S0022-0302(92)77901-6. Lammoglia, M.A., R.E. Short, S.E. Bellows, R.A. Bellows, M.D. MacNeil, and H.D. Hafs. 1998. Induced and synchronized estrus in cattle: dose titration of estradiol benzoate in peripubertal heifers and postpartum cows after treatment with an intravaginal progesterone-releasing insert and prostaglandin F2alpha. J. Anim. Sci. 76:1662–1670. doi:10.2527/1998.7661662x. Laporte, H.M., H. Hogeveen, Y.H. Schukken, and J.P.T.M. Noordhuizen. 1994. Cystic ovarian disease in Dutch dairy cattle, I. Incidence, risk factors and consequences. Livest. Prod. Sci. 38:191–197. doi:10.1016/0301-6226(94)90170-8. Larson, S.F., W.R. Butler, and W.B. Currie. 1997. Reduced fertility associated with low progesterone postbreeding and increased milk urea nitrogen in lactating cows. J. Dairy Sci. 80:1288–1295. doi:10.3168/jds.S0022-0302(97)76058-2. Laurinčík, J., J. Pícha, D. Píchová, and J. Pivko. 1991. Effect of LH-RH on meiotic maturation of preovulatory oocytes in superovulated heifers. Anim. Reprod. Sci. 25:189–197. doi:10.1016/0378-4320(91)90014-Q. Lee, C.N., J.K. Critser, and R.L. Ax. 1985. Changes of luteinizing hormone and progesterone for dairy cows after gonadotropin-releasing hormone at first postpartum breeding. J. Dairy Sci. 68:1463–1470. doi:10.3168/jds.S0022-0302(85)80984-X. Lee, L.A., J.D. Ferguson, and D.T. Galligan. 1989. Effect of disease on days open assessed by survival analysis. J. Dairy Sci. 72:1020–1026. doi:10.3168/jds.S0022-0302(89)79197-9.  188   Lee, Y., J.D. Bok, H.J. Lee, H.G. Lee, D. Kim, I. Lee, S.K. Kang, and Y.J. Choi. 2016. Body temperature monitoring using subcutaneously implanted thermo-loggers from Holstein steers. Asian-Australasian J. Anim. Sci. 29:299–306. doi:10.5713/ajas.15.0353. Lehrer, A.R., G.S. Lewis, and E. Aizinbud. 1992. Oestrus detection in cattle: recent developments. Anim. Reprod. Sci. 28:355–362. doi:10.1016/0378-4320(92)90121-S. LeRoy, C.N.S., J.S. Walton, and S.J. LeBlanc. 2018. Estrous detection intensity and accuracy and optimal timing of insemination with automated activity monitors for dairy cows. J. Dairy Sci. 101:1638–1647. doi:10.3168/jds.2017-13505. Lewis, G.S., E. Aizinbud, and A.R. Lehrer. 1989. Changes in electrical resistance of vulvar tissue in Holstein cows during ovarian cycles and after treatment with prostaglandin F2α. Anim. Reprod. Sci. 18:183–197. doi:10.1016/0378-4320(89)90020-1. Lewis, G.S., D.W. Caldwell, C.E. Rexroad, H.H. Dowlen, and J.R. Owen. 1990. Effects of gonadotropin-releasing hormone and human chorionic gonadotropin on pregnancy rate in dairy cattle. J. Dairy Sci. 73:66–72. doi:10.3168/jds.S0022-0302(90)78647-X. López-Gatius, F., M. López-Béjar, M. Fenech, and R.H.F. Hunter. 2005a. Ovulation failure and double ovulation in dairy cattle: Risk factors and effects. Theriogenology 63:1298–1307. doi:10.1016/j.theriogenology.2004.06.010. López-Gatius, F., P. Santolaria, A. Martino, F. Delétang, and F. De Rensis. 2006. The effects of GnRH treatment at the time of AI and 12 days later on reproductive performance of high producing dairy cows during the warm season in northeastern Spain. Theriogenology 65:820–830. doi:10.1016/J.THERIOGENOLOGY.2005.07.002. López-Gatius, F., P. Santolaria, I. Mundet, and J.L. Yániz. 2005b. Walking activity at estrus and subsequent fertility in dairy cows. Theriogenology 63:1419–1429. doi:10.1016/j.theriogenology.2004.07.007. Lopez, H., L.. Satter, and M.. Wiltbank. 2004. Relationship between level of milk production and estrous behavior of lactating dairy cows. Anim. Reprod. Sci. 81:209–223. doi:10.1016/J.ANIREPROSCI.2003.10.009. Løvendahl, P., and M.G.G. Chagunda. 2010. On the use of physical activity monitoring for  189   estrus detection in dairy cows. J. Dairy Sci. 93:249–259. doi:10.3168/jds.2008-1721. Lucy, M.C. 2001. Reproductive loss in high-producing dairy cattle: Where will it end?. J. Dairy Sci. 84:1277–1293. doi:10.3168/JDS.S0022-0302(01)70158-0. Lucy, M.C., and J.S. Stevenson. 1986. Gonadotropin-releasing hormone at estrus: Luteinizing hormone, estradiol, and progesterone during the periestrual and postinsemination periods in dairy cattle. Biol. Reprod. 35:300–311. doi:10.1095/biolreprod35.2.300. Lyimo, Z.C., M. Nielen, W. Ouweltjes, T.A.M. Kruip, and F.J.C.M. van Eerdenburg. 2000. Relationship among estradiol, cortisol and intensity of estrous behavior in dairy cattle. Theriogenology 53:1783–1795. doi:10.1016/S0093-691X(00)00314-9. Mader, T.L., M.S. Davis, and T. Brown-Brandl. 2006. Environmental factors influencing heat stress in feedlot cattle. J. Anim. Sci. 84:712–719. Madureira, A., B. Silper, T. Burnett, L. Polsky, L. Cruppe, D. Veira, J. Vasconcelos, and R. Cerri. 2015a. Factors affecting expression of estrus measured by activity monitors and conception risk of lactating dairy cows. J. Dairy Sci. 98:7003–7014. doi:10.3168/jds.2015-9672. Madureira, A.M.L., L.B. Polsky, T.A. Burnett, B.F. Silper, S. Soriano, A.F. Sica, K.G. Pohler, J.L.M. Vasconcelos, and R.L.A. Cerri. 2019. Intensity of estrus following an estradiol-progesterone-based ovulation synchronization protocol influences fertility outcomes. J. Dairy Sci. 102:3598–3608. doi:10.3168/jds.2018-15129. Madureira, A.M.L., B.F. Silper, L.B. Polsky, T.A. Burnett, E.L.D. Filho, S. Soriano, A.F. Sica, J.L.M. Vasconcelos, and R.L.A. Cerri. 2015b. Effects of expression of estrus measured by activity monitors on ovarian dynamics and conception risk in Holstein cows. J. Dairy Sci. 98(Suppl.):874. Mann, G.E., G.E. Lamming, R.S. Robinson, and D.C. Wathes. 