Protection of livestock against bluetongue virus vector Culicoides imicola using insecticide-treated netting in open areas


Dr C. Calvete, Centro de Investigación y Tecnología Agroalimentaria (CITA-Gobierno de Aragón), Unidad de Sanidad Animal, Ctra Montañana 930, 50059 Zaragoza, Spain. Tel.: +34 76 716453; Fax: +34 76 716335; E-mail:


The protection of livestock against Culicoides species (Diptera: Ceratopogonidae) using physical barriers or chemically treated barriers is difficult owing to the small size of these biting midges and animal welfare concerns associated with the reduction of air flow. Culicoides imicola Kieffer is the main bluetongue virus vector in the Mediterranean basin, including the southern Iberian peninsula, where livestock is mainly housed in open pens or sheds which offer no physical protection against C. imicola. In this study we assessed the efficacy of surrounding yearling ewe pens with a canvas barrier or a cypermethrin-treated canvas barrier in reducing the entry of Culicoides spp. and C. imicola. Analyses were based on comparisons of Culicoides catches in traps in pens with and without barriers, and in traps located outside pens. Although there was no clear reduction in the abundance of Culicoides other than C. imicola in pens with either barrier, the C. imicola presence was markedly reduced by the insecticide-treated barrier compared with the untreated barrier; the latter did not reduce the abundance of this species in pens. Estimates of the protection conferred against C. imicola by the treated barrier differed depending on whether catch comparisons were based on outside traps or on traps located inside no-barrier pens. The results suggest that the use of insecticide-treated barriers may reduce contact between livestock and C. imicola in open areas or sheds. More research is necessary to assess the degree of protection as a function of barrier height, C. imicola abundance, and the size of the area to be protected.


Culicoides biting midges (Diptera: Ceratopogonidae) are the main vectors implicated in the transmission of bluetongue virus (BTV), which affects domestic ruminants (Mellor et al., 2000). The efficacy of chemical treatment of animals (Mullens et al., 2000, 2001; Doherty et al., 2001; Mehlhorn et al., 2008), farm premises or surrounding areas (Satta et al., 2004; Calvete et al., 2007) with repellent or insecticide compounds as a way to protect livestock from Culicoides attack has been investigated with equivocal results. The protection of livestock from Culicoides spp. during BT epizootics by increasing shelter or stabling has also been proposed. However, depending on the endophilic/exophilic character of the species implicated in BTV transmission, the level of protection has been highly variable and in some situations the necessary degree of confinement has animal welfare implications (Bishop et al., 1994; Meiswinkel et al., 2000; Doherty et al., 2004). Only when the BTV vector is exophilic does increased stabling or shelter apparently confer any protection against the disease, as has been recorded in some areas where Culicoides imicola Kieffer is the main disease vector (Theiler, 1921; Barnard, 1997).

Culicoides imicola is considered the major field vector of BTV in many areas of the world (Mellor et al., 2000), including the Mediterranean basin (Mellor et al., 1983; Mellor, 1996; Mellor & Wittmann, 2002). This species is thought to have been the main vector in recent BT epizootics in the southern and central Iberian peninsula (Miranda et al., 2003, 2004; Calvete et al., 2008) and the confinement of livestock in stables has been recommended to reduce the risk for BTV transmission. However, because climatic conditions are mild in this area, livestock are usually confined in sheds or open yards, and many farmers have no facilities in which to protect animals by stabling. In addition, it has recently been shown that, by contrast with previous studies suggesting that this species is exophilic, C. imicola in this area apparently actively enters farm buildings through wall openings and reaches higher abundances inside than outside (Calvete et al., 2009). This suggests that, as for other Culicoides species with endophilic tendencies (Meiswinkel et al., 2000), the stabling necessary for protection against BTV is likely to create animal welfare problems.

