POPULATION DYNAMICS AND OVERWINTERING CAPABILITIES OF THE
STABLE FLY STOMOXYS CALCITRANS (L.) (DIPTERA: MUSCIDAE) AND THEIR
PUPAL PARASITOIDS (HYMENOPTERA: PTEROMALIDAE, ICHNEUMONIDAE)
ON DAIRY OPERATIONS IN SOUTHERN MANITOBA
BY
GINA E. C. KARAM
A Thesis submitted to the Faculty of Graduate Studies of
The University of Manitoba
in partial fulfilment of the requirements of the degree of
MASTER OF SCIENCE
Department of Entomology
University of Manitoba
Winnipeg, Manitoba
Copyright © 2020 by Gina E. C. Karam
ABSTRACT
The stable fly Stomoxys calcitrans (L.) (Diptera: Muscidae) has been a severe biting pest in North America for the last 250 years. Severe seasonal infestation is a common occurrence on
Canadian livestock production sites. The projected effects of climate warming within the next 30 years may lead to even larger infestations of longer duration, increasing both economic losses and concerns regarding animal welfare.
To understand the relationship between stable fly population dynamics and the environmental factors that drive population fluctuations, stable fly activity was monitored at three dairy operations near Glenlea, Manitoba for two years (June to October 2017 and May to
October 2018). Adult stable flies were trapped using Coroplast® panels for 24-hours once weekly and changes in adult abundance, sex ratios, and age structure were examined. During both years, adult abundance followed a unimodal trend with the highest abundance recorded in July.
Pupal stable flies were collected in autumn (November 2017 and September 2018) to determine the potential for overwintering survival under laboratory and field conditions. In the lab, pupae were held at 0°C, 2°C, or 5°C and sampled every two weeks from 24 November 2017 to 8 June 2018, and 26 September 2018 to 22 May 2019 to examine adult emergence, pupal development, and parasitism. Two adults emerged from the 1350 pupae observed in 2017 and
256 adults emerged from the 1440 pupae observed in 2018. There were 166 pupal parasitoids recovered in 2017 and 17 pupal parasitoids recovered in 2018. Significantly more adult flies emerged when held at 5°C, and intrapuparial development was observed under 10°C. As evidenced by the results of this study, large numbers of adult stable flies are unlikely to overwinter unsheltered in Manitoba, but pupal parasitoids are able to.
ii Stable fly population dynamics and overwintering survival of stable flies and pupal parasitoids have important implications for producers looking to reduce economic loss and animal welfare concerns, while also preventing environmental impacts from overusing chemical controls.
iii ACKNOWLEDGMENTS
First, thank you to my advisor, Dr. Kateryn Rochon. You have advised me on much more than just science. You helped cultivate within me new understandings, different perspectives, and a more in-depth way of thinking about the natural world that surrounds me. Words just cannot express how grateful I am that you let me learn in my own way and make my own mistakes yet you were available to help me when things got too difficult. There were times when
I honestly did not know if I was capable of finishing this project, but you were always there to talk me out of my self-doubt. Your patience and consistency truly helped to bring out the best in me, even if I could not see the best in myself at times. I genuinely appreciate all the time and effort you put into shaping me into a better scientist.
I thank my committee members, Dr. Emma McGeough and Dr. Mahmood Iranpour, for your guidance during this project. I greatly appreciate all of your assistance and valuable input during our time together, as well as your helpful recommendations on improving this thesis.
Jordan Bannerman aka “The Stats/R Wizard”, thank you for your in-depth and patient assistance with R and statistics, which I can actually say I have come to enjoy thanks to your tutelage.
To my lab mates Phoenix Nakagawa, Lilianne Tran, Madelin Pineau, Alyssa Coopland, and Evan Warren: thank you for your help with processing one hundred thousand stable flies and numerous parasitoids. You all made my project so much easier and my lab work so much fun!
Thank you to technicians Dave Holder and Lisa Babey for all of your assistance during last-minute fieldwork, designing and creating lab equipment, and going above and beyond for myself and all the students in the Entomology department. We are lucky to have your creativity and help in our Department.
iv This project could not have happened without Mike Hummel and Tracy Gilson at Glenlea
Research Station, Ron and the Van Walleghem family, and Reg and the Wiens Family, as well as all your precious Holstein girls.
Thank you Timi Ojo from Manitoba Agriculture for providing me with accurate weather station data reports from the Kelburn Farm.
Where would I be without Rachel Cherka, Arielle Nagy, Christie Lavallee, Seriki
Muhammad, Conny Willing, Brian Moons, Gabrielle Grenier and all of my Entomology and
Biology friends. Thanks to your consistent support and encouragement, I made it! You are all such genuine people that I am lucky to surround myself with. You guys kept me working hard, laughing and upbeat, thinking critically, and buzzed on caffeine 24/7. You have helped me stay as sane as possible during these last few years and I am so thankful to have you all in my life.
To my huge and wonderful family: Mom, Dad, Karl, Heather, Erik, Sydney, Shaun,
Bailey, Gran, Gramps, Mama, Baba, Sara, Nabil, Leila, Nora, and all the Salos, Campbells,
Karams, and Bouabdillahs. THANK YOU SO MUCH! Your love, understanding, and kindness is exactly what I needed to complete this project. I could never have done this without knowing all of you have my back when “stuff” hits the fan and I feel like quitting. I am so lucky to have such a tight-knit family like I do. You complete my circle and I am so grateful for your encouragement and support, even if you are still not sure what I have been working on this whole time.
And finally, to the love of my life, Yassin: you are my rock, my shining light, my listening ear, and my inspiration to better myself each day. I am so unbelievably grateful for your seemingly endless patience and support during all these years. Your passion, wisdom, and kindness are unmatched. I am thankful for you in every single way and I love you.
v For generously providing funding assistance, I thank the University of Manitoba
Graduate Fellowship, the Manitoba Graduate Research Grant Program, the Prairie Improvement
Network Graduate Fellowship and the Canadian Dairy Commission Scholarship.
vi TABLE OF CONTENTS
ABSTRACT ...... ii
ACKNOWLEDGMENTS ...... iv
TABLE OF CONTENTS ...... vii
LIST OF TABLES ...... ix
LIST OF FIGURES ...... x
CHAPTER 1: REVIEW OF PERTINENT LITERATURE...... 13
Introduction ...... 13
Stable Fly Population Dynamics and Dispersal ...... 25
Biology of Parasitoid Wasps and Use as Biological Control Agents ...... 30
CHAPTER 2: ADULT STABLE FLY STOMOXYS CALCITRANS (L.) (DIPTERA: MUSCIDAE) SEASONAL POPULATION DYNAMICS ON DAIRY FARMS IN SOUTHERN MANITOBA...... 36
Abstract ...... 36
Introduction ...... 36
Materials and Methods ...... 38
Results ...... 49
Discussion ...... 55
CHAPTER 3: WINTER SURVIVAL OF STABLE FLY STOMOXYS CALCITRANS (L.) (DIPTERA: MUSCIDAE) PUPAE AND THEIR PUPAL PARASITOIDS (HYMENOPTERA: ICHNEUMONIDAE, PTEROMALIDAE) UNDER LABORATORY AND FIELD CONDITIONS IN SOUTHERN MANITOBA...... 76
Abstract ...... 76
Introduction ...... 76
vii Materials and Methods ...... 79
Results ...... 86
Discussion ...... 90
CHAPTER 4: GENERAL DISCUSSION ...... 111
Insights ...... 111
Limitations ...... 115
Future Directions ...... 119
LITERATURE CITED ...... 122
APPENDICES ...... 144
Appendix A. Cattle enclosures at three dairy farms near Glenlea, Manitoba...... 144
Appendix B. Design of Coroplast® sticky traps...... 147
Appendix C. Correlation matrices for examining strength of correlation among
environmental variables ...... 148
Appendix D. Cross-correlation plots evaluating the relationship between trapped adult
stable fly abundance and environmental variables ...... 150
viii LIST OF TABLES
Table 2.1. Cross-correlation factors of environmental variables by site and year...... 63
Table 2.2. Negative binomial regression analysis of the relationship between the number of adult flies captured on sticky traps and significant environmental conditions on three southern Manitoba dairy farms in 2017 and 2018...... 64
Table 2.3. Monthly measurements of significant environmental conditions from St. Adolphe weather station in 2017 and 2018...... 65
Table 2.4. Monthly subsample totals of adult male and female stable flies removed from sticky traps on three southern Manitoba dairy farms in 2017 and 2018...... 66
Table 2.5. Sweep net collected adult stable fly counts from three southern Manitoba dairy farms in 2018...... 66
Table 2.6. Adult female stable fly parity from three southern Manitoba dairy farms in 2017 and 2018...... 67
Table 3.1. Emergence of adult stable flies by sex and incubation temperature from 2018 laboratory overwintered pupae...... 97
Table 3.2. Adult stable fly emergence from laboratory overwintered stable fly pupae held in incubators from October 2018 to May 2019...... 98
Table 3.3. Stable fly intrapuparial development stages observed from unemerged pupae taken from three incubator temperatures during 2017 and 2018 laboratory overwintering experiments...... 99
Table 3.4. Number of adult parasitoids by species collected from laboratory overwintered stable fly pupae from 2017-2018 and 2018-2019 experiments...... 100
Table 3.5. Proportion of stable fly pupae at different stages of development during laboratory and field overwintering experiments in 2017...... 101
Table 3.6. Dissections of stable fly pupae overwintered on three dairy farms in southern Manitoba near Winnipeg. Pupae were buried on 17 November 2017 and retrieved from their sites on 16 June 2018...... 102
ix LIST OF FIGURES
Figure 2.1. Location of Manitoba Agriculture weather station and the three dairy farms used as stable fly population research sites in 2017 and 2018...... 68
Figure 2.2. Total number (± standard error) of adult stable flies captured on sticky traps per sampling date. Six traps were monitored for one 24-hour span every week on three dairy farms in southern Manitoba in 2017 (a) and 2018 (b)...... 69
Figure 2.3. Mean abundance (± standard error) of adult stable flies captured weekly on three southern Manitoba dairy farms in 2017 and 2018...... 70
Figure 2.4. Mean number (± standard error) of subsampled adult female (a) and male (b) stable flies per trap per week in 2017 and 2018...... 71
Figure 2.5. Mean (± standard error) weighted ages of female stable flies trapped across all sites during the trapping season for 2017 (a) and 2018 (b)...... 72
Figure 2.6. Mean (± standard error) weighted ages of female stable flies trapped at Glenlea Research Station dairy farm during the trapping season for 2017 (a) and 2018 (b)...... 73
Figure 2.7. Mean (± standard error) weighted ages of female stable flies trapped at Red River Holsteins dairy farm during the trapping season for 2017 (a) and 2018 (b)...... 74
Figure 2.8. Mean (± standard error) weighted ages of female stable flies trapped at Vancrest Holsteins dairy farm during the trapping season for 2017 (a) and 2018 (b)...... 75
Figure 3.1a. Dissections of laboratory overwintered stable fly pupae in 2017. Developmental stages were categorized according to Fraenkel and Bhaskaran (1973). .. 103
Figure 3.1b. Dissections of stable fly pupae overwintered in an incubator set to 0°C from 24 November 2017 to 8 June 2018...... 104
Figure 3.1c. Dissections of stable fly pupae overwintered in an incubator set to 2°C in the University of Manitoba’s Veterinary Entomology laboratory from 24 November 2017 to 8 June 2018...... 105
x Figure 3.1d. Dissections of stable fly pupae overwintered in an incubator set to 5°C in the University of Manitoba’s Veterinary Entomology laboratory from 24 November 2017 to 8 June 2018...... 106
Figure 3.2a. Dissections of laboratory overwintered stable fly pupae in 2018. Developmental stages were categorized according to Fraenkel and Bhaskaran (1973). .. 107
Figure 3.2b. Dissections of stable fly pupae overwintered in an incubator initially set to 0°C in the University of Manitoba’s Veterinary Entomology laboratory from 26 September 2018 to 22 May 2019...... 108
Figure 3.2c. Dissections of stable fly pupae overwintered in an incubator set to 2°C in the University of Manitoba’s Veterinary Entomology laboratory from 26 September 2018 to 22 May 2019...... 109
Figure 3.2d. Dissections of stable fly pupae overwintered in an incubator set to 5°C in the University of Manitoba’s Veterinary Entomology laboratory from 26 September 2018 to 22 May 2019...... 110
Figure A.1. Outdoor Holstein heifer enclosure at the Glenlea Research Station dairy farm near Glenlea, Manitoba...... 144
Figure A.2. Placement of a Coroplast® sticky trap relative to outdoor Holstein heifer enclosure on Red River Holsteins dairy farm near Glenlea, Manitoba...... 145
Figure A.3. Holstein heifer enclosure at Vancrest Holsteins dairy farm near Glenlea, Manitoba...... 146
Figure B.1. Design of Coroplast® sticky traps for capturing adult stable flies...... 147
Figure C.1. Correlation matrix of environmental variables from 2017...... 148
Figure C.2. Correlation matrix of environmental variables from 2018...... 149
Figure D.1. Cross-correlation plot evaluating the relationship between adult stable fly abundance and average air temperature at Glenlea Research Station in 2017...... 150
Figure D.2. Cross-correlation plot evaluating the relationship between adult stable fly abundance and average air temperature at Red River Holsteins in 2017...... 151
xi Figure D.3. Cross-correlation plot evaluating the relationship between adult stable fly abundance and average air temperature at Vancrest Holsteins in 2017...... 152
xii CHAPTER 1: REVIEW OF PERTINENT LITERATURE
INTRODUCTION
Taxonomy
The stable fly Stomoxys calcitrans (L). (Diptera: Muscidae) cemented its place in veterinary entomology over a century ago as one of the most important livestock pests in the world (Bishopp 1913; Brues 1913; Taylor et al. 2020). Stable flies are taxonomically classified within the family Muscidae and subfamily Stomoxyinae (Zumpt 1973; Malaithong et al. 2019) and there are 18 species within the genus Stomoxys (Zumpt 1973).
Arrival in North America
Members of the genus Stomoxys are found globally today due to their adaptability and following the movement of their hosts (Brues 1913; Bishopp 1913). Of the 18 Stomoxys species,
Stomoxys calcitrans is the only species that is found in North America (Marquez et al. 2007).
Stable flies likely arrived in North America during the 1700s, when European settlers brought their livestock with them across the Atlantic Ocean during colonization of the Eastern coast
(Brues 1913). Thomas Jefferson wrote that the American Declaration of Independence was quickly accepted by Congress in 1776 due to the enormous numbers of stable flies biting attendees at its signing (Brues 1913). His report suggests that stable flies were already well- established biting pests in parts of North America almost 250 years ago (Brues 1913).
Feeding Behaviour
The Stomoxys genus is unique among muscid flies, in that it is comprised almost entirely blood-feeders (Malaithong et al. 2019). Typical adults of the 4000 Muscidae species have sponging mouthparts and cannot bite; however, approximately 50 species within Stomoxyinae have a piercing proboscis (Crosskey 1993) comprised of the sclerotized labrum, labium, and
13 hypopharynx (Brain 1912). The three hardened mouthparts and rows of serrations on another mouthpart, the labellum, enable individuals to easily pierce the skin of their host and ingest blood via the hypopharynx and labrum (Brain 1912; Zumpt 1973).
Both sexes of stable flies require blood meals, meaning every adult in a stable fly population seeks hosts and blood-feeds, unlike other members of the order Diptera, such as mosquitoes (family Culicidae) where only females take blood meals (Cortinas and Jones 2006).
Stable flies are telmophage, or “pool feeders”, meaning they cut the skin, vessels, and capillaries of their host and siphon the blood that pools in the wound (Cortinas and Jones 2006; Wikel et al.
2017). This feeding method is painful to the host because stable flies lack anesthetic compounds in their saliva (Cortinas and Jones 2006; Baldacchino et al. 2013). Therefore, adults are often interrupted during feeding and require multiple attempts to complete a full blood meal (James and Harwood 1969; Foil and Hogsette 1994; Baldacchino et al. 2013).
Stable flies average two full blood meals per day (Cortinas and Jones 2006). A blood meal is 11–15 μL (Schowalter and Klowden 1979; Baldacchino et al. 2013) and takes 2–4 minutes (Salem et al. 2012) to complete. Stable flies survive if fed non-blood sources like flower nectar or sugar-water/honey-water solutions (Müller et al. 2012; Salem et al. 2012; Baldacchino et al. 2013), but without blood, they are unable to reproduce (Anderson 1978; Moobola and
Cupp 1978; Salem et al. 2012), male mating drive decreases (Anderson 1978) and females will not undergo ovarian development (Kuzina 1942). Females must blood-feed at least 3–5 times for ovarian development to begin, and must also take three additional blood meals for every ovarian reproductive cycle after initial ovarian development (Chia et al. 1982). Overall longevity also significantly decreases in individuals fed a sugar-water only diet (Moobola and Cupp 1978).
14 Stable flies are generalist parasites and have been observed feeding on many different host species (e.g., humans, pets, livestock, wildlife, and even pelicans and other large birds
(Friesen and Johnson 2013). Stable flies feed on humans and animals at beaches where sea grasses have washed up and have begun fermenting in the heat (Hogsette et al. 1987). Stable flies are colloquially named “dog flies” in the southern United States, as their penchant for biting dogs’ ears is severe enough to damage and permanently disfigure the sensitive pinnae tissue
(Hogsette et al. 1987). On large-bodied hosts, stable flies show preference for lower parts of the body and aggregate from the hoof or paw up to the hock (including the cannon, pastern, and fetlock) to feed because the short sparse fur on the legs provides easier access to the skin and vessels below (Foil and Hogsette 1994). The lower part of the front legs is also a safer location to feed because the host’s tail is unable to sweep the biting flies away (Foil and Hogsette 1994).
Although they are generalists, stable flies prefer large-bodied, herbivorous mammals as their hosts, especially domesticated ruminant (Artiodactyla: Ruminantia) species like cattle, bison, and goats (Anderson and Tempelis 1970; Hogsette et al. 1987).
Pest Status and Impacts on Agriculture
Stable flies have been heavily researched (Kneeland et al. 2012.) as they are problematic for humans and animals alike, in both commercial and recreational settings. Stable flies are found in especially large numbers around areas with livestock, where hosts and oviposition sites alike are plentiful. They are also found on beaches and lawns (Hogsette and Ruff 1985), and at zoos and parks (Hogsette and Ose 2017) where their persistent biting activity causes pain and distress to animals and humans alike (Cook et al. 1999).
Large stable fly populations are detrimental to producers and their animals, from both welfare and economic perspectives. Stable fly biting pressure forces the hosts to resort to
15 avoidance behaviours like aggregating, running, stamping feet, skin shaking (panniculus reflex), swishing tails, and shaking heads (Campbell et al. 1987). Performing these actions means energy is diverted away from feeding and growing and interferes with resting. In animal production, there are significant decreases in milk and meat yields (Campbell et al. 1987). Aggregating can increase the chance for heat stress (Wieman et al. 1992) and running can have the potential for severe injury (fractures, lacerations, etc.) and death of animals in some cases (Taylor et al. 2017).
Estimated annual economic losses from avoidance behaviours and injury caused from stable fly biting pressure exceeds $2.2 billion on beef and dairy farms in the United States (Taylor et al.
2012) and the only Canadian estimate comes from confined cattle feeding operations [feedlots]
20 years ago, where losses exceeded $25 million in the province of Alberta alone (Colwell
2000). There are no recent estimates on economic losses in Canada.
In North America, large-bodied livestock such as cattle and horses, are severely afflicted by stable fly biting pressure. Outdoor barns hold large numbers of host animals in small spaces for easy access and build-up of ideal oviposition substrate (Anderson and Tempelis 1970;
Hogsette et al. 1987). Specific host “choice” may be regionally based, in that stable flies from various regions of the world have access to different species of large herbivores from which to blood-feed (Warnes and Finlayson 1987). For example, in Egypt, stable flies were observed feeding on donkeys preferably, followed by horses (Hafez & Gamal-Eddin 1959), and stable flies in Chad bit and fed from camels most often (Gruvil & Balis 1966). When presented with a choice of host blood in laboratory experiments, both female and male stable flies preferred cattle over horse or chicken blood (Friesen and Johnson 2012). This preference is likely due to a significant increase in longevity observed for both sexes (Moobola and Cupp 1978), increased male mating drive and increased lifetime egg production in females (Friesen and Johnson 2012).
16 Stable flies have been severe pests of cattle for many decades in confinement settings like feedlots or dairy operations where many animals are in close proximity (Bruce and Decker 1958;
Campbell et al. 1977; Campbell et al. 1987; Wieman et al. 1992). Within the last few decades, stable flies have become a severe pest of pastured cattle and free-range cattle in the Northern
Great Plains ecoregion of the USA and Canada (Campbell et al. 2001). With the increasing popularity of round hay bales for winter feeding, the manure and straw mix to create an ideal stable fly oviposition and development substrate (Hall et al. 1982; Campbell et al. 2001; Broce et al. 2005).
During their biting activity, stable flies can spread disease and illness via mechanical transmission of pathogens. The process of mechanical pathogen transmission involves a stable fly biting an infected host and while feeding, infected blood remains on its proboscis or in its crop (Baldacchino et al. 2013). As the bite is painful, the fly is often interrupted multiple times during feeding and may fly to a new host to continue its feeding (Baldacchino et al. 2013). When the fly begins feeding on the new host, the infected blood is transferred into the new host either through remnants of blood on the proboscis or through regurgitation into the bloodstream while feeding (Baldacchino et al. 2013).
Because stable flies do not remain on a single host to feed, there is an increased risk of pathogen transmission among individuals in a herd. Stomoxys species have the potential to mechanically transmit bacteria, viruses, protozoa, and helminths (Baldacchino et al. 2013;
Baleba et al. 2020). However, in most cases, their implication as primary vectors has not been demonstrated. They have nonetheless been associated with the transmission of equine infectious anemia virus (Retroviridae), lumpy skin disease virus (Poxviridae), Apicomplexan protozoan
Besnoitia besnoiti (bovine besnoitiosis) (Foil et al. 1983; Baldacchino et al. 2013; Sharif et al.
17 2019; Sohier et al. 2019). Stable flies also serve as intermediate host to Habronema microstoma, a stomach worm of horses (Traversa et al. 2008).
Reproduction
Stable flies wait for potential mates to pass by while resting on light-coloured (especially white) vertical surfaces, including painted white barns, fences, and white vehicles parked near host enclosures (Buschman and Patterson 1981). These surfaces where flies gather and wait in large numbers for mates are called either "waiting stations" (Downes 1969) or "encounter sites"
(Parker 1978). Males and females are both observed at waiting stations, but when males are established at a waiting station, they pester the females to the point that females begin to situate themselves farther away from the males on less preferable perches, such as unpainted or dark- coloured fences and wooden objects for reprieve (Buschman and Patterson 1981). Males are often seen at waiting stations with their heads pointed to the ground and rest about 50–100 cm above the ground surface (Buschman and Patterson 1981). When an insect passes a male in a waiting site, he flies out to inspect it, then returns to his same position, suggesting a territorial nature to the behaviour (Buschman and Patterson 1981). The inspection behavior has been recorded in other Diptera species and is known to be related to mating behaviour (Downes 1969;
Parker 1978).
Stable flies develop from egg to adult within two weeks when under ideal conditions where larvae are in good quality substrate and temperatures are high (~30°C) (Florez-Cuadros et al. 2019). Adults mate about 3–5 days after emerging from their pupae and females start laying eggs about 2–3 days after mating (Foil and Hogsette 1994). One female can lay over 100 eggs in as little as eight days after emergence from the pupal stage (Foil and Hogsette 1994). Female stable flies, after 3–5 blood meals, undergo maturation and ovarian development (Chia et al.
18 1982). Stages of ovarian development and the number of times a female has laid eggs, or the female’s parity, can be assessed using a staining technique with neutral red dye (Scholl 1980).
Virgin females are considered nulliparous, meaning they have not laid eggs, uniparous means one oviposition cycle has occurred, biparous means two oviposition cycles, and so on. Parity can be used as a proxy method for aging females (Scholl 1980).
Female stable flies lay their eggs in oviposition substrates that are primarily comprised of moist, decaying and fermenting vegetation (Todd 1964; Foil and Hogsette 1994; Talley et al.
2009; Friesen et al. 2016), including animal feed grasses and legumes like timothy hay and alfalfa, wet animal bedding such as straw and wood shavings, sugarcane cuttings, seaweed and kelp, mowed grass piles, and kitchen vegetable compost (Skoda and Thomas 1993; Foil and
Hogsette 1994; Broce et al. 2005; Cook et al. 2018).
Stable fly oviposition is often associated with animal manure in confined animal operations like dairies or feedlots, but even so, there are regional differences in female preference for an oviposition substrate. In Florida, decomposing and fermenting vegetable matter is the critical, primary oviposition habitat and manure is only important when it has been deposited on hay or straw, increasing the moisture content of the substrate (Hogsette et al. 1987).
Machtinger et al. (2016) found that stable fly pupae were most numerous in a hay substrate mixed with horse manure in Florida, and also noted that increased sanitation efforts reduced adult abundance on site. Over time, if not removed, the stratified layers of aging manure-hay become an ideal habitat for larval development and large numbers of stable fly larvae and pupae are found in the mixed manure-hay substrate (Hogsette et al. 1987; Broce and Haas 1999).
Manure contains moisture and diverse bacterial colonies which are essential for development as larval stable flies feed on the bacteria within the substrate (Lysyk et al. 1999; Rochon et al.
19 2005; Talley et al. 2009). Friesen et al. (2016) recorded large numbers of stable fly larvae in alkaline substrate with elevated levels of moisture, ammonium, and temperature near
23°C. Larvae have not been observed developing in pure manure in Florida, but in California, uncontaminated manure is an extremely productive larval substrate (Meyer and Shultz 1990).
Significantly more adult stable flies emerged from manure alone than manure-vegetation mixtures, even with high competition from house flies (Musca domestica) for the pure manure sites (Meyer and Shultz 1990). It is unknown how long these sites remain appealing to female stable flies for oviposition and good enough quality for the larvae to survive.
On pastures, stable flies can use dung pats that have begun to dry out in the sun and heat, creating a hard crust on the outside of the pat that shelters the developing larvae from desiccation and temperature extremes (Skoda and Thomas 1993; Foil and Hogsette 1994). Stable flies on pastures prefer manure that is at least two weeks old as the ideal age for oviposition (Broce and
Haas 1999). Other dipteran pest species like house flies and horn flies (Haematobia irritans (L.)) use very fresh, undisturbed manure for oviposition (Kunz et al. 1970; Broce and Haas 1999;
Kuramochi 2000), but stable flies have not been found using fresh manure whatsoever on pastures (Hogsette et al. 1987). Waste hay and feed that collects along the edges of round bale feeders or ring feeders in pastures has become a highly productive larval development site
(Broce et al. 2005). Because the ring feeders are not typically moved around on the pasture, the accumulation and stratification of feed, manure, and urine during the winter months, along with spring precipitation creates a damp, warm environment with diverse bacterial colonies; the ideal conditions required for larval development (Hogsette et al. 1987; Broce et al. 2005; Talley et al.
