Arbovirus vectors in municipalities with a high risk of dengue in Cauca, Southwestern Colombia

The Culicidae family has two of the most important disease vector genus: Aedes spp. and Culex spp. Both of these are involved in the transmission of arboviruses. Here, we provide novel data for the geographical distribution of 2,383 specimens in the Culicidae family. We also report the percentage of houses infested with these vectors, and Shannon and Simpson diversity indices in three municipalities located in Cauca, Colombia. This dataset is relevant for research on vector-borne diseases because Aedes spp. can transmit arboviruses such as dengue, Zika and chikungunya, and Culex spp. is a well-known vector of West Nile virus and Venezuelan equine encephalitis.


DATA DESCRIPTION Introduction
Arboviruses like dengue, chikungunya, yellow fever and -more recently -Zika, are expanding, with more cases and fatalities, making them a concern for public health at the international level [1,2]. In 2019, countries in the Americas recorded more than 2.7 million cases of dengue -the highest in history -including 22,127 severe cases and 1,206 deaths reported in October 2019 [3].
The main mosquitoes involved in the transmission of vector-borne disease are Aedes aegypti (Linnaeus), Aedes albopictus (Skuse) and Culex quinquefasciatus (Say). These belong to the Culicidae family (Diptera: Nematocera), which comprises about 3200 recognized species. Tropical rainforests, where fauna is more diverse but less well surveyed than temperate regions, probably house many more as-yet undiscovered species [4]. The Ae. aegypti mosquito is a predominantly domestic species, usually infesting natural or artificial containers in or around dwellings. The female feeds on human or domestic animal blood [5]. Owing to its close relationship with humans, it is essentially an urban mosquito. In Colombia, it has been recorded at altitudes ranging from 0-2302 m above sea level [6] and has also been reported in rural areas [7].
Ae. aegypti females restrict their hematophagous habits to daylight hours, with the peak of biting activity at dawn and shortly before dusk. Diurnal activity has also been recorded in the male, where they copulate and feed on sugary substances. [8,9]. Under optimal conditions of food availability and adequate oviposition sites, the average dispersion of a female Aedes spp. mosquito is estimated to be between 50 and 100 m, which limits its visits to two or three dwellings during its adult life [7,10]. However, fed females have been recorded to disperse as far as 800 m in 6 days [11].
The mosquito Ae. albopictus (Skuse) is more commonly found in forested regions and in open spaces with abundant vegetation, typical of suburban or rural areas. They can be found in environments with extensive vegetation, such as bamboo stumps, tree cavities, plant axils (bromeliads) and water reservoirs in rock crevices as natural breeding sites [12].
When cohabiting with Ae. aegypti in urban areas, it can thrive in artificial reservoirs like flower vases, tires and cans [13]. However, although Ae. albopictus is anthropophilic, the hematophagous behavior of mosquito populations may depend on the availability of alternative food sources (domestic and wild animals) [14].
Finally, Culex quinquefasciatus (Say) is considered a highly anthropophilic species, which is associated with both urban and rural human habitats in Colombia [15]. Its biting activity occurs mostly nocturnally, but it remains indoors during the day [8,16]. It is distributed across Colombia at elevations ranging from 0-3000 m above sea level (masl) [15]. Together with Ae. aegypti, the immature stages of Cx. quinquefasciatus also develop in structures serving as drains and catch-basins for stormwater located in streets and avenues [17].
Here we describe and quantify data associated with Culicidae adult mosquitoes (Diptera: Nematocera) collected with Prokopack aspirators in urban households in three municipalities with a high incidence of dengue in the department of Cauca (southwestern Colombia).

Context
Data from the 2383 specimens (529 occurrences) reported here from the Culicidae family are novel unpublished data from the Cauca department, located in the Pacific region of Colombia. The data were collected in 2021 by a multidisciplinary team made up of environmental health technicians, geographic and environmental engineers, and professionals with extensive experience in medical entomology.
The mosquitoes were collected in three municipalities in the department of Cauca, an area of study chosen for its endemic-epidemic behavior for dengue disease. The event is characterized by focal endemics, variable transmission scenarios and temporal, seasonal and cyclical patterns in at-risk populations.
These municipalities are also part of an ongoing research project, which aims to determine the relationship between environmental, biological, and sociodemographic factors influencing the increase in the burden of dengue and its spatial variation, using geographic information systems in the construction of risk maps.
These data are valuable to the scientific community because they show the spatial location of specimen collection, allowing authorities to take action in those neighborhoods where mosquitoes have a higher percentage of house infestation. In addition, by being identified and geo-referenced, they allow various analyses concerning other entomological and environmental variables.

METHODS Sampling
The collection of species belonging to the Culicidae family was done in the municipalities of Piamonte, Patía, and Miranda. These localities were chosen because they present a high risk of dengue according to incidence rate global cluster analysis (10,000 inhabitants), performed using the High-Low Clustering (Getis-Ord General) analysis in the ArcGIS ® 10.8 software (SCR_011081). To detect hot spots, the incidence rates of dengue per 10,000 inhabitants from 2014-2018 were calculated from cases reported to the Sistema Nacional de Vigilancia en Salud Pública (SIVIGILA; National System of Public Health Surveillance), and the population statistics of the projection system reported by the National Administrative Department of Statistics in Colombia.
Afterwards, sampling was delimited to a neighborhood scale by Kernel density analysis (hot spot analysis) of dengue cases reported between 2015 and 2019 in the urban area of each municipality. The sample size was calculated through the estimated dengue prevalence (10.5%) in the municipality.
In total, 935 houses and 17 neighborhoods were visited during 2021. In Piamonte and Patía, three dengue clusters were identified, therefore three neighborhoods were sampled (n = 180 houses) for the first case, and seven neighborhoods (n = 335 houses) for the second. Finally, in Miranda, six neighborhoods (n = 420 houses) from four clusters with a higher-than-expected prevalence of the disease were sampled.