1999. The regulation of interferon-tau production and uterine hormone receptors during early pregnancy. J. Reprod. Fertil. Suppl. 54:317–328. Meadows, C., P.J. Rajala-Schultz, and G.S. Frazer. 2005. A spreadsheet-based model demonstrating the nonuniform economic effects of varying reproductive performance in Ohio dairy herds. J. Dairy Sci. 88:1244–1254. doi:10.3168/jds.S0022- 190   0302(05)72791-0. Mee, M.O., J.S. Stevenson, B.M. Alexander, and R.G. Sasser. 1993. Administration of GnRH at estrus influences pregnancy rates, serum concentrations of LH, FSH, estradiol-17 beta, pregnancy-specific protein B, and progesterone, proportion of luteal cell types, and in vitro production of progesterone in dairy cows. J. Anim. Sci. 71:185–198. doi:10.2527/1993.711185x. Mee, M.O., J.S. Stevenson, R.K. Scoby, and V. Folman. 1990. Influence of gonadotropin-releasing hormone and timing of insemination relative to estrus on pregnancy rates of dairy cattle at first service. J. Dairy Sci. 73:1500–1507. doi:10.3168/jds.S0022-0302(90)78817-0. Melampy, R.M., M.A. Emmerson, J.M. Rakes, L.J. Hanka, and P.G. Eness. 1957. The effect of progesterone on the estrous response of estrogen-conditioned ovariectomized cows. J. Anim. Sci. 16:967–975. doi:10.2527/jas1957.164967x. Melendez, P., J. Bartolome, L.F. Archbald, and A. Donovan. 2003. The association between lameness, ovarian cysts and fertility in lactating dairy cows. Theriogenology 59:927–937. doi:10.1016/S0093-691X(02)01152-4. Michaelis, I., O. Burfeind, and W. Heuwieser. 2014. Evaluation of oestrous detection in dairy cattle comparing an automated activity monitoring system to visual observation. Reprod. Domest. Anim. 49:621–628. doi:10.1111/rda.12337. Milian-Suazo, F., H.N. Erb, and R.D. Smith. 1989. Risk factors for reason-specific culling of dairy cows. Prev. Vet. Med. 7:19–29. doi:10.1016/0167-5877(89)90032-9. Miura, R., K. Yoshioka, T. Miyamoto, H. Nogami, H. Okada, and T. Itoh. 2017. Estrous detection by monitoring ventral tail base surface temperature using a wearable wireless sensor in cattle. Anim. Reprod. Sci. 180:50–57. doi:10.1016/j.anireprosci.2017.03.002. de Mol, R.M., G. André, E.J.B. Bleumer, J.T.N. van der Werf, Y. de Haas, and C.G. van Reenen. 2013. Applicability of day-to-day variation in behavior for the automated detection of lameness in dairy cows. J. Dairy Sci. 96:3703–3712. doi:10.3168/jds.2012-6305. Morais, R., A. Valente, J.C. Almeida, A.M. Silva, S. Soares, M.J.C.S. Reis, R. Valentim, and J.  191   Azevedo. 2006. Concept study of an implantable microsystem for electrical resistance and temperature measurements in dairy cows, suitable for estrus detection. Sensors Actuators A Phys. 132:354–361. doi:10.1016/J.SNA.2006.04.011. Moreira, F., C. Orlandi, C.A. Risco, R. Mattos, F. Lopes, and W.W. Thatcher. 2001. Effects of presynchronization and bovine somatotropin on pregnancy rates to a timed artificial insemination protocol in lactating dairy cows. J. Dairy Sci. 84:1646–1659. doi:10.3168/jds.S0022-0302(01)74600-0. Moreira, F., C. Risco, M.F. Pires, J.D. Ambrose, M. Drost, M. DeLorenzo, and W.W. Thatcher. 2000. Effect of body condition on reproductive efficiency of lactating dairy cows receiving a timed insemination. Theriogenology 53:1305–1319. Morgan, W.F., and I.J. Lean. 1993. Gonadotrophin-releasing hormone treatment in cattle: a meta-analysis of the effects on conception at the time of insemination. Aust. Vet. J. 70:205–209. doi:10.1111/j.1751-0813.1993.tb03304.x. Morris, D., and M. Diskin. 2008. Effect of progesterone on embryo survival. animal 2:1112–1119. doi:10.1017/S1751731108002474. Morris, M.J., K. Kaneko, S.L. Walker, D.N. Jones, J.E. Routly, R.F. Smith, and H. Dobson. 2011. Influence of lameness on follicular growth, ovulation, reproductive hormone concentrations and estrus behavior in dairy cows. Theriogenology 76:658–668. doi:10.1016/J.THERIOGENOLOGY.2011.03.019. Mosher, M.D., J.S. Ottobre, G.K. Haibel, and D.L. Zartman. 1990. Estrual rise in body temperature in the bovine II. The temporal relationship with ovulation. Anim. Reprod. Sci. 23:99–107. doi:10.1016/0378-4320(90)90052-H. Mwaanga, E., and T. Janowski. 2000. Anoestrus in dairy cows: Causes, prevalence and clinical forms. Reprod. Domest. Anim. 35:193–200. doi:10.1046/j.1439-0531.2000.00211.x. Nanda, A., W. Ward, P. Williams, and H. Dobson. 1988. Retrospective analysis of the efficacy of different hormone treatments of cystic ovarian disease in cattle. Vet. Rec. 122:155–158. doi:10.1136/vr.122.7.155. Nett, T.M., A.M. Turzillo, M. Baratta, and L.A. Rispoli. 2002. Pituitary effects of steroid  192   hormones on secretion of follicle-stimulating hormone and luteinizing hormone. Domest. Anim. Endocrinol. 23:33–42. Neves, R.C., and S.J. LeBlanc. 2015. Reproductive management practices and performance of Canadian dairy herds using automated activity-monitoring systems. J. Dairy Sci. 98:2801–2811. doi:10.3168/jds.2014-8221. Neves, R.C., K.E. Leslie, J.S. Walton, and S.J. LeBlanc. 2012. Reproductive performance with an automated activity monitoring system versus a synchronized breeding program. J. Dairy Sci. 95:5683–5693. doi:10.3168/jds.2011-5264. Nor, N.M., W. Steeneveld, and H. Hogeveen. 2014. The average culling rate of Dutch dairy herds over the years 2007 to 2010 and its association with herd reproduction, performance and health. J. Dairy Res. 81:1–8. doi:10.1017/S0022029913000460. NRC. 2001. Nutrient Requirements of Dairy Cattle. 