bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

1 Diversity and seasonality of horse (Diptera: Tabanidae) in Uruguay 2 Martín Lucasab; Tiago K. Krolowc; Franklin Riet-Correaa; Antonio Thadeu M. Barrosd; 3 Rodrigo F. Krügere; Anderson Saraviaa; Cecilia Miraballesa* 4 5 a Instituto Nacional de Investigación Agropecuaria (INIA), Plataforma de Salud , 6 Estación Experimental INIA Tacuarembó, Ruta 5 km 386, Tacuarembó, Uruguay, CP 7 45000. 8 b Facultad de Veterinaria, Universidad de la República, Alberto Lasplaces 1550, 9 Montevideo, Uruguay, CP 11600. 10 c Universidade Federal do Tocantins – UFT, Rua 03, Qd 17, S/N, Bairro Jardim dos 11 Ipês, Porto Nacional, TO, Brazil. 12 d Embrapa Gado de Corte, Campo Grande, MS, Brazil. 13 eUniversidade Federal de Pelotas, Instituto de Biologia, Departamento de Microbiologia 14 e Parasitologia, Campus Universitário Capão do Leão, s/nº, Pelotas, RS, Brazil. 15 ⁎ Corresponding author at: Instituto Nacional de Investigación Agropecuaria (INIA), 16 Plataforma de Salud Animal, Estación Experimental INIA Tacuarembó, Ruta 5 17 km 386, Tacuarembó, CP 45000, Uruguay. E-mail address: [email protected] (C. 18 Miraballes). 19 20 Abstract 21 Horse flies (Diptera: Tabanidae) are hematophagous that cause direct and 22 indirect losses in livestock production and are important vectors of pathogens. The aim 23 of this study was to determine the diversity and seasonality of horse species at an 24 experimental farm in Tacuarembó and the diversity of species in different departments 25 of Uruguay. For 20 months, systematic collections were performed using Nzi and 26 Malaise traps in two different environments at the experimental farm. Temperature, 27 humidity and rainfall were recorded using a local climatological station. In addition, 28 nonsystematic collections were made at farms located in the departments of Paysandú, 29 Tacuarembó and Colonia. A total of 3,666 horse flies were collected, allowing the 30 identification of 16 species. Three species were recorded for the first time in Uruguay: 31 Dasybasis ornatissima (Brèthes), Dasybasis missionum (Macquart), and Tabanus aff. 32 platensis Brèthes. A species that had not been previously taxonomically described was 33 identified (Tabanus sp.1). In the systematic captures, the most abundant species were bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

34 Tabanus campestris Brèthes, T. aff. platensis and D. missionum, representing 77.6% of 35 the collected specimens. The environment was an important factor related to the 36 abundance of horse flies, as well as the mean temperature. The horse fly season in 37 Tacuarembó started in September and ended in May, with three evident peaks, the most 38 important one during summer. No horse flies were caught during winter. Variations in 39 the prevalence of species in the different departments were observed, indicating the 40 need to carry out new sampling efforts in different areas.

41 Keywords: Horse fly; Uruguay; Systematic collections; NZI traps 42 Introduction 43 Horse flies (Tabanidae) are hematophagous dipterans that cause direct losses to 44 livestock production due to irritation, stress, and blood loss in , mainly in cattle 45 and horses (Baldacchino et al. 2014). A decrease in weight gains of 0.1 kg up to one 46 kilogram per day in cattle has been described due to the direct effect of horse flies (Foil 47 and Hogsette 1994; Baldacchino et al. 2014). Economic losses are directly related to the 48 number of horse flies present in the environment (Perich et al. 1986). In addition to the 49 losses caused by the direct effects of horse flies, they cause indirect losses due to their 50 role as a mechanical vector of numerous pathogens, including those causing bovine 51 leukosis, vesicular stomatitis, equine infectious anemia, swine fever, anthrax, tularemia, 52 various species of trypanosomes and Anaplasma marginale (Krinsky 1976; Foil 1989). 53 These indirect losses may be even more important than the direct losses, but not all 54 members of the Tabanidae family have the same potential for transmitting disease 55 agents because their hematophagous behavior and anatomical characteristics, which 56 determine the amount of blood they can transport, varies (Magnarelli and Anderson 57 1980; Foil 1989; Scoles et al. 2008). 58 Uruguay has a subtropical climate according to the Köppen classification, 59 presenting four marked seasons with a mean annual temperature of 17.29 °C and a mean 60 humidity of 76.03% (INIA 2019). It is included in the Pampa biome, along with the 61 state of Rio Grande do Sul (Brazil) and the province of Buenos Aires (Argentina). 62 The emergence of the first generation of horse flies during the year depends on 63 the latitude and the season (Chvála et al. 1972; Kruger and Krolow 2015). Horse flies 64 are active mostly in warm seasons, when both the relative humidity and temperature are 65 high. Only females are hematophagous, and adult longevity varies from two to three 66 weeks (Baldacchino et al. 2014). Although the females need a blood meal every three or bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