2009).
20 Development and Longevity
The stable fly life cycle is holometabolous, meaning they undergo complete metamorphosis and have four distinct life stages: egg, larva (with three instars), pupa within a puparium, and adult. As ectotherms, stable fly metabolic and developmental processes are strictly influenced by external temperatures. Stable flies are a globally distributed species, therefore accurately assessing the development time among different geographic regions requires a record of degree-days, rather than a measure of calendar time. Stable flies require 224.7 accumulated degree-days at 15°C to complete their life cycle from egg to adult (Gilles et al.
2005), or 232 accumulated degree days at 10°C (Lysyk 1993). Lysyk (1998) recorded an increased speed of development in lab settings up to 31°C, at which point there were diminished returns, with higher temperatures leading to lethal effects. Once a female has located a suitable oviposition site, she begins to lay eggs, which is the first stage of the life cycle. In her lifetime, which is approximately two weeks in the wild (Killough and McKinstry 1965) and up to 2-3 months in laboratory settings (James and Harwood 1969), a female will lay between 60-800 eggs in four or five oviposition events (Foil and Hogsette 1994). Hatching is temperature-dependent; approximately 97% of the variation in duration of development from egg to adult is explained by the temperature in the developmental substrate alone (Florez-Cuadros et al. 2019). Eggs can hatch within one week within 26°C substrate, and in as little as 12 hours at temperatures around
30°C (Foil and Hogsette 1994; Florez-Cuadros et al. 2019). Once hatched, the larvae begin to feed on the bacterial communities present in their habitat (Foil and Hogsette 1994). Larvae undergo three ecdysis (exoskeleton/cuticle moults) events, the duration of which is also temperature and diet dependent; higher temperatures and higher quality diets reduce the time it takes from first to third instar (Florez-Cuadros et al. 2019). Once at the third instar, they prepare
21 for pupation by moving away from the moist areas and into a drier area of the larval substrate
(Foil and Hogsette 1994). Third instar stable fly larvae produce a puparium, a sclerotized, barrel- shaped outer layer of the larval cuticle which encases and protects the living pupa within
(Roseland et al. 1985). Adults emerge from puparia within 1–3 weeks (James and Harwood
1969; Romero et al. 2006; Showler and Osbrink 2015) and begin the life cycle over again, with newly eclosed adults mating about three days after emergence (Foil and Hogsette 1994).
Stable Fly Population Control Methods
Stable flies complete their life cycle very quickly and populations can reach substantial numbers in a matter of weeks. To control stable fly populations, producers can develop and implement an Integrated Pest Management (IPM) program to prevent economic losses and improve animal welfare. Integrated Pest Management programs follow a multi-pronged approach of selecting and integrating a variety of pest control actions, including preventative measures, monitoring pest populations, and making control decisions on the basis of economic loss. The concept revolves around a level of loss or harm that is deemed acceptable, and beyond which the losses are equal to the cost of control measures (i.e., preventing the losses) that point is the economic injury level (EIL). Economic injury levels are difficult to establish as they vary depending on host species, market values, and the cost of the control options available (Axtell
1981). Prevention should be the initial step to inhibit pest populations growth. For example, eliminating potential larval developmental substrate by removing waste and soiled bedding, and keeping silage, hay, straw, and other vegetation dry. As control measures may take some time to take effect, producers should act before pest populations reach the economic injury level, at a point where actions will maintain the losses at an acceptable level – that point is the economic threshold (ET). Consequently, monitoring pest populations is an essential component of IPM.
22 Stable fly bites evoke strong responses from their host (Colwell and Kavaliers 1992) and even a few flies will lead to significant losses; therefore, the ET for stable flies in cattle is established at five stable flies per front leg (Berry et al. 1983, Campbell et al. 1987, 1989). Once this threshold is reached, control measures should be applied to suppress the fly population; control measures can come in the form of mechanical, biological, or chemical options.
There is a considerable body of research on mechanical, chemical, and biological control methods for reducing stable fly populations in livestock production systems. Some methods include traps in and around livestock pens, applying chemical insecticides, and releasing stable fly parasitoids on sites with stable fly infestations. Sanitation is an effective mechanical method of reducing populations. It involves the removal of dirty bedding, waste feed, and manure piles, all of which are ideal larval development sites (Berkebile et al. 1994; Taylor and Berkebile
2011). Additionally, sanitation around ring and cone feeder stations is imperative for reducing stable fly populations on feedlots and in pastures. Field studies have shown that these feeding sites possess critical characteristics that gravid females are drawn to when ready to oviposit: neutral pH levels between 6.5 and 7.0, 65–75% moisture content, a diverse fecal coliform bacterial community, and an optimal substrate depth of 15 to 22 cm composed of an equal ratio of hay-manure (Broce et al. 2005; Talley et al. 2009; Wienhold and Taylor 2012).
Insecticide resistance to commonly used chemicals, including organochlorines and pyrethroids, has been observed in stable fly populations from Kansas (Cilek and Greene 1994),
California (Mullens et al. 1995), Nebraska (Marçon et al. 1997), Florida (Pitzer et al. 2010),
Brazil (Barros et al. 2019), and Germany (Reissert-Oppermann et al. 2019). Insecticide resistance is especially concerning in insect populations with high reproductive rates and quick development times. In light of these findings, mechanical controls are preferred in IPM
23 programs. Producers tend to use insecticides at will, neglecting to calculate the critical information necessary for an ecologically sound IPM program. There has not been research on insecticide resistance in stable fly populations in Canada, but it is expected since stable flies are panmictic and genetic makeup is not truly distinct from one region to another, especially within geographic continents (Kneeland et al. 2016).
Broad-spectrum insecticides are not only overused, they also harm native stable fly predators and parasitoids and pose health risks to humans during applications. Insecticides impact the environment when applied incorrectly, contaminating waterways and bioaccumulating within food webs. Further, the location of insecticide application often happens along the dorsal line or near the ears which poses a problem for effectiveness. Stable flies preferentially bite the legs of large bodied hosts which often stand in manure, urine, and mud in their stalls, or walk in wet grass on pasture, preventing insecticides from persisting where the flies bite most often. If cattle are provided access to water and fly pressure is high, they will stand in the ponds for temporary relief and will wash off the treatment in the process.
Dairy producers have further limited options when it comes to chemical control. They must be cautious of which chemicals are registered for use on lactating dairy cows considering the potential for milk contamination. Sanitation of outdoor confined animal operations allows for removal and disposal of larval stable flies, which are not bothersome to animals or costly to producers and are safe options for pregnant animals. Supplementing mechanical control with biological control could lead to acceptable fly population sizes without risking the harmful effects of chemicals on the environment, people, or animals (e.g., Pickens and Miller 1987;
Lazarus et al. 1989; Schmidtmann 1991; Geden et al. 1992; Miller et al. 1993a; Miller et al.
1993b; Thomas et al. 1996; Kaufman et al. 2005). Control methods on dairy farms are limited,
24 but an integrated approach using a combination of chemical, cultural/mechanical, and biological controls is rather effective.
Overall, controlling stable fly populations has proven to be a challenging task. Adults can migrate long distances passively by means of wind currents and storms (Hogsette and Ruff 1985) and have flown 29.1 km in flight mill studies (Bailey et al. 1973). With these life history traits, stable fly populations can reach economic injury levels on beef heifers in approximately three weeks (Campbell et al. 1987; Showler and Osbrink 2015) and remain a severe pest for the entire duration of the temperate summer season. One way to combat animal welfare concerns and economic losses from stable fly biting activity is to further understand stable fly phenology, such as knowing when flies are emerging, when exponential population growth occurs, and implementing controls between the times of emergence and exponential growth. To create a stronger IPM program, increasing our knowledge of stable fly population dynamics is necessary.
STABLE FLY POPULATION DYNAMICS AND DISPERSAL
Importance of Climatic Differences Between Temperate and Tropical Zones
Stable flies are globally distributed, therefore differences in climate and weather patterns across geographic regions must be considered when assessing stable fly phenology. As with many insect pest populations, stable fly abundance fluctuates within a geographic region over the course of a year. In tropical and subtropical zones, stable flies can remain a year-round outdoor pest due to mild climates with temperatures consistently above 10°C (Beresford and Sutcliffe
2009b; Showler and Osbrink 2015). Other factors such as precipitation and development site availability may be more insightful when examining stable fly populations in these regions. For example, drought conditions in California significantly reduced stable fly counts during the first
25 year, but normal rainfall the following spring led to much higher fly counts the second year
(Mullens and Meyer 1987). This is expected, as stable fly populations are strongly influenced by temperature and accumulated precipitation in the two to four weeks preceding observed population growth (Taylor et al. 2017), where warm temperatures and high moisture levels lead to large populations. Populations grow rapidly when developmental substrate becomes available.
Females are drawn to the new substrate and populations infest surrounding areas where they were not established before. During harvesting season on pineapple and sugarcane plantations, the residues left behind are chopped into the soil and start to ferment, providing superior quality developmental sites (Solórzano et al. 2015; Jelvez Serra et al. 2017). Stable fly outbreaks are commonly observed on plantations and the neighbouring areas during harvesting, with fly population estimates exceeding 700 stable flies per cow (Solórzano et al. 2015). When the harvest ends, the fly populations decline, but the residues have the potential to be active development sites as long as the moisture and temperature in the residues remain high enough.
Stable Flies as a Seasonal Concern in Temperate Zones
Stable flies are considered a seasonal concern in temperate regions including Canada
(Lysyk 1993; Beresford and Sutcliffe 2009a; Beresford and Sutcliffe 2012). Populations tend to fluctuate over the course of a year, where fly biting pressure on livestock is low in early spring and late fall due to cool temperatures, but becomes severe in mid-summer with high temperatures (e.g., Khumalo and Galloway 1996; Lysyk 1993; Broce et al. 2005; Beresford and
Sutcliffe 2012). Stable fly populations are largest and therefore most damaging during the summer months, with some regional variation. In Canada and the United States, adult emergence is observed in April, May, and June, with population peaks usually in June, July, and August.
The farther north stable fly populations are located, the later first emergence and subsequent
26 population peaks will occur. For example, in southern Alberta, first emergence is in May with peaks in August and September (Lysyk 1993), whereas in California, first emergence is in April with the peak in June (Mullens and Meyer 1987). As summer transitions into fall and winter, stable fly activity begins to decline. Adult stable flies are limited by cold temperatures as they lack a diapause stage in their life cycle (Berry et al. 1978; Jones and Kunz 1997). As temperatures drop, feeding activity slows then ceases altogether at approximately 12°C (Bailey and Meifert 1973), typically reducing stable fly populations to zero by the end of October (Lysyk
1993).
Effects of Climate Change on Stable Fly Populations
Stable flies may become a more permanent, year-round pest in northern countries due to the effects of climate change. Stable fly populations are more likely to grow larger due to warming temperatures and also survive for a longer season (Gilles et al. 2008). Stable flies are capable fliers (Hogsette et al. 1987) and may quickly infest newly available areas.
In the province of Manitoba, there are no environmental buffers such as mountain ranges or ocean coasts, and stable flies must cope with temperature extremes as the seasons change. The disparity between July (mid-summer) and January (mid-winter) air temperatures in the province is approximately 36°C not including the windchill factor (Environment and Climate Change
Canada 2019). Temperature extremes of +40.6°C to -45.0°C have been recorded in Manitoba, creating a temperature difference of 85°C (Environment and Climate Change Canada 2019).
Typically, Manitoba’s ambient summer temperatures are warm (23°C to 26°C average;
Environment and Climate Change Canada 2019), but do not surpass the adult stable fly upper lethal temperature limit of 38.3°C (Berry and Kunz 1977; Berry and Kunz 1978; Lysyk 1998;
Skovgård and Nachman 2017). Most of the winter is lethally cold, with over 100 days of the year
27 at temperatures below -10°C, not including the windchill factor (Environment and Climate
Change Canada 2019).
Stable Fly Overwintering
Stable flies have been found overwintering as larvae, pupae, and adults in different areas of North America (Berkebile et al. 1994; Beresford and Sutcliffe 2009a). Because there is no diapause stage in the stable fly life cycle, the immature stages need shelter that is cool enough to slow development, yet warm enough to prevent freezing (Berry et al. 1978). Beresford and
Sutcliffe (2009) suggest that northern temperate populations are re-populated by a few locally overwintered individuals, and evidence of local overwintering in North America has been observed at southern latitudes where winters tend to be very mild and short in duration, such as
Texas (Bishopp 1913) and Florida (Simmons 1944). In central and northern regions of North
America, Beresford and Sutcliffe (2009) found evidence of adult flies overwintering in animal confinement buildings in southern Ontario, yet Berkebile et al. (1994) found no overwintering adults indoors in southern Nebraska.
An alternative hypothesis is that populations are enhanced by individuals passively transported via weather fronts (Hogsette and Ruff 1985). A combination of physical and environmental conditions, including powerful flight physiology and high winds, can transport adult flies far from their emergence sites. In Florida, adult flies were dusted with fluorescent dyes and after a storm, recovered on traps located 225 km away (Hogsette and Ruff 1985). The majority of the migrating stable flies were females, which is expected of migrating flying insects
(Johnson 1966; Hogsette and Ruff 1985). Although there are two hypotheses each with evidence to support them, the extent to which locally overwintered or migrating individuals enhance populations in northern temperate regions in spring is still unknown.
28 Status on Global Stable Fly Population Dynamics Research
Adult population dynamics studies have been conducted in North America and around the world. Studies have occurred in Manitoba (Khumalo and Galloway 1996), Alberta (Lysyk
1993), Nebraska (Thomas et al. 1990; Taylor et al. 2017), California (Mullens and Meyer 1987),
Florida (Hogsette and Ruff 1985; Pitzer et al. 2011; Machtinger et al. 2016), England (Ball 1984;
Parravani et al. 2019), Slovakia (Semelbauer e et al. 2018), France (Jacquiet et al. 2014), and
Australia (Urech et al. 2012; Godwin e et al. 2017). These regions are quite distinct in terms of climate and environmental conditions, but similar trends appear in regions with comparable climates across the world. For example, bimodal seasonal population distributions have been observed in cooler temperate zones like Alberta (Lysyk 1993), Nebraska (Taylor et al. 2007), and France (Jacquiet et al. 2014), while unimodal seasonal populations distributions tend to occur in warmer tropical regions like Thailand (Masmeatathip et al. 2006) and Ethiopia (Sinshaw et al. 2006).
Population Structure: Sex Ratio and Age Cohorts
Population studies focus on seasonality and how the environment affects the abundance of stable flies. Other important aspects of population assessment are sex and age distribution.
The number of each sex and age cohort provides critical information on the status of population growth. Large numbers of females present at certain times of year can allow for estimates of population growth, while assessing the age status can provide insight into recruitment and survival from the juvenile to adult life stage, informing on the suitability of the surrounding larval substrate. Scholl (1980) conducted the first thorough assessment of female stable fly reproductive structures and how to accurately age females/assess parity using an ovarian staining
29 technique with neutral red dye mixed with saline. Sexing can also be done externally using eye- width, where females have eyes set wider apart than males.
Beresford and Sutcliffe (2012) analyzed stable fly fecundity and population structure in southern Ontario using female ovarian aging and parity to assess the populations. There is no known way to assess male ages in wild populations. The male to female ratio over the 5-year study was near 1:1 (Beresford and Sutcliffe 2012). On average, 42.4% of the annually captured females were nulliparous (Beresford and Sutcliffe 2012). These authors examined their field- caught flies using the Krafsur et al. (1994) pteridine aging/survival model and found less than half of females survive to be uniparous, but three quarters of uniparous females survive to be multiparous. These results indicate a higher mortality in younger adult females than older females.
BIOLOGY OF PARASITOID WASPS AND USE AS BIOLOGICAL CONTROL AGENTS
Parasitoid Biology
In order to combat stable fly population growth in an environmentally sound manner, parasitoid wasps may be used as biological control methods in IPM strategies (Axtell 1986;
Machtinger et al. 2015). The most used and studied parasitoids are from the families
Pteromalidae and Ichneumonidae and they parasitize stable fly pupae. Parasitoid wasp females lay their eggs directly into the stable fly puparium and the parasitoid offspring will develop using the developing fly’s body as a source of nutrition and using the puparium as shelter (Jervis et al.
2001). As the parasitoid develops into an adult within the body tissues, the host ultimately dies
(Jervis et al. 2001), which provides a form of natural fly population control.
30 Biological control is preferable to chemical control, whenever possible, because of environmental concerns surrounding insecticide use, the development of chemical resistance in pest populations (Meyer et al. 1987; Cilek and Greene 1994), and the dwindling number of options as old products are removed from the market and few new, environmentally safe pesticides are developed (Hogsette 1999; Romero et al. 2010). Increased resistance to permethrin
(a commonly used insecticide), has been identified on a molecular level. An allele called kdr has mutated into the kdr-his allele in Florida stable fly populations, with a frequency ranging from
0.46 to 0.78 (Olafson et al. 2011). This implies that stable flies in North America are already resistant to the most common insecticides, therefore new methods of population control must be designed and implemented.
There are many studies documenting the use of parasitoid wasps as stable fly biological control agents and the natural presence of these wasps on livestock farms (e.g. Machtinger et al.
2015; Pitzer et al. 2011; Geden and Moon 2009; Loera-Gallardo et al. 2008; Floate and
Skovgård 2004; Seymour and Campbell 1993: McKay and Galloway 1999); however, the effectiveness and ultimately the success of biological control by these wasps is highly variable.
Research on nuisance fly parasitoids has shown some pteromalid wasp species, especially
Muscidifurax raptor Girault & Sanders, Muscidifurax zaraptor Kogan & Legner, Spalangia cameroni Perkins, and Spalangia nigroaenea Curtis, can provide acceptable levels of control under specific conditions typically involving indoor facilities (Seymour and Campbell 1993).
Limiting Factors of Parasitoid Success
Studies of ichneumonid and pteromalid wasp parasitism conducted in laboratories often result in high levels of pupal parasitism (Skovgård and Nachman 2004). Unfortunately, laboratory settings are less variable than field conditions and using parasitoids for fly control in
31 cattle operations under field conditions has proven to be challenging (Skovgård and Nachman
2004). For example, in laboratory settings, parasitoids have no option to leave the release site and temperatures indoors typically remain constant (Skovgård and Nachman 2004). As well, parasitoids have all the required resources for their populations to grow unchecked in laboratory settings, lacking predation, competition, and illness. In outdoor livestock production, parasitoid wasp populations almost always remain too small to control stable fly populations sufficiently
(Axtell 1986). This is likely due to the fast speed at which stable flies develop and reproduce, where a new stable fly generation is produced much faster than the parasitoids can control
(Legner and Brydon 1966). Therefore, biological control success truly relies on the knowledge and efforts of the producer.
Timing and consistency of releases are critical components to biological control success and are completely reliant on the schedule of the producer (Skovgård and Nachman 2004). Other producer-related factors such as cost, correct parasitoid species chosen, and the amount of parasitoids purchased may also impact the efficacy of biocontrol (McKay and Galloway 1999).
The location of the wasp release also affects the effectiveness of control. Machtinger et al.
(2015) measured the distance parasitoids will fly in order to find pupal filth fly hosts and found it is usually just a few metres from the release site. If released too far from the target site, the wasps will not successfully parasitize pupae or establish on the site. Parasitoid behavioural differences when locating stable fly pupae in the substrate may affect the parasitoids’ success at control (Rueda and Axtell 1985; Meyer et al. 1991; Pitzer et al. 2011). Spalangia spp. tend to dig down into substrate to attack pupae buried within, whereas Muscidifurax spp. prefer pupae shallower in the substrate (Rueda and Axtell 1985; Meyer et al. 1991; Pitzer et al. 2011).
32 Sanitation used simultaneously with parasitoid release removes important layers of substrate and the fly pupae located within the substrate that may not be parasitized due to overabundance.
Parasitoid Species Abundance and Overwintering
Parasitoid species from the genus Spalangia make up the majority of the species composition in certain regions (Gibson and Floate 2004; Loera-Gallardo et al. 2008; Romero et al. 2010; Pitzer et al. 2011) but can be almost non-existent in other regions. In Florida dairies,
85.7% of parasitoids recovered from house fly and stable fly pupae were from the genus
Spalangia (Romero et al. 2010). On Florida horse farms, Pitzer et al. (2011) found that 99.9% of parasitoid species recovered from pupae were Spalangia spp., with one Phygadeuon species and one Muscidifurax raptor recovered from two separate pupae. In Norway, 95% of the naturally occurring parasitoids were Spalangia cameroni (Birkemoe et al. 2009). Conversely, in Manitoba,
Phygadeuon fumator parasitized 79.9% of the naturally occurring pupae on dairy farms and
Spalangia cameroni was found in extremely low abundance (McKay and Galloway 1999). In southern Alberta as well, Spalangia spp. were seldom reported in parasitoid surveys of the region, and when found, it was in very low numbers (Lysyk 1995; Floate et al. 1999).
Although not totally understood, it is likely that on-site management practices and environmental conditions play a role in parasitoid species composition, abundance, and survival.
There does not appear to be a relationship between parasitoid activity and temperature, precipitation, nor the abundance of viable host pupae in Florida (Romero et al. 2010). Fly breeding site disturbance and moisture levels, as well as sanitation practices on dairies each had an impact on parasitoid abundance. Undisturbed moist fly developmental sites had a positive effect on parasitoid abundance, and high levels of disturbance and sanitation decreased parasitoid abundance (Romero et al. 2010).
33 The duration of sub-zero temperatures in temperate regions seems to limit overwintering survival stable fly pupae and their pupal parasitoids (Guzman and Petersen 1986; Rivers et al.
2000), as does the life stage the parasitoid is at during the winter months (Floate and Skovgård
2004). In southern Alberta, Floate and Skovgård (2004) found 100% mortality of Spalangia cameroni (the most common commercially available parasitoid) within four weeks of field trials started in November, yet Spalangia cameroni has been observed digging at least 10 cm below the substrate surface to reach host pupae, a strategy that should increase its overwintering survival (Smith and Rutz 1991; Floate and Skovgård 2004). Other parasitoid species, including
Nasonia vitripennis (Walker) and Urolepis rufipes (Ashmead) had higher levels of overwintering field survival (Floate and Skovgård 2004). Nasonia vitripennis is known to undergo larval diapause, making this species more cold hardy and adaptable to winter field environments
(Schneiderman and Horwitz 1958; Rivers et al. 2000). Egg stages had the highest overwintering mortality both in barn and field conditions, while pupal stages had the highest survival (Floate and Skovgård 2004).
Ultimately, the species and abundance of parasitoids in a region does not appear to closely follow any single trend or pattern, resulting in fluctuations that are difficult to delineate.
Biological control, using dynamic living organisms like parasitoid wasps, requires more research in order to lead to successful stable fly population management.
INTRODUCTION TO THE RESEARCH PROJECT
Although highly researched, stable flies still pose major economic and animal welfare concerns around the world. Because stable fly populations are so strongly affected by climatic conditions, IPM models must have detailed information from various geographic regions to
34 create holistic strategies for predicting population patterns and outbreaks. Much is known about stable fly biology in general, but information from the Canadian Prairie provinces, at north latitude with extreme seasonal conditions, is mostly unavailable at this time. Stable fly population dynamics data, as well as parasitoid species distribution data have been collected and analyzed in neighbouring provinces, but current information from Manitoba is scarce and climate change has the potential to alter behavioural and phenological patterns that have been observed in the past. Additionally, the climate of Manitoba is typically colder and drier for a longer duration in the winter months than that of southern Ontario and Alberta, which creates different overwintering pressures on both stable flies and their pupal parasitoids.
The first goal of this research project was to determine wild stable fly population dynamics and phenology, including emergence, peaks, and cessation, in the province of
Manitoba. We examined how environmental conditions affected stable fly abundance, as well as the population structure, including sex ratios and age structure. The second goal of this research project was to determine the overwintering potential of stable flies, as well as their pupal parasitoids, in both field and laboratory settings. We examined stable fly developmental status under different temperatures and durations and identified parasitoid wasp species and abundance.
35 CHAPTER 2: ADULT STABLE FLY STOMOXYS CALCITRANS (L.) (DIPTERA: MUSCIDAE) SEASONAL POPULATION DYNAMICS ON DAIRY FARMS IN SOUTHERN MANITOBA, CANADA.
ABSTRACT The relationship between adult stable fly population abundance and environmental variables was examined at dairy farms over two years. Coroplast sticky traps were used to monitor changes in adult abundance at each site. Changes in abundance, age structure, and sex ratio were evaluated. Adult abundance followed unimodal distribution and populations peaked in mid-July. Age structure differences between years support the migratory hypothesis of recolonization. Changes in the sex ratio over the seasons were due to trap biases and dispersal.
This study aimed to determine which environmental factors lead to rapid adult stable fly population growth in southern Manitoba and provide increased precision when designing and implementing control methods to target stable flies.
INTRODUCTION
Stable flies Stomoxys calcitrans (L.) (Diptera: Muscidae) have been considered one of the most economically important insect pests for over a century (Newstead 1906; Bishopp 1913).
For cattle especially, severe biting activity from both sexes of stable fly causes decreased weight gain (Campbell et al. 1977; Campbell et al. 2001) and decreased milk production (Bruce and
Decker 1958; Taylor et al. 2012), as well as the potential for physical injury and health issues.
Stable flies cause significant economic losses for producers and the cattle industry as a whole
(Kunz et al. 1991; Taylor and Berkebile 2006; Taylor et al. 2012).
Stable flies reproduce rapidly under warm temperatures (Lysyk 1998; Florez-Cuadros et al. 2019). Due to the environmental effects of climate change, stable flies may eventually
36 become year-round pests as far north as Canada. With warming temperatures, populations are likely to reach injurious levels at an earlier date and individuals may survive for a longer portion of the year (Gilles et al. 2008). Compounding this issue, stable flies are capable fliers (Bailey et al. 1973; Hogsette et al. 1987) and may quickly take advantage of the changing climate by infesting new areas. Currently, severe seasonal infestation is a common occurrence on Canadian livestock production sites (Lysyk 1993; Beresford and Sutcliffe 2009a). Within the next few decades, under the effects of climate change, stable fly infestations may become a larger threat to economic productivity and animal welfare. A deeper understanding of stable fly population dynamics is required to plan and execute successful pest management programs to mitigate the risks associated with increased stable fly biting activity.