7th ed. Natl. Acad. Sci., Washington, DC. O’Callaghan, K.A., P.J. Cripps, D.Y. Downham, and R.D. Murray. 2003. Subjective and objective assessment of pain and discomfort due to lameness in dairy cattle. Anim. Welf. 12:605–610. Olori, V.E., S. Brotherstone, W.G. Hill, and B.J. McGuirk. 1997. Effect of gestation stage on milk yield and composition in Holstein Friesian dairy cattle. Livest. Prod. Sci. 52:167–176. doi:10.1016/S0301-6226(97)00126-7. Oltenacu, P.A., T.R. Rounsaville, R.A. Milligan, and R.H. Foote. 1981. Systems analysis for designing reproductive management programs to increase production and profit in dairy herds. J. Dairy Sci. 64:2096–2104. doi:10.3168/jds.S0022-0302(81)82813-5. Padmanabhan, Y., K. Leung, and E.M. Convey. 1982. Ovarian steroids modulate the self-priming effect of luteinizing hormone-releasing hormone on bovine pituitary cells in vitro. Endocrinology 110:717–721. doi:10.1210/endo-110-3-717. Palmer, M.A., G. Olmos, L.A. Boyle, and J.F. Mee. 2010. Estrus detection and estrus characteristics in housed and pastured Holstein–Friesian cows. Theriogenology 74:255–264. doi:10.1016/j.theriogenology.2010.02.009.  193   Palmer, M.A., G. Olmos, L.A. Boyle, and J.F. Mee. 2012. A comparison of the estrous behavior of Holstein-Friesian cows when cubicle-housed and at pasture. Theriogenology 77:382–388. doi:10.1016/j.theriogenology.2011.08.010. Panin, M., L. Corain, S. Montelli, B. Cozzi, and A. Peruffo. 2015. Gene expression profiles of estrogen receptors α and β in the fetal bovine hypothalamus and immunohistochemical characterization during development. Cell Tissue Res. 359:619–626. doi:10.1007/s00441-014-2023-5. Pennington, J.A., J.L. Albright, and C.J. Callahan. 1986. Relationships of sexual activities in estrous cows to different frequencies of observation and pedometer measurements. J. Dairy Sci. 69:2925–2934. doi:10.3168/jds.S0022-0302(86)80748-2. Pennington, J.A., and C.J. Callahan. 1986. Use of mount detectors plus chalk as an estrous detection aid for dairy cattle. J. Dairy Sci. 69:248–252. doi:10.3168/JDS.S0022-0302(86)80393-9. Peralta, O.A., R.E. Pearson, and R.L. Nebel. 2005. Comparison of three estrus detection systems during summer in a large commercial dairy herd. Anim. Reprod. Sci. 87:59–72. doi:10.1016/j.anireprosci.2004.10.003. Pereira, M.H.C., A.D.P. Rodrigues, R.J. De Carvalho, M.C. Wiltbank, and J.L.M. Vasconcelos. 2014. Increasing length of an estradiol and progesterone timed artificial insemination protocol decreases pregnancy losses in lactating dairy cows. J. Dairy Sci. 97:1454–1464. doi:10.3168/JDS.2013-7287. Pereira, M.H.C., A.D.P. Rodrigues, T. Martins, W.V.C. Oliveira, P.S.A. Silveira, M.C. Wiltbank, and J.L.M. Vasconcelos. 2013. Timed artificial insemination programs during the summer in lactating dairy cows: Comparison of the 5-d Cosynch protocol with an estrogen/progesterone-based protocol. J. Dairy Sci. 96:6904–6914. doi:10.3168/jds.2012-6260. Pereira, M.H.C., M.C. Wiltbank, and J.L.M. Vasconcelos. 2016. Expression of estrus improves fertility and decreases pregnancy losses in lactating dairy cows that receive artificial insemination or embryo transfer. J. Dairy Sci. 99:2237–2247. doi:10.3168/JDS.2015-9903. Perry, G.A., O.L. Swanson, E.L. Larimore, B.L. Perry, G.D. Djira, and R.A. Cushman. 2014. Relationship of follicle size and concentrations of estradiol among cows exhibiting or  194   not exhibiting estrus during a fixed-time AI protocol. Domest. Anim. Endocrinol. 48:15–20. doi:10.1016/J.DOMANIEND.2014.02.001. Peter, A.T., W.T. Bosu, and R.J. DeDecker. 1989. Suppression of preovulatory luteinizing hormone surges in heifers after intrauterine infusions of Escherichia coli endotoxin. Am. J. Vet. Res. 50:368–373. Pieper, L., M.G. Doherr, and W. Heuwieser. 2016. Consumers’ attitudes about milk quality and fertilization methods in dairy cows in Germany. J. Dairy Sci. 99:3162–3170. doi:10.3168/jds.2015-10169. Plaizier, J.C.B., G.J. King, J.C.M. Dekkers, and K. Lissemore. 1997. Estimation of economic values of indices for reproductive performance in dairy herds using computer simulation. J. Dairy Sci. 80:2775–2783. doi:10.3168/jds.S0022-0302(97)76240-4. Platz, S., F. Ahrens, J. Bendel, H.H.D. Meyer, and M.H. Erhard. 2008. What happens with cow behavior when replacing concrete slatted floor by rubber coating: A case study. J. Dairy Sci. 91:999–1004. doi:10.3168/jds.2007-0584. Polsky, L.B., A.M.L. Madureira, E.L.D. Filho, S. Soriano, A.F. Sica, J.L.M. Vasconcelos, and R.L.A. Cerri. 2017. Association between ambient temperature and humidity, vaginal temperature, and automatic activity monitoring on induced estrus in lactating cows. J. Dairy Sci. 100:8590–8601. doi:10.3168/jds.2017-12656. Portaluppi, M.A., and J.S. Stevenson. 2005. Pregnancy rates in lactating dairy cows after presynchronization of estrous cycles and variations of the Ovsynch protocol. J. Dairy Sci. 88:914–921. doi:10.3168/jds.S0022-0302(05)72758-2. Pursley, J.R., M.R. Kosorok, and M.C. Wiltbank. 1997a. Reproductive management of lactating dairy cows using synchronization of ovulation. J. Dairy Sci. 80:301–306. doi:10.3168/jds.S0022-0302(97)75938-1. Pursley, J.R., R.W. Silcox, and M.C. Wiltbank. 1998. Effect of time of artificial insemination on pregnancy rates, calving rates, pregnancy loss, and gender ratio after synchronization of ovulation in lactating dairy cows. J. Dairy Sci. 81:2139–2144. doi:10.3168/JDS.S0022-0302(98)75790-X. Pursley, J.R., M.C. Wiltbank, J.S. Stevenson, J.S. Ottobre, H.A. Garverick, and L.L. Anderson.  