An option for reducing the contact of C. imicola with livestock in these areas involves using chemically treated net barriers placed around sheds or yards where livestock is confined nightly or permanently. This approach parallels the use of net barriers to prevent the entry of some pests into crop areas, or to protect people from biting midges (Schreck & Kline, 1983). Adequate ventilation is afforded by the absence of a roof, so the use of chemically treated net barriers may provide an inexpensive way of enhancing the protection of livestock against Culicoides attack without putting animal welfare at risk. The current study presents the results of a trial carried out to evaluate the efficacy of an insecticide-treated canvas barrier to prevent the entry of C. imicola into pens housing yearling ewes.

Materials and methods

The study was performed in August and September 2007 in Badajoz province, in the southwest of peninsular Spain. Three temporary pens of 62 m2 each were constructed 100 m apart using trellis dividers 2 m high. Each pen was equipped with an automatic water dispenser and the soil surface was free of vegetation. On each night of the study, 50 yearling ewes were placed in each of the three pens. The yearling ewes were let out 1 h after sunrise each morning and spent the day grazing in the field surrounding each pen. One hour before sunset they were returned to their pens. Each group of yearling ewes was always housed in the same pen.

Fifteen days after the pens were constructed; a canvas barrier 2.6 m high was added to the outer side of the trellis dividers in two of the pens and the third pen was retained as a reference without a barrier. Each canvas barrier completely surrounded its pen and was fixed to the soil with stakes and stones to prevent Culicoides entering beneath it. The barriers were kept closed at all times other than when the yearling ewes were let out from or returned to the pens. One day before the barrier installation, one of the canvas barriers was treated in the laboratory by spraying with a 0.5-g/L solution of cypermethrin until the canvas was thoroughly impregnated; it was then left to dry. As a consequence of rainfall within several days of installation, the canvas was re-sprayed with cypermethrin solution (0.5 g/L) 4 days after its initial installation, using a compressed-air hand-pump sprayer to completely impregnate the outer surface of the canvas as before.

Culicoides sampling was carried out using 4-W ultraviolet light traps fitted with a suction fan and a collecting vessel containing ethanol and ethylene glycol in water to preserve the samples (Miniature Blacklight model 1212; John Hock Co., Gainesville, FL, U.S.A.). Two traps were placed inside each pen at a height of 1.5 m and outwith yearling ewe access. Additionally, four traps were placed outside the pens to estimate the Culicoides population in the area. One trap was located 50 m from the experimental pens and the other three were placed within 30 m of farm buildings or enclosures housing livestock (pigs, goats and horses, respectively). These buildings were 500–900 m from the experimental pens (Fig. 1).

Figure 1.

Map of the sampling area showing the typical wind direction. Grey squares represent the test pens: 1a, no-barrier pen; 1b, untreated barrier pen; 1c, insecticide-treated barrier pen. (2) Enclosures in which yearling ewes grazed during the day. (3) Sheep stables. (4) Pig-fattening unit. (5) Goat enclosures. (6) Horse enclosures. inline image, traps located outside pens.

The traps were operated from 1 h before dusk until 1 h after dawn for 10 nights during the first 15 days after pen construction (pre-treatment period), and for 8 nights during the first 10 days after barrier installation (post-treatment period). The trapped Culicoides were collected and preserved in ethanol until they were processed in the laboratory, where specimens of Culicoides were identified on the basis of their wing pattern, as described by Rawlings (1996). In addition, all C. imicola specimens captured were sexed and females were age-graded as nulliparous, parous, gravid or freshly engorged (Dyce, 1969).

