**2. The vector of Schmallenberg virus in Portugal: lessons learned from the past and applications to the future**

Blue tongue virus and African horse sickness virus (AHSV) are arboviruses that have circulated in Portugal in the past [11]. Both BTV and AHSV are double-stranded RNA viruses from the family Reoviridae that cause infectious, non-contagious illness, included in List A diseases by the Office International des Epizooties. BTV affects all species of ruminants [12], whereas

AHSV affects equines and occasionally dogs [10]. These arboviruses are transmitted by the bites of vector species of *Culicoides* [13]. Thus, so as with Schmallenberg, the incidence and geographical distribution of BTV and AHSV are associated to the distribution and abundance of *Culicoides. Culicoides imicola* constitutes the only field vector of AHSV while being the main vector of BTV in Europe and Africa [11].

Both BTV and AHSV have sporadically emerged into the southern European countries of the Mediterranean basin with the largest epidemic of BTV occurring between 1998 and 2002, and affecting Bulgaria, Greece, Turkey, Italy, Macedonia, Yugoslavia, Spain, France, Montenegro, Serbia and Bosnia and Herzegovina [14, 15]. Interestingly, BTV northern spread has shown novel territories of *C. imicola* expansion, which is believed to have been influenced by global warming [16].

Given the epizootic features of BTV, there was the constant need to record the distribution of *C. imicola* and to identify whether other potential vector species of the *Culicoides* genus are sufficiently distributed and in sufficient numbers to act as vectors for sustained arbovirus transmission, and also to map areas of higher risk for endemicity of BTV and ASHV due to the constant presence of adult *Culicoides* vectors [11]. These goals have for the past years been achieved by vector surveillance systems across Mediterranean Europe and bordering countries, thus producing detailed predictive risk maps of *Culicoides*-borne disease.

Discussing this topic shows the need to highlight two reports of *Culicoides* vector surveillance carried out in Portugal [11, 17], which mostly cover all territories and provide detailed information on the temporal distribution of *Culicoides* species in Portugal from 2000 to 2010, as independent entomological surveys.

Both studies have assumed similar sampling schemes, and divided mainland Portugal into 45 quadrats (or geographical units) each 50 km × 50 km so as to cover all territories in detail (**Figure 10**) and performed similar trapping strategies.

Capela et al. [11] sampled a total of 87 sites (including at least two livestock holdings or farms) within almost all geographical units. The authors took into account for variations in environmental conditions between sampling years, randomly dividing into two equal groups and sampling the first in 2000 and the second in 2001. Farms were included in the study if fulfilling the following criteria: located at least 10 km apart and 2.5 km from the coast, contained a minimum of five large livestock animals and did not use insecticides. Trapping was performed using Onderstepoort-type black light traps [18, 19] with 8-W UV light bulbs and downdraught suction, set on each night between 1 h before sunset to approximately 08.00 h the following morning. Traps were set outside but within 25 m of livestock. Ribeiro et al. [17] sampled a total of 212 sites within all geographical units between 2005 and 2010. Farms were included in the study if fulfilling the following criteria: located at a minimum distance of 10 km from other sampled holdings and at least 2.5 km from the coast, and contain a minimum of five horses or ruminants (preferably cattle). Recruited farms were also not permitted to enforce insecticide application for the duration of the survey. Trapping was performed using Center for Disease Control (CDC) miniature light traps (model 1212; John W. Hock, Gainesville, FL, USA) with a 4-W UV light and a suction fan, set from dusk to dawn. Traps were set outside but within 30 m from livestock.

**Figure 10.** Sampling frames for *Culicoides* entomological surveys in Portugal since 2000 [11, 17].

#### **2.1. Step one 2000–2001**

AHSV affects equines and occasionally dogs [10]. These arboviruses are transmitted by the bites of vector species of *Culicoides* [13]. Thus, so as with Schmallenberg, the incidence and geographical distribution of BTV and AHSV are associated to the distribution and abundance of *Culicoides. Culicoides imicola* constitutes the only field vector of AHSV while being the

66 Epidemiology of Communicable and Non-Communicable Diseases - Attributes of Lifestyle and Nature on Humankind

Both BTV and AHSV have sporadically emerged into the southern European countries of the Mediterranean basin with the largest epidemic of BTV occurring between 1998 and 2002, and affecting Bulgaria, Greece, Turkey, Italy, Macedonia, Yugoslavia, Spain, France, Montenegro, Serbia and Bosnia and Herzegovina [14, 15]. Interestingly, BTV northern spread has shown novel territories of *C. imicola* expansion, which is believed to have been influenced by global

Given the epizootic features of BTV, there was the constant need to record the distribution of *C. imicola* and to identify whether other potential vector species of the *Culicoides* genus are sufficiently distributed and in sufficient numbers to act as vectors for sustained arbovirus transmission, and also to map areas of higher risk for endemicity of BTV and ASHV due to the constant presence of adult *Culicoides* vectors [11]. These goals have for the past years been achieved by vector surveillance systems across Mediterranean Europe and bordering coun-

Discussing this topic shows the need to highlight two reports of *Culicoides* vector surveillance carried out in Portugal [11, 17], which mostly cover all territories and provide detailed information on the temporal distribution of *Culicoides* species in Portugal from 2000 to 2010,

Both studies have assumed similar sampling schemes, and divided mainland Portugal into 45 quadrats (or geographical units) each 50 km × 50 km so as to cover all territories in detail