195   1997b. Pregnancy rates per artificial insemination for cows and heifers inseminated at a synchronized ovulation or synchronized estrus. J. Dairy Sci. 80:295–300. doi:10.3168/jds.S0022-0302(97)75937-X. Rabiee, A.., K.. Macmillan, and F. Schwarzenberger. 2001. The effect of level of feed intake on progesterone clearance rate by measuring faecal progesterone metabolites in grazing dairy cows. Anim. Reprod. Sci. 67:205–214. doi:10.1016/S0378-4320(01)00130-0. Rajamahendran, R., J. Robinson, S. Desbottes, and J.S. Walton. 1989. Temporal relationships among estrus, body temperature, milk yield, progesterone and luteinizing hormone levels, and ovulation in dairy cows. Theriogenology 31:1173–1182. doi:10.1016/0093-691X(89)90086-1. Randi, F., M. McDonald, P. Duffy, A.K. Kelly, and P. Lonergan. 2018. The relationship between external auditory canal temperature and onset of estrus and ovulation in beef heifers. Theriogenology 110:175–181. doi:10.1016/J.THERIOGENOLOGY.2018.01.001. Reames, P.S., T.B. Hatler, S.H. Hayes, D.L. Ray, and W.J. Silvia. 2011. Differential regulation of estrous behavior and luteinizing hormone secretion by estradiol-17β in ovariectomized dairy cows. Theriogenology 75:233–240. doi:10.1016/J.THERIOGENOLOGY.2010.08.009. Redden, K.D., A.D. Kennedy, J.R. Ingalls, and T.L. Gilson. 1993. Detection of estrus by radiotelemetric monitoring of vaginal and ear skin temperature and pedometer measurements of activity. J. Dairy Sci. 76:713–721. doi:10.3168/jds.S0022-0302(93)77394-4. Reimers, T.J., R.D. Smith, and S.K. Newman. 1985. Management factors affecting reproductive performance of dairy cows in the northeastern United States. J. Dairy Sci. 68:963–972. doi:10.3168/jds.S0022-0302(85)80916-4. Reith, S., H. Brandt, and S. Hoy. 2014. Simultaneous analysis of activity and rumination time, based on collar-mounted sensor technology, of dairy cows over the peri-estrus period. Livest. Sci. 170:219–227. doi:10.1016/J.LIVSCI.2014.10.013. Reith, S., and S. Hoy. 2012. Relationship between daily rumination time and estrus of dairy cows. J. Dairy Sci. 95:6416–6420. doi:10.3168/jds.2012-5316.  196   Reith, S., and S. Hoy. 2018. Review: Behavioral signs of estrus and the potential of fully automated systems for detection of estrus in dairy cattle. Animal 398–407. doi:10.1017/S1751731117001975. Rhodes, F.M., C.R. Burke, B.A. Clark, M.L. Day, and K.L. Macmillan. 2002. Effect of treatment with progesterone and oestradiol benzoate on ovarian follicular turnover in postpartum anoestrous cows and cows which have resumed oestrous cycles. Anim. Reprod. Sci. 69:139–150. doi:10.1016/S0378-4320(01)00141-5. Rhodes, F.M., S. McDougall, C.R. Burke, G.A. Verkerk, and K.L. Macmillan. 2003. Invited review: Treatment of cows with an extended postpartum anestrous interval. J. Dairy Sci. 86:1876–1894. doi:10.3168/jds.S0022-0302(03)73775-8. Ribadu, A.Y., K. Nakada, M. Moriyoshi, W.C. Zhang, Y. Tanaka, and T. Nakao. 2000. The role of LH pulse frequency in ACTH-induced ovarian follicular cysts in heifers. Anim. Reprod. Sci. 64:21–31. Ribadu, A.Y., K. Nakada, Y. Tanaka, M. Moriyoshi, W.C. Zhang, and T. Nakao. 1999. Lack of LH response to exogenous estradiol in heifers with ACTH-induced ovarian follicular cysts. J. Vet. Med. Sci. 61:979–981. doi:10.1292/jvms.61.979. Rivera, F., C. Narciso, R. Oliveira, R.L.A. Cerri, A. Correa-Calderón, R.C. Chebel, and J.E.P. Santos. 2010. Effect of bovine somatotropin (500mg) administered at ten-day intervals on ovulatory responses, expression of estrus, and fertility in dairy cows. J. Dairy Sci. 93:1500–1510. doi:10.3168/jds.2009-2489. Rivera, F.A., L.G.D. Mendonça, G. Lopes, J.E.P. Santos, R. V Perez, M. Amstalden, A. Correa-Calderón, and R.C. Chebel. 2011. Reduced progesterone concentration during growth of the first follicular wave affects embryo quality but has no effect on embryo survival post transfer in lactating dairy cows. Reproduction 141:333–342. doi:10.1530/REP-10-0375. Rivera, H., H. Lopez, and P.M. Fricke. 2004. Fertility of Holstein dairy heifers after synchronization of ovulation and timed AI or AI after removed tail chalk. J. Dairy Sci. 87:2051–2061. doi:10.3168/JDS.S0022-0302(04)70023-5. Rivier, C., and S. Rivest. 1991. Effect of stress on the activity of the hypothalamic-pituitary-gonadal axis: peripheral and central mechanisms. Biol. Reprod. 45:523–532. doi:10.1095/biolreprod45.4.523.  197   Roche, J.F. 1996. Control and regulation of folliculogenesis--a symposium in perspective. Rev. Reprod. 1:19–27. doi:10.1530/ROR.0.0010019. Roche, J.R., N.C. Friggens, J.K. Kay, M.W. Fisher, K.J. Stafford, and D.P. Berry. 2009. Invited review: Body condition score and its association with dairy cow productivity, health, and welfare. J. Dairy Sci. 92:5769–5801. doi:10.3168/jds.2009-2431. Roche, J.R., K.A. Macdonald, C.R. Burke, J.M. Lee, and D.P. Berry. 2007. Associations among body condition score, body weight, and reproductive performance in seasonal-calving dairy cattle. J. Dairy Sci. 90:376–391. doi:10.3168/jds.S0022-0302(07)72639-5. Rodrigues, W.B., A.S. Silva, J.C.B. Silva, N.A. Anache, K.C. Silva, C.J.T. Cardoso, W.R. Garcia, P. Sutovsky, and E. Nogueira. 2019. Timed artificial insemination plus heat II: gonadorelin injection in cows with low estrus expression scores increased pregnancy in progesterone/estradiol-based protocol. animal 1–6. doi:10.1017/S1751731119000454. Rodtian, P., G. King, S. Subrod, and P. Pongpiachan. 