Status of Stable Fly Population Dynamics
Stable fly biology and physiology have been studied extensively; however, population dynamics and factors driving the observed fluctuations remain poorly understood. Stable flies are a year-round concern in tropical and subtropical geographic regions (Masmeatathip et al. 2006;
Muenworn et al. 2010; Phasuk et al. 2013), where consistent warm temperatures coupled with high annual precipitation create ideal developmental conditions for stable flies (Taylor et al.
2017). Conversely, in temperate zones such as Canada, pronounced seasonal changes limit stable fly populations and lead to extirpation of stable fly populations during the cold and dry winter conditions. As spring approaches and temperatures begin to rise, adults begin to emerge and recolonize similar sites.
Stable Fly Sex Ratios and Age Cohorts
Knowing the age of individual flies in a population can improve predictions regarding population change based on the number of reproductive individuals present (Beresford and
37 Sutcliffe 2012) and evaluating the sex ratio of populations may increase the accuracy of predicting population growth given the number of each sex present. Higher numbers of females may indicate the potential for strong population growth, whereas fewer females may signal a potential decline. If sex biases in wild stable fly populations exist, it is due to factors such as sampling methods, or environmental conditions acting on both juvenile and adult life stages. It is currently unclear how environmental conditions affect the population structure in southern
Manitoba, including the age cohorts and sex ratio.
Goals of the Study
The main objective of this study was to assess stable fly seasonal activity on dairy farms in southern Manitoba to determine how environmental conditions affect adult emergence, abundance, and population structure. Trends in emergence dates, population peaks, age structure from female stable flies, and sex ratios were evaluated during this two-year study. These data aimed to determine which environmental factors lead to rapid population growth in southern
Manitoba and improve the design and implementation of control methods targeting stable flies.
MATERIALS AND METHODS
Study Sites
Three dairy farms in southern Manitoba were chosen as the field study sites for the two years of research in 2017 and 2018. All farms were located within 2 km of each other, which acted as a control to minimize differences in environmental conditions among sites (Figure 2.1).
The first site was the Glenlea Research Station farm (49.6429, -97.1735), which housed approximately 60 Holstein dairy heifers and approximately 10 Angus beef cattle each year in outdoor concrete-lined confinement pens (Appendix A, Figure A.1). Glenlea Research Station
38 housed cattle in covered outdoor pens with concrete floors and straw for bedding. Enclosure sanitation was conducted approximately every four to six weeks, with a full strip cleaning of the enclosures, mechanically removing manure but not washing. The soiled straw bedding was scraped out using a skid-steer loader and fresh straw was laid. The soiled bedding was driven to the onsite waste pile and deposited until the waste pile was removed in mid-August. This farm did not appear to have a formal pest management protocol in place during either study year, but sanitation was maintained on a relatively consistent schedule. Angus beef cattle were not present at Glenlea Research Station after approximately mid-September both years.
The second site was Red River Holsteins (49.6272, -97.1646), which housed approximately 70 Holstein heifers each year in two soil bedded confinement pens (Figure A.2).
Red River Holsteins housed all heifers in fully open outdoor enclosures. Both outdoor enclosures had shelter on the north side in the form of trees acting as wind blocks, but no human-made shelter was present in either enclosure. The substrate was bare soil with approximately five areas of straw bedding each covering roughly 2 m x 2 m. Enclosure maintenance in the form of manure removal was not observed in either enclosure, but fresh straw was added to the already existing mounds as it was soiled, creating stratified layers of straw. Red River Holsteins housed approximately 15 to 20 milking cows in a fully enclosed barn, which had concrete floors and straw bedding. During milking barn sanitation, the manure and soiled straw were removed using a skid-steer loader and dumped into a sewage pit approximately 100 m from each of the barn and two outdoor enclosures. Red River did not appear to have a formal pest management protocol in place during either study year, but milking barn sanitation was maintained on a relatively consistent schedule, with a full strip clean occurring approximately one time every four to five weeks.
39 The third site was Vancrest Holsteins (49.5953, -97.1426), which kept approximately 50
Holstein heifers on soil bedding confinement pens and approximately 20 calves held individually in hutches on straw bedding in 2017 (Figure A.3). Vancrest Holsteins housed their cattle in an outdoor enclosure similar to Red River Holsteins, with bare soil substrate and no human-made shelter within the enclosure. Enclosure maintenance in the form of substrate removal was not observed at this dairy, but fresh straw was added to the existing straw mounds as it was soiled, creating stratified layers of straw. Vancrest Holsteins housed approximately 25 to 30 milking cows in a fully enclosed barn, which had concrete floors and straw bedding. During milking barn sanitation, the manure and soiled straw was removed using a skid-steer loader. The manure and straw bedding were dumped approximately 150 m behind the calf hutches onto an onsite waste pile and continually deposited over the summer. Vancrest did not appear to have a formal pest management protocol in place during either study year; however, on days when producers noticed fly numbers were high, a pyrethrin aerosol (KONK® Residual Farm and Home,
AirGuard Control, Concord, ON) was sprayed the outside of all the calf hutches. Milking barn sanitation was maintained on a relatively consistent schedule, with a full strip clean occurring approximately one time every four to six weeks. After the completion of the 2017 field season,
Vancrest Holsteins underwent construction of two new barns to keep all milking cows, heifers, and calves entirely indoors. Hence, for the 2018 field season there were no heifers in outdoor pens.
Adult Stable Fly Trapping and Preservation
All trapping in both years occurred around the perimeter of outdoor animal enclosures.
We set up six white Coroplast® panels (30.5 cm x 22 cm) covered with a thin film of adhesive
(Tanglefoot™, The Scotts Company LLC, Marysville OH) to monitor adult stable fly abundance
40 (Beresford and Sutcliffe 2006). The adhesive was applied on both sides of the panel using a 6.35 cm paintbrush. Wooden stakes approximately 40 cm long were driven into the ground, then panels were stapled to the stakes, so the bottoms of the panels were approximately 30 cm above the ground surface (Appendix B, Figure B.1). Traps were set up weekly at the three study sites from 16 June to 24 October 2017, and from 16 May to 4 October 2018. Trapping started earlier in 2018 to make sure we recorded the first week of adult activity. The trapping season ended when the air temperature dropped below 10°C. Traps were positioned at least 10 metres apart along the perimeter of the outdoor cattle confinement pens and left for 24 hours before they were brought back to the lab where the number of stable flies was recorded from each trap. Trapping was moved to the next day if it was raining or strongly windy, as adult stable flies are not active during adverse weather conditions. Cattle were present at all farms throughout the duration of trapping except for Vancrest Holstein farm in 2018, when cattle were no longer kept outdoors.
Although cattle were removed from outdoor pens, traps were still set up in 2018 in the same locations as the previous year.
Depending on the number of adult stable flies captured during the sampling period, subsamples of 100 adults were randomly taken from each of the six traps from each site. Up to a total of 600 adults were subsampled per site with up to 1800 subsampled every week. The subsampled stable flies were cleaned of adhesive using Klean Strip® Green® Paint Thinner (W.
M. Barr, Memphis, TN) and 70% ethanol. A 1-minute wash in 15 ml of paint thinner followed by a 1-minute wash in 15 ml of 70% ethanol removed all traces of Tanglefoot™. The cleaned stable flies were then placed in glass vials filled with 70% ethanol to preserve the adult’s internal organs and structures and finally placed in a 5°C incubator for storage until sexing and aging.
41 Sex Determination and Female Gonotrophic Assessment
The preserved adult stable flies were used to determine the population age structure and sex ratio. Adult flies were visually sexed based on the distance between the compound eyes, where females have wider set eyes than males. After sexing and recording the number of each sex, the male flies were discarded while the females were kept for assessing gonotrophic development and parity. There is currently no method for aging male stable flies using internal structures. Following the methods of Scholl (1980), ovaries were removed from the abdomens and graded from stage 1 (small size; no egg development) to 4 (large size; advanced egg development). Ovaries were also examined for follicular relics, which are structures that provide physical evidence of a previous oviposition cycle. Ovaries with one, two, or more follicular relics were further classified into uniparous, biparous or multiparous categories respectively, while ovaries lacking follicular relics were classified as nulliparous.
Although Coroplast® sticky traps catch stable flies at twice the rate of other traps, they are biased towards capturing more males than females, with an expected ratio of two males to one female (Beresford and Sutcliffe 2006). The traps also catch more nulliparous females, as well as more unfed stable flies of both sexes (Beresford and Sutcliffe 2006). To better represent the wild stable fly sex ratio and mitigate biases, we used timed sweep netting sessions. After setting up the sticky traps, we used sweep nets and swept at passing and resting stable flies until we captured 30 stable flies near the cattle or until 20 minutes had elapsed. We transferred the collected flies into labelled plastic bags and sexed them in the lab using the distance between compound eyes as before.
42 Weather Data Collection
Weather data were obtained from the St. Adolphe weather station (49.6958, -97.1178) located on
Kelburn Farm (Figure 2.1) and monitored by the Government of Manitoba Agriculture Branch.
The station is located 5.1 km north of the Glenlea Research Station farm, the most northern of the three dairy farms used during this research project. The weather conditions were recorded daily and included wind speed (km/h), wind direction (degrees), air temperature (°C; taken as maximum, average, and minimum), total precipitation (mm), soil temperature (°C) and soil moisture (%) both measured at 5 cm below the soil surface (taken as maximums, averages, and minimums), and relative humidity (%). From the air temperature maximum and minimum values, we calculated cumulative degree-days using the standardized base10 (Pruess 1983; Lysyk
1993) for each year using the modified sine wave method from Allen (1976).
Statistical Analyses
All statistical analyses were conducted using R Software Versions 4.0.0 - 4.0.3 (R Core
Team 2020) and the following R packages: astsa (Stoffer 2020), car (Fox and Weisberg 2019), countreg (Zeileis and Kleiber 2020; Zeileis et al. (2008); Kleiber and Zeileis (2016), dplyr
(Wickham et al. 2018), emmeans (Lenth 2020), MASS (Venables and Ripley 2002), multcomp
(Hothorn et al. 2008), multcompView (Graves et al. 2019), PerformanceAnalytics (Peterson and
Carl 2020), plyr (Wickham 2011), pscl (Zeileis et al. 2008; Jackman 2020); RColourBrewer
(Neuwirth 2014), RcppRoll (Ushey 2018), reshape2 (Wickham 2007), and sciplot (Morales
2020).
43 Adult Abundance between Years and among Sites
Trapped adult stable fly data were expressed as total counts. A Chi-square goodness-of- fit test was used to determine if there were significant differences in the total number of trapped adults between the two sampling years.
Correlation Matrices for Environmental Variables
Environmental variables were recorded by Manitoba Agriculture at the St. Adolphe weather station. Each variable was recorded as a daily maximum, average, and minimum. The daily air temperature maxima and minima were used to calculate degree-days. Degree days are based on an insect’s developmental rate at a given temperature between the upper and lower limits. Taking the difference of daily maximum and minimum temperatures, units of heat are measured over time and corresponding juvenile growth and development can be inferred
(Murray 2008). The degree-day base10 was used, which is the approximate developmental threshold temperature of stable flies (Lysyk 1998). Degree-days were calculated for the 24-hour period each day starting at midnight (2400 hours) with temperature data starting from 1 January of both years.
To determine if multicollinearity was present among the environmental (explanatory) variables, two Pearson product-moment correlation analyses were conducted and the values were placed into Pearson correlation matrices; one for each of the sampling years. The Pearson product-moment correlation determines how strong the linear relationship is between two variables. When more than two variables are analyzed, the output is in the form of the Pearson cross-correlation matrix (Appendix C), which provides Pearson’s correlation coefficient r. The value of the Pearson’s coefficient (-1 to +1) depicts how well each pair of all the variables fits the model, as well as the direction of the relationship. The two Pearson cross-correlation matrices
44 allowed us to assess variables that were strongly correlated with each other and helped to evaluate the risk of multicollinearity among the variables before generating the models. The weather variables chosen for future analyses were based on the results of the cross-correlation matrices. The air temperature and precipitation variables were correlated with other variables, but due to their known critical importance in stable fly development (Taylor et al. 2017), these variables were still included in model construction. Air temperature, relative humidity, soil moisture and soil temperature were taken as maximum, average, and minimum, and then all three values of the same variable were placed into the correlation matrices. Therefore, the three values of the same weather variable were all strongly correlated.
Cross-Correlation Plots
The weather variables chosen from the Pearson cross-correlation matrices were average wind speed (km/h), average air temperature (°C), total daily precipitation (mm), average soil temperature measured at 5 cm below the soil surface (°C), average soil moisture measured at 5 cm below the soil surface (%), and relative humidity (%). These variables were chosen because they were the least correlated with each other or were critically important to stable fly development and survival, regardless of correlation.
Adult stable fly emergence is largely dependent on the survival and development of larval and pupal stages, as well as the environmental conditions the juveniles are subjected to during their development. We created a time series with lags and graphical output in the form of cross-correlation plots (Appendix D) to determine the relationship between trapped adult stable fly abundance and weather variables in the preceding weeks. The cross-correlation plots depict weekly lags on the x-axis and the strength of correlation on the y-axis. The dashed lines above and below the x-axis indicate the significant correlation threshold (+0.500 and -0.500), above
45 and below which correlation factors are statistically significant. Because weather data were recorded daily, but adult flies were trapped for a single 24-hour period each week, we averaged the weather variables values per week to match the scale of trapping. We then grouped the weather variables into approximate 7-day weeks (infrequently only 6 days due to sampling dates that avoided adverse weather conditions) up to four weeks prior to any given sampling date to identify when each weather variable was most strongly correlated to trapped adult abundance over the entire trapping period (June to October 2017 and May to October 2018).
Cross-correlation plots (ccf function in R) with time lags of the number of days between sampling (typically seven days) enabled us to visually examine relationships between one weather variable and the number of adult stable flies captured per sampling week. Cross- correlation assumes there is no autocorrelation between variables. Examining cross-correlation plots within a time series analysis led us to observe the relationship between adult stable fly abundance per trapping date and environmental variables up to nine weeks prior to the sampling date. The time series analysis provided information on the number of weeks prior to week zero that a weather variable influenced adult abundance. We were only interested in weather data four weeks prior sampling (i.e., four time lags) because anything further than one month into the past becomes difficult to interpret and is not biologically relevant. We extracted the correlation factors from the ccf vector created in R to confirm the strength of the correlation (r-values) between weather variables in each preceding week and adult fly abundance on the sampling date.
Model Selection and Model Simplification
To examine which environmental variables had significant effects on adult stable fly abundance each year (expressed as total counts), we used generalized linear models (GLM) (glm
R function, MASS package (Venables and Ripley 2002)). Datasets with a count response
46 variable are typically analyzed using GLMs. Preliminary data exploration results of summary statistics, graphs, plots, and histograms revealed the count data did not follow a normal distribution and the variances were heterogeneous. To account for these characteristics, we chose models that could tolerate count data with heterogeneous variances and non-normal distributions rather than implementing logarithmic or square root data transformations. Data transformations often do not behave well with count data and non-parametric tests have assumptions that would still be violated by our data characteristics (Sileshi 2006; O’Hara and Kotze 2010; St. Pierre et al. 2018).
We fitted both Poisson and quasi-Poisson models which are both able to handle the non- normal characteristics mentioned above, but found that after assessing model fit we had overdispersion, meaning the residual deviances were greater than the residual degrees of freedom. Overdispersion violates an assumption of Poisson models, where the mean and variance are assumed to be equal. To minimize overdispersion, we fitted a negative binomial model to the data and found it fit the data well from both 2017 and 2018. Negative binomial regressions are the strongest model choice when the dependent variable is count data, independent variables are continuous, observations are independent, and counts follow a negative binomial distribution. Negative binomial regression follows similar assumptions to
Poisson regression, as it is a more general version of the Poisson, but negative binomial regression is applicable when data is overdispersed (i.e. when the variance is unequal to the mean).
Variable selection began by applying Akaike’s Information Criterion (AIC; Akaike 1974) to select the best-fitting negative binomial regression models for each year. The function stepAIC
(MASS R package) was used on the negative binomial regression models with only single terms
47 to automate the first steps of model selection, choosing all the single variables in a forward selection and adding variables until the AIC score could not be reduced further. Once the single term models were reduced fully, interaction terms (up to two terms) between the single variables already present within the fully reduced models were manually added to the reduced model, and using backward selection, removed one-by-one based on the results of an ANOVA-like test called the Analysis of Deviance test. This test is specifically designed for negative binomial- distributed data and compares the likelihood ratios between the reduced model (model 1) and the model with all interaction terms present (model 2). If the p-value from the likelihood comparison was not significant, indicating the reduced model was accepted, the highest-order interaction term was removed, and a new model containing the next highest interaction term (model 3) was tested against the previous model (model 2) until the current model was not rejected.
Once a final model was developed for each year, we re-assessed the fit of each model and verified assumptions using various basic plots in R for visual assessment. Rootograms were created to visually examine the fit of the negative binomial models. Rootograms are graphical tools similar to a line plot that assess model goodness of fit in count data regression analyses
(Tukey 1977; Kleiber and Zeileis 2016). Pseudo-R2 statistics are R2-like measures, used when analyzing Poisson, negative binomial, and zero-inflated model regressions as a method of estimating model goodness-of-fit (Heinzl and Mittlböck 2003; Martin and Hall 2015). Pseudo-R2 values falling between 0.2 and 0.4 represent a very strong model fit to the data (Louviere et al.
2000). McFadden’s Pseudo-R2 values for both models were calculated as a measure of model goodness-of-fit, along with the AIC values.
48 G-test Goodness of Fit on Sex Ratio and Generalized Linear Models on Age Structure
To see if the sex ratios differed between sweep-net captured and sticky trap-collected adults, we conducted a G-test of Goodness of Fit (also called the Likelihood Ratio Test). The G- test is the appropriate choice when testing if observations of a categorical variable fit the expected theoretical values, assuming that the observations are independent. In order to avoid the loss of information such as variation among sites, the G-test was a more appropriate test than
Pearson’s chi-square test. The G-test does not inappropriately pool data prior to the statistical analysis, thereby avoiding sacrificial pseudoreplication. We calculated the trapping and sweep netting sex ratio of each site between years, then compared the observed ratios to the expected sweep net and sticky trap ratios (Beresford and Sutcliffe 2006).
To assess the female age structure, we calculated the mean weighted age of females by parity status where nulliparous, uniparous and biparous females were assigned an age of 1, 2 and
3 respectively on the y-axis. A greater number on the y-axis indicated a larger proportion of older flies. When examining differences in female age between years and among sites, negative binomial regression analyses were used to account for overdispersion in the data, which violates a fundamental assumptions of Poisson models. An ANOVA-like test called Analysis of
Deviance was used to determine if significant changes from the null regression hypotheses were observed.
RESULTS
Adult Abundance between Years and among Sites
A total of 96,291 adult stable flies were captured at all three sites during the two sampling years. Sampling began on 16 June 2017 and the first flies were collected on that date.
49 The following year, sampling began 16 May 2018, the first flies were caught on 7 June 2018.
Adult trapping abundance followed the same general pattern: an exponential increase to mid-
July, then a steady decline to a relatively consistent level until the air temperature dropped below the adult flight threshold temperature of approximately 10oC (Lysyk 1998; Figure 2.2). Sampling ceased on the day of first snowfall both years (24 October 2017 and 4 October 2018).
Adult stable flies were more numerous in 2017 (n = 53,545) than 2018 (n = 42,746)
(Figure 2.2; χ2 = 12.25, df = 1, P <0.001). Annual population distributions were unimodal with the population peak at the end of July (21 July 2017 and 26 July 2018) in both years. In 2017, the mean number of stable flies caught per trap was 164 ± 12, with a range from zero flies to 1361, and in 2018 was 113 ± 8, ranging from zero flies to 918. The largest total number of stable flies per trap occurred at Red River Holsteins both years.
Adult stable fly abundance was significantly different among the three dairy farms
(Figure 2.3; χ2 = 20.17, df = 2, P < 0.001), with a post hoc Sidak’s t-test revealing that more flies were caught at Red River Holstein than at Glenlea Research Station or Vancrest Holsteins, which in turn did not differ in both years. Stable fly abundance was also significantly different between
2 years (χ = 964.30, df = 5, P <0.001). Red River Holsteins (Padj = 0.005) and Vancrest Holsteins
(Padj < 0.001) had significantly higher fly abundance in 2017 than in 2018. Trapped adult abundance was not significantly different between years at Glenlea Research Station (Padj =
0.98).
Correlation Matrices for Weather Variables
The weather variables considered for modeling were average air temperature, precipitation, wind speed, average soil moisture, average soil temperature, relative humidity, and cumulative degree-days, due to their importance in juvenile stable fly development and adult
50 activity levels. Explanatory variables with multiple measurements (maximum, average, and minimum values) were assessed and only the averages were retained in model design. The air temperature variables were strongly correlated (0.50 or higher) with the soil temperature and soil moisture measurements in 2017, but there was a smaller correlation between air temperature and soil moisture in 2018. The correlation between soil temperature and soil moisture was stronger in
2017 than in 2018 (-0.73 and 0.09, respectively).
Time Series Cross-Correlation Plots 2017 and 2018
Correlation coefficient factors expressed the strength of the relationship between adult stable fly abundance and each weather variable prior to a given sampling date (Table 2.1), indicating if the variable may have contributed to either an increase or decrease in the number of trapped adult flies. Weather variables had a significant correlation with adult stable fly abundance when the correlation coefficient factor crossed the significance threshold (a = 0.5 or greater, a = -0.5 or below).
Overall, there were no unifying trends between weather variables and fly abundance between years. Precipitation, wind speed, and humidity were included in the analysis but were not correlated with adult stable fly abundance. In 2017, air temperature, soil temperature and daily degree days were positively correlated with adult stable fly numbers the week of sampling
(week 0) at Red River Holstein, indicating that as the units of these variables increased, the number of adult flies trapped during the same week also increased. Soil moisture was negatively correlated, meaning that as soil moisture increased, the number of adult flies trapped during the same week decreased. There was a negative correlation between cumulative degree days and adult stable fly abundance in the two to four weeks preceding sampling. This is because as summer progressed to autumn, an increased number of degree days meant that it was nearing the
51 end of the stable fly season, leading to fewer trapped adult flies. A similar trend was observed at
Vancrest Holstein. There was no correlation between weather variables and adult stable fly abundance at the Glenlea Research Station.
In 2018, soil temperature was the only environmental variable that was significantly correlated with trapped adult abundance (Table 2.1). An increase in soil temperature in the weeks prior to sampling led to an increased number of adult flies trapped at all three sites. At Red River
Holstein and Glenlea Research Station, this correlation was significant for up to three weeks before sampling, and at Vancrest Holstein an increase in soil temperature one to two weeks prior to the week of sampling led to increased adult fly abundance.
Model Selection and Model Simplification
The general model equation for negative binomial regression model with the lowest AIC score for 2017 is as follows:
(trapped adult flies) = ß0 + ß1 (relative humidity) + ß2 (soil moisture) +
ß3 (cumulative degree days) + ß4 (soil moisture*relative humidity) + (Eq.1)
where ß0 is the intercept, ß1–4 are parameter estimates, and is the error. The general model equation for negative binomial regression model with the lowest AIC score for 2018 is as follows:
(trapped adult flies) = ß0 + ß1 (soil temperature) + ß2 (soil moisture) +
ß3 (cumulative degree days) + (Eq. 2)
where ß0 is the intercept, ß1–3 are parameter estimates, and is the error. Model parameters for the negative binomial regression analyses were different between years and all the variables
52 included in the models were significant (Table 2.2). The negative binomial regression is interpreted as such: as a weather variable increases or decreases by one increment, the difference in the logs of expected counts for the number of total trapped stable flies will change by its respective regression coefficient, as long as the other weather variables used are held constant
(UCLA Statistical Consulting Group n.d.). For example, for each one-unit increase in relative humidity in 2017, the expected log count of the number of total adult stable flies captured increases by 0.3139 (or approximately 1.4 adult flies). For the 2017 model (Eq. 1, Table 2.2), we noticed that soil moisture had the strongest influence on total stable flies captured, followed by humidity, a negative relationship with the interaction term, and degree days. For the 2018 model
(Eq. 2, Table 2.2), soil temperature had the strongest influence, followed by soil moisture and degree days.
Soil moisture and cumulative degree days were significant predictor variables in both models. Average weather and environmental conditions between the two years differed in terms of reported daily precipitation and soil moisture (Table 2.3). The year 2018 had almost double the soil moisture in June and July than 2017, as well as higher total daily precipitation. Monthly air temperature averages, soil temperatures, and relative humidity were similar across years
(Table 2.3). The models contained higher residual deviance compared to the residual degrees of freedom in both models suggesting that overdispersion is only partially dealt with. The rootograms indicate underfitting of zeros and overfitting of subsequent counts. These qualities mean the models are lacking some variables that may explain the system in greater detail.
Sex Ratio
A total of 41,458 stable flies were subsampled from the total number of stable flies caught each week during the sampling season over the two-year study. Of the 21,966 flies
53 subsampled from traps in 2017, 4910 (22.4%) were female and 17,056 (77.6%) were male (Table
2.4). A total of 19,492 flies were subsampled from traps in 2018, 7441 (38.2%) were female and
12,051 (61.8%) were male. Wild stable fly sex ratios follow a 1:1 male-female ratio (Beresford and Sutcliffe 2006), but Coroplast® sticky traps follow a 2:1 male-female ratio (Beresford and
Sutcliffe 2006). The mean sex ratio (male:female) from sticky trap sampling was 3.4:1 in 2017, and 1.6:1 in 2018. The sex ratio decreased over the duration of sticky trap sampling in both years but noticeably higher numbers of males were recovered in 2017, starting with a remarkably high male: female ratio in June (5.8:1) and decreasing to 1.9:1 by the end of the season (Table 2.4).
Although stable fly abundance was lower in 2018, there was a notable 1.5-fold increase in the number of females trapped (Figure 2.4). By site, Red River Holstein had a 20% increase in the number of females trapped, while at Glenlea Research Station and at Vancrest Holstein, the number of trapped females increased by 11% and 17%, respectively (Figure 2.4).
Sweep-net collection yielded a total of 1491 adult stable flies across both years (Table
2.5), of which 866 were males (58.1%) and 625 were females (41.9%). When we compared the expected and observed sex ratios from flies collected from traps (expected 2:1 ratio) and flies collected from sweep-nets (expected 1:1 ratio), the sex ratios were significantly different than the expected ratios for both types of collection (Table 2.5; Gsweep = 320.14, df = 54, P < 0.001; Gtrap =
5562.29, df = 111, P < 0.001;).