Statistical analysis was carried out by fitting generalized linear models (GLMs) with a Poisson distribution error and log-link function (McCullagh & Nelder, 1997) to capture variation in the six Culicoides groups analysed, which included: Culicoides other than C. imicola (non-imicola Culicoides); C. imicola, and the four age-graded classes of C. imicola females (nulliparous, parous, gravid and engorged). The location of traps (no-barrier pen, untreated barrier pen, treated barrier pen, and outside), period (pre- and post-treatment periods), and their interactions were used as predictor variables. The interaction term was introduced to model the slope of variation in Culicoides catch across period as a function of trap location (i.e. to model the effect of barriers on the entry of Culicoides into pens). Because the attraction of Culicoides might differ between traps outside and inside pens as a result of the proximity of the yearling ewes to the latter (Mands et al., 2004), two different models were fitted to each dependent variable. In the first model, data from outside pens were used as a reference (control level) for the location variable; in the second model, only data from traps inside pens were considered and data from the no-barrier pen were used as the reference. For all models, pre-treatment data were used as the reference level for the period variable. Wald tests were used to test the significance of predictor variables. Model estimates corrected for over-dispersion were generated by estimating the dispersion parameter using Pearson's chi-squared test (McCullagh & Nelder, 1997). Statistical differences were considered significant at P < 0.05.


A total of 11 990 Culicoides midges comprising nine Culicoides species were captured during the study and 4940 (41.2%) of these were identified as C. imicola. Of these, 565 were male and 4375 were female. Females included 885 nulliparous, 2685 parous, 650 gravid and 155 engorged specimens. The pattern of capture was characterized by a general decrease in mean trap catch from the pre- to post-treatment period for the six groups of Culicoides analysed, with the exception of the traps located inside the no-barrier and untreated barrier pens, where the mean catch of total C. imicola and nulliparous, gravid and engorged females increased (Table 1).

Table 1.  Mean number (standard deviation) and range of trap catches as a function of trap location and sampling period (pre- and post-treatment) for non-imicola Culicoides, total C. imicola, and nulliparous, parous, gravid and engorged C. imicola females.
 Non-imicola CulicoidesTotal C. imicolaNulliparous femalesParous femalesGravid femalesEngorged females
Outside pen62.616.041.715.36.33.724.
 0–3460–2000–3820–1140–490–21 0–2420–290–570–560–230–3
No-barrier pen50.636.924.141.02.810.417.317.
Untreated barrier pen45.23.118.324.02.37.711.
Treated barrier pen76.46.342.

The GLMs fitted to the data revealed no significant difference in mean trap catches of non-imicola Culicoides, total C. imicola or any C. imicola age-graded female class as a function of trap location in either the first model type (using outside traps as the reference level) (Table 2) or the second model type (using the no-barrier pen as the reference) (Table 3). A significant decrease was found from the pre- to post-treatment period for n in both model types, but this effect of period was not detected for C. imicola or the age-graded female classes, except for a reduction in the catch of parous females in the model fitted with outside traps as the reference level (Table 2).

Table 2.  Parameters ± standard error (P-values) for generalized linear models fitted using outside trap catches as the reference level for the trap location variable.
 Non-imicola CulicoidesTotal C. imicolaNulliparous femalesParous femalesGravid femalesEngorged females
  1. Pre-treatment data were used as the reference level for the period variable. P < 0.05 values are in bold.

Post-treatment period−0.80 ± 0.21−0.21 ± 0.150.004 ± 0.18−0.31 ± 0.14−0.02 ± 0.19−0.03 ± 0.11
No-barrier pen0.52 ± 0.280.26 ± 0.240.25 ± 0.280.26 ± 0.220.05 ± 0.330.29 ± 0.18
Untreated barrier pen−0.64 ± 0.49−0.13 ± 0.280.04 ± 0.31−0.20 ± 0.26−0.22 ± 0.36−0.12 ± 0.21
Treated barrier pen−0.10 ± 0.40−0.19 ± 0.32−0.42 ± 0.40−0.08 ± 0.260.03 ± 0.34−0.14 ± 0.21
Post- × no-barrier pen0.65 ± 0.280.46 ± 0.240.54 ± 0.280.31 ± 0.220.49 ± 0.330.41 ± 0.18
Post- × untreated barrier pen−0.41 ± 0.490.34 ± 0.280.47 ± 0.310.25 ± 0.260.23 ± 0.360.30 ± 0.21
Post- × treated barrier pen−0.38 ± 0.40−0.54 ± 0.32−0.79 ± 0.40−0.27 ± 0.26−0.44 ± 0.34−0.37 ± 0.21
Table 3.  Parameters ± standard error (P-values) for generalized linear models fitted using inside no-barrier pen trap catches as the reference level for the trap location variable.
 Non-imicola CulicoidesTotal C. imicolaNulliparous femalesParous femalesGravid femalesEngorged females
  1. Pre-treatment data were used as the reference level for the period variable. P < 0.05 values are in bold.