67 four days, their hematophagous habits are painful, and animals try to remove them, 68 which may cause the interruption of feeding. This interruption causes females to seek 69 other animals to complete a meal (Hornok et al. 2008); they are able to fly several 70 kilometers, reaching speeds of five m/s, which highlights their potential as a mechanical 71 vector (Hornok et al. 2008). 72 To study the diversity and abundance of horse fly species in different 73 environments, a wide variety of traps have been used (Thorsteinson et al. 1965; 74 Thompson 1969). However, the Nzi trap was designed specifically to catch biting flies 75 (Mihok 2002), such as horse flies and stable flies (Stomoxyinae). This trap has been 76 shown to capture a greater number of horse flies and species than other traps (Van 77 Hennekeler et al. 2008), but some genera of horse flies tend to be captured in higher 78 numbers by other traps (Baldacchino et al. 2014). 79 Approximately 4300 species belonging to 137 genera of horse flies have been 80 described worldwide (Coscarón and Papavero 2009; Coscarón and Martínez 2019). The 81 Tabanidae family is subdivided into four subfamilies: Chrysopsinae, Pangoniinae, 82 Scepsidinae and . The subfamilies Chrysopsinae and Tabaninae are 83 considered the most relevant in the mechanical transmission of pathogens (Hawkins et 84 al. 1982; Scoles et al. 2008). In the Neotropical region, 71 genera and 1205 species have 85 been described (Coscarón and Papavero 2009; Henriques et al. 2012; Coscarón and 86 Martínez 2019). Regionally, 350 species of Tabanidae have been described in Argentina 87 (Coscarón 1998) and 480 in Brazil (Taxonomic Catalog of Fauna of Brazil 2019). 88 In Uruguay, 43 species belonging to 14 genera have been reported (Coscarón 89 and Martínez 2019); however, most of these captures were performed in the XIX and 90 XX centuries, so it is unknown whether this diversity of horse flies still exists in the 91 country. Additionally, there have been no studies in the country regarding the 92 abundance, ecology and seasonality of the Tabanidae family. 93 The objectives of this study were to evaluate the seasonality of the horse fly 94 species present at an experimental farm in Tacuarembó and the diversity of species in 95 different departments of Uruguay. 96 97 Materials and Methods 98 For 20 months, between October 2017 and June 2019, systematic collections 99 were performed using four Nzi traps and two Malaise traps at the Experimental Farm of 100 INIA La Magnolia (EFILM) in the department of Tacuarembó (31°42'29.0" S, 55° bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

101 48'09.1" W). Two Nzi traps and one Malaise trap were placed in a "lowland” habitat 102 with native forest and creeks, while the other two Nzi traps and one Malaise trap were 103 placed in "highland" habitat, which were clearer environments, near a cultivated 104 eucalyptus forest. All these traps remained active during the 20 months of the study. 105 According to a conventional climatological station located at the EFILM, the following 106 experimental parameters were recorded on a monthly basis: average maximum, average 107 minimum and average mean temperatures; average mean relative humidity and 108 accumulative rainfall. 109 In addition, nonsystematic collections were performed on farms located at the 110 departments of Paysandú, Tacuarembó and Colonia using traps and/or manually. These 111 captures were carried out with the aim of expanding the diversity of horse fly species 112 collected. A defined collection time pattern was not established, and the capture 113 methods varied, so these data were not considered for the analysis of seasonality and 114 abundance of the horse fly species. 115 Of the total number of horse flies captured, samples from one to 10 specimens of 116 each suspected species were prepared for taxonomic identification. 117 118 Data collection 119 Systematic collections 120 Between October and December 2017, two Nzi traps (Rincon-Vitova, USA) 121 (Nzi 1 and Nzi 4) and one Malaise trap (Rincon-Vitova, USA) (Mal 6) were placed in a 122 clear field environment close to artificial eucalyptus forests (highland), and two Nzi 123 traps (Nzi 2 and Nzi 3) and one Malaise trap (Mal 5) were placed in an environment 124 close to creeks and native forest (lowland). Because alcohol was not used in the 125 collectors for the conservation of horse flies, during summer (December to February), 126 when the populations of horse flies were highest, the collections were carried out at 127 least three times a week to prevent desiccation of the captured specimens. During winter 128 (June to August), collections were performed at least once a week, while in spring 129 (September to November) and autumn (March to May), the collections were made from 130 one to three times a week depending on the number of specimens collected. 131 132 Nonsystematic collections 133 To identify species in different areas of the country or species that might not be 134 captured by traps, sporadic samples were performed in the departments of Tacuarembó bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