Female Age Structure and Parity Status
Of the 12,351 female stable flies subsampled from Coroplast® sticky traps, 12,100 were suitable to assess parity (Table 2.6). The remaining 251 were unusable due to abdominal damage or were identified as female by their head, but no abdomen was present. Small bird feathers were removed from some traps at the end of the season indicating potential predation off traps.
54 A total of 4880 females in 2017 and 7220 females in 2018 were dissected and sorted into one of three parity categories: nulliparous, uniparous, or multiparous. In both years at all sites, over 90% of the females dissected were nulliparous and less than 1% were multiparous (Table
2.6). More females were sampled in 2018 (2 = 53.85, df = 1, P < 0.001) and more females of all ages were trapped at Glenlea Research Station in both years (2 = 16.00, df = 2, P < 0.001).
Greater numbers of nulliparous females were recorded at all sites in 2018 than 2017 (2 = 14.24, df = 2, P = 0.007), and more uniparous (2 = 7.32, df = 2, P < 0.001) and multiparous females
(2 = 4.72, df = 2, P = 0.041) were recorded in 2017.
The mean weighted age of females changed over time during the trapping season (Figure
2.5). There was greater variation in 2017 when there were three periods with peaks of older females. In contrast, the proportion of older flies only increased on 2 August 2018 and 25
September 2018. To determine if the variation observed in 2017 was common throughout the population or restricted to one site, the mean weighted female age by site was plotted (Figures
2.6–2.8) and it was found that the first peak in the overall plot (Figure 2.5) reflected that there were older flies at all sites by the end of June, with most older females located at Glenlea
Research Station (Figure 2.6). Older females were also collected at all sites in early August, but this time the majority were collected from Red River Holsteins. Near the end of the sampling period in mid-October, a greater proportion of older flies were collected at Vancrest Holsteins
(Figure 2.5). In 2017, older female flies were recovered more frequently than in 2018, and a greater proportion were observed at the beginning and at the end of the season.
DISCUSSION
Within the next few decades, environmental conditions may favour even faster growth and
55 development of stable fly populations (Prairie Climate Centre 2019). Economic losses and animal welfare are already major concerns in livestock production and may become even more so as climate shifts continue. An important aspect of managing stable flies is understanding how populations change over the fly season and how they are influenced by environmental conditions. The goal of this study was to assess stable fly seasonal activity on dairy farms in southern Manitoba to determine how environmental conditions affect adult emergence, abundance, and population structure. Monitoring populations and examining data regarding emergence dates, population peaks, and population structure will assist in determining which environmental factors lead to significant changes. This information aims to increase precision in stable fly control procedures.
Environmental Conditions and Site-by-Site Variation
The three dairy farms (Glenlea Research Station, Red River Holsteins and Vancrest
Holsteins) were chosen for their nearness to one another, as a means of controlling some of the environmental variation that can occur when conducting large-scale outdoor studies. Although the environmental conditions the premises were subjected to were similar, the farms differed in cattle housing and management, and pest management practices. Adult stable fly abundance at a site is influenced by host activity, host abundance, and the availability of oviposition sites
(Gersabeck and Merritt 1983), therefore, the observed differences between the years may be explained by these factors.
Year was a significant factor in adult fly abundance at Vancrest Holsteins because cattle management on that farm changed from 2017 to 2018. In autumn of 2017, cattle housing on
Vancrest Holsteins changed. After the fly trapping season ended, the producers built a fully enclosed barn with concrete floors with a layer of sand on top of the concrete and transferred all
56 the cattle indoors. Straw bedding was used as an additional layer on top of the sand. No cattle remained outdoors for the entirety of the 2018 sampling year; therefore, adult flies had limited access to cattle housed inside the barn.
Red River Holsteins had significantly higher numbers of flies per trap in both study years. Soiled straw bedding was dumped into an in-ground sewage pit approximately 100 m from the two outdoor enclosures rather than piled into a mound like at Glenlea Research Station and Vancrest Holsteins. By storing it in-ground, the run-off from rainfall mixed with the straw and manure, creating an ideal juvenile development site. The temperature around the edges of the pit, when felt by gloved hands, was warmer than skin temperature and a hardened crust of dried straw formed on top of the pit. Though larval development site features were not measured, this hardened crust may have kept heat and moisture at suitable levels for development and emergence. Examining beneath the hardened crust exposed dense populations of fly larvae and pupae around the edges. When above-ground manure mounds were examined with trowels at
Glenlea Research Station and Vancrest Holsteins, both larvae and pupae were found sparsely distributed at much lower densities than observed at Red River Holsteins.
Although environmental conditions were assumed to be similar across sites, differences at the microscale, though unmeasured in this study, may have influenced the significant difference in adult abundance between years. The complete stable fly life cycle takes between
225 degree-days (Gilles et al. 2005) and 232 degree-days (Lysyk 1993). There was the potential for four generations in 2017 based on a developmental time of 225 degree-day development time: one in June, one in July and two in August, and there was the potential for five generations in 2018, with one each month from May to September. Even with the extra generation of stable flies in 2018, fewer overall adults were captured.
57 A difference of 10, 799 adult stable flies was recorded between years. This difference may be partially explained by ambient air temperatures in the month of October in 2017 and
2018. October temperatures became cold early in 2018 and the trapping season ended two weeks earlier in 2018 due to snow fall. Within that same two-week time frame at the end of the season in 2017, there were 2359 adult stable flies trapped, which could have potentially accounted for
21.8% of the total difference in adult abundance observed between years.
Trap and Sweep Net Sex Ratios
Knowing the sex ratio within a population can increase the accuracy of predicting population growth. Populations with higher female abundance may indicate stronger potential for population growth, whereas fewer females may signal the potential for smaller populations.
Population sex ratios at the three dairy farms were assessed using both trapping and sweep net collections. Male and female stable flies have been observed waiting for mates on white or light-coloured vertical objects (Buschman and Patterson 1981) known as waiting stations (Downes 1969) or encounter sites (Parker 1978). When males gather in high numbers at a waiting station, females avoid these sites and remain on less favourable, darker coloured objects (Buschman and Patterson 1981). Sweep net collection occurred along the cattle enclosures, closer to the enclosures than the traps. We sweep net-collected where flies were observed resting on posts and gates of any colour, but far enough away to avoid startling the cattle with the nets.
There were almost twice as many males in July than what would be expected normally in the population (Table 2.5). Later in the season, the reverse was observed: there were half as many males caught in October compared to what would normally be expected, although much fewer flies were caught in October compared to the other months (Table 2.5). There were
58 significant differences between the expected and the observed sex ratios of both sweep net- collected (1M:1F) and trapped (2M:1F) adults.
Males outnumber females in most cases on farm sites (Hogsette and Ruff 1985) and this was observed in this study in both years. However, the ratios of males to females were significantly higher than expected. One explanation may be increased male activity. In house flies (Musca domestica L.), males are more active than females at medium densities, which may be attributed to increased aggressive behaviour observed in male flies towards other males and females (Bahrndorff et al. 2012). Trap and sweep collection occurred as near to the cattle as possible. The increased number of males collected is partially due to the known bias in sticky traps but may also be because males were using sticky traps as waiting stations nearer the cattle and keeping the females away with their aggressive harassing behaviour.
Female Parity and Age Cohorts
Similar patterns in female age structure were expected between years, with greater numbers of older females early on, as they would disperse from other farms. This trend was observed in 2017, with older females captured earlier in the year; however in 2018, young females were captured during the earliest weeks of trapping. This could indicate the first flies caught overwintered locally; however, stable fly development time requirements suggest otherwise. Stable fly development requires 225 degree-days from egg to adult at 10°C, which is considered the lower development threshold. If stable flies overwintered locally, it is expected that they would be trapped before the accumulation of 225 degree-days because they would have already been at either the larval or pupal stage of the lifecycle. The spring of 2018 was warm and had accumulated 225 degree-days above 10°C by 31 May 2018, one week before trapping the first adult stable fly. Therefore, it cannot be assumed that the young stable flies were
59 overwintering on the dairy farms, but rather that warm spring conditions led to more rapid development of immatures than in 2017, generating greater amounts of young female recruited into the population early in the season.
Most of the dissected females were nulliparous, which was expected (Beresford and
Sutcliffe 2012). Sticky traps caught more nulliparous females than uniparous or multiparous, which follows previous studies (Beresford and Sutcliffe 2006). The sticky traps likely catch more nulliparous females because they are virgins looking to mate with males, therefore they are attracted to encounter sites such as sticky traps where males are more numerous. Sticky trap catches are biased toward males, where more males are expected to be captured than females, potentially because the males are looking for encounter sites to gain a vantage point to inspect passing flies as mates or competitors (Buschman and Patterson 1981). Although more flies were caught in 2017, almost double the number of females were caught and dissected in 2018. This may be due to the greater number of young flies in the population throughout the year.
In 2017, greater numbers of older females were caught three times in the season: a smaller peak on 7 July, a large peak on 3 August, and at the very end of the season on October
24. These peaks indicate fewer young flies were emerging and being recruited into the population during these weeks. Juvenile development site characteristics were not measured at each site, but soil temperature and soil moisture measurements were taken by the St. Adolphe weather station. Three weeks prior to 7 July 2017, almost 28 mm of rain fell within 24-hours, followed by an additional 30 mm of rainfall sporadically leading up to 7 July 2017. Similarly, in the three weeks prior to 3 August 2017, 40.5 mm of rain fell sporadically. Large rain events may flood the juvenile development sites and limit oxygen availability in the soil (Rozendaal 1997), causing juveniles to die off in large numbers. At the end of the season, cooler temperatures
60 would lead to slower development of the larvae and fewer young flies emerging. In Nebraska during August, female stable flies appear to avoid oviposition even when presented with high quality developmental substrate (Friesen et al. 2016). The egg life stage has the lowest tolerance to cold injury, followed by larvae then pupae (Beerwinkle et al. 1978). An observed increase in mean female age near the last weeks of adult trapping in both years suggest decreased recruitment from the juvenile population. This may indicate that females may hypothetically allot time for egg development into larvae or pupae, increasing the chances of offspring survival at low temperatures.
Model Selection
The environmental variables that best predicted adult abundance in the 2017 model were soil moisture, humidity, and degree-days. For the year 2018, the best predictive variables in the model were soil temperature, soil moisture, and degree-days. The models chosen were acceptable in explaining which environmental variables influenced stable fly populations each year based on the low AIC scores of the final models; however, comparison of the residual deviance to the residual degrees of freedom suggests that overdispersion in both models is only partially dealt with and the rootograms indicate underfitting of zeros and overfitting of subsequent counts (Kleiber and Zeileis 2016). These attributes mean there are additional factors influencing fly abundance that are not accounted for, and the models are missing some variables that would explain the system in greater detail.
Soil temperature is key for juvenile flies, providing the necessary warmth for larvae and pupae to develop. Soil moisture is critical for developing stable flies, preventing desiccation and transporting nutrients and bacteria which larvae feed upon (Ranjard and Richaume 2001).
Average soil moisture was higher in 2018 than in 2017. There were no measurements taken of
61 the larval site characteristics, but larval development sites may have been often too wet in 2018, limiting juvenile survival based on the lack of oxygen available in water-logged soil (Rozendaal
1997).
The results from this study indicate that stable flies in Manitoba become active in June, peak in mid-July and, based on the population structure, stable flies did not overwinter at these dairy farms. As milder and shorter winters are expected, stable fly activity could begin earlier and last longer. The models indicated that the variables that best predicted adult abundance during the season were associated with adult flight and juvenile development, and while some of the variation in adult stable fly abundance is addressed, overall adult stable fly population dynamics on these farms may be better explained by measuring on-site variables associated with larval development habitat were not measured at the farm level. The explanatory variables retained in the model appear to impact the juvenile stages more directly than emerged adults. The presence of soil moisture and soil temperature as important predictors are an indication that adult abundance is largely driven by the quality of the larval environment and the survival of juveniles to the adult stage more than immigration of adults from surrounding areas. Understanding how environmental conditions affect larval habitat may be more informative and may provide information for control.
62 Table 2.1. Cross-correlation factors of environmental variables by site and year. Significant values above (0.500 or below -0.500) are starred. Daily degree-days were accumulated above 10°C for the week between trapping events. Cumulative degree-days were accumulated from 1 January of each study year.
Location Week Air Soil Soil Daily Cumulative Temperature Moisture Temperature Degree Days Degree Days 2017 Glenlea 0 0.277 -0.402 0.218 0.244 0.138 Glenlea 1 0.292 -0.308 0.307 0.312 -0.002 Glenlea 2 0.225 -0.232 0.266 0.230 -0.172 Glenlea 3 -0.001 0.099 0.104 -0.045 -0.284 Glenlea 4 0.119 0.112 0.151 0.105 -0.380 Red River 0 0.610* -0.517* 0.524* 0.582* -0.385 Red River 1 0.464 -0.401 0.399 0.440 -0.493 Red River 2 0.236 -0.157 0.216 0.189 -0.599* Red River 3 0.008 0.162 0.029 -0.057 -0.675* Red River 4 -0.107 0.292 -0.090 -0.192 -0.670* Vancrest 0 0.516* -0.409 0.451 0.500 -0.241 Vancrest 1 0.390 -0.316 0.322 0.372 -0.387 Vancrest 2 0.247 0.224 0.183 0.204 -0.535* Vancrest 3 0.034 0.072 0.035 -0.039 -0.610* Vancrest 4 -0.103 0.246 -0.088 -0.178 -0.610* 2018 Glenlea 0 0.220 -0.077 0.451 0.164 0.351 Glenlea 1 0.266 -0.016 0.513* 0.225 0.258 Glenlea 2 0.322 0.112 0.558* 0.321 0.155 Glenlea 3 0.327 0.220 0.523* 0.342 0.039 Glenlea 4 0.283 0.251 0.421 0.271 -0.072 Red River 0 0.434 0.174 0.637* 0.427 0.121 Red River 1 0.362 0.331 0.634* 0.348 0.025 Red River 2 0.353 0.443 0.621* 0.360 -0.077 Red River 3 0.316 0.493 0.538* 0.324 -0.185 Red River 4 0.274 0.450 0.402 0.277 -0.286 Vancrest 0 0.308 -0.099 0.481 0.250 0.283 Vancrest 1 0.298 0.029 0.509* 0.291 0.190 Vancrest 2 0.346 0.179 0.548* 0.356 0.082 Vancrest 3 0.267 0.326 0.465 0.291 -0.036 Vancrest 4 0.246 0.333 0.405 0.241 -0.119
63
Table 2.2. Negative binomial regression analysis of the relationship between the number of adult stable flies captured on sticky traps and significant environmental conditions on three southern Manitoba dairy farms in 2017 and 2018.
DF Estimate Std. Error z P 2xLog McFadden AIC Score Likelihood Pseudo R2
2017 Variables Intercept 322 -14.82 2.41 -6.14 8.18e-10 -3849.23 0.033 3861.2 Humidity 322 0.313 0.035 9.057 < 2.0e-16 (df=6) Soil moisture 322 0.451 0.068 6.601 4.09e-11 Degree days 322 0.001 0.0002 4.719 2.37E-06 Humidity*Soil moisture 322 -0.008 0.001 -7.917 2.42E-15
2018 Variables Intercept 374 -7.056 0.522 -13.51 < 2.0e-16 -3675.21 0.077 3685.2 Soil temperature 374 0.384 0.016 24.443 < 2.0e-16 (df=5) Soil moisture 374 0.035 0.007 4.784 4.09E-11 Degree days 374 0.004 0.0002 18.433 < 2.0e-16
64
Table 2.3. Monthly measurements of significant environmental conditions from St. Adolphe weather station in 2017 and 2018. Degree days were calculated from the first to last date of each month except October (until 24 October 2017 and 4 October 2018) to correspond to the end of trapping each year.
Month Avg. Air Total Avg. Relative Avg. Soil Avg. Soil Cumulated Degree Temperature (°C) Precipitation (mm) Humidity (%) Temperature (°C) Moisture (%) Days 2017 June 17.6 58.5 65.3 18.0 27.9 221.9 July 19.9 57.0 74.4 20.3 24.0 294.3 August 18.5 24.6 70.0 20.6 25.6 262.4 September 14.3 67.9 70.1 15.9 29.2 154.2 October 7.5 16.7 66.7 9.4 40.5 55.3
2018 May 15.4 33.5 51.0 12.1 40.4 203.7 June 19.6 69.1 65.2 18.9 40.8 278.6 July 20.2 69.2 72.7 21.4 41.1 305.0 August 19.3 34.6 67.7 18.2 23.8 288.9 September 11.8 70.7 73.9 13.1 25.3 110.1 October 2.2 22.2 78.2 6.3 41.5 1.7
65 Table 2.4. Monthly subsample totals of adult male and female stable flies removed from sticky traps on three southern Manitoba dairy farms in 2017 and 2018.
Month Males Females Total Sex Ratio (M:F)
2017 June 1111 193 1304 5.8 July 5574 941 6515 5.9 August 5810 1397 7207 4.2 September 2907 1494 4401 1.9 October 1654 885 2539 1.9
2018 May 0 0 0 0 June 1280 353 1633 3.6 July 3902 2429 6331 1.6 August 4447 3062 7509 1.5 September 2323 1521 3844 1.5 October 99 76 175 1.3
Table 2.5. Sweep net collected adult stable fly counts from three southern Manitoba dairy farms in 2018.
Month Males Females Total Sex Ratio (M:F)
May 0 0 0 0.0 June 168 119 287 1.4 July 233 127 360 1.8 August 263 188 451 1.4 September 191 168 359 1.1 October 11 23 34 0.5
66
Table 2.6. Adult female stable fly parity from three southern Manitoba dairy farms in 2017 and 2018.
Female Parity Status Location Nulliparous Uniparous Multiparous Total
Glenlea 1963 137 8 2108 2017 Red River 1145 61 3 1209 Vancrest 1472 87 4 1563
Glenlea 2523 83 6 2612 2018 Red River 2484 55 1 2540 Vancrest 2034 34 0 2068
67
Figure 2.1. Location of Manitoba Agriculture weather station and the three dairy farms used as stable fly population research sites in 2017 and 2018: Glenlea Research Station, Red River Holsteins, and Vancrest Holsteins.
68
a
b
Day of Year
Figure 2.2. Total number (± standard error) of adult stable flies captured on sticky traps per sampling date. Six traps were monitored for one 24-hour span every week on three dairy farms in southern Manitoba in 2017 (a) and 2018 (b).
69
Week
Glenlea Red River Vancrest
Figure 2.3. Mean abundance (± standard error) of adult stable flies captured weekly on three southern Manitoba dairy farms in 2017 and 2018. Error bars represent the standard error of the mean.
70
a
eek Glenlea
W Red River Vancrest
b
Glenlea eek Red River W Vancrest
Figure 2.4. Mean number (± standard error) of subsampled adult female (a) and male (b) stable flies per week per week in 2017 and 2018.
71
a
b
Figure 2.5. Mean (± standard error) weighted ages of female stable flies trapped across all sites during the trapping season for 2017 (a) and 2018 (b).
72
a
b
Figure 2.6. Mean (± standard error) weighted ages of female stable flies trapped at Glenlea Research Station dairy farm during the trapping season for 2017 (a) and 2018 (b).
73
a
b
Figure 2.7. Mean (± standard error) weighted ages of female stable flies trapped at Red River Holsteins dairy farm during the trapping season for 2017 (a) and 2018 (b).
74
a
b
Figure 2.8. Mean (± standard error) weighted ages of female stable flies trapped at Vancrest Holsteins dairy farm during the trapping season for 2017 (a) and 2018 (b). There is missing data in the space between Julian Date 278 to 293 in 2017, leading to the detached data point.
75 CHAPTER 3: WINTER SURVIVAL OF STABLE FLY STOMOXYS CALCITRANS (L.) (DIPTERA: MUSCIDAE) PUPAE AND THEIR PUPAL PARASITOIDS (HYMENOPTERA: ICHNEUMONIDAE, PTEROMALIDAE) UNDER LABORATORY AND FIELD CONDITIONS IN SOUTHERN MANITOBA
ABSTRACT
Stable fly pupae survival at low temperatures was examined in laboratory and field experiments. Approximately 3000 pupae were collected from substrate on dairy farms and divided among three temperature treatments in incubators. Pupae were removed from incubators every two weeks and after three weeks in a growth chamber at 26°C, were examined for adult emergence, intrapuparial development and parasitism. During the field component, field- collected pupae were placed into nylon stockings with soil and buried within three different substrates at each dairy farm. The stockings remained on-site for the duration of winter and were removed once the first adult stable flies were trapped and examined for adult emergence, intrapuparial development, and parasitism. The results of this study support the adult stable fly migratory hypothesis of recolonization and hint that stable fly pupae may develop under temperatures lower than the estimated developmental threshold.
INTRODUCTION
Stable flies Stomoxys calcitrans (L.) (Diptera: Muscidae) are blood-sucking dipterans that frequently inhabit livestock production areas. Cattle are particularly affected by stable flies, with adult fly bites causing major economic losses annually due to reduced feeding and resting by cattle, resulting in weight loss (Campbell et al. 2001; Taylor et al. 2012) and decreased milk production in dairy cows (Bruce and Decker 1958; Taylor et al. 2012).
76 As global climate change brings about warming temperatures, the resulting shorter and milder winters will lead to a longer period of stable fly activity (Gilles et al. 2008). In Canada, stable flies are a seasonal pest that can nonetheless cause considerable concern while active
(Lysyk 1993; Beresford and Sutcliffe 2009a). In Manitoba, the temperature extremes between the summer and winter months are typically much larger than in southern Alberta and southern
Ontario due to the lack of geographic buffers in Manitoba (Environment and Climate Change
Canada 2019). Most of the winter months in Manitoba are lethally cold to most insects, with over
100 days of the year with recorded temperatures below -10°C (Environment and Climate Change
Canada 2019). Cold injury poses a real and lethal threat to stable flies in Manitoba. Adult stable flies die if exposed to temperatures below 4°C for a few hours with no refuge (Jones and Kunz
1997). Stable flies maintain an internal temperature warm enough to avoid freezing by finding warm refugia when the temperature drops (Salt 1961). Two strategies to avoid freezing are migration and overwintering.
Stable Fly Migration
In North America, stable flies can disperse over long distances, transported to new locations at high altitudes by wind and storms (Wellington 1945; Hogsette and Ruff 1985; Isard et al. 2001). Marked stable flies transported by a high-pressure storm system were recaptured up to 225 km from the release site within 24 hours (Hogsette and Ruff 1985). As there is no scientific evidence supporting active migration, it is likely that stable flies recolonize northern areas each spring through passive movement north on weather fronts.
Stable Fly Overwintering
There is evidence that stable flies survive the winter months in southern and central regions of the United States, but limited evidence from Canada. There is no evidence that stable
77 flies overwinter in large numbers, but certain management practices and livestock housing on dairy and beef farms may provide suitable habitat to support some overwintering adult stable flies. In Canada, many adult stable flies were observed in barns in January and some in February in southern Ontario (Beresford and Sutcliffe 2009a). Stable flies locally overwinter in southern
Alberta, as indicated by adult stable fly presence on traps prior to noticeable biting activity
(Lysyk 1993), but which life stage overwinters is unclear. Overwintering sites must be warm enough to avoid freezing but cool enough to slow immature development until environmental conditions are suitable. Outdoor overwintering sites can have internal temperatures changing dramatically (Salt 1961). Cold temperatures slow intrapuparial development and most stable flies will not reach the adult stage in the late autumn in Canada. Overwintering research on stable flies of any life stage has not been conducted in the province of Manitoba.
Parasitoids of Stable Flies
Stable flies have a variety of predators and parasites that inhibit their development and affect their survival. In North America, muscoid flies such as stable flies, house flies (Musca domestica L.), face flies (Musca autumnalis De Geer), and horn flies (Haematobia irritans L.), are attacked mostly by parasitoid wasps in the family Pteromalidae, which target the pupal life stage (Greene et al. 1989). Commonly found pupal parasitoid genera within the North American
Pteromalidae include Spalangia Latreille, Muscidifurax Girault and Sanders, Urolepis Walker, and Nasonia Ashmead. Wasps in the family Ichneumonidae, especially the genus Phygadeuon
Gravenhörst, are also common stable fly pupal parasitoids in North America (McKay and
Galloway 1999).
Because pupal parasitoids develop within the puparium of their host, they are subjected to the same environmental conditions as the stable flies. In Canada, pupal parasitoids overwinter as
78 juveniles (Floate and Skovgård 2004). Pupal parasitoid studies are often conducted during the spring and summer months, when stable fly populations are at their peak, but published data on the survival of pupal parasitoids subjected to winter environmental conditions are lacking.
Understanding the overwintering survivorship and activity of pupal parasitoids, when stable fly populations are small, is important for creating a complete biological control program. In
Manitoba, there is little information on parasitoid wasps, and even less on stable fly pupal parasitoids specifically.
Goals of the Study
The main objective of this study was to assess the survival of stable fly pupae at low temperatures for various periods of time in conditions typical of Manitoba winters.
Complementary field and laboratory overwintering experiments were conducted to collect data from a temperature-controlled environment and a naturally-occurring environment. The laboratory component allowed access to pupae for dissections, providing information on intrapuparial development as experimental weeks progressed under three distinct temperature treatments. The concurrent field component provided information on intrapuparial development within three different outdoor substrate treatments on three dairy farms in southern Manitoba.
MATERIALS AND METHODS
Pupae Collection
Stable fly pupae were collected from soil-manure substrate on Vancrest Holsteins dairy farm on November 4, 2017. Suitable locations were examined for pupae using trowels and gloved hands to dig down to about 15 cm below the substrate surface. Suitable substrates examined were a mixture of approximately 50% hay-grain feed mixed with 50% cattle manure
79 directly beside a feed bunk within an outdoor heifer pen, as well as 90% straw bedding mixed with 10% cattle manure within an area reserved for calf hutches. Most of the pupae were found under a concrete feed bunk in an empty outdoor pen that contained approximately 40 to 50
Holstein heifers during the summer. The cattle were removed from the pen approximately 72 hours prior to pupal collection. Substrate containing pupae was placed into three 20-litre pails and transported back to the laboratory. Similar procedures were used again for pupal collection in 2018, with two differences: all pupae were collected from Glenlea Research Station farm, and pupae were collected on September 26, 2018.