1996. Oestrous behaviour of Holstein cows during cooler and hotter tropical seasons. Anim. Reprod. Sci. 45:47–58. Roelofs, J., F. López-Gatius, R.H.F. Hunter, F.J.C.M. van Eerdenburg, and C. Hanzen. 2010. When is a cow in estrus? Clinical and practical aspects. Theriogenology 74:327–344. doi:10.1016/j.theriogenology.2010.02.016. Roelofs, J.B., E.G. Bouwman, S.J. Dieleman, F.J.C.M. Van Eerdenburg, L.M.T.E. Kaal-Lansbergen, N.M. Soede, and B. Kemp. 2004. Influence of repeated rectal ultrasound examinations on hormone profiles and behaviour around oestrus and ovulation in dairy cattle. Theriogenology 62:1337–1352. doi:10.1016/j.theriogenology.2004.02.002. Roelofs, J.B., F.J.C.M. Van Eerdenburg, N.M. Soede, and B. Kemp. 2005b. Various behavioral signs of estrous and their relationship with time of ovulation in dairy cattle. Theriogenology 63:1366–1377. doi:10.1016/j.theriogenology.2004.07.009. Roelofs, J.B., F.J.C.M. Van Eerdenburg, N.M. Soede, and B. Kemp. 2005a. Pedometer readings for estrous detection and as predictor for time of ovulation in dairy cattle. Theriogenology 64:1690–1703. doi:10.1016/j.theriogenology.2005.04.004.  198   Roelofs, J.B., and E. van Erp-van der Kooij. 2015. Estrus detection tools and their applicability in cattle: recent and perspectival situation. Anim. Reprod.  12:498–504. Roelofs, J.B., E.A.M. Graat, E. Mullaart, N.M. Soede, W. Voskamp-Harkema, and B. Kemp. 2006. Effects of insemination–ovulation interval on fertilization rates and embryo characteristics in dairy cattle. Theriogenology 66:2173–2181. doi:10.1016/j.theriogenology.2006.07.005. Roelofs, J.B., C. Krijnen, and E. van Erp-van der Kooij. 2017. The effect of housing condition on the performance of two types of activity meters to detect estrus in dairy cows. Theriogenology 93:12–15. doi:10.1016/j.theriogenology.2017.01.037. Rorie, R.W., T.R. Bilby, and T.D. Lester. 2002. Application of electronic estrus detection technologies to reproductive management of cattle. Theriogenology 57:137–148. Rutherford, A.J., G. Oikonomou, and R.F. Smith. 2016. The effect of subclinical ketosis on activity at estrus and reproductive performance in dairy cattle. J. Dairy Sci. 99:4808–4815. doi:10.3168/JDS.2015-10154. Rutllant, J., M. Lopez-Bejar, and F. Lopez-Gatius. 2005. Ultrastructural and rheological properties of bovine vaginal fluid and its relation to sperm motility and fertilization: a review. Reprod. Domest. Anim. 40:79–86. doi:10.1111/j.1439-0531.2004.00510.x. Ryan, D.P., E. Kopel, M.P. Boland, and R.A. Godke. 1991. Pregnancy rates in dairy cows following the administration of a GnRH analogue at the time of artificial insemination or at mid-cycle post insemination. Theriogenology 36:367–377. doi:10.1016/0093-691X(91)90465-P. Ryan, D.P., S. Snijders, T. Condon, M. Grealy, J. Sreenan, and K.J. O’Farrell. 1994. Endocrine and ovarian responses and pregnancy rates in dairy cows following the administration of a gonadotrophin releasing hormone analog at the time of artificial insemination or at mid-cycle post insemination. Anim. Reprod. Sci. 34:179–191. doi:10.1016/0378-4320(94)90015-9. Saacke, R.G., J.C. Dalton, S. Nadir, R.L. Nebel, and J.H. Bame. 2000. Relationship of seminal traits and insemination time to fertilization rate and embryo quality. Anim. Reprod. Sci. 60–61:663–677.  199   Saint-Dizier, M., and S. Chastant-Maillard. 2012. Towards an automated detection of oestrus in dairy cattle. Reprod. Domest. Anim. 47:1056–1061. doi:10.1111/j.1439-0531.2011.01971.x. Sakatani, M., M. Takahashi, and N. Takenouchi. 2016. The efficiency of vaginal temperature measurement for detection of estrus in Japanese Black cows. J. Reprod. Dev. 62:201–207. doi:10.1262/jrd.2015-095. Sangsritavong, S., D.K. Combs, R. Sartori, L.E. Armentano, and M.C. Wiltbank. 2002. High feed intake increases liver blood flow and metabolism of progesterone and estradiol-17beta in dairy cattle. J. Dairy Sci. 85:2831–2842. doi:10.3168/jds.S0022-0302(02)74370-1. Santos, J.E.., R.L.. Cerri, M.. Ballou, G.. Higginbotham, and J.. Kirk. 2004. Effect of timing of first clinical mastitis occurrence on lactational and reproductive performance of Holstein dairy cows. Anim. Reprod. Sci. 80:31–45. doi:10.1016/S0378-4320(03)00133-7. Santos, V.G., P.D. Carvalho, C. Maia, B. Carneiro, A. Valenza, and P.M. Fricke. 2017. Fertility of lactating Holstein cows submitted to a Double-Ovsynch protocol and timed artificial insemination versus artificial insemination after synchronization of estrus at a similar day in milk range. J. Dairy Sci. 100:8507–8517. doi:10.3168/jds.2017-13210. Saumande, J., and P. Humblot. 2005. The variability in the interval between estrus and ovulation in cattle and its determinants. Anim. Reprod. Sci. 85:171–182. doi:10.1016/J.ANIREPROSCI.2003.09.009. Sawyer, G.J., I.D. Russell-Brown, and J.K. Silcock. 1986. A comparison of three methods of oestrus detection in commercial dairy herds verified by serum progesterone analysis. Anim. Reprod. Sci. 10:1–10. doi:10.1016/0378-4320(86)90135-1. Scanavez, A.L.A., B. Fragomeni, and L.G.D. Mendonça. 2018. Animal factors associated with core body temperature of nonlactating dairy cows during summer. J. Anim. Sci. 96:5000–5009. doi:10.1093/jas/sky353. Schefers, J.M., K.A. Weigel, C.L. Rawson, N.R. Zwald, and N.B. Cook. 2010. Management practices associated with conception rate and service rate of lactating Holstein cows in large, commercial dairy herds. J. Dairy Sci. 93:1459–1467. doi:10.3168/jds.2009-2015.  