135 (three farms), Paysandú (one farm) and Colonia (two farms) (Table 1). These samplings 136 were performed manually and/or with Nzi traps. In the manual sampling, some horse 137 flies were caught while feeding on animals, while others were manually caught in the 138 environment. 139 140 Statistical analysis 141 To perform descriptive and inferential analyses, two databases were created in 142 Excel, one for the systematic captures and the other for the nonsystematic captures and 143 were then imported into STATA 2014 (Stata Corp 2015) after identifying any data 144 errors. 145 The datasets generated during and/or analyzed during the current study are 146 available at Miraballes et al (2019). 147 Descriptive analysis 148 Descriptive analyses were performed based on the total number of individuals 149 captured by location, date, season, trap type, type of manual capture (environmental or 150 on animals) and species. 151 Inferential analysis 152 The influence of mean temperature (mt) and environment (lowland and 153 highland) on the abundance of horsefly species was tested using generalized linear 154 models (GLMs). The models were obtained by extraction of significant terms (p<0.05) 155 from the full model, which included all of the variables mt, maximum temperature, 156 minimum temperature, relative humidity, rainfall, and environment and their 157 interactions, as suggested by Crawley (2007). Each term was analyzed followed by 158 ANOVA and chi-square tests (Chi) to recalculate the deviation explained by the other 159 terms. 160 We tested the hypothesis that the mt in different environments alters the 161 abundance of Tabanidae using models with response variables that assume integer 162 values with regard to abundance. The explanatory variables that were used were mt and 163 environment without interaction terms. The complete models that were used for 164 hypothesis testing are as follows: 165 166 Abundance=mt+environment 167 bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

168 The different years in which the experiments were conducted were not used 169 as explanatory variables because there was autocorrelation between mt and year. In 170 the models, a plus sign (+) denotes the addition of a variable. A quasi-Poisson error 171 distribution and log link function were used to model the estimated abundance. 172 173 Results 174 During the study, 3,666 horse flies were collected: 3,211 through systematic 175 collections and 455 by nonsystematic ones. These specimens represented two 176 subfamilies, three tribes, six genera and 16 species. Fourteen species were definitively 177 identified taxonomically, and one species was preliminarily identified as Tabanus 178 affinity platensis Brèthes (T. aff. platensis); an undescribed species of Tabanus 179 (Tabanus sp.1) was found. In addition, it was not possible to taxonomically identify 11 180 individuals belonging to the genus Tabanus that were likely members of the same 181 species, and they were classified as Tabanus spp. 182 183 Systematic collections 184 Of the 3,211 specimens captured by this method, a total of 15 species were 185 identified, of which three were Tabanus campestris Brèthes; T. aff. platensis and 186 Dasybasis missionum (Macquart) represented 77.6% of the total captures (Table 2). 187 Other species of varying relative abundance identified were a new species to be 188 described (Tabanus sp.1), Tabanus triangulum Wiedemann, Tabanus acer Brèthes and 189 Poeciloderas quadripunctatus (Fabricius). The remaining species only accounted for 190 3.0% of the captures. Additionally, 3.4% of the individuals could not be identified 191 because they were in poor condition. 192 Traps located in the "lowland" habitat were responsible for 92.5% of the total 193 catches during the study, and within this habitat the Nzi 3 trap captured 84.5% of the 194 total specimens (Table 2). The most abundant species in the “lowland” were T. 195 campestris (30.0%), T. aff. platensis (28.3%) and D. missionum (22.9%). Captures in 196 the "highland" habitat represented only 7.5% of the total and the three most abundant 197 species were: Tabanus sp.1 (32.2%), T. triangulum (18.4%) and D. missionum (14.2%). 198 Figure 1 depicts the seasonality of the horse flies in the "lowland" habitat, where 199 most individuals were caught, and shows that they were active from the end of 200 September to the beginning of May, with no activity from June to August. It also shows 201 the presence of three peaks: one at the end of spring, another one in mid-summer (the bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

202 most pronounced one), and a third one during the fall. A decrease in the total captures 203 during the second season was also observed. 204

205 206 Fig. 1. Number of horse flies caught in the “lowland” environment by systematic 207 captures during the study period. 208 Of the total captures in both habitats, 2,336 individuals were captured during the 209 first season of horse fly activity (September 2017-May 2018), while 875 horse flies 210 were captured during the second season (September 2018-May 2019). Figure 2 shows 211 the different seasonality of the three most prevalent species. It can be observed that D. 212 missionum showed a different behavior, presenting a peak during late spring and 213 autumn. bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