Approximately 3000 muscid fly pupae were collected each year. In 2017, the pails containing the substrate and pupae were placed in 4°C storage for approximately one week until further processing; in 2018, the pupae were processed on the day of collection. Fly pupae were removed from the substrate using gloved hands and rinsed off in small batches with the help of sieves. Once rinsed and dried slightly on paper towels, the pupal spiracular plates (respiratory organs) were examined under a light microscope to confirm the species, where triangular and far apart spiracles identified stable flies, and round, close together spiracles identified house flies
(Cumming 2006; Friesen et al. 2015) which were discarded. Each year, 10 randomly selected stable fly pupae were placed into each of 10 Petri dishes lined with damp brown paper towel and placed into a 26°C growth chamber for three weeks to determine the viability and emergence of the collected pupae. In 2017, the pails containing substrate and pupae held at 4°C were exposed to ambient air temperatures of up to 44°C for a few hours following an incubator malfunction during storage.
80 Laboratory Preparation and Incubation
To determine the effect of overwintering temperature on survival, stable fly pupae were incubated at three different temperatures. Thirty stable fly pupae were counted and placed into each of 45 Petri dishes containing 20 mL of dry potting mix (Sunshine® Professional Growing
Mix #4 Peat/Perlite) to act as a sterile substrate for incubation and to prevent any pathogens within the natural substrate from interfering with the experiment (1350 pupae in total). The potting mix was sprayed with tap water a few times with a spray bottle until damp to prevent pupae desiccation, but not saturated, to prevent decomposition or fungal growth. The Petri dish lids were taped on and three randomly selected dishes were placed directly into a growth chamber set at 26°C as an experimental control for temperature. The remaining 42 Petri dishes were numbered and randomly assigned to each of three incubators set to 0, 2 and 5°C using a random number generator. The same random number generator was also used when removing the Petri dishes from incubation. These temperatures were chosen specifically, as subnivean
(beneath snow cover) environments reach and maintain temperatures of around 0°C to 2°C when snow cover reaches approximately 15 cm (Merriam et al.1983), and 5°C to factor in slightly warmer temperatures related to decomposition or fermentation of vegetation in most natural developmental sites. The experiment was repeated in 2018 and 48 Petri dishes were prepared and distributed to the same treatment groups (1440 pupae in total).
The 2°C and 5°C incubators were Percival fridge-type incubators (Percival #I-35VL,
Percival Scientific, Perry, IA); however, maintaining a Percival incubator at 0°C was not possible due to the potential for ice damage, so a 0°C incubator was constructed using a large, insulated expanded polystyrene foam box with Ziploc® sandwich bags filled with ice. The box was filled with its allotment of 15 Petri dishes and bags of ice, sealed with the lid, and placed
81 into a 4°C walk-in cooler. The ice was changed every second or third day and the temperature within the polystyrene incubator was monitored with a waterproof HOBO® U23 Pro v2
Temperature/Relative Humidity data logger (Onset Computer Corporation, Bourne, MA). There was a chamber failure on 21 October 2018 when the temperature in the 2°C Percival incubator holding pupae-filled Petri dishes increased to 38°C and could have been at that temperature for
24 to 48 hours, the time in between misting the Petri dishes.
The Petri dishes were left in their respective incubators for two weeks, the only disturbance being a heavy misting of water (approximately 25-30 sprays from a spray bottle) into the incubators every 24 to 48 hours to maintain the moisture levels in the Petri dishes. One dish from each of the three incubators was removed using the random number generator and the pupae were taken out and placed into 96-well plates. The well plates were then set in a 26°C growth chamber for three weeks to allow time for development at a near optimal temperature.
After the three weeks in the growth chamber, the plates were opened and examined under a light microscope for emerged adult flies or parasitoids. Intact pupae with no signs of emergence were dissected with forceps and the pupal stages within were categorized into three groups based on the developmental stage, loosely following Fraenkel and Bhaskaran (1973). Upon dissection, puparia full of amorphous white tissue without discernable features were described
“Undeveloped”, and may have contained either larval-apolysis stage or cryptocephalic stage flies. Puparia containing an individual pupa with discernable features but white in colour were described “Juvenile White” to correspond with the phanerocephalic pupa stage. The final stage of development within the puparium was called the “Juvenile Black” stage, which corresponds to the pharate stage with a black-coloured, fully formed fly that had yet to emerge. Different development category names were chosen rather than those presented in Fraenkel and Bhaskaran
82 (1973) because it was difficult to distinguish the larval-pupal apolysis and the cryptocephalic stages, as both are white amorphous masses of tissue which can only be determined under electron microscopes.
The presence of parasitoids within the puparia was recorded. Larval and pupal stages were labelled “Juvenile Parasitoids,” as they are difficult to identify without genetic testing.
Parasitoid eggs were never observed within pupae, either due to lack of presence or failure to distinguish eggs from stable fly pupal tissue. Emerged adults were identified using the Illustrated
Key to the Native and Introduced Chalcidoid Parasitoids of Filth Flies in America North of
Mexico (Gibson 2000), and the Key to Subfamilies of Holarctic and Neotropical Ichneumonidae
(Wahl 1993). Adult parasitoids were preserved in vials filled with 70% ethanol until they could be shipped to the Canadian Centre for DNA Barcoding (CCDB) in Guelph, Ontario for species confirmation.
Field Preparation and On-site Burial
A complementary field portion of the overwintering experiment was conducted by placing stable fly pupae in nylon stockings buried on dairy farms during the 2017-2018 winter, with the goal of providing insight into low temperature survival under naturally occurring conditions. Four hundred and fifty stable fly pupae were collected from under a cattle feed bunk on Vancrest Holsteins dairy farm and extracted from the substrate as described previously. Nine
100% nylon stockings (Joe Fresh™ reinforced pantyhose, Loblaw Companies Ltd., Brampton,
ON) were filled with 500 mL of potting mix (Sunshine® Professional Growing Mix #4
Peat/Perlite, Sun Gro Horticulture, Agawam MA) and 50 stable fly pupae. Nylon stockings were chosen because the mesh weave is fine enough to keep emerged stable fly adults and parasitoids inside, yet also allows for natural oxygen and moisture saturation.
83 Glenlea Research Station, Red River Holsteins, and Vancrest Holsteins were the three dairy farms chosen for this research project. At each farm, three pupae-filled nylon stockings were buried 10 cm (the approximate depth the pupae were originally found) below each of three substrate surfaces and covered back over with the same substrate. The stockings were buried in three consistent substrate types located near the cattle enclosures. At each dairy farm, one stocking was buried in substrate consisting of nearly 100% straw bedding, the second stocking was buried in a substrate that was approximately 80/20 cattle manure and soil respectively, and the third stocking was buried in substrate comprised of approximately 50/40/10 cattle manure, fine-particulate sand, and soil, respectively. The straw and sand substrates were chosen based on research outlining known filth fly developmental sites (straw - Schmidtmann 1991; sand -
Hogsette 1996) and based on observed substrate similarities among the three farms. The stockings were buried on 17 November 2017 and the burial locations were marked with poles and flagging tape for retrieval in spring. The stockings were left undisturbed until their retrieval on 6 June 2018, the date the first adult stable flies were recorded during sticky trap monitoring.
Upon retrieval, the stockings were brought to the laboratory and processed as described for the pupae removed from the incubators. All pupae were removed from the nylon stockings and placed into 96-well plates, separated by farm and substrate type. The well plates were then placed in a 26°C growth chamber for three weeks to allow time for development at a near optimal temperature. After the three weeks in the growth chamber, the plates were opened and examined under a light microscope for emerged adult flies or parasitoids. Intact pupae were dissected with forceps and the pupal stages within the puparia were categorized into three groups based on the developmental stage following Fraenkel and Bhaskaran (1973). The presence of
84 parasitoids was recorded, and adult wasps were identified and preserved in 70% ethanol until shipped to the CCDB as described previously.
Statistical Analyses
Adult stable fly emergence was expressed as proportion data for the initial 100-pupae viability experiment and as counts for subsequent analyses. Intrapuparial developmental stage data were expressed as count data, as were juvenile and adult parasitoid data. Data collected from the 0°C incubator after 13 February 2019 were removed from the analyses due to a six- week delay in processing between 13 February and 27 March. All statistical analyses were conducted using R Software Versions 4.0.0 - 4.0.3 (R Core Team 2020) and the following R packages: car (Fox and Weisberg 2019), countreg (Zeileis and Kleiber 2020; Zeileis et al. 2008;
Kleiber and Zeileis 2016), emmeans (Lenth 2020), MASS (Venables and Ripley 2002), multcomp (Hothorn et al. 2008), multcompView (Graves et al. 2019), plyr (Wickham 2011), pscl (Zeileis et al. 2008; Jackman 2020), and rcompanion (Mangiafico 2020).
A two proportions z-test was conducted on the 100 pupae initially removed to assess viability in both years to examine differences in adult emergence prior to starting the experiments. Adult emergence and pupal developmental stages from both the laboratory and field overwintering experiments were count data which followed non-normal distributions not improved through logarithmic or square root transformations; therefore tests allowing for non- normal distributions were employed. A Poisson regression analysis was conducted to determine if there was a relationship between the total number of adult stable flies that emerged from pupae held at different incubation temperatures (0, 2, and 5°C). A second Poisson regression analysis was conducted to examine relationships between the number of each pupal stage within each incubation temperature and experimental year. During the analyses, an ANOVA-like test called
85 Analysis of Deviance was used to further assess changes in deviance from the null regression hypotheses and if those changes were significant. A Chi-square test of independence and post- hoc pairwise Fisher’s exact tests were used to determine if there were differences in the abundance of each pupal development stage, and those which were parasitized, in each of the three field overwintering sites and substrates. A second Chi-square test of independence and post-hoc pairwise Fisher’s exact tests were used to determine if there were differences in the abundance of pupal development stages, including those parasitized, between laboratory and field studies in 2017.
RESULTS
Laboratory Overwintering Experiment
In 2017, the 100 randomly selected pupae used to determine viability resulted in an average emergence of 33.5% ± 2.5% recorded for each of the 10 Petri dishes. The 100 pupae were removed from the substrate collection pails prior to the temperature malfunction in the walk-in storage incubator and therefore provided an accurate emergence baseline for the project.
A total of 1344 out of 1350 stable fly pupae were dissected for the laboratory experiment (Figure
3.1a). Four of the pupae were damaged and therefore were not dissected. During the control week where none of the 90 pupae underwent any temperature treatments (week 0), one emerged
(0.01%), eight were undeveloped (8.9%), 13 were juvenile white (14.3%), 56 were juvenile black (62.2%) and 12 were parasitized (13.3%).
Overall, two adults partially emerged where their heads had fully emerged from the cephalic cap of the puparium, but the thorax and abdomen remained trapped inside. These two flies were considered to have emerged. One adult stable fly emerged during the control week on
86 24 November 2017 within the 96-well plate in the 26°C growth chamber. The second adult had emerged then died within its Petri dish in the 5°C incubator. It is unknown when this second adult emerged during the overwintering experiment as it was still covered by a layer of substrate, but the dish was randomly selected for dissections on 11 May 2018, therefore the emergence was considered as part of experimental week 24 dissections even if the fly likely emerged before that time. Over half (54.4%) of the pupal dissections revealed pharate (i.e., fully formed) flies within the puparia (Figure 3.1a). Juvenile white flies made up 26% of the dissections and undeveloped flies comprised 7.1%.
Emergence results from the initial viability test were similar in 2018, where the 100 randomly selected pupae resulted in an average emergence of 31.5% ± 3.5% recorded for each of the 10 Petri dishes. There was no difference in the proportions of emerged adults from the initial viability experiments (2 = 0.087, df = 1, P = 0.768). A total of 1442 stable fly pupae were dissected for the second year of the laboratory overwintering experiment (Figure 3.2a). Similar to 2017, juvenile black pupal stage was the most common developmental stage observed, comprising 46% of the total pupal dissections for the year. Juvenile white pupae made up 25.5% of the dissections. There were 130 undeveloped flies and 261 adult flies emerged.
Of the 261 flies that emerged in 2018, 133 were females and 117 were males (Table 3.1).
Six adults that emerged were damaged and unable to be sexed. Four females and one male that emerged during the first week (used as a control) were removed from the summary as they did not undergo any temperature treatment. The number of adults that emerged differed between the three incubator temperatures (2 = 19.11, df =2, P = 0.001). A post-hoc Sidak’s t-test revealed that significantly more flies emerged from pupae held at 5°C (Padj = 0.016) compared to pupae held at 0°C. Adult emergence steadily increased until the maximum emergence at experiment
87 week 16 (16 January 2019; Table 3.2), followed by a decline. On two separate occasions in the
5°C incubator, an adult stable fly was found fully emerged and alive within its Petri dish prior to the dish’s selection for dissections. Because the Petri dishes were randomly selected for dissections, the Petri dishes with the emerged adults were marked with the date of observed emergence and examined externally for sex. One male was observed on 8 October 2018, and the dish was selected for dissections on 5 December 2018. One female was observed on 9 January
2019 and was selected for dissections on 10 April 2019.
There were significant differences in the number of each pupal developmental stage recorded among the three incubator temperatures during the two years of laboratory experiments
(Table 3.3; 2 = 30.27, df = 5, P < 0.001). More undeveloped stage pupae were observed from
0°C incubator in 2018 (Padj =0.021) than the 0°C incubator in 2017, while juvenile white stage pupae were observed more frequently in the 0°C incubator in 2018 (Padj = 0.001). Significantly more juvenile white stage pupae were observed in 2017 from both the 2°C (Padj = 0.024) and
5°C (Padj < 0.001) incubators. Higher numbers of juvenile black stage pupae were recovered from the 0°C incubator in 2017 (Padj < 0.001) than in 2018. Figures 3.1c and 3.1d show that, as experimental weeks progressed, the number of undeveloped pupae decreased in the 2°C and 5°C treatments and the number of pupae in the juvenile white stage increased. No undeveloped pupae were recovered from week 14 to week 26 in the 2°C treatment. This trend of increased advanced development was also observed in the 0°C treatment (Figure 3.1b) with fewer undeveloped pupae as the experiment progressed and an increase in the number of juvenile white and black pupae. Similar trends were observed in 2018 within the 0°C and 2°C incubators (Figures 3.2b and 3.2c), where the number of pupae in more advanced intrapuparial development increased and the number of undeveloped pupae decreased as experiment weeks progressed.
88 A total of 166 stable fly pupae (12.35%) were parasitized in 2017. Of the 166 pupal parasitoids found, 34 were adults and were identified using morphological keys. The remaining
132 were larval and pupal stages of parasitoids that could not be identified with morphological keys (Table 3.4). Spalangia was the most commonly found genus and Spalangia cameroni
Perkins as the most common species. Other parasitoid species identified, in order of most to least numerous, were Spalangia nigra Latreille, Urolepis rufipes (Ashmead), Nasonia vitripennis,
Spalangia endius Walker, and Phygadeuon fumator (Table 3.4). Only 18 parasitoids were recovered from the 2018 pupae, compared to 166 in 2017. Of the 18 parasitoids only five were adults and were all identified as Phygadeuon fumator; the remainder were juvenile stages of unknown species. The adult parasitoid identification results from the genetic analysis conducted by the CCDB did not yield further information. Of 78 adult parasitoids sent for identification, 19 individuals that were identified only to family level were uploaded to the Barcode of Life Data
System (BOLD), 39 individuals were contaminated with Wolbachia and could not be identified without further processing (i.e. failure tracking on the plate for amplifying insect DNA over
Wolbachia), and the remaining 20 failed at the PCR amplification stage.
Field Overwintering Experiment 2017-2018
Nylon stockings were removed from the field and the stable fly pupae they contained were dissected. Of 450 pupae that were left to overwinter, 445 were recovered. No adult stable flies had emerged over the course of the winter (Table 3.5). Although there were no differences in intrapuparial development among the three farm locations (Table 3.6; 2 = 12.99, df = 6, P =
0.052), there were significant differences in the number of juvenile stages recovered among the straw and sand-soil substrate types (2 = 12.89, df = 6, P = 0.04). Marginally fewer juvenile white stages were found in straw (Padj = 0.048) and more pupae were parasitized (Padj = 0.019).
89 There was no difference with manure-soil (Padj = 0.232). Undeveloped pupae comprised 3.8% of field pupal dissection results, 44.9% were juvenile white, 45.4% were juvenile black and 5.8% were parasitized (Table 3.5).
Significant differences were observed in the abundance of juvenile stages between the laboratory and field overwintering experiments of 2017 (Table 3.5; 2 =62.90, df = 3, P <
0.001). The proportion of juvenile white stage pupae recovered from the field overwintering experiment was significantly higher (Padj < 0.001) and parasitoids were recovered in significantly higher proportions from the laboratory overwintering experiment (Padj < 0.001).
There were 26 parasitoids recovered from the field overwintered pupae. One adult emerged after three weeks in the 26°C growth chamber and was identified as Spalangia cameroni. The remaining 25 were unidentified juveniles (larval and pupal stages).
DISCUSSION
Stable fly activity begins in early summer, and as temperatures increase, population sizes grow rapidly. As our global climate shifts, environmental conditions may favour an earlier onset of activity and later extirpation date, creating a longer stable fly season. An essential aspect of managing stable fly population growth in Canada, as a temperate zone, is understanding how seasonal environmental changes influence stable flies. In the literature there is evidence supporting two hypotheses of recolonization by stable flies: overwintering and passive migration.
The objective for these experiments was to determine if stable fly pupae could survive both the approximate duration and subnivean temperatures of southern Manitoba winters to emerge as viable adults in spring by conducting laboratory and field overwintering experiments.
90 Adult Stable Fly Emergence
Significantly more adult flies emerged from pupae during the laboratory experiment in
2018. Based on the initial viability tests, we could have expected approximately 30% adult emergence from the pupae collected for the experiments. Overall, adult emergence was less than
1% in 2017 and 18.1% in 2018. The sex ratio upon emergence was almost exactly 1:1 (54% female, 46% male), which is the expected result from wild stable fly populations (Beresford and
Sutcliffe 2012).
There were some procedural differences that could explain the discrepancy in adult emergence between the two years of the experiment. First and foremost, the pupae collected in
2017 were stored at 4°C before being allocated to their respective treatments, whereas the pupae collected in 2018 were immediately processed. Storage at low temperatures may have affected the survival of the pupae but it is unlikely the impact was significant considering the pupae remained in the substrate, offering some insulation similar to outdoor conditions. However, the cold room malfunction that brought ambient air temperatures up to 44°C for as long as 24 hours conceivably affected the pupae. The fact that the viability tests were not significantly different between years would support that hypothesis.
Developmental thresholds for stable fly development have been calculated and different values have been presented: 10.7°C (Lysyk 1998), 11.4°C (Florez-Cuadros et al. 2019), and
11.8°C (Larsen and Thomsen 1940). These values are rounded down to approximately 10°C when calculating degree-days. Theoretically, the DD10 calculation suggests that stable fly juvenile development should not occur if ambient temperatures fall below 10°C. Contrary to this, two adults were found alive within the 5°C incubators: one male on 8 October 2018 and one female on 9 January 2019. Minor fluctuations in temperature occurred during misting, but the
91 incubator doors were kept open as little as possible in order to prevent the temperatures from increasing. During this study year, there were no malfunctions prior to incubation or within the
5°C incubator and temperature was consistent, suggesting that development may take place in cooler temperatures than 10°C, albeit very slowly.
In 2018, more adults emerged from pupae that were held at 5°C than at 0°C during the laboratory experiment, which was expected given that very low temperatures (below 0°C) greatly increase both development time and pupal mortality (Beerwinkle et al. 1978). With that in mind, we also expected a higher proportion of adult emergence at the beginning of the incubation period and a decline as the experiment progressed; however, we observed the opposite trend, where the greatest proportion of adults emerged within dishes removed from incubators at weeks 12 and 16 with 33.3% and 41.1% adult emergence, respectively. The 5°C treatment was chosen to mimic heat released from decomposing vegetation present in most known natural developmental sites (Todd 1964; Foil and Hogsette 1994; Talley et al. 2009;
Friesen et al. 2016), although these studies did not research overwintering or survival under cold environmental conditions. The temperature treatment of 5°C was also chosen with the assumption that stable fly development would not occur under 10°C and all emergence would occur after time within the 26°C growth chamber or not at all; however, based on the two separate incidences of adult emergence in the 5°C prior to Petri dish removal, it is possible that
5°C provides stable fly pupae with enough heat to slowly develop.
Intrapuparial Development
Significant differences were observed in the number of each juvenile stage within incubation temperature treatments between years. The timing of pupae collection may explain some of these differences in development. We collected pupae on 4 November 2017 and 26
92 September 2018; a difference of approximately six weeks. The pupae that were collected earlier
(2018) had less time to develop within the natural substrate than those collected later. There were, overall, more undeveloped and white stage pupae recorded from the 0°C treatment in 2018 compared to 2017, probably because these younger pupae were then placed into the 0°C temperature treatment, slowing development even further. Along the same lines, there were more pharate pupae (juvenile black stage) in the 0°C treatment in 2017, when the pupae were collected later. The greater proportion of juvenile black pupae may also be attributed to the lack of adult emergence in 2017, whereas there was 13.6% adult emergence from pupae held in the 0°C treatment in 2018.
There were significantly more pupae in the white stage recorded from the 2°C and 5°C treatments in 2017. These results would suggest that less development occurred compared to
2018. In 2018, development in the 2°C incubator may have been accelerated due to the temperature increase up 38°C for up to 48 hours; however, the 5°C incubator suffered no malfunction and the same trend was observed. In 2017, there was an increase in the proportion of white pupae as undeveloped pupae decreased over time at 2°C and 5°C, but the proportion of black pupae did not change. Something occurred that inhibited adult emergence, but also the transition from juvenile white to juvenile black. In contrast, the proportion of juvenile white pupae decreased over time in 2018 while the proportion of juvenile black pupae increased, and adult flies emerged.
The chamber malfunction that increased ambient air temperatures up to 44°C for up to 24 hours in 2017 conceivably could have led to pupal mortality; however, the consistent change in development stages observed among the pupae suggests that some pupae survived but, for some reason, failed to emerge. The same pattern occurred the following year except some flies
93 emerged successfully, suggesting that future studies should investigate intrapuparial development at temperatures below the calculated developmental threshold of 10°C.
Proportions of pupal development stages between the laboratory and field experiments were similar, with the exception of juvenile white stage and parasitoid abundance. Because pupae were inaccessible during the field overwintering experiment, it is difficult to determine how the pupae were developing relative to the lab experiment. However, there were more white stage pupae recovered from the field, and no adults emerged which suggests that overall, less development occurred in the field.
Although adult stable fly emergence was not recorded from field overwintered pupae, relationships between the pupal development stage and substrate type were observed. The nylon stockings placed in straw and manure-soil bedding had a significantly larger proportion of pupae in advanced stages of development compared to those overwintered in sand-soil substrate, which contained a larger proportion of less developed pupae. Straw and manure-soil bedding may produce more exothermic reactions and remain at warmer temperatures longer than a sand-soil substrate, leading to increased development for the pupae placed in these locations. Immature stable flies are abundant in straw substrate laid on packed soil, indicating suitability for oviposition and juvenile development (Schmidtmann 1991).
Stable Fly Pupal Parasitism
Five adult individuals of Phygadeuon fumator were the only parasitoids that emerged in
2018. Phygadeuon fumator is common across Canada and it copes well with cool weather
(Mason and Gillespie 2013). Phygadeuon fumator was recorded from almost 80% wild-collected pupae twenty years ago at dairy farms in Manitoba (McKay and Galloway 1999). Therefore, we expected to find Phygadeuon fumator at higher abundance than other species of parasitoids, but
94 this was not the case in 2017 when Phygadeuon fumator was only recorded one time. Spalangia, a cosmopolitan parasitoid genus (Moon et al. 1982), was the most numerous parasitoid recovered from pupae in 2017. Other North American studies of parasitoid diversity on dairy farms from
Florida (Romero et al. 2010) and California (Meyer et al. 1990) also found the majority of pupal parasitoids to be Spalangia species.
In the field overwintering experiment, more parasitoids were recovered from pupae overwintered in straw bedding. Given that the pupae were parasitized prior to their random allotment into nylon bags, this may have been random chance. Conversely, straw may have supported parasitoid development as it seemed to favour stable fly pupae development.
Parasitoid wasp eggs are approximately 0.3-0.4 mm long, and were never observed within stable fly pupae, either because they were absent or because they could not be distinguished from pupal tissue. If straw bedding provides better insulation from cold temperatures, parasitoids may have developed from eggs to the more clearly visible larval and pupal stage, thereby increasing the number of observations of parasitoids in pupae overwintered in straw.
There were 144 immature parasitoids in total from both the laboratory and field that remained unidentified. It is possible that the actual species composition is different from what was recorded using only the adult parasitoid data. Further assessment of juvenile parasitoid stages would provide a more accurate description of species diversity in southern Manitoba.
Surprisingly, one Spalangia cameroni that was buried in the stocking on Glenlea within the manure-soil substrate was the only adult parasitoid wasp to emerge, surviving the entire 30- week duration of the experiment from 17 November 2017 to 6 June 2018. Floate and Skovgård
(2004) classified Spalangia cameroni as a Category 1 parasitoid species, meaning it has very low cold-hardiness and usually dies early in winter. The manure-soil bedding creates an exothermic
95 reaction during decomposition and may have provided a suitable the microhabitat for survival and development. Spalangia cameroni burrow deeper into substrate than other parasitoid wasp species to parasitize fly pupae (Rueda and Axtell 1985), assisting Spalangia cameroni in avoiding competition (Smith and Rutz 1991) and potentially increasing winter survival (Floate and Skovgård 2004). The emergence of this parasitoid suggests that at least some stable fly pupal parasitoids are able to survive the full duration of Manitoba winters in a natural setting.
This overwintering study, the first of its kind in Manitoba, provides information on the potential for pupal stable fly overwintering in a natural outdoor environment, and, serendipitously, pupal parasitoid diversity and abundance within a small region of 5 square km.
In the next few decades, as winters become milder and shorter in Manitoba, there is the potential for large numbers of stable fly pupae to survive winter in microhabitats that provide a warm enough temperature for slow development until spring. This is supported by the fact that adult stable flies emerged from pupae held at low temperatures for many consecutive weeks in laboratory studies. The results of this study created questions regarding pupal development below 10°C that should be investigated. If there is development at temperatures as low as 2°C or
5°C, milder winters may mean that some of the pupae observed on sites in fall will survive to emerge in spring.
96
Table 3.1. Emergence of adult stable flies by sex and incubation temperature from 2018 laboratory overwintered pupae. 4 females and 1 male that emerged during the first control week were removed from the summary as they did not undergo any temperature effects. Six adults were damaged and unable to be sexed.