Capela et al. [11] sampled a total of 87 sites (including at least two livestock holdings or farms) within almost all geographical units. The authors took into account for variations in environmental conditions between sampling years, randomly dividing into two equal groups and sampling the first in 2000 and the second in 2001. Farms were included in the study if fulfilling the following criteria: located at least 10 km apart and 2.5 km from the coast, contained a minimum of five large livestock animals and did not use insecticides. Trapping was performed using Onderstepoort-type black light traps [18, 19] with 8-W UV light bulbs and downdraught suction, set on each night between 1 h before sunset to approximately 08.00 h the following morning. Traps were set outside but within 25 m of livestock. Ribeiro et al. [17] sampled a total of 212 sites within all geographical units between 2005 and 2010. Farms were included in the study if fulfilling the following criteria: located at a minimum distance of 10 km from other sampled holdings and at least 2.5 km from the coast, and contain a minimum of five horses or ruminants (preferably cattle). Recruited farms were also not permitted to enforce insecticide application for the duration of the survey. Trapping was performed using Center for Disease Control (CDC) miniature light traps (model 1212; John W. Hock, Gainesville, FL, USA) with a

tries, thus producing detailed predictive risk maps of *Culicoides*-borne disease.

main vector of BTV in Europe and Africa [11].

as independent entomological surveys.

(**Figure 10**) and performed similar trapping strategies.

warming [16].

During summer, 166 samples were collected containing 55,937 *Culicoides* spp. Individuals [11]. *Culicoides imicola* was the most frequently observed species, accounting for 66% of all individuals, followed by *C. obsoletus* (17.3%) and *C. pulicaris* (10.7%), and with *C. puncticollis* and other *Culicoides* complexes accounting for a very low proportion among all individuals. Despite being found at higher numbers, *C. imicola* was less prevalent across geographical units (found in 64%) than either *C. obsoletus* (found in 82%) or *C. pulicaris* (found in 93%). *Culicoides imicola* was significantly more prevalent in south Portugal (91% of southern geographical units) than north Portugal (42% of northern geographical units). The most northern site positive for *C. imicola* in this study was at 41°38.4′ N. *Culicoides imicola* was collected until the maximum altitude of 850 m above sea level. *Culicoides imicola* appeared to be absent from the north-west corner of Portugal and along the north-west coast. On the contrary, *C. imicola* was found to be highly abundant in the central eastern Portugal. During winter, 22,883 individuals of *Culicoides* spp. were collected with *C. pulicaris* accounting for 47% of the total *Culicoides* spp. catch, followed by *C. obsoletus* (6%) and *C. imicola* (1%).

#### **2.2. Step two 2005–2010**

Of the total 5800 catches, 3632 contained *Culicoides* species [17]. *Culicoides imicola* was the most frequently observed species, accounting for 74.8% of the individuals, followed by *C. obsoletus* (7.7%). The central region of the country accounted for the highest catches of *C. imicola. Culicoides imicola* was found to be less prevalent than *C. obsoletus* and comparing the distribution data with the one reported in 2000–2001 [20], *C. imicola* was found in five more geographical units, mainly in the northern regions. *Culicoides imicola* prevalence (per geographical unit) was higher in both central and southern regions when compared to the north. The most northern site positive for *C. imicola* in this study was at 41°92′ N. *Culicoides imicola* was collected until the maximum altitude of 1694 m above sea level. *Culicoides imicola* was found to be highly abundant in the central region of the country. The largest collections of *C. imicola* occurred during the summer months of July, August and September, and the lowest during the winter months of December, January and February.

#### **2.3. Overall analysis 2000–2010**

When combined, both studies provide robust evidence that *C. imicola* has been the most prevalent *Culicoides* species in Portugal for the decade between 2000 and 2010, followed by members of the *Obsoletus* group, clearly showing the sustained presence of Schmallenberg virus vectors across the territory [11, 17].

Both studies also provide strong support to the notion that *C. imicola* is more prevalent in the central and south of Portugal, while the *Obsoletus* group is more widespread throughout the territory. Preferences in vector distribution have been related to different climate and habitat particularities, which are markedly distinct between the north and the central/south regions in Portugal [21]. Mainland Portugal geography is clearly demarcated by both the Atlantic at the north and the Mediterranean at the south with a borderline set across the territory and defined by Tagus, dividing the north with its forests, valleys and mountains, and the south with its vast lowlands where typical Mediterranean vegetation grows [17]. Climate is also diverse with maritime features and sharp differences between seasons in the north, and dry hot climate in the south. Higher prevalence of *C. imicola* in the south has been associated to the vector preference for breeding in moist nutrient-rich soils with high exposure to sun, typical features of southern regions of Portugal [17], and a preference not observed for *Obsoletus* group [20].

Both reports also show that over the 10-year time frame of 2000–2010, *C. imicola* has been detected more to the north but also at higher altitudes (850 m vs 1645 m) supporting that the vector is adapting and spreading to newer territories. Nonetheless, the authors report that the low numbers found suggest that these locations may be of borderline suitability, and the specimens caught could potentially have been wind-borne from more suitable regions [17, 22]. In conclusion, combined entomological data from both Capela et al. [11] and Ribeiro et al. [17] increase the understanding of the ecology of *Culicoides* vectors and *Culicoides* activity in Portugal. They provide important data on vectors that are known to have a significant impact on ruminants, in particular and within the scope of this review, of interest to Schmallenberg virus epidemiology in Portugal and in the support to design strategies to prevent disease spread in Portugal.