214 215 Fig. 2. Seasonality of the 3 species most captured in the systematic collection: T. 216 campestris (Fig. 2A), T. aff. platensis (Fig. 2B) and D. missionum (Fig. 2C). 217 218 The Nzi traps were responsible for 97.1% of the collections, while the Malaise 219 traps captured 2.9%. In the “lowlands”, the Malaise trap captured 1.4% of the total 220 individuals, while in the "highlands", it was responsible for 21.7% of the collections. Of 221 the 15 species captured in the systematic collections, the Nzi traps only did not capture 222 Tabanus pungens Wiedemann, while the Malaise traps did not capture four of the 15 bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

223 species: P. quadripunctatus, Tabanus fuscofasciatus Macquart, Tabanus claripennis 224 (Bigot) and Dasybasis ornatissima (Brèthes). 225 Throughout the study period, the monthly average temperature ranged from 10.4 226 °C to 25.0 °C, the average minimum temperatures were between 6.0 °C and 19.4 °C, 227 and the average maximum temperature ranged between 15.2 °C and 31.4 °C. The 228 accumulated rainfall during this period was 2715.3 millimeters (mm), and the average 229 relative humidity varied between 56% and 90% (Figure 3; Supplementary Table 1). 230 During the first season (September 2017-May 2018), the accumulated rainfall was 944.8 231 mm, while in the second season (September 2018-May 2019), the accumulated rainfall 232 was 1309.6 mm, marking a clear difference in rainfall, especially during the summer 233 capture peak (December-January).

234 235 Fig. 3. Monthly values of average maximum temperature, average minimum 236 temperature and accumulated rainfall throughout the study period 237 238 Considering the response variable of the abundance of horse flies captured by 239 the traps, during the entire study period, the average minimum and maximum 240 temperatures, relative humidity and rainfall did not have a significant effect (p > 0.05) 241 and were removed from the final model. The significant differences found were due to

242 the increase in the mt and the environment. The mt (Chi1;129= 3,989.3, p<0.001) and

243 environment (Chi1;128= 2,675.1, P<0.001) influenced the variation in the abundance of 244 horseflies (Figure 4), and there was no interaction between these variables. According to 245 the abundance models, there were 12 times more horseflies in the lowland environment bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

246 than in the highland, and individuals were predicted to be present at average 247 temperatures of 18 °C in the highland and 12 °C in the lowland. The models predicted 248 approximately 200 individuals per trap at 25 °C in the lowland, while in the highland 249 habitat, approximately 15 individuals were predicted.

250

251 Fig. 4. Abundance of horseflies in Tacuarembo, Uruguay, as a function of mean 252 temperature (°C) (mt), month, and year according to GLM analysis with a quasi-Poisson 253 distribution.

254 Nonsystematic collections 255 A total of 455 individuals were collected in the departments of Colonia (n= 343), 256 Tacuarembó (n= 74) and Paysandú (n= 38) (Figure 5; Table 3).

257 258 Fig. 5. Location and capture method of nonsystematic collections bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

259 In Colonia, 78.7% were Poeciloderas lindneri (Kröber), while in Paysandú, the 260 two most captured species were T. fuscofasciatus (47.4%) and P. lindneri (42.1%). 261 Regarding the manual captures in the Tacuarembó department, the most abundant 262 species was the undescribed Tabanus sp.1 (60.8% of the captures). 263 Through this sampling, a total of 12 species were collected, with Dichelacera 264 unifasciata Macquart being the only species not caught during the systematic captures. 265 Eleven individuals classified as Tabanus spp. were also captured, but taxonomic 266 identification was not achieved. 267 268 Discussion 269 Of the 14 species taxonomically identified, D. ornatissima and D. missionum 270 were recorded for the first time in Uruguay, although both species have also been 271 recorded in Argentina, and D. missionum has been recorded in Brazil (Coscarón and 272 Papavero 2009; Kruger and Krolow 2015). It is necessary to achieve the final 273 identification of T. aff. platensis by comparison with "type" specimens. Tabanus 274 platensis has not been reported in Uruguay, but its presence is expected, since its 275 distribution in Argentina includes Chaco, Santa Fe, Entre Ríos and Buenos Aires 276 (Coscarón and Papavero 2009). 277 Two of the three new records for Uruguay (D. missionum and T. aff. platensis) 278 were among the three most prevalent species found during the systematic captures and 279 had not been described as abundant in similar studies in southern Brazil (Kruger and 280 Krolow 2015). An undescribed species (Tabanus sp.1) was also identified, which will 281 be described posteriorly. 282 The fact that captures were concentrated between September and May, without 283 horse fly activity during winter, differs from what was observed in southern Brazil, 284 where, despite a marked decrease in captures during winter, the presence of horse flies 285 still occurred (Kruger and Krolow 2015). In southern Brazil, there is a peak in horse fly 286 abundance during spring and a gradual decrease as winter approaches, but it is still 287 possible to capture horse flies during this season (Kruger and Krolow 2015). In the 288 Tacuarembó department, which belongs to the same Pampa biome as southern Brazil, 289 the largest number of horse flies was observed during summer, with lower peaks during 290 spring and autumn. This variation can be explained by climatological differences 291 between zones or between years as well as differences in the species of horse flies 292 captured in the two regions (Burnett and Hays 1974; Gorayeb 1995; Kruger and Krolow bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