Post-treatment period−0.85 ± 0.28−0.12 ± 0.190.08 ± 0.25−0.21 ± 0.160.07 ± 0.220.08 ± 0.12
Untreated barrier pen−0.57 ± 0.46−0.11 ± 0.270.09 ± 0.34−0.20 ± 0.23−0.17 ± 0.33−0.13 ± 0.18
Treated barrier pen−0.03 ± 0.39−0.17 ± 0.30−0.38 ± 0.41−0.07 ± 0.230.08 ± 0.32−0.15 ± 0.18
Post- × untreated barrier pen0.36 ± 0.460.25 ± 0.270.40 ± 0.340.16 ± 0.230.14 ± 0.330.19 ± 0.18
Post- × treated barrier pen−0.33 ± 0.39−0.63 ± 0.30−0.87 ± 0.41−0.37 ± 0.23−0.53 ± 0.32−0.49 ± 0.18

Parameters estimated for interaction terms in the first model type showed that the decrease in trap catches after treatment were more marked for outside traps than for the no-barrier pen traps for the six Culicoides groups analysed. This relationship was statistically significant for catches of non-imicola Culicoides and engorged C. imicola females, and approached significance for catches of total C. imicola and nulliparous C. imicola females (Table 2). Conversely, the decrease in trap catches was more marked for traps in the treated barrier pen than outside traps for the six Culicoides groups. In this case, the statistical relationship was significant only for nulliparous C. imicola females, but approached significance for total C. imicola and engorged females. For the interaction between period and the untreated barrier pen, estimated parameters showed that the decrease in catch of non-imicola Culicoides was more marked in traps inside the untreated barrier pen than in traps outside the pen; by contrast, the decrease was less marked for total C. imicola and the four age-graded female classes. Nevertheless, no significant or near-significant relationship was detected.

When no-barrier pen traps were used as the reference level (Table 3), the fitted models did not show any significant or near-significant difference for the interaction term between period and the untreated barrier pen traps for any of the six Culicoides groups. However, the model revealed a significantly greater decrease from the pre- to post-treated period in treated barrier pen traps compared with no-barrier pen traps for total C. imicola, and nulliparous and engorged female catches; for gravid females, the decrease approached significance.


Only partial or no protection of livestock and humans from Culicoides have been achieved by physical barriers alone, or by combined physical and chemical barriers. The small size of biting midges allows them to enter through conventional nets used against other biting insects, such as mosquitoes. Moreover, the use of fine nets does not prevent entry, restricts air flow (Porter, 1959) and seems to increase the Culicoides population inside the net (Calvete et al., 2007); these nets are no longer considered suitable for use on buildings. Only by increasing shelter in open areas using dense and impenetrable physical barriers such as tarpaulins (Doherty et al., 2004) or by using repellent-treated netting (Zaugg, 1978; Schreck & Kline, 1983) has a degree of protection against Culicoides been achieved. In partial agreement with previous studies, the current results show that the use of an impenetrable physical barrier alone did not completely prevent the entry of Culicoides into pens, but treatment of the barrier with an insecticide reduced the exposure of livestock to C. imicola.