Incubator Temperature Females Males n (%) n (%) 0°C 21 (15.8) 24 (20.5%)
2°C 54 (40.6) 34 (29.1%)
5°C 58 (43.6) 59 (50.4%)
Total 133 117
97 Table 3.2. Adult stable fly emergence from pupae during laboratory overwintering experiment from October 2018 to May 2019. Stable fly pupae were held in Petri dishes of potting soil and maintained at a constant temperature of either 0°C, 2°C or 5°C. Petri dishes containing pupae were randomly chosen for removal from incubators at two-week intervals. Pupae dishes were transferred to a 26°C growth chamber for another two weeks then examined for adult emergence.
Incubator Temperature Removal Date 0°C 2°C 5°C Emerged Adults 10 Oct 2018 0 1 8 9 24 Oct 2018 2 1 3 6 7 Nov 2018 1 6 6 13 21 Nov 2018 4 7 6 17 05 Dec 2018 2 9 13 24 19 Dec 2018 7 12 11 30 02 Jan 2019 4 2 13 19 16 Jan 2019 10 11 16 37 30 Jan 2019 4 7 4 15 13 Feb 2019 2 8 10 20 27 Mar 2019a 3 0 6 9 10 Apr 2019 3 13 11 27 24 Apr 2019 3 6 6 15 08 May 2019 1 7 6 14 22 May 2019 0 0 1 1 a Six week delay between 13 February 2018 and 27 March 2018 removal.
98
Table 3.3. Stable fly intrapuparial development stages observed from intact, unparasitized pupae taken from three incubator temperatures during 2017 and 2018 laboratory overwintering experiments. The 90 pupae from “week 0” controls were not included from either year as the pupae did not undergo incubator temperature treatments.
Incubator Developmental Stage Temperature (n) Undeveloped Juvenile White Juvenile Black Emerged 0°C (365) 32 88 245 0 2017 2°C (376) 27 132 217 0 5°C (362) 30 116 214 2 0°C (330a) 61 120 104 45 2018 2°C (442) 29 93 232 88 5°C (445) 15 64 249 117 a Six-week delay between 13 February and 27 March led to removal of 0°C data subsequent to 13 February.
99
Table 3.4. Number of adult parasitoids by species collected from laboratory overwintered stable fly pupae from 2017-2018 and 2018- 2019 experiments.
Parasitoid Species Spalangia Spalangia Spalangia Urolepis Phygadeuon Nasonia Juvenile Total cameroni nigra endius rufipes fumator vitripennis Parasitoids 2017 21 5 1 4 1 2 132 166 2018 0 0 0 0 5 0 12 17
100
Table 3.5. Percentage of stable fly pupae at different stages of development during laboratory and field overwintering experiments in 2017. Data expressed as percentages of the totals: 1195 pupae from laboratory and 445 pupae from field.
Stage of Pupal Development
Undeveloped Juvenile White Juvenile Black Emerged Parasitized Laboratory 7.1 26.1 54.4 0.1 12.3
Field 3.8 44.9 45.5 0.0 5.8
101 Table 3.6. Dissections of stable fly pupae overwintered on three dairy farms in southern Manitoba near Winnipeg. Pupae were buried on 17 November 2017 and retrieved from their sites on 16 June 2018.
Location Substrate Type Undevelopeda Juvenile Juvenile Emerged Parasitized a a White Black
Glenlea Straw Bedding 2 14 31 0 3
Glenlea Manure-Soil 5 14 27 0 3*
Glenlea Sand-Soil 3 26 19 0 2
Red River Straw Bedding 0 25 22 0 3
Red River Manure-Soil 1 26 21 0 2
Red River Sand-Soil 2 24 18 0 1
Vancrest Straw Bedding 0 21 21 0 8
Vancrest Manure-Soil 1 25 23 0 2
Vancrest Sand-Soil 3 25 20 0 2
a Developmental stages of stable fly pupae were assessed and categorized based on body development and colour according to Fraenkel and Bhaskaran (1973). * The single adult parasitoid that emerged from Glenlea manure-soil was identified as Spalangia cameroni using the Illustrated Key to the Native and Introduced Chalcidoid Parasitoids of Filth Flies in America North of Mexico (Gibson 2000).
102
100
90
80
70
60
50
40
Number of Pupae Numberof 30
20
10
0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 Experimental Week
Undeveloped Juvenile White Juvenile Black Emerged Parasitized
Figure 3.1a. Dissections of stable fly pupae overwintered in the University of Manitoba’s Veterinary Entomology laboratory from 24 November 2017 to 8 June 2018. Developmental stages were assessed and categorized based on body development and colour according to Fraenkel and Bhaskaran (1973).
103
35
30
25
20
15 Number of Pupae Numberof 10
5
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 Experimental Week
Undeveloped Juvenile White Juvenile Black Emerged Parasitized
Figure 3.1b. Dissections of stable fly pupae overwintered in an incubator set at 0°C in the University of Manitoba’s Veterinary Entomology laboratory from 24 November 2017 to 8 June 2018. Data from the control week (experimental week 0) was removed as the pupae did not undergo any temperature treatments. Developmental stages were assessed and categorized based on body development and colour according to Fraenkel and Bhaskaran (1973).
104
35
30
25
20
15 Number of Pupae Numberof 10
5
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 Experimental Week
Undeveloped Juvenile White Juvenile Black Emerged Parasitized
Figure 3.1c. Dissections of stable fly pupae overwintered in an incubator set at 2°C in the University of Manitoba’s Veterinary Entomology laboratory from 24 November 2017 to 8 June 2018. Data from the control week (experimental week 0) was removed as the pupae did not undergo any temperature treatments. Developmental stages were assessed and categorized based on body development and colour according to Fraenkel and Bhaskaran (1973).
105
35
30
25
20
15 Number of Pupae Numberof 10
5
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 Experimental Week
Undeveloped Juvenile White Juvenile Black Emerged Parasitized
Figure 3.1d. Dissections of stable fly pupae overwintered in an incubator set at 5°C in the University of Manitoba’s Veterinary Entomology laboratory from 24 November 2017 to 8 June 2018. Data from the control week (experimental week 0) was removed as the pupae did not undergo any temperature treatments. Developmental stages were assessed and categorized based on body development and colour according to Fraenkel and Bhaskaran (1973).
106
100
90
80
70
60
50
40
Number of Pupae Numberof 30
20
10
0 0 2 4 6 8 10 12 14 16 18 20 26 28 30 32 34 Experimental Week
Undeveloped Juvenile White Juvenile Black Emerged Parasitized
Figure 3.2a. Dissections of stable fly pupae overwintered in the University of Manitoba’s Veterinary Entomology laboratory from 26 September 2018 to 22 May 2019. Developmental stages were assessed and categorized based on body development and colour according to Fraenkel and Bhaskaran (1973). Dissections between weeks 20 and 26 are six weeks delayed (Week 20 is 13 February 2019 and week 26 is 27 March 2019). The 0°C incubator was not maintained during the delay thus the temperature did not remain at 0°C. All pupae that were held in the 0°C incubator after week 20 were removed from analyses due to the temperature increase, therefore only pupal data from 2°C and 5°C are present from week 26 and subsequent weeks.
107
35
30
25
20
15 Number of Pupae Numberof 10
5
0 2 4 6 8 10 12 14 16 18 20 Experimental Week
Undeveloped Juvenile White Juvenile Black Emerged Parasitized
Figure 3.2b. Dissections of stable fly pupae overwintered in an incubator initially set at 0°C in the University of Manitoba’s Veterinary Entomology laboratory from 26 September 2018 to 22 May 2019. Developmental stages were assessed and categorized based on body development and colour according to Fraenkel and Bhaskaran (1973). Subsequent to week 20, there was a six- week delay in dissections. The 0°C incubator was not maintained during the delay thus the temperature did not remain at 0°C and the data were excluded from the results.
108
35
30
25
20
15 Number of Pupae Numberof 10
5
0 2 4 6 8 10 12 14 16 18 20 26 28 30 32 34 Experimental Week
Undeveloped Juvenile White Juvenile Black Emerged Parasitized
Figure 3.2c. Dissections of stable fly pupae overwintered in an incubator set at 2°C in the University of Manitoba’s Veterinary Entomology laboratory from 26 September 2018 to 22 May 2019. Developmental stages were assessed and categorized based on body development and colour according to Fraenkel and Bhaskaran (1973). There was a chamber failure on experimental week 2 when the temperature in the 2°C Percival incubator holding pupae-filled Petri dishes increased to 38°C and could have been at that temperature for 24 to 48 hours, the time in between misting the Petri dishes. Subsequent to week 20, there was a six-week delay in dissections, hence the shift in experimental week order.
109
35
30
25
20
15 Number of Pupae Numberof 10
5
0 2 4 6 8 10 12 14 16 18 20 26 28 30 32 34 Experimental Week
Undeveloped Juvenile White Juvenile Black Emerged Parasitized
Figure 3.2d. Dissections of stable fly pupae overwintered in an incubator set at 5°C in the University of Manitoba’s Veterinary Entomology laboratory from 26 September 2018 to 22 May 2019. Developmental stages were assessed and categorized based on body development and colour according to Fraenkel and Bhaskaran (1973). Subsequent to week 20, there was a six week delay in dissections, hence the shift in experimental week order.
110 CHAPTER 4: GENERAL DISCUSSION
Insights
The factors that influence stable fly population dynamics are of interest to both scientists and producers alike due to the profound, global impact stable flies have on agricultural economics and animal welfare (Campbell et al. 1987; Campbell et al. 2001; Taylor et al. 2012;
Taylor et al. 2017). Temperature and precipitation have been described as the key environmental drivers of stable fly population dynamics in Nebraska (Taylor et al. 2017), yet the applications of this information may not accurately predict population changes in other geographic regions based on the local climate. The global climate is changing and the need for understanding stable fly population changes based on regional environmental factors is pressing, as increased warming may lead to a longer duration of stable fly activity. Assessing changes in stable fly populations based on the climates of additional geographic regions will improve predictive models and enhance their applicability.
Stable fly colonization occurs either via passive migration (Hogsette and Ruff 1985) or overwintering (Lysyk 1993; Beresford and Sutcliffe 2009a) and passive migration is more probable in Manitoba based on the findings from this study. It is evident that stable flies are present in Canada from late spring to early fall and reach detrimental numbers mid-summer but there is little research explaining the factors enabling such large populations sizes, and the method allowing stable flies to recolonize Canadian sites each year. Chapters two and three of this thesis examined how changes in environmental conditions over all four seasons in southern
Manitoba impact stable fly populations and their pupal parasitoids.
The research in chapter two focused on examining the link between spring-to-autumn environmental conditions and adult stable fly population dynamics in southern Manitoba. This is
111 one of the first studies examining stable fly population dynamics based on environmental factors in Manitoba. Using negative binomial regression models containing weekly adult stable fly trapping data and local environmental variables, the results of this study support previous research describing temperature and precipitation as key factors influencing stable fly population growth (Taylor et al. 2017). Unexpectedly, soil temperature, relative humidity, and soil moisture were important predictive variables in my models, and including air temperature and precipitation values into these models did not improve the explanatory capabilities. The retained variables are related to air temperature and precipitation, yet it appears that specific environmental predictors of stable fly populations may vary slightly among years, as the model variables were not the same between years. This may be due in part to the inconsistent nature of the juvenile habitat, where the quality and composition of the habitat changes over time and under different conditions (Talley et al. 2009). Juvenile recruitment into the adult population is difficult to determine in this study as we did not examine juvenile habitat characteristics, though it is straightforward to assume that high quality developmental substrate increases the survival of juvenile stable flies, leading to a larger adult population. Similar to Khumalo and Galloway
(1996), stable fly populations on the farms in this study were unimodally distributed and populations were largest near cattle confinement in mid-to-late July. A sex ratio of 2 males:1 female was expected for trap collected adults and 1 male:1 female was expected for sweep-net collections, but the number of males collected from both trapped and sweep-net collected adults were surprisingly high. Taylor et al. (2010) found that males and females dispersed equally in both distance and direction from juvenile development sites, but it is understood that on a smaller scale, males move more often than females as a result of mate searching and deterring male
112 competitors (Buschman and Patterson 1981). Sex ratios provide information on the potential for population growth.
Aging techniques using ovarian relics provided information on the female age structure of stable fly populations in southern Manitoba, and this data was extrapolated to both sexes, as it is difficult to age wild male stable flies. Female age structure varied between the two years and among sites. In 2017, a higher proportion of older females (i.e. more ovarian relics from oviposition events) captured earlier in the year provides evidence supporting the passive movement “migration” hypothesis, where adults emerged from southerly sites are transported north via winds from high pressure systems. In 2018, the mean female age consisted mainly of younger females early and throughout the year. A similar number of spring degree days accumulated in both years at the time the first adults were captured on sticky traps (274 degree- days on 16 June 2017; 266 degree-days on 7 June 2018), suggesting that factors other than air temperature and degree days are driving the differences in older early season females in 2017.
Information on the male age structure was lacking from this study, and it is possible that females and males have different life history traits regarding aging and mortality. Although no sex-based differences in longevity were observed under controlled laboratory conditions (Salem et al.
2012), theoretically, male stable flies in the wild may undergo higher mortality rates and shorter longevity based on their tendencies of increased movement and engaging in risky aggressive behaviours.
The research presented in chapter three addressed the winter survival capabilities of stable fly pupae and their pupal parasitoids in both laboratory and field settings. These results led to the conclusions that stable fly pupae do not survive winter (approximately 27 to 30 weeks below the approximate developmental threshold of 10°C) in southern Manitoba based on the lack
113 of adult emergence from pupae buried outside in the field study, and the emergence of a single adult during the last week of laboratory experiments on 22 May 2019. Therefore, it is more likely that currently, the majority of stable flies arrive in spring by passive movements via northward storm systems. Research on northward movements by stable flies has not been examined in
North America since Hogsette and Ruff (1985) and has not been conducted in Canada.
Additionally, there have been few overwintering studies conducted in Canada. Adult stable flies are overwintering in sheltered reservoir populations, based on research from southern Ontario
(Beresford and Sutcliffe 2009a). As there is the potential for stable fly overwintering in southern
Manitoba within the next few decades if the climate continues to warm, the physiological and behavioural requirements for successful stable fly overwintering in Canada should be examined prior to this occurrence.
Some parasitoids within overwintered stable fly pupae successfully completed their development. This may be due to the biology of pupal parasites, where parasitoids overwinter as larvae or pupae within the fly puparium (Floate and Skovgård 2004) and are known to possess cold hardiness (Petersen and Meyer 1983) and diapause physiology (Legner and Gerling 1967;
DeLoof et al. 1979), unlike the stable fly host. A single individual of Spalangia cameroni emerged as an adult from stable fly pupae buried at the Glenlea Research Station farm in 2017.
After seven months within manure-soil substrate, this is a surprising result, given that Spalangia cameroni are thought to have low cold tolerance. Spalangia cameroni burrow at least 5–10 cm under the ground surface to parasitize fly pupae (Rueda and Axtell 1985). By digging deeper into the substrate, Spalangia cameroni are thought to avoid competition from other parasitoids (Smith and Rutz 1991) and concurrently increase winter survival through protection from cold temperatures (Floate and Skovgård 2004). The Canadian parasitoid monitoring studies conducted
114 two decades ago (Lysyk 1995; McKay and Galloway 1999; Floate et al. 2000) recorded notably low Spalangia cameroni abundance yet Spalangia species were the most numerous adult parasitoid wasps recovered from stable fly pupae during this 2017 overwintering study.
Manitoba has extremely harsh winters, but climate warming trends are increasing, therefore so may the survival rates of parasitoid wasps. The presence of five adult Phygadeuon fumator in
2018 was expected, given their life history traits and previous research on abundance in southern
Manitoba (McKay and Galloway 1999).
The results of this study contribute to the understanding that stable flies are adaptable and thus able to populate areas under various climatic conditions. This study is one of a few studies directly analyzing pupal stable fly overwintering potential in North America, as most stable fly research is revolves around the control of growing stable fly populations. Overall, these findings describe relatively predictable patterns of stable fly and parasitoid wasp activity in southern
Manitoba. This information is important for livestock producers wishing to reduce stable fly- related economic losses, improve animal welfare, and employ an ecologically sound option for stable fly population control. With increasing awareness of the environmental conditions required for optimal stable fly population growth, producers may prepare for and target populations prior to reaching economic injury levels.
Limitations
This research project had a number of limitations. Firstly, though adult stable flies are the damaging life stage, it is important to understand the recruitment rate of juveniles into the adult population in order to create a fully accurate population model. It is known that certain habitat features (e.g. high measures of fecal coliform bacteria and warm substrate temperature) increase adult stable fly emergence (Talley et al. 2009). Juvenile habitat characteristics were not
115 measured, therefore the study is missing a key component of both population dynamics and overwintering biology. Recording on-site juvenile habitat characteristics such as substrate temperature, and substrate moisture would allow us to design a stronger model, rather than relying on data from a weather station located 5.1 km from the farms. In addition, data on pupal depth, density and diversity of fecal coliform bacteria, and pH could have described the relationship between environmental variables, changes in developmental substrates and the growth of adult populations.
Secondly, based on the overdispersion in the models from both years, not all of the variables that influence adult stable fly abundance were accounted for. Besides the described above, other management practices such as insecticide use, or biotic factors like natural enemies may play a role in population growth that has not been accounted for in this study.
Thirdly, 144 juvenile parasitoid wasps could not be identified. The species diversity and relative abundance were only reported for the adult parasitoids, and identifying the juveniles could improve understanding of parasitoids and their use as biological control agents in southern
Manitoba. The status of development within the stable fly pupae prior to beginning experiments is unknown as the pupae were collected from wild populations and not laboratory reared.
Parasitoids can be purchased through suppliers and though the species provided may not be as advertised (McKay and Galloway 1999), incorporating biological control to limit chemical use is an ecologically sound option. Parasitoid research in Manitoba is lacking and new parasitoid species records may be discovered, which is important for establishing accurate species ranges and creating conservation baselines.
Fourth, the overwintering experiments were based on the assumptions that stable flies might overwinter as pupae. Sites where high densities of pupae are found that could serve as
116 overwintering sites are located in substrate and should be warm enough for pupae to survive but cool enough to slow development (below 10°C). The temperatures of 0°C, 2°C, and 5°C were specifically chosen for the incubators but may not reflect the actual temperatures within a natural juvenile stable fly overwintering site. The temperatures inside the juvenile habitat may be cooler or warmer than those selected based on factors such as substrate type, snow cover, or vegetative cover. The incubators maintained a constant temperature for the most part, unlike subnivean habitats which are subject to temperature fluctuations (Danks 1978). During natural fluctuation, periods of increased warm temperatures would accelerate development within the substrate. A field study examining the effects of fluctuating temperatures using hourly HOBO® temperature and moisture logs would lead to more accurate natural overwintering habitat data, rather than attempting to replicate natural effects in a laboratory setting, as geographic differences in latitude and microhabitat types lead to varied results under natural conditions (Danks 1978). The ability to access pupae during the field study would be an asset to determining if intrapuparial development does indeed occur under natural conditions. Weekly random removals of pupae of known age for dissection could provide insight into development at low temperatures.
The overwintering substrate in both the field and laboratory experiments was sterile in order to remove any potential pathogens or parasites. Substrate moisture is required to prevent desiccation and also to carry nutrients and bacteria which larvae require (Ranjard and Richaume
2001). Pupae also require moisture to prevent desiccation and facilitate adult emergence, but it is unknown whether pupae might absorb water containing extra nutrients or beneficial microorganisms from the substrate that improve the conditions for the developing pupae and for emergence. Pupae from the 2017 laboratory overwintering experiment appeared to continue to develop over time, but only two partially succeeded in emerging. Although the incubators were
117 misted with water up to two times weekly during the laboratory overwintering, it is possible that desiccation may have led to unsuccessful emergence. Improvements to these aspects of experimental design would be to increase the size of the pupal overwintering apparatus to allow space for natural substrate as well as increased volume to act as an extra layer of protection from minor temperature fluctuations and desiccation. Adding a timed water mister into the incubators would provide consistent humidity levels and prevent desiccation.
During this study, significant differences were reported in the expected and observed sex ratios. An overall improvement of stable fly trap design could result in more accurate natural assessment of populations. All currently available biting fly traps, including the Olson trap
(Hogsette and Ruff 1990), Broce Alsynite trap (Broce 1988), Nzi (Mihok 2002), EZ Trap®
(Farnam Companies Inc., Phoenix, AZ), Coroplast® sticky trap, and BiteFree Trap® (Farnam
Companies Inc., Phoenix, AZ) tend towards a biased sex ratio of approximately 2 male:1 female
(Taylor and Berkebile 2006), which requires the monitor to use additional less biased trapping methods, such as sweep net collection or aspirator collection to disentangle the true sex ratio from trapping biases. Our knowledge of stable fly eye biology and subsequent visual stimulation suggests that stable flies are strongly attracted to traps based on trap colour and reflectivity
(approximately 450 to 500 nm wavelengths; Agee and Patterson 1983; Zacks and Loew 1989), but the exact mechanism of attractance remains unknown.
Lastly, the incubator temperature increasing above the lethal upper limit likely contributed to the lack of emergence in the 2017 overwintering study. As there were no alarm systems in place to alert for technical issues, it is impossible to know exactly how long the pupae were exposed to the high temperatures. Larsen and Thomsen (1940) observed that pupal mortality did not increase during a transition from warm rearing temperatures to at least 35°C
118 and they suggest the vulnerable stages are likely larvae or the larval-pupal transition stage.
However, the pupae from this experiment would have had no time to acclimate to such high temperatures. Since adults emerged from the 2018 experiment where no malfunctions occurred prior to placing pupae within the designated chambers, it is assumed that the 44°C temperature had an impact on the pupae in 2017, resulting in only two adults partially emerging.
Improvements to this design flaw included processing and preparing the pupae on the collection date as was done in 2018, removing the need for storage. HOBO® temperature data loggers were used in incubators both years, but daily physical temperature checks would ensure the consistency of the incubator temperatures. Ideally, obtaining an incubator that can maintain 0°C without frosting would make it possible to study specimens at that temperature, as polystyrene containers filled with ice do not remain consistently at 0°C unless they are constantly monitored.
Future Directions
While the findings from this thesis revealed how stable flies and their pupal parasitoids are influenced by seasonal environmental conditions in southern Manitoba, further studies are required to better characterize these relationships. Stable flies are a severe seasonal pest, therefore it is important to continue to improve our knowledge of stable fly biology to develop innovative ways of controlling populations to improve animal welfare and limit the economic damage the flies can cause. Parasitoids wasps are natural enemies that may be used more judiciously in stable fly control. Parasitoid wasps are highly diverse, therefore assessing host and habitat requirements, and the species response to climate change is difficult (Shaw and Hochberg
2001; Shaw 2006). Conservation of these important natural enemies requires a clearer understanding of how and where they overwinter, and how producers could adapt their sanitation protocols to support resident parasitoids.
119 There is a need for long-term year-round monitoring in areas where stable flies reach and exceed economic injury levels to fully assess the impacts of changing environmental conditions and husbandry practices. Research is lacking on specific life stage behavioural and physiological strategies when responding to seasonal conditions. Stable fly populations are affected by autumn, winter, and spring conditions but how these distinct conditions affect each life stage is unclear, particularly the effects on larvae and pupae. For example, future research should determine the degree-day requirements for each stage of intrapuparial development, which could provide information needed to evaluate overwintering potential under changing conditions
The developmental threshold for stable flies is between 10.7°C (Lysyk 1998), 11.4°C
(Florez-Cuadros et al. 2019) and 11.8°C (Larsen and Thomsen 1940) and theoretically, development should not occur until near 10°C. Surprisingly, findings from the 2018 laboratory overwintering experiment report stable fly development and adult emergence within a 5°C incubator. Régnière et al. (2012) observed the majority of developmental threshold temperatures are theoretical data and are typically not measured. Though the linear relationship between temperature and stable fly development is a valuable tool for approximating population growth rates, there are a multitude of factors that contribute to development that are not measured when basing results on calculated values (Bergant and Trdan 2006). Experiments to improve accuracy of developmental thresholds should be conducted using live stable flies under various conditions including low and high temperatures, nutritional quality, substrate type, and collection from different geographic regions (Bergant and Trdan 2006).
Vertebrate species migration involves behavioural changes based on environmental cues followed by undistracted directional movements of the majority of the population and finally a return journey by the same generation. Insect migration, defined very generally as “long-range
120 movements” (Dingle 1996; Woiwood et al. 2001) is similar, but there has not been a documented case of a return journey by the same generation (Holland et al. 2006). Stable flies follow visual cues (Zacks and Loew 1989) and chemostimuli (Jeanbourquin and Guerin 2007), and are known to passively migrate, yet there is little evidence of stable fly behavioural changes based on environmental cues, other than the expected decrease in activity as temperatures drop. Friesen et al. (2016) postulated that female stable flies may be sensing and acting upon unknown environmental cues to cease mating and oviposition, based on the lack of larvae present in known developmental substrate. To support these findings further in a future study, a mark- recapture study could determine if stable flies exhibit true migratory capabilities; that is, if stable flies perceive token environmental stimuli or cues such as decreasing photoperiod or ambient temperatures in autumn and attempt movements towards different habitat types or sheltered sites.
Mark-recapture studies are notorious for underrepresentation of marked individuals, but studies conducted on stable flies in Nebraska (Taylor et al. 2010) and Florida (Hogsette and Ruff 1985) recovered enough adults to derive meaningful conclusions. Using Nzi cloth-target traps along with a mark-recapture would be beneficial, as Nzi traps attract flies in a directional bias, assisting in determining directionality (Taylor and Berkebile 2006). The results of this study could provide further support for the stable fly migratory hypothesis in Canada.
121 LITERATURE CITED
Agee, H.R. and Patterson, R.S. 1983. Spectral sensitivity of stable, face, and horn flies and behavioral responses of stable flies to visual traps (Diptera: Muscidae). Environmental Entomology, 12: 1823-1828.
Akaike H. 1974. A new look at statistical model identification. IEEE Transactions on Automatic Control, 19: 716–723.
Allen, J.C. 1976. A modified sine wave method for calculating degree-days. Environmental Entomology, 5: 388–396.
Anderson, J.R. 1978. Mating behavior of Stomoxys calcitrans: Effects of a blood meal on the mating drive of males and its necessity as a prerequisite for proper insemination of females. Journal of Economic Entomology, 71: 379–386.
Anderson, J.R. and Tempelis, C.H. 1970. Precipitin test identification of blood meals of Stomoxys calcitrans (L.) caught on California poultry ranches, and observations of digestion rates of bovine and citrated human blood. Journal of Medical Entomology, 7: 223–229.
Axtell, R.C. 1981. Livestock integrated pest management (IPM): Principles and prospects. Systems Approach to Animal Health and Production, 31-40.