293 2015). In Tacuarembó, the peak recorded in autumn was mostly due to D. missionum. 294 This peak was not recorded in Rio Grande do Sul, where D. missionum was not among 295 the most abundant species (Kruger and Krolow 2015). It is important to note that 296 different species respond differently to the variation in climatic factors (Van Hennekeler 297 et al. 2008). 298 Throughout the study period, only the mean temperature and environment had 299 statistically significant effects on the number of individuals captured. In contrast, some 300 studies have related the abundance of horse flies with an increase in relative humidity 301 (Cardenas et al. 2012) interacting with the mean temperature (Kruger and Krolow 2015) 302 and highlight that the influence of this variable differs according to species (Van 303 Hennekeler et al. 2008; Kruger and Krolow 2015). 304 The environment was a determining factor related to the number of horse flies 305 collected. The traps located in “lowland” habitats, with native forest and creeks, 306 captured 12 times more horse flies on average than the traps located in the “highland” 307 habitat. These findings would allow a more efficient use of traps as a measure to 308 mitigate horse fly attack on commercial farms through their use in environments with a 309 greater number of horse flies. In addition, grazing management patterns that preclude 310 the animals from grazing in the most infested paddocks during population peaks could 311 be established to prevent the hosts and parasites from being simultaneously present 312 (Foil and Hogsette 1994; Baldacchino et al. 2014). Barros (2001) did not report relevant 313 differences in the number of horse flies in the Pantanal biome (Brazil) when comparing 314 captures carried out in areas of native forests with captures carried out in grasslands. 315 However, Gorayeb (2000) found differences in the abundance and diversity of species 316 in different environments in Santa Barbara, Pará (Brazil), after performing captures in 317 two environments: near forest and in areas of clear fields (grassland). This author 318 captured 71.5% of the specimens in the environment near the forest and 28.5% in the 319 clear field environment. In addition, 23 of the 47 species were present in both 320 environments, while 14 were exclusively found in the environment near forest, and 9 321 were only found in the grassland environment. 322 Even when the first season of captures occurred during a dry summer and the 323 second during a rainy summer, a similar pattern in population peaks was observed. 324 However, during the second season of systematic captures, a remarkable decrease was 325 observed in the number of individuals caught. This decrease in captures during the 326 second season may have been influenced by the constant presence of the traps; however, bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

327 no remarkable decrease in the density of horse fly populations in the environment has 328 been evident in previous studies using traps as a control method (Vale et al. 1988; Foil 329 and Hogsette 1994; Baldacchino et al. 2014). Another possibility is that this decrease 330 was related to climatological and/or environmental factors, such as the amount of 331 rainfall during the second year, which could affect the abundance and activity of the 332 horse flies, as well as trap efficiency. 333 The variations in the prevalence and the species captured in different regions 334 denote the need to continue the collection of horse flies in different areas of the country. 335 As an example, in the department of Colonia, the most prevalent species found was P. 336 lindneri, while in the systematic collections performed in the department of 337 Tacuarembó, this species accounted for less than one percent of the total captures. 338 Additionally, Coscarón and Martínez (2019) highlighted the importance of carrying out 339 new captures in Uruguay, especially in rural areas, to assess whether the diversity of 340 species reported is still maintained after urbanization. 341 Horse flies are detrimental to livestock production due to their direct effects on 342 animals and their ability to act as vectors for at least 35 pathogens (Krinsky 1976, Foil 343 1989, Baldacchino et al. 2014). Relevant agents of disease potentially transmissible by 344 horse flies in Uruguay include A. marginale, bovine leucosis virus, Brucella abortus, 345 Leptospira spp, and equine infectious anemia virus. More studies need to be performed 346 to determine the role of the Tabanidae species found in Uruguay on the epidemiology of 347 these agents. 348 349 Acknowledgments 350 We would like to acknowledge Marcelo Alfonso for help with the artwork preparation 351 and Gonzalo Escayola for helping with the data collection. 352 353 Conflict of interest statement 354 On behalf of all authors, the corresponding author states that there is no conflict of 355 interest. 356 357 References 358 Baldacchino F, Desquesnes M, Mihok S, Foil L, Duvallet G, Jittapalapong S (2014) 359 Tabanids: Neglected subjects of research, but vectors of disease agents bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