200   Senger, P.L. 1994. The estrus detection problem: New concepts, technologies, and possibilities. J. Dairy Sci. 77:2745–2753. doi:10.3168/JDS.S0022-0302(94)77217-9. Sheldon, I.M., G.S. Lewis, S. LeBlanc, and R.O. Gilbert. 2006. Defining postpartum uterine disease in cattle. Theriogenology 65:1516–1530. doi:10.1016/j.theriogenology.2005.08.021. Sheldon, I.M., D.E. Noakes, and H. Dobson. 2000. The influence of ovarian activity and uterine involution determined by ultrasonography on subsequent reproductive performance of dairy cows. Theriogenology 54:409–419. doi:10.1016/S0093-691X(00)00358-7. Shephard, R.W., J.M. Morton, and S.T. Norman. 2014. Effects of administration of gonadotropin-releasing hormone at artificial insemination on conception rates in dairy cows. Anim. Reprod. Sci. 144:14–21. doi:10.1016/j.anireprosci.2013.11.004. Silper, B.F., A.M.L. Madureira, L.B. Polsky, S. Soriano, A.F. Sica, J.L.M. Vasconcelos, and R.L.A. Cerri. 2017. Daily lying behavior of lactating Holstein cows during an estrus synchronization protocol and its associations with fertility. J. Dairy Sci. 100:8484–8495. doi:10.3168/JDS.2016-12160. Silper, B.F., L. Polsky, J. Luu, T.A. Burnett, J. Rushen, A.M. de Passillé, and R.L.A. Cerri. 2015. Automated and visual measurements of estrous behavior and their sources of variation in Holstein heifers. II: Standing and lying patterns. Theriogenology 84:333–341. doi:10.1016/J.THERIOGENOLOGY.2014.12.030. de Silva, A.W.M.V., G.W. Anderson, F.C. Gwazdauskas, M.L. McGilliard, and J.A. Lineweaver. 1981. Interrelationships with estrous behavior and conception in dairy cattle. J. Dairy Sci. 64:2409–2418. doi:10.3168/JDS.S0022-0302(81)82864-0. Silva, C.C., and P.G. Knight. 2000. Effects of androgens, progesterone and their antagonists on the developmental competence of in vitro matured bovine oocytes. J. Reprod. Fertil. 119:261–269. Silva, E., R.A. Sterry, and P.M. Fricke. 2007. Assessment of a practical method for identifying anovular dairy cows synchronized for first postpartum timed artificial insemination. J. Dairy Sci. 90:3255–3262. doi:10.3168/jds.2006-779.  201   Simerly, R.B. 1998. Organization and regulation of sexually dimorphic neuroendocrine pathways. Behav. Brain Res. 92:195–203. doi:10.1016/S0166-4328(97)00191-5. Simerly, R.B., A.M. Carr, M.C. Zee, and D. Lorang. 1996. Ovarian steroid regulation of estrogen and progesterone receptor messenger ribonucleic acid in the anteroventral periventricular nucleus of the rat. J. Neuroendocrinol. 8:45–56. doi:10.1111/j.1365-2826.1996.tb00685.x. Solano, L., H.W. Barkema, E.A. Pajor, S. Mason, S.J. LeBlanc, J.C. Zaffino Heyerhoff, C.G.R. Nash, D.B. Haley, E. Vasseur, D. Pellerin, J. Rushen, A.M. de Passillé, and K. Orsel. 2015. Prevalence of lameness and associated risk factors in Canadian Holstein-Friesian cows housed in freestall barns. J. Dairy Sci. 98:6978–6991. doi:10.3168/JDS.2015-9652. Sood, P., and A.S. Nanda. 2006. Effect of lameness on estrous behavior in crossbred cows. Theriogenology 66:1375–1380. doi:10.1016/J.THERIOGENOLOGY.2006.04.031. Souza, A.H., A. Gümen, E.P.B. Silva, A.P. Cunha, J.N. Guenther, C.M. Peto, D.Z. Caraviello, and M.C. Wiltbank. 2007. Supplementation with estradiol-17β before the last gonadotropin-releasing hormone injection of the Ovsynch protocol in lactating dairy cows. J. Dairy Sci. 90:4623–4634. doi:10.3168/jds.2007-0172. Souza, A.H., E.P.B. Silva, A.P. Cunha, A. Gümen, H. Ayres, D.J. Brusveen, J.N. Guenther, and M.C. Wiltbank. 2011. Ultrasonographic evaluation of endometrial thickness near timed AI as a predictor of fertility in high-producing dairy cows. Theriogenology 75:722–733. doi:10.1016/j.theriogenology.2010.10.013. Sterry, R.A., M.L. Welle, and P.M. Fricke. 2006. Treatment with gonadotropin-releasing hormone after first timed artificial insemination improves fertility in noncycling lactating dairy cows. J. Dairy Sci. 89:4237–4245. doi:10.3168/jds.S0022-0302(06)72469-9. Stevenson, J.S. 2001. Reproductive management of dairy cows in high milk-producing herds. J. Dairy Sci. 84:E128-143. doi:10.3168/jds.S0022-0302(01)70207-X. Stevenson, J.S., and J.H. Britt. 2017. A 100-Year Review: Practical female reproductive management. J. Dairy Sci. 100:10292–10313. doi:10.3168/jds.2017-12959. Stevenson, J.S., E.P. Call, R.K. Scoby, and A.P. Phatak. 1990. Double insemination and  202   gonadotropin-releasing hormone treatment of repeat-breeding dairy cattle. J. Dairy Sci. 73:1766–1772. doi:10.3168/jds.S0022-0302(90)78855-8. Stevenson, J.S., S.L. Hill, R.L. Nebel, and J.M. DeJarnette. 2014. Ovulation timing and conception risk after automated activity monitoring in lactating dairy cows. J. Dairy Sci. 97:4296–4308. doi:http://dx.doi.org/10.3168/jds.2013-7873. Stevenson, J.S., G.C. Lamb, Y. Kobayashi, and D.P. Hoffman. 1998. Luteolysis during two stages of the estrous cycle: subsequent endocrine profiles associated with radiotelemetrically detected estrus in heifers. J. Dairy Sci. 81:2897–2903. doi:10.3168/jds.S0022-0302(98)75850-3. Stevenson, J.S., and A.P. Phatak. 2005. Inseminations at estrus induced by presynchronization before application of synchronized estrus and ovulation. J. Dairy Sci. 88:399–405. doi:http://dx.doi.org/10.3168/jds.S0022-0302(05)72700-4. Stevenson, J.S., and S.L. Pulley. 