The absence of a clear effect of the treated barrier was not surprising because Culicoides species differ in their behaviour and host/shelter preferences (Barnard, 1997; Meiswinkel et al., 2000; Mands et al., 2004). Results for non-imicola Culicoides probably reflect the diverse behaviours of the species included in this group (Doherty et al., 2004). Little is known about the altitude at which C. imicola usually flies when seeking hosts and how it overcomes any barriers (e.g. by flying above or by flying up and over barriers). The marked reduction in C. imicola catch in the treated barrier pens indicates that C. imicola individuals may have flown up the outer face of the barrier, contacted the insecticide and been killed before clearing the barrier. In addition, although it might be assumed that the Culicoides specimens caught inside the barrier pens had flown over the barrier, it is possible that some or all entered when the yearling ewes were let out from or returned to the pens. In this case, the treated barrier would not have reduced the entry of Culicoides, but would have killed those that had already entered the pen. These possible mechanisms have different implications for disease prevention.

The statistical analysis of the degree to which the C. imicola catch was reduced by the insecticide barrier was highly influenced by which trap location variable was selected as the reference. The installation of the canvas barriers around the pens coincided with adverse climatic conditions, including a decrease in temperature and the occurrence of rain. These conditions probably reduced the activity of Culicoides during the post-treatment period, causing a general decrease in specimen catch in traps (Mellor et al., 2000), including in those traps located outside. However, this marked decrease was not observed in catches of all Culicoides groups in the no-barrier pen, nor in catches of total C. imicola and most of the C. imicola age-graded female classes in the untreated barrier pen. Several studies have demonstrated that the trap response of Culicoides is host odour-dependent and that C. imicola is attracted into stables by the odour of manure (Barnard, 1997; Mands et al., 2004). Hence, it may be that climatic conditions during the post-treatment period reduced Culicoides activity around outside traps without an animal bait to a higher degree than around traps inside pens with an animal bait present. In this case, the model fitted with the no-barrier pen traps as the reference should provide a more reliable estimate of the reduction in catch (and therefore of the degree of protection) attributable to barriers because the baiting conditions among pens were equivalent.

The results of this study suggest that the use of treated barriers should be further explored as an inexpensive option to protect livestock from C. imicola attack in open yards, to increase the shelter provided by sheds, or to increase the level of protection in buildings. However, although the insecticide-treated barrier seemed to reduce the abundance of C. imicola in pens, the effect was not complete because some specimens were caught inside. This suggests that combining the use of treated barriers with other protection methods, such as livestock treatment, will be necessary to ensure protection against this species (Mullens et al., 2000, 2001; Doherty et al., 2001; Mehlhorn et al., 2008).

The extent to which the degree of reduction in C. imicola abundance observed in this survey might be affected by the height of the barrier, the abundance of C. imicola and the size of the protected area remains unclear. The height of the barrier may have a significant influence on the passage of C. imicola above it. Similarly, greater abundance of C. imicola outside than inside the study area may result in lower protection because the pressure on biting midges to enter the protected area will be greater. Furthermore, an increase in the size of the area to be protected may decrease the degree of protection as a result of the increase in the ratio between the protected surface area and the barrier length. In addition, an increase in the size of the protected area may result in this species breeding inside the barrier if the conditions are appropriate. Therefore, further studies will be necessary to determine the efficacy of these types of barrier in protecting livestock against the transmission of BTV and other arboviruses by C. imicola. Given that it is unclear how the numbers, species composition and physiological status of catches in a light trap relate to those feeding on a natural host, it is desirable that further studies using actual sheep as bait animals are performed (Carpenter et al., 2008).


Funding for this study was provided by TRAGSEGA under the project: ‘Assessment of the efficacy of different methods of preventing bluetongue and controlling its vectors.’ We thank A. Boluda, M. Díaz-Molina and C. Díez-De la Varga for their help and assistance in fieldwork, and convey special thanks to R. Calero for granting permission to work on the Selection and Breeding Animal Centre (CENSYRA) property in Badajoz and for assistance.