360 Important!. Infect Genet Evol 28: 596-615. 361 http://dx.doi.org/10.1016/j.meegid.2014.03.029 362 363 Barros ATM (2001) Seasonality and relative abundance of Tabanidae (Diptera) 364 captured on horses in the Pantanal, Brazil. Mem Inst Oswaldo Cruz 96: 917-923. 365 http://dx.doi.org/10.1590/S0074-02762001000700006 366 Burnett AM, Hays KL (1974) Some influences of meteorological factors on flight 367 activity of female horse flies (Diptera: Tabanidae). Environ Éntomol 3: 515-521. 368 https://doi.org/10.1093/ee/3.3.515 369 Cardenas RE, Hernandez NL, Barragán AR, Dangles O (2012) Differences in 370 morphometry and activity Among tabanid fly assemblages in an Andean tropical 371 montane cloud forest: indication of altitudinal migration?. Biotropica 45: 63-72. 372 https://doi.org/10.1111/j.1744-7429.2012.00885.x 373 Catálogo Taxonômico da Fauna do Brasil (2019) 374 http://fauna.jbrj.gov.br/fauna/listaBrasil/ConsultaPublicaUC/ConsultaPublicaUC. 375 do Accessed 04 August 2019. 376 Chvála M, Lyneborg L, Moucha J (1972) The Horse Flies of Europe (Diptera, 377 Tabanidae). Entomological Society of Copenhagen, Copenhagen. ISBN: 978-09- 378 00-84857-5 379 Coscarón S, Martinez M (2019) Checklist of tabanidae (Insecta: Diptera) from Uruguay. 380 J Éntomol Soc Argentina 78: 40-46. https://doi.org/10.25085/rsea.780105 381 Coscarón S, Papavero N (2009) Catalogue of Neotropical Diptera. Tabanidae. 382 Neotropical Diptera 16: 1-199. 383 Coscarón S (1998) Tabanidae (Diptera-Insecta). In: Morrone JJ, Coscarón S (ed) 384 Biodiversidad de artrópodos argentinos: una perspectiva biotaxonómica Sur, La 385 Plata, pp 341- 352 386 Foil LD (1989) Tabanids as vectors of disease agents. Parasitol Today 5: 88-96. 387 https://doi.org/10.1016/0169-4758(89)90009-4 388 Foil LD, Hogsette JA (1994) Biology and Control of Tabanids, stable flies and horn 389 flies. Rev Sci Tech 13: 1125-1158. 390 Gorayeb IS (1995) Tabanidae (Diptera) da Amazônia. XI. Sazonalidade espécies das da 391 Amazônia Oriental e correlação com fatores climáticos. Bol Mus Paraense Emílio 392 Goeldi 9: 241-281. 393 http://repositorio.museu-goeldi.br:8080/jspui/handle/mgoeldi/1004 bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

394 Gorayeb IS (2000) Tabanidae (Diptera) da Amazônia. XVI – Atividade diurna de 395 hematofagia de espécies da Amazônia Oriental, em áreas de mata e pastagem, 396 correlacionada com fatores climáticos. Bol Mus Paraense Emílio Goeldi 16: 23- 397 63. http://repositorio.museu-goeldi.br:8080/jspui/handle/mgoeldi/1021 398 Hawkins JA, Love JN, Hidalgo RJ (1982) Mechanical transmission of Anaplasmosis by 399 tabanids (Diptera, Tabanidae). Am J Vet Res 43: 732–734. 400 Henriques AL, Krolow TK, Rafael JA (2012) Corrections and additions to Catalog of 401 Neotropical Diptera (Tabanidae) of Coscarón & Papavero (2009). Revista 402 Brasileira de Entomologia 56: 277-280. 403 http://dx.doi.org/10.1590/S0085-56262012005000042 404 Hornok S, Földvári G, Elek V, Naranjo V, Farkas R, de la Fuente J (2008) Molecular 405 identification of Anaplasma marginale and rickettsial endosymbionts in blood- 406 sucking flies (Diptera: Tabanidae, Muscidae) and hard ticks (Acari: Ixodidae). Vet 407 Parasitol 54: 354-359. https://doi.org/10.1016/j.vetpar.2008.03.019 408 INIA (2019). GRAS INIA data. http://www.inia.uy/gras/Clima/Banco-datos- 409 agroclimatico). 410 Krinsky WL (1976) Animal-disease agents transmitted by horse flies and deer flies 411 (Diptera, Tabanidae). J Med Éntomol 13: 225-275. 412 https://doi.org/10.1093/jmedent/13.3.225 413 Krüger RF, Krolow TK (2015) Seasonal patterns of horse fly richness and abundance in 414 the Pampa biome of southern Brazil. J Ecol Vector 40: 364-372. 415 https://doi.org/10.1111/jvec.12175 416 Magnarelli LA, Anderson JF (1980) Feeding-behavior of Tabanidae (Diptera) on cattle 417 and serological analysis of partial blood meals. Environ Éntomol 9: 664-667. 418 https://doi.org/10.1093/ee/9.5.664 419 Mihok S (2002) The development of a multipurpose trap (the Nzi) for tsetse and other 420 biting flies. Bull Éntomol Res 92: 385-403. https://doi.org/10.1079/BER2002186 421 Miraballes C, Lucas M, Krolow T, Riet-Correa F, Medeiros de Barros AT, Kruger R, 422 Saravia A (2019) data for Diversity and seasonality of horse flies (Diptera: 423 Tabanidae) in Uruguay”, Mendeley Data, v2 424 http://dx.doi.org/10.17632/zstbdddgv8.2 425 Perich MJ, Wright RE, Lusby KS (1986) Impact of horse flies (Diptera: Tabanidae) on 426 beef-cattle. J Econ Éntomol 79: 128-131. https://doi.org/10.1093/jee/79.1.128 bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