2016. Feedback effects of estradiol and progesterone on ovulation and fertility of dairy cows after gonadotropin-releasing hormone-induced release of luteinizing hormone. J. Dairy Sci. 99:3003–3015. doi:10.3168/jds.2015-10091. Stronge, A.J.H., J.M. Sreenan, M.G. Diskin, J.F. Mee, D.A. Kenny, and D.G. Morris. 2005. Post-insemination milk progesterone concentration and embryo survival in dairy cows. Theriogenology 64:1212–1224. doi:10.1016/j.theriogenology.2005.02.007. Strüve, K., K. Herzog, F. Magata, M. Piechotta, K. Shirasuna, A. Miyamoto, and H. Bollwein. 2013. The effect of metritis on luteal function in dairy cows. BMC Vet. Res. 9:244–253. doi:10.1186/1746-6148-9-244. Sturman, H., E.A.B. Oltenacu, and R.H. Foote. 2000. Importance of inseminating only cows in estrus. Theriogenology 53:1657–1667. doi:10.1016/S0093-691X(00)00305-8. Suthar, V., O. Burfeind, S. Bonk, R. Voigtsberger, C. Keane, and W. Heuwieser. 2012. Factors associated with body temperature of healthy Holstein dairy cows during the first 10 days in milk. J. Dairy Res. 79:135–142. doi:10.1017/S0022029911000896. Suthar, V.S., O. Burfeind, J.S. Patel, A.J. Dhami, and W. Heuwieser. 2011. Body temperature around induced estrus in dairy cows. J. Dairy Sci. 94:2368–2373.  203   doi:10.3168/jds.2010-3858. Sutherland, S., and D. Lindsay. 1991. Ovariectomized does do not require progesterone priming for oestrous behaviour. Reprod. Fertil. Dev. 3:679–684. doi:10.1071/RD9910679. Swanson, L. V., and S.K. McCarthy. 1978. Estradiol treatment and luteinizing hormone (LH) response of prepubertal Holstein heifers. Biol. Reprod. 18:475–480. doi:10.1095/biolreprod18.3.475. Talukder, S., K.L. Kerrisk, L. Ingenhoff, P.C. Thomson, S.C. Garcia, and P. Celi. 2014. Infrared technology for estrus detection and as a predictor of time of ovulation in dairy cows in a pasture-based system. Theriogenology 81:925–935. doi:10.1016/j.theriogenology.2014.01.009. Talukder, S., P.C. Thomson, K.L. Kerrisk, C.E.F. Clark, and P. Celi. 2015. Evaluation of infrared thermography body temperature and collar-mounted accelerometer and acoustic technology for predicting time of ovulation of cows in a pasture-based system. Theriogenology 83:739–748. doi:10.1016/j.theriogenology.2014.11.005. Tenhagen, B.-A., M. Drillich, R. Surholt, and W. Heuwieser. 2004a. Comparison of timed AI after synchronized ovulation to AI at estrus: Reproductive and economic considerations. J. Dairy Sci. 87:85–94. doi:10.3168/jds.S0022-0302(04)73145-8. Tenhagen, B.A., R. Surholt, M. Wittke, C. Vogel, M. Drillich, and W. Heuwieser. 2004b. Use of Ovsynch in dairy herds--differences between primiparous and multiparous cows. Anim. Reprod. Sci. 81:1–11. Tenhagen, B.A., C. Vogel, M. Drillich, G. Thiele, and W. Heuwieser. 2003. Influence of stage of lactation and milk production on conception rates after timed artificial insemination following Ovsynch. Theriogenology 60:1527–1537. Thatcher, W.W., M. Drost, J.D. Savio, K.L. Macmillan, K.W. Entwistle, E.J. Schmitt, R.L. De la Sota, and G.R. Morris. 1993. New clinical uses of GnRH and its analogues in cattle. Anim. Reprod. Sci. 33:27–49. doi:10.1016/0378-4320(93)90105-Z. Thorup, V.M., L. Munksgaard, P.E. Robert, H.W. Erhard, P.T. Thomsen, and N.C. Friggens. 2015. Lameness detection via leg-mounted accelerometers on dairy cows on four  204   commercial farms. Animal 9:1704–1712. doi:10.1017/S1751731115000890. Vailes, L.D., and J.H. Britt. 1990. Influence of footing surface on mounting and other sexual behaviors of estrual Holstein cows. J. Anim. Sci. 68:2333–2339. Vailes, L.D., S.P. Washburn, and J.H. Britt. 1992. Effects of various steroid milieus or physiological states on sexual behavior of Holstein cows. J. Anim. Sci. 70:2094–2103. doi:10.2527/1992.7072094x. Valenza,  a, J.O. Giordano, G. Lopes, L. Vincenti, M.C. Amundson, and P.M. Fricke. 2012. Assessment of an accelerometer system for detection of estrus and treatment with gonadotropin-releasing hormone at the time of insemination in lactating dairy cows. J. Dairy Sci. 95:7115–7127. doi:10.3168/jds.2012-5639. Vanholder, T., G. Opsomer, and A. de Kruif. 2006. Aetiology and pathogenesis of cystic ovarian follicles in dairy cattle: a review. Reprod. Nutr. Dev. 46:105–119. doi:10.1051/rnd:2006003. Vasudevan, N., and D.W. Pfaff. 2008. Non-genomic actions of estrogens and their interaction with genomic actions in the brain. Front. Neuroendocrinol. 29:238–257. doi:10.1016/j.yfrne.2007.08.003. Veerkamp, R.F., J.K. Oldenbroek, H.J. Van der Gaast, and J.H. Van der Werf. 2000. Genetic correlation between days until start of luteal activity and milk yield, energy balance, and live weights. J. Dairy Sci. 83:577–583. Van Vliet, J.H., and F.J.C.M. Van Eerdenburg. 1996. Sexual activities and oestrus detection in lactating Holstein cows. Appl. Anim. Behav. Sci. 50:57–69. doi:10.1016/0168-1591(96)01068-4. De Vries, A., J.D. Olson, and P.J. Pinedo. 2010. Reproductive risk factors for culling and productive life in large dairy herds in the eastern United States between 2001 and 2006. J. Dairy Sci. 93:613–623. doi:10.3168/jds.2009-2573. Walker, S., R. Smith, D. Jones, J. Routly, M. Morris, and H. Dobson. 2010. The effect of a chronic stressor, lameness, on detailed sexual behaviour and hormonal profiles in milk and plasma of dairy cattle. Reprod. Domest. Anim. 45:109–117. doi:10.1111/j.1439-0531.2008.01263.x.  205   Walker, S.L., R.F. Smith, D.N. Jones, J.E. Routly, and H. Dobson. 