Axtell, R.C. 1986. Status and potential of biological control agents in livestock and poultry pest management systems. In Biological control of muscoid flies. Edited by R.S. Patterson and D. A. Rutz. Miscellaneous Publications of the Entomological Society of America, 61: 1-9.
Axtell, R.C., and Rutz, D.A. 1986. Role of parasites and predators as biological fly control agents in poultry production facilities. In Biological control of muscoid flies. Edited by R.S. Patterson and D.A. Rutz. Miscellaneous Publications of the Entomological Society of America, 61: 88-100.
Bahrndorff, S., Kjærsgaard, A., Pertoldi, C., Loeschcke, V., Schou, T.M., Skovgård, H., and Hald, B. 2012. The effects of sex-ratio and density on locomotor activity in the house fly, Musca domestica. Journal of Insect Science, 12: 71.
Bailey, D.L., and Meifert, D.W. 1973. Feeding incidence of stable flies on beef cattle, as influenced by temperature, relative humidity and light. Environmental Entomology, 2: 1125– 1126.
122 Bailey, D.L., Whitfield, T.L., and Smittle, B.J. 1973. Flight and dispersal of the stable fly. Journal of Economic Entomology, 66: 410–411.
Baldacchino, F., Muenworn, V., Desquesnes, M., Desoli, F., Charoenviriyaphap, T., and Duvallet, G. 2013. Transmission of pathogens by Stomoxys flies (Diptera, Muscidae): a review. Parasite, 20: 26.
Baleba, S.B.S., Baldwyn, T., Masiga, D., Getahun, M.N., and Weldon, C.W. 2020. Stable flies, Stomoxys calcitrans L. (Diptera: Muscidae), improve offspring fitness by avoiding oviposition substrates with competitors or parasites. Frontiers in Ecology and Evolution, 8: 5.
Ball, S.G. 1984. Seasonal abundance during the summer months of some cattle-visiting Muscidae (Diptera) in northeast England. Ecological Entomology, 9: 1-10.
Barros, A.T.M, Rodrigues, V.D., Cançado, P.H.D., and Domingues, L.N. 2019. Resistance of the stable fly, Stomoxys calcitrans (Diptera: Muscidae), to cypermethrin in outbreak areas in Midwestern Brazil. Brazilian Journal of Veterinary Parasitology, 28: 802-806.
Beerwinkle, K.R., Berry, I.L., and Kunz, S.E. 1978. Prediction models for mortality of immature stable flies caused by cold temperatures. Environmental Entomology, 7: 273-277.
Beresford, D.V. and Sutcliffe, J.F. 2006. Studies on the effectiveness of coroplast sticky traps for sampling stable flies (Diptera: Muscidae), including a comparison to Alsynite. Journal of Economic Entomology, 99: 1025-35.
Beresford, D.V. and Sutcliffe, J.F. 2009a. Local infestation or long distance migration? The seasonal re-colonization of dairy farms by Stomoxys calcitrans (Diptera: Muscidae) in south- central Ontario. Journal of Economic Entomology, 102: 788-798.
Beresford, D.V. and Sutcliffe, J.F. 2009b. The effect of Macrocheles muscaedomesticae and M. subbadius (Acarina: Macrochelidae) phoresy on the dispersal of Stomoxys calcitrans (Diptera: Muscidae). Systems of Applied Acarology Special Publications, 23: 1–30.
Beresford, D.V. and Sutcliffe, J.F. 2012. Field measurements of stable fly (Diptera: Muscidae) demography, fecundity, and survival based on daily trap catches at a beef farm in southern Ontario over a 5-yr period. Journal of Medical Entomology, 49: 1262–1269.
Bergant, K. and Trdan, S. 2006. How reliable are thermal constant for insect development when estimated from laboratory experiments? Entomologia Experimentalis et Applicata, 120: 251-256.
123 Berkebile, D.R., Thomas, G.D., and Campbell, J.B. 1994. Overwintering of the stable fly (Diptera: Muscidae) in southeastern Nebraska. Journal of Economic Entomology, 87: 1555– 1563.
Berry, I.L. and Kunz, S.E. 1977. Mortality of adult stable flies. Environmental Entomology, 6: 569–574.
Berry, I.L. and Kunz, S.E. 1978. Oviposition of stable flies in response to temperature and humidity. Environmental Entomology, 7: 213–216.
Berry, I.L., Foerster, K.W., and Campbell, J.B. 1978. Overwintering behavior of stable flies in manure mounds. Environmental Entomology, 7: 67-72.
Berry, I.L., Stage, D.A., and Campbell, J.B. 1983. Populations and economic impacts of stable flies on cattle. Transactions of the American Society of Agricultural Engineers, 26: 873-877.
Birkemoe, T., Soleng, A., and Aak, A. 2009. Biological control of Musca domestica and Stomoxys calcitrans by mass releases of the parasitoid Spalangia cameroni on two Norwegian pig farms. Biocontrol, 54: 425-436.
Bishopp, F. C. 1913. The stable fly (Stomoxys calcitrans L.), an important livestock pest. Journal of Economic Entomology, 6: 112-126.
Brain, C.K. 1912. Stomoxys calcitrans. Annals of the Entomological Society of America, 5: 421- 430.
Broce, A.B. 1988. An improved Alsynite cylinder trap for stable flies (Diptera: Muscidae). Journal of Medical Entomology, 25: 406-409.
Broce, A.B., and Haas, M.S. 1999. Relation of cattle manure age to colonization by stable fly and house fly (Diptera: Muscidae). Journal of the Kansas Entomological Society, 72: 60–72.
Broce, A.B., Hogsette, J.A., and Paisley, S. 2005. Winter feeding sites of hay in round bales as major developmental sites of Stomoxys calcitrans (Diptera: Muscidae) in pastures in spring and summer. Journal of Economic Entomology, 98: 2307–2312.
Bruce, W.N. and Decker, G.C. 1958. The relationship of stable fly abundance to milk production in dairy cattle. Journal of Economic Entomology, 51: 269–274.
Brues, C.T. 1913. The geographical distribution of the stable fly, Stomoxys calcitrans. Journal of Economic Entomology, 6: 460-477.
124
Buschman, L.L. and Patterson, R.S. 1981. Assembly, mating, and thermoregulating behavior of stable flies under field conditions. Environmental Entomology, 10: 16-21.
Campbell, J.B., White, R.G., Wright, J.E., Crookshank, R., and Clanton, D.C. 1977. Effects of stable flies on weight gains and feed efficiency of calves on growing or finishing rations. Journal of Economic Entomology, 70: 592-594.
Campbell, J.B., Berry, I.L., Boxler, D.J., Davis, R.L., Clanton, D.C., and Deutscher, G.H. 1987. Effects of stable flies (Diptera: Muscidae) on weight gain and feed efficiency of feedlot cattle. Journal of Economic Entomology, 80: 117–119.
Campbell, J.B., Berry, I.L., Petersen, J.J., and Greene, G.L. 1989. Economic threshold for stable flies on confined livestock. Miscellaneous Publications of the Entomological Society of America, 74: 18-22.
Campbell, J.B., Skoda, S.R., Berkebile, D.R., Boxler, D.J., Thomas, G.D., Adams, D.C., and Davis, R. 2001. Effects of stable flies (Diptera: Muscidae) on weight gains of grazing yearling cattle. Journal of Economic Entomology, 94: 780–783.
Chia, L.S., Baxter, J.A., and Morrison, P.E. 1982. Quantitative relationship between ingested blood and follicular growth in the stable fly, Stomoxys calcitrans. Canadian Journal of Zoology, 60: 1917–1921.
Cilek, J.E. and Greene, G.L. 1994. Stable Fly (Diptera: Muscidae) Insecticide resistance in Kansas cattle feedlots. Journal of Economic Entomology, 87: 275–279.
Colwell, D.D. 2000. A guide for the control of flies in Alberta confined feeding operations. Alberta Agriculture and Rural Development. Retrieved 18 October 2018 from https://www1.agric.gov.ab.ca/$department/deptdocs.nsf/all/epw12257/$file/guide.pdf?OpenElem ent
Colwell, D.D. and Kavaliers, M. 1992. Evidence for activation of endogenous opioid systems in mice following short exposure to stable flies. Medical and Veterinary Entomology, 6: 159-164.
Cook, D.F., Dadour, I.R., and Keals, N.J. 1999. Stable fly, house fly (Diptera: Muscidae), and other nuisance fly development in poultry litter associated with horticultural crop production. Journal of Economic Entomology. 92: 1352–1357.
125 Cook, D.F., Telfer, D.V., Lindsey, J.B., and Deyl, R.A. 2018. Substrates across horticultural and livestock industries that support the development of stable fly, Stomoxys calcitrans (Diptera: Muscidae): Substrates that support stable fly. Australian Entomology, 57: 344–348.
Cortinas, R. and Jones, C.J. 2006. Ectoparasites of cattle and small ruminants. Veterinary Clinics Food Animal Practice, 22: 673–693.
Crosskey R.W. 1993. Stable-flies and horn-flies (bloodsucking Muscidae). In Medical Insects and Arachnids. Edited by R.P. Lane and R.W. Crosskey. Springer, Dordrecht.
Cumming, J.E. 2006. Diptera associated with livestock dung: stable fly Stomoxys calcitrans (L.). Retrieved 20 July 2020 from http://www.nadsdiptera.org/FFP/stable.htm
Danks, H.V. 1978. Modes of seasonal adaptation in the insects I: winter survival. The Canadian Entomologist, 110: 1167-1205.
Danks, H.V. 2004. Seasonal adaptations in arctic insects. Integrative and Comparative Biology, 44: 85-94.
DeLoof, A., Van Voon, J., and Vanderroost, C. 1979. Influence of ecdysterone, precocene and compounds with juvenile hormone activity on induction, termination and maintenance of diapause in the parasitoid wasp, Nasonia vitripennis. Physiological Entomology, 4: 319–328.
Dingle, H. 1996. Migration: the biology of life on the move. Oxford University Press, Oxford, UK.
Downes, J.A. 1969. The swarming and mating flight of Diptera. Annual Review of Entomology, 14: 271-98.
Eddy, G.W., Roth, A.R., and Plapp Jr., F.W. 1962. Studies on the flight habits of some marked insects. Journal of Economic Entomology, 55: 603-607.
Environment and Climate Change Canada. 2019. Canadian Climate Normals 1981-2010 Station Data. Retrieved 16 July 2020 from: https://climate.weather.gc.ca/climate_normals/results_1981_2010_e.html?searchType=stnProv& lstProvince=MB&txtCentralLatMin=0&txtCentralLatSec=0&txtCentralLongMin=0&txtCentral LongSec=0&stnID=3698&dispBack=0
Floate, K. and Skovgård, H. 2004. Winter survival of nuisance fly parasitoids (Hymenoptera: Pteromalidae) in Canada and Denmark. Bulletin of Entomological Research, 94: 331-340.
126 Floate, K.D, Khan, B., and Gibson, G.A.P. 1999. Hymenopterous parasitoids of filth fly (Diptera: Muscidae) pupae in cattle feedlots. The Canadian Entomologist, 131: 347-362.
Floate, K.D., Coghlin, P., and Gibson, G.A.P. 2000. Dispersal of the filth fly parasitoid Muscidifurax raptorellus (Hymenoptera: Pteromalidae) following mass-releases in cattle confinements. Biological Control, 18: 172–178.
Florez-Cuadros, M., Berkebile, D., Brewer, G., and Taylor, D.B. 2019. Effects of diet quality and temperature on stable fly (Diptera: Muscidae) development. Insects, 10: 207.
Foil, L.D. and Hogsette, J.A. 1994. Biology and control of tabanids, stable flies, and horn flies. Revue Scientifique et Technique (International Office of Epizootics), 13: 1125–1158.
Foil, L.D., Meek, C.L., Adams, W.V., and Issel, C.J. 1983. Mechanical transmission of equine infectious anemia virus by deer flies (Chrysops flavidus) and stable flies (Stomoxys calcitrans). American Journal of Veterinary Research, 44:155-166.
Fox, J. and Weisberg, S. 2019. An R Companion to Applied Regression, Third Edition. Sage, Thousand Oaks, California.
Fraenkel, G. and Bhaskaran, G. 1973. Pupariation and pupation in cyclorraphous flies (Diptera): terminology and interpretation. Annals of the Entomological Society of America, 66: 418–422.
Friesen, K.M. and Johnson, G.D. 2012. Reproductive potential of stable flies (Diptera: Muscidae) fed cattle, chicken, or horse blood. Journal of Medical Entomology, 49: 461-466.
Friesen, K.M. and Johnson, G.D. 2013. Stable fly phenology in a mixed agricultural-wildlife ecosystem in northeast Montana. Environmental Entomology, 42: 49–57.
Friesen, K., Chen, H., Zhu, J., and Taylor, D.B. 2015. External morphology of stable fly (Diptera: Muscidae) larvae. Journal of Medical Entomology, 52: 626–637.
Friesen K., Berkebile, D.R., Wienhold, B.J., Durso, L., Zhu, J.J., and Taylor, D.B. 2016. Environmental parameters associated with stable fly (Diptera: Muscidae) development at hay feeding sites. Environmental Entomology, 45: 570–576.
Geden, C.J. and Moon, R.D. 2009. Host ranges of gregarious muscoid fly parasitoids: Muscidifurax raptorellus (Hymenoptera: Pteromalidae), Tachinaephagus zealandicus (Hymenoptera: Encyrtidae), and Trichopria nigra (Hymenoptera: Diapriidae). Environmental Entomology, 38: 700-707.
127 Geden, C.J., Rutz, D.A., Miller, R.W., and Steinkraus, D.C. 1992. Suppression of house flies (Diptera: Muscidae) on New York and Maryland dairies using releases of Muscidifurax raptor (Hymenoptera: Pteromalidae) in an integrated management program. Environmental Entomology, 21: 1419-1426.
Gerling, D. and Legner, E.F. 1968. Developmental history and reproduction of Spalangia cameroni, parasite of synanthropic flies. Annals of the Entomological Society of America, 61: 1436–1443.
Gersabeck, E.F. and Merritt, R.W. 1983. Vertical and temporal aspects of Alsynite panel sampling for adult Stomoxys calcitrans (L.) (Diptera: Muscidae). Florida Entomologist, 66: 222- 227.
Gibson, G.A.P. 2000. Illustrated key to the native and introduced Chalcidoid parasitoids of filth flies in America north of Mexico (Hymenoptera: Chalcidoidea). http://res2.agr.ca/ecorc/apss/chalkey/keyintro.htm
Gibson, G.A.P. and Floate, K.D. 2004. Filth fly parasitoids on dairy farms in Ontario and Quebec, Canada. The Canadian Entomologist, 136: 407-417.
Gilles, J., David, J-F., and Duvallet, G. 2005. Effects of temperature on the rate of increase of Stomoxys calcitrans and Stomoxys niger niger (Diptera: Muscidae) from La Réunion Island. Journal of Medical Entomology, 42: 959–965.
Gilles, J., David, J., Duvallet, G., and Tillard, E. 2008. Potential impacts of climate change on stable flies, investigated along an altitudinal gradient. Medical and Veterinary Entomology, 22: 74–81.
Godwin, R.M., Mayer, D.G., Brown, G.W., Leemon, D.M., and James, P.J. 2017. Predicting nuisance fly outbreaks on cattle feedlots in subtropical Australia. Animal Production Science, 58: 343-349.
Graves, S., Piepho, H-P, Selzer, L., and Dorai-Raj, S. 2019. multcompView: Visualizations of paired comparisons. R package version 0.1-8.
Greene, G. L. 1989. Seasonal population trends of adult stable flies. In Current status of stable fly (Diptera: Muscidae) research. Edited by J.J. Petersen and G.L. Greene. Miscellaneous Publications of the Entomological Society of America, 74: 12-17.
128 Greene, H.W. 2005. Organisms in nature as a central focus for biology. Trends in Ecology and Evolution, 20: 23–27.
Greene, G.L., Hogsette, J.A., and Patterson, R.S. 1989. Parasites that attack stable fly and house fly (Diptera: Muscidae) puparia during the winter on dairies in northwestern Florida. Journal of Economic Entomology, 82: 412-415.
Gruvil, J. and Balis, J. 1966. Trypanosomiasis caused by Trypanosoma evansi in the camel in Chad and its chief vectors. Revue d'Élevage et de Médecine Vétérinaire des Pays Tropicaux (NS), 18: 233-231.
Guzman, D.R. and Petersen, J.J. 1986. Overwintering of filth fly parasites (Hymenoptera: Pteromalidae) in open silage in eastern Nebraska. Environmental Entomology, 15: 1296-1300.
Hafez, M. and Gamal-Eddin, F.M. 1959. On the feeding habits of Stomoxys calcitrans L. and sitens Rond. With special reference to their biting cycle in nature. Bulletin Société Entomologique d'Egypte, 43: 291-301.
Hall, R.D., Thomas, G.D., and Morgan, C.E. 1982. Stable flies, Stomoxys calcitrans (L.), breeding in large round hay bales: initial associations (Diptera: Muscidae). Journal of Kansas Entomology, 55: 617–620.
Heinzl, H. and Mittlböck, M. 2003. Pseudo R-squared measures for Poisson regression models with over- or underdispersion. Computational Statistics and Data Analysis, 44: 253-271.
Helfrich, L.A., Weigmann, D.L., Hipkins, P., and Stinson, E.R. 2009. Pesticides and aquatic animals: a guide to reducing impacts on aquatic systems. In Virginia Polytechnic Institute and State University. Retrieved 17 July 2020 from https://pubs.ext.vt.edu/420/420-013/420-013.html
Hoffmann, A.A., Chown, S.L., and Clusella-Trullas, S. 2013. Upper thermal limits in terrestrial ectotherms: how constrained are they? Functional Ecology, 27: 934–949.
Hogsette, J.A. 1996. Development of house flies (Diptera: Muscidae) in sand containing varying amounts of manure solids and moisture. Journal of Economic Entomology, 89: 940-945.
Hogsette, J.A. 1999. Management of ectoparasites with biological control organisms. International Journal of Parasitology, 29: 147-151.
129 Hogsette, J.A. and Ose, G.A. 2017. Improved capture of stable flies (Diptera: Muscidae) by placement of knight stick sticky fly traps protected by electric fence inside animal exhibit yards at the Smithsonian’s National Zoological Park. Zoo Biology, 36: 382-386.
Hogsette, J.A. and Ruff, J.P. 1985. Stable fly (Diptera: Muscidae) migration in northwest Florida. Environmental Entomology, 14: 170–175.
Hogsette, J.A. and Ruff, J.P. 1990. Comparative attraction of four different fiberglass traps to various age and sex classes of stable fly (Diptera: Muscidae) adults. Journal of Economic Entomology, 83: 883-886.
Hogsette, J.A., Ruff J.P., and Jones, C.J. 1987. Stable fly biology and control in northwest Florida. Journal of Agricultural Entomology, 4: 1–11.
Holland, R.A., Wikelski, M., and Wilcove, D.S. 2006. How and who do insects migrate? Science, 313: 794-796.
Hothorn, T., Bretz, F. and Westfall, P. 2008. Simultaneous inference in general parametric models. Biometrical Journal, 50: 346-363.
Isard, S.A., Kristovich, D.A.R., Gage, S.H., Jones, C.J., and Laird, N.F. 2001. Atmospheric motion systems that influence the redistribution and accumulation of insects on the beaches of the Great Lakes in North America. Aerobiologica, 17: 275-291.
Jackman, S. 2020. pscl: Classes and methods for R developed in the political science computational laboratory. United States Studies Centre, University of Sydney. Sydney, New South Wales, Australia. R package version 1.5.5.
Jacquiet, P., Rouet, D., Bouhsira, E., Salem, A., Lienard, E., and Franc, M. 2014. Population dynamics of Stomoxys calcitrans (L.) (Diptera: Muscidae) in southwestern France. Revue de Médecine Vétérinaire, 165: 267-271.
James, M.T., and Harwood, R.F. 1969. Herms’s Medical Entomology. 6th ed. Collier- MacMillan Canada Limited, Don Mills, Ontario. pp. 484.
Jeanbourquin, P. and Guerin, P.M. 2007. Chemostimuli implicated in selection of oviposition substrates by the stable fly Stomoxys calcitrans. Medical and Veterinary Entomology, 21: 209- 216.
130 Jelvez Serra, N.S., Goulart, H.F., Triana, M.F., Dos Santos Tavares, S., Almeida, C.I.M., da Costa, J.G., Santana, A.E.G., and Zhu, J.J. 2017. Identification of stable fly attractant compounds in vinasse, a by-product of sugarcane–ethanol distillation. Medical and Veterinary Entomology, 31: 381-391.
Jervis, M.A., Heimpel, G.E., Ferns, P.N., Harvey, J.A., and Kidd, N.A.C. 2001. Life-history strategies in parasitoid wasps: a comparative analysis of 'ovigeny'. Journal of Animal Ecology, 70: 442-458.
Johnson, C.G. 1966. A functional system of adaptive dispersal by flight. Annual Review of Entomology, 11: 233-260.
Jones, S.R. and Kunz, S.E. 1997. Importance of supercooling points in the overwintering of the horn fly and stable fly (Diptera: Muscidae). Journal of Medical Entomology, 43: 426–429.
Kaufman, P.E., Rutz, D.A., and Frisch, S. 2005. Large sticky traps for capturing house flies and stable flies in dairy calf greenhouse facilities. Journal of Dairy Science, 88: 176-181.
Khumalo, W.V. and Galloway, T.D. 1996. Seasonal abundance of stable flies, Stomoxys calcitrans (L.) (Diptera: Muscidae), at Glenlea, Manitoba. Proceedings of the Entomological Society of Manitoba, 52: 4-17.
Killough, R.A. and McKinstry, D.M. 1965. Mating and oviposition studies of the stable fly. Journal of Economic Entomology, 58: 489–491.
King, B.H. 1998. Host age response in the parasitoid wasp Spalangia cameroni. Journal of Insect Behavior, 11: 103–117.
King, B.H. 2001. Parasitization site on the host of the parasitoid wasp Spalangia endius (Hymenoptera: Pteromalidae). Environmental Entomology, 30: 346–349.
Kleiber, C. and Zeileis, A. 2016. Visualizing count data regressions using rootograms. The American Statistician, 70: 296-303.
Kneeland, K.M., Skoda, S.R., Hogsette, J.A., Li, A.Y., Molina-Ochoa, J., Lohmeyer, K.H., & Foster, J.E. 2012. A century and a half of research on the stable fly, Stomoxys calcitrans (L.) (Diptera: Muscidae), 1862-2011: an annotated bibliography. ARS-173. U.S. Department of Agriculture, Agricultural Research Service, Washington, DC.
131 Kneeland, K.M., Skoda, S.R., and Foster, J.E. 2016. Genetic variability of the stable fly assessed on a global scale using amplified fragment length polymorphism. Insect Science, 23: 695–703.
Krafsur, E.S., Moon, R.D., and Lysyk, T.J. 1994. Adult age and reproductive success in stable fly populations (Diptera: Muscidae). The Canadian Entomologist, 126: 239-249.
Kunz, S.E., Blume, R.R., Hogan, B.F., and Matter, J.J. 1970. Biological and ecological investigations of horn flies in central Texas: influence of time of manure deposition on oviposition. Journal of Economic Entomology, 63: 930–933.
Kunz, S.E., Murrel, K.D., Lambert, G., James, L.F., and Terrill, C.E. 1991. Estimated losses of livestock to pests, pp. 69-98 In CRC Handbook of Pest Management in Agriculture, vol. I. Edited by D. Pimentel. CRC Press, Boca Raton, Florida.
Kuramochi, K. 2000. Ovipositional behavior of the horn fly (Diptera: Muscidae) in the field. Journal of Medical Entomology, 37:461-6.
Kuzina, O.S. 1942. On the gonotrophic relationship in Stomoxys calcitrans L. and Haematobia irritans L. Meditsinskaia Parazitologiia I Parazitarnye Bolezni, 11: 70–78.
Larsen, E.B. and Thomsen, M. 1940. The influence of temperature on the development of some species of Diptera. Videnskabelige Meddelelser fra Dansk Naturhistorisk Forening i Kjøbenhavn, 104: 1-75.
Lazarus, W.F., Rutz, D.A., Miller, R.W., and Brown, D.A. 1989. Costs of existing and recommended manure management practices for house fly and stable fly (Diptera: Muscidae) control on dairy farms. Journal of Economic Entomology, 82: 1145-1151.
Legner, E.F. and Brydon, H.W. 1966. Suppression of dung inhabiting fly populations by pupal parasites. Annals of the Entomological Society of America, 59: 638–651.
Legner, E.F. and Gerling, D. 1967. Host-feeding and oviposition on Musca domestica by Spalangia cameroni, Nasonia vitripennis, and Muscidifurax raptor (Hymenoptera: Pteromalidae) influences their longevity and fecundity. Annals of the Entomological Society of America, 60: 678–691.
Lenth, R. 2020. emmeans: Estimated marginal means, aka least-squares means. R package version 1.5.2-1.
132 Loera-Gallardo, J., Luna-Salas, J.F., and Gibson, G.A.P. 2008. First report of pupal parasitoids of filth-breeding flies (Diptera) from bovine manure in northeastern Mexico. The Canadian Entomologist, 140: 682-689.
Louviere, J.J., Hensher, A.D., and Swait, D.J. 2000. Stated choice methods. New York: Cambridge University Press.
Lysyk, T.J. 1993. Seasonal abundance of stable flies and house flies (Diptera: Muscidae) in dairies in Alberta, Canada. Journal of Medical Entomology, 30: 888-95.
Lysyk, T.J. 1995. Parasitoids (Hymenoptera; Pteromalidae, Ichneumonidae) of filth fly (Diptera: Muscidae) pupae at dairies in Alberta. Journal of Economic Entomology, 88: 659–665.
Lysyk, T.J. 1998. Relationships between temperature and life-history parameters of Stomoxys calcitrans (Diptera: Muscidae). Journal of Medical Entomology, 35: 107-119.
Lysyk, T.J., Kalischuk-Tymensen, L., Selinger, L.B., Lancaster, R.C., Wever, L., and Cheng, K.J. 1999. Rearing stable fly larvae (Diptera: Muscidae) on an egg yolk medium. Journal of Medical Entomology, 36: 382-388.
Machtinger, E.T., Geden, C.J., Kaufman, P.E., and House, A.M. 2015. Use of pupal parasitoids as biological control agents of filth flies on equine facilities. Journal of Integrated Pest Management, 6: 16.
Machtinger, E.T., Leppla, N.C., and Hogsette, J.A. 2016. House and stable fly seasonal abundance, larval development substrates, and natural parasitism on small equine farms in Florida. Neotropical Entomology, 45: 433-440.