427 Scoles G, Miller A, Foil L (2008) Comparison of the Efficiency of Biological 428 Transmission of Anaplasma marginale (Rickettsial: Anaplasmataceae) by 429 Dermacentor andersoni Stiles (Acari: Ixodidae) With Mechanical Transmission 430 by the Horse Fly, Tabanus fuscicostatus Hine (Diptera: Muscidae). J 431 Méd Éntomol 45: 109-114. https://doi.org/10.1093/jmedent/45.1.109 432 StataCorp. (2015) Stata Statistical Software: Release 14. College Station, TX: StataCorp 433 LP. 434 Thompson, PH (1969) Methods for Collecting Tabanidae (Diptera) Ann Éntomol Soc 435 Am 62: 50-57. https://doi.org/10.1093/aesa/62.1.50 436 Thorsteinson AJ, Bracken GK, Hanec W (1965) The orientation of horse flies and deer 437 flies (Tabanidae, Diptera). III. The use of traps in the study of orientation of 438 Tabanids in the field. Éntomol Exp Appl 8: 189-192. 439 https://doi.org/10.1111/j.1570-7458.1965.tb00853.x 440 Vale GA, Lovemore DF, Flint S, Cockbill GG (1988) Odour-baited targets to control 441 tsetse flies, Glossina spp. (Diptera: Glossinidae), in Zimbabwe. Bull Éntomol Res 442 78: 31–49. https://doi.org/10.1017/S0007485300016059 443 Van Hennekeler K, Jones RE, Skerratt LF, Fitzpatrick LA, Reid SA, Bellis GA (2008). 444 A comparison of trapping methods for Tabanidae (Diptera) in North Queensland, 445 Australia. Méd Vet Éntomol 22: 26-31. 446 https://doi.org/10.1111/j.1365-2915.2007.00707.x 447 448 List of figures 449 Fig. 1. Number of horse flies caught in the “lowland” environment by systematic 450 captures during the study period. 451 Fig. 2. Seasonality of the 3 species most captured in the systematic collection: T. 452 campestris (Fig. 2A), T. aff. platensis (Fig. 2B) and D. missionum (Fig. 2C). 453 Fig. 3. Monthly values of average maximum temperature, average minimum 454 temperature and accumulated rainfall throughout the study period 455 Fig. 4. Abundance of horseflies in Tacuarembo, Uruguay, as a function of mean 456 temperature (°C) (mt), month, and year according to GLM analysis with a quasi-Poisson 457 distribution. The models for each environment are

458 and 459 Fig. 5 Location and capture method of nonsystematic collections bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

460

461

462 463 464 465 Table 1. Location and capture method of nonsystematic horse flies collections 466 Identifier Location Department Capture method 31°42'29.0"S; Manually (from Manually 7 Tacuarembó 55°48'09.1"W environment) 31°21'37.8"S Manually (feeding on Manually 8 Tacuarembó 56°05'14.6"W cattle) 31°39'59.8"S Manually (from Manually 9 Tacuarembó 55°57'32.5"W environment) 31°28'29.4"S Manually (feeding on Manually 10 Paysandú 57°53'44.4"W cattle) 31°28'29.4"S NZI 11 Paysandú NZI trap 57°53'44.4"W 34°17'30.2"S NZI 12 Colonia NZI trap 57°37'41.4"W 34°18'14.1"S Manually (feeding on Manually 13 Colonia 57°31'42.7"W cattle) 34°18'14.1"S NZI 14 Colonia NZI trap 57°31'42.7"W 467 468 469 470 471 472 473 474 475 476 477 478 479 bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