Species collection
The entomological inspection was performed between 8:00 and 17:00 hours, in an average time of 10 minutes per house. In each house, a Prokopack aspirator (see Figure 1) was used to catch the adult mosquitoes present in the living room, dining room, bathrooms, kitchens, laundry yard and others, searching especially in shaded areas or near water containers. After the inspection was complete, collection cups were inspected to verify whether they had caught adult mosquitos. If so, then the collection cup was placed in a plastic bag along with a piece of chloroform-soaked cotton to immobilize the specimens, facilitating posterior differentiation and storage.

Species classification and spatial characterization
After collecting the adult mosquitos, taxonomic identification and classification of the species Ae. aegypti, Ae. albopictus and Cx. quinquefasciatus was carried out by expert entomologists. Specialized taxonomic keys, such as those developed by Forattini [8] and Rueda [18] (Table 1) were used. Males were also distinguished from females.
Following the protocol for the fieldwork development [19], once females of the species Ae. aegypti were identified, they were stored in 0.2-ml vials for posterior processing using molecular biology techniques ( Figure 2). Other species were stored dry for their subsequent entry into the collection by the Entomology group at Instituto Nacional de Salud (Bogota, Colombia).
Each vial was assigned a code corresponding to the house in which its specimens were collected. This code was associated with a sociodemographic survey applied to the same house using the application ArcGIS ® Survey 123, which established the geographical position of the specimen collection. As well as metadata related to location, altitude values were obtained from the National Aeronautics and Space Administration (NASA) platform

DATA VALIDATION AND QUALITY CONTROL
A total of 2383 individuals corresponding to three species of the Culicidae family, Ae. aegypti (n = 572), Ae. albopictus (n = 36) and Cx. quinquefasciatus (n = 1775) were collected in 529 of 935 houses located in Cauca, Colombia. Table 2 presents house infestation percentage per municipality, house index and additional descriptive measures of the sampled mosquitos at each location. The number of adult Ae. aegypti and Cx. quinquefasciatus per house is shown as the percentage of total screened houses in Figure 3. It is worth highlighting that in three houses in Patía, more than 15 Ae. aegypti individuals were found, and the highest number of adults collected in a single house was 60. In Miranda, more than 10 individuals were found in four houses, and the highest number of adults in a single house was 20. Lastly, Piamonte presented the lowest number of adults per house, and the highest number in a single house was seven.
As a measure of community heterogeneity, Shannon and Simpson diversity indices were used [21]. The average diversity index for the three municipalities ranged from 0.27 to 0.66 for the Shannon index, and from 0.12 to 0.41 for the Simpson index. Both indices were highest in Patía and lowest in Piamonte. Graphically, location diversity can be better visualized using the Margalef index. For Miranda, the sample is very diverse, while for the other groups, a sample of diverse characteristics is observed. In general, these values are not high, being in the range of 0.28 to 0.35. (Figure 4).
The highest diversity index was recorded for Ae. aegypti (0.1488), followed by Cx. quinquefasciatus (0.0953), and least values were recorded in Ae. albopictus (0.0275). The dominance index of all mosquitoes sampled at three different municipalities was estimated (Table 3); the highest values were recorded in Cx. quinquefasciatus (0.5548) followed by Ae. aegypti (0.0576) and least in Ae. albopictus (0.0002). These results are similar to those of other previous studies [8].  Table 2. Descriptive entomological measures by sampling locality. The total of positive screened houses for each mosquito species is shown as a total and as a percentage for each municipality. Further information regarding sex-specific relative density and ratios females: males are also included.     geographical coordinates, sex, stage of development, and taxonomy. The minimum information required to include a specimen in the collection is related to the standard Darwin Core and is the same as the minimum information required for publication in the Global Biodiversity Information Facility (GBIF).

RE-USE POTENTIAL
The database and vector distribution map provide a novel resource for understanding the abundance and behavior of mosquito-borne diseases for the entomology and vector-borne diseases community. To improve the accessibility and usability of these data, they have been included in the GBIF. These data will be useful for museums setting up similar displays, and the data may be used for other research purposes such as predictive models and species geographic distributions. The Ae. aegypti and Ae. albopictus database may also be helpful for similar projects elsewhere. We suggest to others to make their data similarly available.

DATA AVAILABILITY
The datasets supporting this article are available in the GBIF repository [22].

EDITOR'S NOTE
This paper is part of a series of Data Release articles working with GBIF and supported by TDR, the Special Programme for Research and Training in Tropical Diseases, hosted at the World Health Organization [23].

AUTHORS' CONTRIBUTIONS
CM and ES conceptualised the study, with CAM made the project administration. CM, ES, MCL, and SM drafted the manuscript. CAM, PF, ES and MCL generated the field data. PF, HC, CAM and AAD reviewed and edited the draft. All authors made comments on the manuscript.