Malaithong, N., Duvallet, G., Ngoen-Klan, R., Bangs, M.J., and Chareonviriyaphap, T. 2019. Stomoxyinae flies in Thailand: A précis, with abridged taxonomic key to the adult species. Vector Borne Zoonotic Diseases, 19: 385–394.
Mangiafico, S. 2020. rcompanion: Functions to support extension education program evaluation. R package version 2.3.26.
Marçon, P.C.R.G., Thomas, G.D., Siegfried, B.D., and Campbell, J.B. 1997. Susceptibility of stable flies (Diptera: Muscidae) from southeastern Nebraska beef cattle feedlots to selected insecticides and comparison of 3 bioassay techniques. Journal of Economic Entomology, 90: 293–298.
133 Marquez, J. G., Cummings, M.A., and Krafsur, F.S. 2007. Phylogeography of stable fly (Diptera: Muscidae) estimated by diversity at ribosomal 16S and cytochrome oxidase I mitochondrial genes. Journal of Medical Entomology, 44: 998-1008.
Martin, J. and Hall, D.B. 2015. R2 measures for zero-inflated regression models for count data with excess zeros. Journal of Statistical Computation and Simulation, 86: 3777-3790.
Masmeatathip, R., Gilles, J., Ketavan, C., and Duvallet, G. 2006. First survey of seasonal abundance and daily activity of Stomoxys spp. (Diptera: Muscidae) in Kamphaengsaen Campus, Nakornpathom Province, Thailand. Parasite, 13: 245-250.
Mason, P.G. and Gillespie, D.R. (Eds.). 2013. Biological control programmes in Canada 2001- 2012. CABI.
Matthysse, J.G. 1945. Observations of house fly overwintering. Journal of Economic Entomology, 38: 493-495.
McKay, T. and Galloway, T.D. 1999. Survey and release of parasitoid wasps (Hymenoptera: Pteromalidae, Ichneumonidae) attacking house flies and stable flies (Diptera: Muscidae) in dairy operations in Manitoba. The Canadian Entomologist, 131: 743–756.
Merriam, G., Wegner, J., and Caldwell, D. 1983. Invertebrate activity under snow in deciduous woods. Holarctic Ecology, 6: 89-94.
Meyer, J.A. and Schultz, T.A. 1990. Stable fly and house fly breeding sites in dairies. California Agriculture, 44: 28-29.
Meyer, J.A., Georghiou, G.P., and Hawley, M.K. 1987. House fly (Diptera: Muscidae) resistance to permethrin on southern California dairies. Journal of Economic Entomology, 80: 636-640.
Meyer, J.A., Mullens, B.A., Cyr, T.L., and Stokes, C. 1990. Commercial and naturally occurring fly parasitoids (Hymenoptera: Pteromalidae) as biological control agents of stable flies and house flies (Diptera: Muscidae) on California dairies. Journal of Economic Entomology, 83: 799‐806.
Meyer, J.A., Shultz, T. A., Collar, C., and Mullens, B.A. 1991. Relative abundance of stable fly and house fly (Diptera: Muscidae) pupal parasites (Hymenoptera: Pteromalidae, Coleoptera: Staphylinidae) on confinement dairies in California. Environmental Entomology, 20: 915-921.
Mihok, S. 2002. The development of a multipurpose trap (the Nzi) for tsetse and other biting flies. Bulletin of Entomological Research, 92: 385–403.
134 Miller, R.W., Rutz, D.A., Pickens, L.G., and Geden, C.J. 1993. Evaluation of traps and the parasitoid Muscidifurax raptor Girault and Sanders to manage house flies and stable flies on dairy farms. Journal of Agricultural Entomology, 10: 9-19.
Miller, R.W., Pickens, L.G., and Potts, W.E. 1993a. Comparison of traps and an integrated program to manage house flies and stable flies on dairy farms. Journal of Agricultural Entomology, 10: 189-196.
Moobola, S. and Cupp, E.W. 1978. Ovarian development in the stable fly, Stomoxys calcitrans, in relation to diet and juvenile hormonal control. Physiological Entomology, 3: 317–321.
Moon, R.D., Berry, I.L., and Petersen, J.J. 1982. Reproduction of Spalangia cameroni Perkins (Hymenoptera: Pteromalidae) on stable fly (Diptera: Muscidae) in the laboratory. Journal of the Kansas Entomological Society, 55: 77-85.
Morales, M. 2020. sciplot: scientific graphing functions for factorial designs. R package version 1.2-0
Morgan, P.B. 1980. Sustained releases of Spalangia endius (Hymenoptera: Pteromalidae) for the control of Musca domestica L. and Stomoxys calcitrans (L.) (Diptera: Muscidae). Journal of the Kansas Entomological Society, 53: 367–372.
Morgan, P.B., Patterson, R.S., Labrecque, G.C., Weidhaas, D.E., Benton, A., and Whitfield, T. 1975. Rearing and release of the house fly pupal parasite Spalangia endius Walker. Environmental Entomology, 4: 609–611.
Muenworn, V., Duvallet, G., Thainchum, K., Tuntakom, S., Tanasilchayakul, S., Prabaripai, A., Akratanakul, P., Sukonthabhirom, S., and Chareonviriyaphap, T. 2010. Geographic distribution of stomoxyine flies (Diptera: Muscidae) and diurnal activity of Stomoxys calcitrans in Thailand. Journal of Medical Entomology, 47: 791–797.
Mullens, B.A. and Meyer, J.A. 1987. Seasonal abundance of stable flies (Diptera: Muscidae) on California dairies. Journal of Economic Entomology, 80: 1039-1043.
Mullens, B.A., Peterson, N.G., Dada, C.E., and Velten, R.K. 1995. Octenol fails to lure stable fly to insecticide. California Agriculture, 49: 16-18.
Müller, G.C., Hogsette, J.A., Beier, J.C., Traore, S.F., Toure, M.B., Traore, M.M., Bah, S., Doumbia, S., and Schlein Y. 2012. Attraction of Stomoxys sp. to various fruits and flowers in Mali. Medical and Veterinary Entomology, 26: 178-87.
135 Murray, M.S. 2008. Using degree days to time treatments for insect pests. Utah State University Extension. Retrieved 5 December 2020 from https://climate.usu.edu/includes/pestFactSheets/degree-days08.pdf
Neuwirth, E. 2014. RColourBrewer: ColourBrewer palettes. R package version 1.1-2.
Newstead, R. 1906. On the life history of Stomoxys calcitrans (L.). Journal of Economic Biology, 1: 157-166.
O’Hara, R.B. and Kotze, D.J. 2010. Do not log-transform count data. Methods in Ecology and Evolution, 1: 118-122.
Olafson, P.U., Pitzer, J.B., and Kaufman, P.E. 2011. Identification of a mutation associated with permethrin resistance in the para-type sodium channel of the stable fly (Diptera: Muscidae). Journal of Economic Entomology, 104: 250-257.
Parker, G.A. 1978. Evolution of competitive mate searching. Annual Review of Entomology, 23: 173-96.
Parravani, A., Chivers, C.A., Bell, N., Long, S., Burden, F., and Wall, R. 2019. Seasonal abundance of the stable fly Stomoxys calcitrans in southwest England. Medical and Veterinary Entomology, 33: 485-490.
Petersen, J.J. and Meyer, J.A. 1983. Host preference and seasonal distribution of pteromalid parasites (Hymenoptera: Pteromalidae) of stable flies and house flies (Diptera: Muscidae) associated with confined livestock in Eastern Nebraska. Environmental Entomology, 12: 567– 571.
Petersen, J.J., Meyer, J.A., Stage, D.A., and Morgan, P.B. 1983. Evaluation of sequential releases of Spalangia endius (Hymenoptera: Pteromalidae) for control of house flies and stable flies (Diptera: Muscidae) associated with confined livestock in eastern Nebraska. Journal of Economic Entomology, 76: 283–286.
Peterson, B.G., and Carl, P. 2020. PerformanceAnalytics: econometric tools for performance and risk analysis. R package version 2.0.4.
Phasuk, J., Prabaripai, A., and Chareonviriyaphap, T. 2013. Seasonality and daily flight activity of stable flies (Diptera: Muscidae) on dairy farms in Saraburi Province, Thailand. Parasite, 20: 17.
136 Pickens, L.G., and Miller, R.W. 1987. Techniques for trapping flies on dairy farms. Journal of Agricultural Entomology, 4: 305-313.
Pitzer, J.B., Kaufman, P.E., and Tenbroeck, S.H. 2010. Assessing permethrin resistance in the stable fly (Diptera: Muscidae) in Florida by using laboratory selections and field evaluations. Journal of Economic Entomology, 103: 2258-2263.
Pitzer, J.B., Kaufman, P.E., Hogsette, J.A., Geden, C.J., and Tenbroeck, S.H. 2011. Seasonal abundance of stable flies and filth fly pupal parasitoids (Hymenoptera: Pteromalidae) at Florida equine facilities. Journal of Economic Entomology, 104: 1108-1115.
Pitzer, J. B., Kaufman, P.E., Tenbroeck, S.H., and Maruniak, J.E. 2014. Host blood meal identification by multiples polymerase chain reaction for dispersal evidence of stable flies (Diptera: Muscidae) between livestock facilities. Journal of Medical Entomology, 48: 53-60.
Prairie Climate Centre. 2019. The climate atlas of Canada (version 2, July 10, 2019). Retrieved from https://climateatlas.ca
Pruess, K. P. 1983. Day-degree methods for pest management. Environmental Entomology, 80: 1039–1043.
R Core Team. 2020. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/.
Ranjard, L., and Richaume, A.S. 2001. Quantitative and qualitative microscale distribution of bacteria in soil. Research in Microbiology, 152: 707–716.
Régnière, J., Powell, J., Bentz, B. and Nealis, V. Effects of temperature on development, survival and reproduction of insects: Experimental design, data analysis and modeling. Journal of Insect Physiology, 58: 634-647.
Reissert-Oppermann, S., Bauer, B., Steuber, S., and Clausen, P-H. 2019. Insecticide resistance in stable flies (Stomoxys calcitrans) on dairy farms in Germany. Parasitology Research, 118: 2499– 2507.
Rivers, D.B., Lee Jr, R.E., and Denlinger, D.L. 2000. Cold hardiness of the fly pupal parasitoid Nasonia vitripennis is enhanced by its host Sarcophaga crassipalpis. Journal of Insect Physiology, 46: 99-106.
137 Robacker, D.C. and Bartelt, R.J. 1997. Chemicals attractive to Mexican fruit fly from Klebsiella pneumoniae and Citrobacter freundii cultures sampled by solid‐phase microextraction. Journal of Chemical Ecology, 23: 2897– 2915.
Rochon, K., Lysyk, T.J., and Selinger, L.B. 2005. Retention of Escherichia coli by house fly and stable fly (Diptera: Muscidae) during pupal metamorphosis and eclosion. Journal of Medical Entomology, 42: 397-403.
Romero, A., Broce, A.B., and Zurek, L. 2006. Role of bacteria in the oviposition behaviour and larval development of stable flies. Medical and Veterinary Entomology, 20: 115-121.
Romero, A., Hogsette, J.A, and Coronado, A. 2010. Distribution and abundance of natural parasitoid (Hymenoptera: Pteromalidae) populations of house flies and stable flies (Diptera: Muscidae) at the University of Florida Dairy Research Unit. Neotropical Entomology, 39: 424- 429.
Roseland, C.R., Grodowitz, M.J., Kramer, K.J., Hopkins, T.L., and Broce, A.B. 1985. Stabilization of mineralized and sclerotized puparial cuticle of muscid flies. Insect Biochemistry, 15: 521-528.
Rozendaal, J.A. 1997. Vector Control: Methods for use by individuals and communities. World Health Organization, Geneva, Switzerland. pp. 412.
Rueda, L.M., and Axtell, R.C. 1985. Effect of depth of house fly pupae in poultry manure on parasitism by six species of Pteromalidae (Hymenoptera). Journal of Entomological Science, 20: 444–449.
Salem, A., Franc, M., Jacquiet, P., Bouhsira, E., and Liénard, E. 2012. Feeding and breeding aspects of Stomoxys calcitrans (Diptera: Muscidae) under laboratory conditions. Parasite, 19: 309–317.
Salt, R.W. 1961. Principles of insect cold-hardiness. Annual Review of Entomology, 6: 55-74.
Schmidtmann, E.T. 1991. Suppressing immature house and stable flies in outdoor calf hutches with sand, gravel, and sawdust bedding. Journal of Dairy Science, 74: 3956-3960.
Schneiderman, H.A., and Horwitz, J. 1958. The induction and termination of facultative diapause in the chalcid wasps Mormoniella vitripennis (Walker) and Tritneptis klugii (Ratzeburg). Journal of Experimental Biology, 35: 520-551.
138 Scholl, P.J. 1980. A technique for physiologically age-grading female stable flies, Stomoxys calcitrans (L.). Institute of Agricultural and Natural Resources, University of Nebraska Research Bulletin. 198.
Scholl, P.J. 1986. Field population studies of Stomoxys calcitrans (L.) in eastern Nebraska. Southwestern Entomologist, 11: 155–160.
Scholl, P.J., Petersen, J.J., Stage, D.A., and Meyer, J.A. 1981. Open silage as an overwintering site for immature stable flies in eastern Nebraska. Southwestern Entomologist, 6: 253-258.
Schowalter, T.D. and Klowden, M.J. 1979. Blood meal size of the stable fly, Stomoxys calcitrans, measured by the HiCN method. Mosquito News, 39: 110–112.
Semelbauer, M., Mangová, B., Barta, M., and Kozánek, M. 2018. The factors influencing seasonal dynamics and spatial distribution of stable fly Stomoxys calcitrans (Diptera, Muscidae) within stables. Insects, 9:142.
Seymour, R.C. and Campbell, J.B. 1993. Predators and parasitoids of house flies and stable flies (Diptera, Muscidae) in cattle confinements in West Central Nebraska. Environmental Entomology, 22: 212-219.
Sharif, S., Jacquiet, P., Prevot, F., Grisez, C., Raymond‐Letron, I., Semin, M.O., Geffré, A., Trumel, C., Franc, M., Bouhsira, É. and Liénard, E. 2019. Stomoxys calcitrans, mechanical vector of virulent Besnoitia besnoiti from chronically infected cattle to susceptible rabbit. Medical and Veterinary Entomology, 33: 247-255.
Shaw, M.R. 2006. Habitat considerations for parasitic wasps (Hymeoptera). Journal of Insect Conservation, 10: 117-127.
Shaw, M.R. and Hochberg, M.E. 2001. The neglect of parasitic Hymenoptera in insect conservation strategies: the British fauna as a prime example. Journal of Insect Conservation, 5: 253-263.
Showler, A.T. and Osbrink, W.L. 2015. Stable fly, Stomoxys calcitrans (L.), dispersal and governing factors. International Journal of Insect Science, 7: 19-25.
Sileshi, G.W. 2006. Selecting the right statistical model for analysis of insect count data by using information theoretic measures. Bulletin of Entomological Research, 96:479-88.
139 Simmons, S.W. 1944. Observations on the biology of the stable fly in Florida. Journal of Economic Entomology, 37: 680-686.
Sinshaw, A., Abebe, G., Desquesnes, M., and Yoni, W. 2006. Biting flies and Trypanosoma vivax infection in three highland districts bordering Lake Tana, Ethiopia. Veterinary Parasitology, 142: 35-46.
Skoda, S.R. and Thomas, G.D. 1993. Breeding sites of stable flies and house flies. In Rural Flies in the Urban Environment. Edited by G.D. Thomas G.D. and S.R. Skoda. Lincoln, Nebraska: Institute of Agriculture and Natural Resources, University of Nebraska. Research Bulletin No. 317: 61–69.
Skovgård, H. and Nachman, G. 2004. Biological control of house flies Musca domestica and stable flies Stomoxys calcitrans (Diptera: Muscidae) by means of inundative releases of Spalangia cameroni (Hymenoptera: Pteromalidae). Bulletin of Entomological Research, 94: 555–567.
Skovgård, H. and Nachman, G. 2017. Modeling the temperature- and age-dependent survival, development, and oviposition rates of stable flies (Stomoxys calcitrans (Diptera: Muscidae). Environmental Entomology, 46: 1130–1142.
Smith, L. and Rutz, D.A. 1991. Microhabitat associations of hymenopterous parasitoids that attack house fly pupae at dairy farms in Central New York. Environmental Entomology, 20: 675- 684.
Sohier, C., Haegeman, A., Mostin, L., De Leeuw, I., Van Campe, W., De Vleeschauwer, A., Tuppurainen, E.S.M., van den Berg, T., De Regge, N., and De Clerq, K. 2019. Experimental evidence of mechanical lumpy skin disease virus transmission by Stomoxys calcitrans biting flies and Haematopota spp. horseflies. Scientific Reports, 9:20076.
Solórzano, J.A., Gilles, J., Bravo, O., Vargas, C., Gomez-Bonilla, Y., Bingham, G.V., and Taylor, D.B. 2015. Biology and trapping of stable flies (Diptera: Muscidae) developing in pineapple residues (Ananas comosus) in Costa Rica. Journal of Insect Science, 15: 145.
Somme, L. 1961. On the overwintering of house flies (Musca domestica L.) and stable flies (Stomoxys calcitrans (L.)) in Norway. Norsk Entomologiskl Tidsskrift, 11: 191-223.
St. Pierre, A.P., Shikon, V., and Schneider, D.C. 2018. Count data in biology- data transformation or model reformation? Ecology and Evolution, 8: 3077-3085.
140 Stoffer, D. 2020. astsa: applied statistical time series analysis. R package version 1.10.
Talley, J., Broce, A., and Zurek, L. 2009. Characterization of stable fly (Diptera: Muscidae) larval developmental habitat at round hay bale feeding sites. Journal of Medical Entomology, 46: 1310–1319.
Talley, J.L., Broce, A.B., and Zurek, L. 2009. Characterization of stable fly (Diptera: Muscidae) larval development habitat at round hay bale feeding sites. Journal Medical Entomology, 46: 1310–1319.
Taylor, D.B. and Berkebile, D.R. 2006. Comparative efficiency of six stable fly (Diptera: Muscidae) traps. Journal of Economic Entomology, 99: 1415-1419.
Taylor, D.B. and Berkebile, D.R. 2011. Phenology of stable fly (Diptera: Muscidae) larvae in round bale hay feeding sites in eastern Nebraska. Environmental Entomology, 40: 184–193.
Taylor, D. B., Berkebile, D.R., and Scholl, P.J. 2007. Stable fly population dynamics in eastern Nebraska in relation to climatic variables. Journal of Medical Entomology, 44: 765-771.
Taylor, D.B., Moon, R.D., Campbell, J.B., Berkebile, D.R., Scholl, P.J., Broce, A.B., and Hogsette, J.A. 2010. Dispersal of stable flies (Diptera: Muscidae) from larval development sites in a Nebraska landscape. Environmental Entomology, 39: 1101-1110.
Taylor, D.B., Moon, R.D., and Mark, D.R. 2012. Economic impact of stable flies (Diptera: Muscidae) on dairy and beef cattle production. Journal of Medical Entomology, 49: 198–209.
Taylor, D.B., Friesen, K., and Zhu, J. 2017. Precipitation and temperature effects on stable fly (Diptera: Muscidae) population dynamics. Environmental Entomology, 46: 434–439.
Taylor, D.B., Harrison, K., and Zhu, J.J. 2020. Methods for surveying stable fly populations. Journal of Insect Science, 20: 17; 1-8.
Thomas, G.D., Berry, I.L., Skoda, S. R., and Berkebile, D.R. 1990. Seasonal ovipositional status of field populations of female stable flies (Diptera: Muscidae) in cattle feedlots as measured by two sampling methods. Environmental Entomology, 19: 597-604.
Thomas, G.D., Skoda, S.R., Berkebile, D.R., and Campbell, J.B. 1996. Scheduled sanitation to reduce stable fly (Diptera: Muscidae) populations in beef cattle feedlots. Journal of Economic Entomology, 89: 411-414.
141 Todd, D.H. 1964. The biting fly Stomoxys calcitrans (L.) in dairy herds in New Zealand. New Zealand Journal of Agricultural Research, 7: 60–79.
Traversa, D., Otranto, D., Iorio, R., Carluccio, A., Contri, A., Baoletti, B., Bartolini, R., and Giangaspero, A. 2008. Identification of the intermediate hosts of Habronema microstoma and Habronema muscae under field conditions. Medical and Veterinary Entomology, 22: 283–287.
Tukey, J.W. 1977. Exploratory data analysis. Addison-Wesley. Reading, Massachusetts, USA. UCLA Statistical Consulting Group. Negative binomial regression. UCLA: Institute for Digital Research and Education. Retrieved 16 March 2020 from https://stats.idre.ucla.edu/stata/output/negative-binomial-regression/
Urech, R., Bright, R.L., Green, P.E., Brown, G.W., Hogsette, J.A., Skerman, A.G., Elson-Harris, M.M., and Mayer, D.G. 2012. Temporal and spatial trends in adult nuisance fly populations at Australian cattle feedlots. Australian Journal of Entomology, 51: 88–96.
Ushey, K. 2018. RcppRoll: efficient rolling/windowed operations. R package version 0.3.0.
Venables, W.N. and Ripley, B.D. 2002. Modern applied statistics with S (4th ed.). Springer, New York.
Wahl, D.B. 1993. Key to subfamilies of Holarctic and Neotropical Ichneumonidae. In Hymenoptera of the world: An identification guide to families. Edited by H. Goulet, and J.T. Huber. Agriculture Canada, Ottawa.
Warnes, M.L. and Finlayson, L.H. 1987. Effect of host behaviour on host preference in Stomoxys calcitrans. Medical and Veterinary Entomology, 1: 53-57.
Wellington, W.G. 1945. Conditions governing the distribution of insects in the free atmosphere. The Canadian Entomologist, 77: 7-15.
Wickham, H. 2007. Reshaping data with the reshape package. Journal of Statistical Software, 21: 1–20. http://www.jstatsoft.org/v21/i12/
Wickham, H. 2011. The split-apply-combine strategy for data analysis. Journal of Statistical Software, 40: 1–29. http://www.jstatsoft.org/v40/i01/
Wickham, H., François, R., Henry, L., and Müller, K. 2018. dplyr: A grammar of data manipulation. R package version 1.0.2.
142 Wieman, G.A., Campbell, J.B., DeShazer, J.A., and Berry I.L. 1992. The effects of stable flies (Diptera: Muscidae) and heat stress on weight gain and feed efficiency of feeder cattle. Journal of Economic Entomology, 85: 1835–1842.
Wienhold, B.J. and Taylor, D.B. 2012. Substrate properties of stable fly (Diptera: Muscidae) developmental sites associated with round bale hay feeding sites in eastern Nebraska. Environmental Entomology, 41: 213–221.
Wikel, S.K., Aksoy, S., and Dimopoulos, G. [Eds.]. 2017. Arthropod vector: controller of disease transmission, volume 2: Vector saliva-host-pathogen interactions. London, UK: Academic Press. pp. 332.
Williams, D.F. 1973. Sticky traps for sampling populations of Stomoxys calcitrans. Journal of Economic Entomology, 66: 1279–1280.
Woiwood, I.P., Reynolds, D.R., and Thomas, C.D. 2001. Insect movement: mechanisms and consequences. CABI, Wallingford, UK.
Zacks, D.N. and Loew, E.R. 1989. Why is Alsynite fiber glass sheet attractive to stable flies? Optical and behavioural studies. Experimental Biology, 48: 215-222.
Zeileis, A. and Kleiber, C. 2020. countreg: Count data regression. R package version 0.2-1.
Zeileis, A., Kleiber, C., and Jackman, S. 2008. Regression models for count data in R. Journal of Statistical Software, 27: 1-25.
Zumpt, F. 1973. The Stomoxyine biting flies of the world: Diptera, Muscidae: taxonomy, biology, economic importance and control measures. Stuttgart: Gustav Fischer Verlag, 175.
143 APPENDICES
APPENDIX A. Cattle enclosures at three dairy farms near Glenlea, Manitoba.
© Gina Karam Figure A.1. Outdoor Holstein heifer enclosure at the Glenlea Research Station dairy farm near Glenlea, Manitoba. Heifers were kept outdoors year-round in concrete-lined enclosures with straw bed pack laid within the enclosure.
144
© Gina Karam
Figure A.2. Placement of a Coroplast® sticky trap relative to outdoor Holstein heifer enclosure on Red River Holsteins dairy farm near Glenlea, Manitoba. Heifers were kept outdoors on bare soil with a patchy distribution of straw bedding scattered within the enclosure.
145 © Gina Karam
Figure A.3. Holstein heifer enclosure at Vancrest Holsteins dairy farm near Glenlea, Manitoba. Heifers were kept outdoors on bare soil with a patchy distribution of straw bedding scattered within the enclosure.
146 APPENDIX B. Design of Coroplast® sticky traps.
© Kateryn Rochon
Figure B.1. Coroplast® sticky traps were comprised of a white 30.5 cm x 22 cm Coroplast® panel covered with a thin film of Tanglefoot™ adhesive on both sides of the panel using a paintbrush. Wooden stakes approximately 40 cm long were driven into the ground, then panels were stapled to them so the bottoms of the panels were approximately 30 cm above the ground.
147
APPENDIX C. Correlation matrices for examining strength of correlation among environmental variables.
Figure C.1. Correlation matrix of environmental variable measurements taken by Manitoba Agriculture from the St. Adolphe weather station in 2017.
148
Figure C.2. Correlation matrix of environmental variable measurements taken by Manitoba Agriculture from the St. Adolphe weather station in 2018.
149 APPENDIX D. Examples of cross-correlation plots for evaluating the relationship between trapped adult stable fly abundance and environmental variables in the preceding weeks. The cross-correlation plots depict weekly lags on the x-axis and the strength of correlation on the y- axis. The dashed lines above and below the x-axis indicate the significant correlation threshold (+0.500 and -0.500), above and below which correlation factors are statistically significant.
Figure D.1. Cross-correlation plot evaluating the relationship between trapped adult stable fly abundance and average air temperature in the preceding weeks at Glenlea Research Station in 2017. There is no significant relationship observed in this figure.
150
Figure D.2. Cross-correlation plot evaluating the relationship between trapped adult stable fly abundance and average air temperature in the preceding weeks at Red River Holsteins in 2017. Note the significance threshold is crossed at lag 0, indicating a significant relationship between the variables during the week of trapping.
151
Figure D.3. Cross-correlation plot evaluating the relationship between trapped adult stable fly abundance and average air temperature in the preceding weeks at Vancrest Holsteins in 2017. Note the significance threshold is crossed at lag 0, indicating a significant relationship between the variables during the week of trapping.
152