480 Table 2. Number of individuals and species of horse flies caught by traps in systematic collections between October 481 2017 and June 2019 Lowland Highland 482 Traps 483 NZI 2 NZI 3 MAL 5 NZI 1 NZI 4 MAL 6 TOTAL (%) Species 484 Number of individuals Catachlorops circumfusus 0 1 0 0 2 1 4 (0.1) 485 Chrysops brevifascia 1 7 2 1 0 0 11 (0.3) 486 Dasybasis missionum 96 564 21 2 14 18 715 (22.3) 487 D. ornatissima 0 1 0 0 0 0 1 (0.0) Poeciloderas lindneri 0 17 0 0 0 1 18 (0.6) 488 P. quadripunctatus 3 30 0 0 0 0 33 (1.0) 489 Tabanus acer 9 31 1 6 12 3 62 (1.9) T. aff. platensis 3 837 0 14 11 2 867 (27.0) 490 T. campestris 28 863 2 5 6 7 911 (28.4) 491 T. claripennis 1 17 0 0 5 0 23 (0.7) 492 T. fuscofasciatus 3 16 0 0 1 0 20 (0.6) T. fuscus 1 17 1 0 0 0 19 (0.6) 493 T. pungens 0 0 0 0 0 1 1 (0.0) 494 T. sp.1 59 203 3 23 40 14 342 (10.7) 495 T. triangulum 5 24 3 27 14 3 76 (2.4) Unidentified 6 87 9 1 3 2 108 (3.4) 496 497 TOTAL (%) 215 (6.7) 2715 (84.5) 42 (1.3) 79 (2.5) 108 (3.4) 52 (1.6) 3211 (100)

498 499 bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

500 Table 3. Number of individuals and species collected in nonsystematic collections using traps or manually in the departments of Paysandú, Colonia and Tacuarembó Tacuarembó Paysandú Colonia Capture method Manually 7 Manually 8 Manually 9 Manually 10 NZI 11 NZI 12 Manually 13 NZI 14 Total Species Number of individuals (%) Chrysops brevifascia 0 0 1 (7,7) 0 0 0 0 0 1 Dasybasis missionum 1 (2,0) 10 (100,0) 0 0 0 0 0 15 (27,3) 26 Dichelacera unifasciata 0 0 5 (38,4) 0 0 0 0 0 5 Poeciloderas lindneri 0 0 0 3 (75,0) 13 (38, 2) 256 (90,2) 4 (100) 10 (18,2) 286 P. quadripunctatus 0 0 1 (7,7) 0 0 0 0 0 1 Tabanus acer 7 (13,7) 0 0 0 0 0 0 0 7 T. aff. platensis 0 0 2 (15,4) 0 2 (5,9) 0 0 1 (1,8) 5 T. campestris 2 (3,9) 0 0 0 1 (3,0) 0 0 0 3 T. claripennis 0 0 0 0 0 4 (1,4) 0 2 (3,6) 6 T. fuscofasciatus 0 0 0 0 18 (52,9) 0 0 0 18 T. sp.1 41(80,4) 0 4 (30,8) 1 (25,0) 0 0 0 0 46 T. spp. 0 0 0 0 0 6 (2,1) 0 5 (9,1) 11 T. triangulum 0 0 0 0 0 0 0 22 (40,0) 22 Unidentified 0 0 0 0 0 18 (6,3) 0 0 18 TOTAL 51 10 13 4 34 284 4 55 455 501 502 bioRxiv preprint doi: https://doi.org/10.1101/794479; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

503 Supplementary table 1. Climatic variables that were recorded and evaluated throughout the study period Average Average Average maximum minimum Accumulated Year Month mean relative temperature temperature rainfall mm humidity % ºC °C 2017 Sep 20.5 12.6 210.4 85 2017 Oct 21.9 12.1 214.8 75 2017 Nov 25.5 12.2 45.1 64 2017 Dec 30.6 16.2 38.6 58 2018 Jan 31.4 18.2 37.3 56 2018 Feb 30.3 16.6 64.9 61 2018 Mar 27.4 13.7 64.0 65 2018 Apr 27.0 16.2 62.1 77 2018 May 19.5 11.1 207.6 83 2018 Jun 15.2 6.0 29.3 83 2018 Jul 15.2 8.0 193.0 85 2018 Ago 16.2 6.8 147.3 82 2018 Sep 21.3 12.7 250.7 81 2018 Oct 22.3 11.7 35.0 73 2018 Nov 26.2 14.8 99.1 71 2018 Dec 27.2 15.4 177.2 73 2019 Jan 28.0 19.4 362.7 80 2019 Feb 27.9 18.6 111.8 76 2019 Mar 26.0 16.2 40.3 78 2019 Apr 24.1 15.1 133.7 81 2019 May 19.0 11.8 99.1 90 2019 Jun 20.4 10.5 91.3 86 504 505