2008. Chronic stress, hormone profiles and estrus intensity in dairy cattle. Horm. Behav. 53:493–501. doi:10.1016/j.yhbeh.2007.12.003. Walker, W.L., R.L. Nebel, and M.L. McGilliard. 1996. Time of ovulation relative to mounting activity in dairy cattle. J. Dairy Sci. 79:1555–1561. doi:10.3168/jds.S0022-0302(96)76517-7. Walsh, S.W., E.J. Williams, and A.C.O. Evans. 2011. A review of the causes of poor fertility in high milk producing dairy cows. Anim. Reprod. Sci. 123:127–138. doi:10.1016/j.anireprosci.2010.12.001. Weck, J., P.C. Fallest, L.K. Pitt, and M.A. Shupnik. 1998. Differential gonadotropin-releasing hormone stimulation of rat luteinizing hormone subunit gene transcription by calcium influx and mitogen-activated protein kinase-signaling pathways. Mol. Endocrinol. 12:451–457. doi:10.1210/mend.12.3.0070. Weller, J.I., and Y. Folman. 1990. Effects of calf value and reproductive management on optimum days to first breeding. J. Dairy Sci. 73:1318–1326. doi:10.3168/jds.S0022-0302(90)78798-X. Welsh, T.H., L. Zhuang, and A.J.W. Hsueh. 1983. Estrogen augmentation of gonadotropin-stimulated progestin biosynthesis in cultured rat granulosa cells. Endocrinology 112:1916–1924. doi:10.1210/endo-112-6-1916. West, J.W. 2003. Effects of heat-stress on production in dairy cattle. J. Dairy Sci. 86:2131–2144. doi:10.3168/JDS.S0022-0302(03)73803-X. Whay, H.R., A.E. Waterman, A.J. Webster, and J.K. O’Brien. 1998. The influence of lesion type on the duration of hyperalgesia associated with hindlimb lameness in dairy cattle. Vet. J. 156:23–29. Whay, H.R., A.J.F. Webster, and A.E. Waterman-Pearson. 2005. Role of ketoprofen in the modulation of hyperalgesia associated with lameness in dairy cattle. Vet. Rec. 157:729–733. doi:10.1136/vr.157.23.729. Williams, E.J., D.P. Fischer, D.E. Noakes, G.C.W. England, A. Rycroft, H. Dobson, and I.M. Sheldon. 2007. The relationship between uterine pathogen growth density and  206   ovarian function in the postpartum dairy cow. Theriogenology 68:549–559. doi:10.1016/J.THERIOGENOLOGY.2007.04.056. Williamson, N.B., R.S. Morris, D.C. Blood, C.M. Cannon, and P.J. Wright. 1972. A study of oestrous behaviour and oestrus detection methods in a large commercial dairy herd. II. Oestrous signs and behaviour patterns. Vet. Rec. 91:58–62. Wilson, S.J., R.S. Marion, J.N. Spain, D.E. Spiers, D.H. Keisler, and M.C. Lucy. 1998. Effects of controlled heat stress on ovarian function of dairy cattle. 1. Lactating cows. J. Dairy Sci. 81:2124–2131. doi:10.3168/JDS.S0022-0302(98)75788-1. Wiltbank, M., H. Lopez, R. Sartori, S. Sangsritavong, and A. Gümen. 2006. Changes in reproductive physiology of lactating dairy cows due to elevated steroid metabolism. Theriogenology 65:17–29. doi:10.1016/j.theriogenology.2005.10.003. Wiltbank, M.C., A. Gümen, and R. Sartori. 2002. Physiological classification of anovulatory conditions in cattle. Theriogenology 57:21–52. doi:10.1016/S0093-691X(01)00656-2. Wiltbank, M.C., and J.R. Pursley. 2014. The cow as an induced ovulator: Timed AI after synchronization of ovulation. Theriogenology 81:170–185. doi:10.1016/j.theriogenology.2013.09.017. Wiltbank, M.C., A.H. Souza, J.O. Giordano, A.B. Nascimento, J.M. Vasconcelos, M.H.C. Pereira, P.M. Fricke, R.S. Surjus, F.C.S. Zinsly, P.D. Carvalho, R.W. Bender, and R. Sartori. 2012. Positive and negative effects of progesterone during timed AI protocols in lactating dairy cattle. Anim. Reprod. 9:231–241. Wise, M.E., D.V. Armstrong, J.T. Huber, R. Hunter, and F. Wiersma. 1988. Hormonal alterations in the lactating dairy cow in response to thermal stress. J. Dairy Sci. 71:2480–2485. doi:10.3168/jds.S0022-0302(88)79834-3. Woelders, H., T. van der Lende, A. Kommadath, M.F.W. te Pas, M.A. Smits, and L.M.T.E. Kaal. 2014. Central genomic regulation of the expression of oestrous behaviour in dairy cows: a review. Animal 8:754–764. doi:10.1017/S1751731114000342. Wrenn, T.R., J. Bitman, and J.F. Sykes. 1958. Body temperature variations in dairy cattle during the estrous cycle and pregnancy. J. Dairy Sci. 41:1071–1076. doi:10.3168/jds.S0022-0302(58)91053-1.  207   Wubishet, A., C.N. Graves, S.L. Spahr, D.J. Kesler, and R.J. Favero. 1986. Effects of GnRH treatment on superovulatory responses of dairy cows. Theriogenology 25:423–427. doi:10.1016/0093-691X(86)90050-6. Xu, Z.Z., D.J. McKnight, R. Vishwanath, C.J. Pitt, and L.J. Burton. 1998. Estrus detection using radiotelemetry or visual observation and tail painting for dairy cows on pasture. J. Dairy Sci. 81:2890–2896. doi:10.3168/jds.S0022-0302(98)75849-7. Yániz, J.L., P. Santolaria, A. Giribet, and F. López-Gatius. 2006. Factors affecting walking activity at estrus during postpartum period and subsequent fertility in dairy cows. Theriogenology 66:1943–1950. doi:10.1016/j.theriogenology.2006.05.013. Zaffino Heyerhoff, J.C., S.J. LeBlanc, T.J. DeVries, C.G.R. Nash, J. Gibbons, K. Orsel, H.W. Barkema, L. Solano, J. Rushen, A.M. de Passillé, and D.B. Haley. 2014. Prevalence of and factors associated with hock, knee, and neck injuries on dairy cows in freestall housing in Canada. J. Dairy Sci. 97:173–184. doi:10.3168/JDS.2012-6367. Zebari, H.M., S.M. Rutter, and E.C.L. Bleach. 2018. Characterizing changes in activity and feeding behaviour of lactating dairy cows during behavioural and silent oestrus. Appl. Anim. Behav. Sci. 206:12–17. doi:10.1016/J.APPLANIM.2018.06.002.   

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0388204/manifest

Comment

Related Items