Review: Artificial container-breeding mosquitoes and cemeteries: a perfect match

Authors


Corresponding Author Dario Vezzani, Unidad de Ecología de Reservorios y Vectores de Parásitos, Departamento de Ecología, Genética y Evolución, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad Universitaria, Pabellón II, 4 ° piso, C1428EHA Buenos Aires, Argentina. Tel./Fax: +54 11 4576 3354; E-mail: vezzani@ege.fcen.uba.ar, vzztato@fibertel.com.ar

Summary

Artificial container-breeding mosquitoes, such as Aedes aegypti, Ae. albopictus, and Culex pipiens, are well-recognized vectors of diseases throughout the world. Cemeteries are considered major sources of mosquitoes and the results of more than 30 studies concerning mosquitoes in cemeteries have been published over the last decade. The characteristics of these environments in regard to the availability of resources for mosquito development were discussed. Also, studies about early detection of Aedes vectors, ecological issues, and mosquito control performed in cemeteries were reviewed. Among 31 mosquito species found breeding in cemeteries from 16 countries, the invasive Ae. aegypti and Ae. albopictus were the most frequent ones. Species of the genus Ochlerotatus, Culex, Toxorhynchites, Culiseta, Armigeres, Lutzia, Uranotaenia, and Tripteroides were also reported. Overall, cemeteries are highly suitable habitats for artificial container-breeding mosquitoes due to the great availability of the different resources that they need (i.e. sugar substances, blood, shelter and water-filled containers). In addition, these places are mostly ideal settings to perform studies in urbanized areas because of high mosquito abundance, heterogeneity of macro- and microhabitats, and an easier access in comparison with private premises. However, the feasibility of a cemetery as a study area must be evaluated in each case considering the objectives of the study and cemetery characteristics.

Abstract

Les moustiques qui se multiplient dans des enceintes artificiels, tels que Aedes aegypti, Ae. albopictus et Culex pipiens sont des vecteurs bien identifiés de maladies dans le monde. Les cimetières sont considérés comme des sources principales des moustiques et plus de 30 études sur des moustiques dans les cimetières ont été publiées au cours de la dernière décennie. Cet article discute les caractéristiques des cimetières en tant que sources de multiplication de moustiques et analyse les études sur le dépistage précoce des vecteurs d’Aedes, les questions écologiques et le contrôle des moustiques. Sur 31 espèces de moustiques qui se multiplient dans les cimetières dans 16 pays, les espèces invasives Ae. aegypti et Ae. Al-bopictusétaient les plus courantes. Les espèces du genre Ochlerotatus, Culex, Toxorhynchites, Culiseta, Armigeres, Lutzia, Uranotaenia et Tripteroides ont également été rapportées. Les cimetières sont des habitats très convenables pour les moustiques qui se multiplient dans des enceintes artificielles à cause de la disponibilité de ressources nécessaires aux moustiques telles que des sources de sucre, de sang, d'abri, et des enceintes remplies d'eau. Les cimetières sont également des sites idéaux pour effectuer des études dans des secteurs urbains parce les moustiques y sont abondants, les macro et micro habitats hétérogènes sont d'accès plus faciles que les sites privés. Cependant, la praticabilité d'un cimetière comme site d’étude doit être évaluée au cas par cas en tenant compte des objectifs de l’étude et des caractéristiques du cimetière.

Abstract

Los mosquitos que utilizan contenedores artificiales para reproducirse, tales como Aedes aegypti, Ae. albopictus, y Culex pipiens, son vectores de enfermedades reconocidos en todo el mundo. Los cementerios son considerados como una de las mayores fuentes de mosquitos, habiéndose publicado más de 30 estudios sobre el tema durante la última década. En este artículo se discuten las características de los cementerios en términos de recursos para el desarrollo de mosquitos y se hace un repaso a los estudios de detección temprana de vectores Aedes, de cuestiones ecológicas y de control de mosquitos. Entre las 31 especies de mosquito que se han encontrado reproduciéndose en los cementerios de 16 países, las invasivas Ae. aegypti y Ae. Albopictus son las más comunes. Especies del género Ochlerotatus, Culex, Toxorhynchites, Culiseta, Armigeres, Lutzia, Uranotaenia, y Tripteroides también han sido reportados. Los cementerios son hábitats muy adecuados para aquellos mosquitos que se reproducen en contenedores artificiales, debido a la disponibilidad de recursos que estos insectos requieren, tales como sustancias azucaradas, sangre, lugares de refugio y contenedores llenos de agua. Los cementerios son también lugares ideales para realizar estudios en áreas urbanas, puesto que en ellos se encuentran mosquitos en abundancia, macro y microhábitats heterogéneos, y son más asequibles que las propiedades privadas. Sin embargo, la viabilidad de los cementerios como áreas de estudio debe ser evaluada en cada caso, considerando tanto los objetivos del estudio como las características del cementerio.

Introduction

Mosquitoes are bothersome, but more importantly, they are no doubt the most medically important insect vectors of diseases. Within this insect suborder of about 3500 species (Service 1995), mosquitoes that breed in containers are a well-defined group characterized by the properties of its habitats. Although the mosquito aquatic habitats encompass a broad and complex spectrum, the containers comprise a distinct group with unique ecological properties (Washburn 1995). Compared with pools, ponds, swamps, and rice fields, containers could be described as follows: (i) they are significantly smaller, and therefore will support fewer species with smaller population sizes and will exhibit higher rates of extinction, (ii) they are habitats with almost no internal productivity, based mainly on the decomposition of leaf litter and other detritus, and (iii) their larval mosquito populations are regulated by food limitation or competitive interactions rather than predation (Washburn 1995; Sunahara et al. 2002; Arrivillaga & Barrera 2004). Container-type habitats are usually divided into natural (tree-holes, bamboo, leaf axils, rock-pools, etc.) and man-made or artificial (water tanks, bottles, tires, flower vases, etc.) (Service 1995). Among mosquitoes that breed (exclusively or not) in artificial containers, three species are cosmopolitan and highly abundant, namely Aedes aegypti, Ae. albopictus, and Culex pipiens complex. These species are recognized as important vectors of diseases worldwide. For example, Ae. aegypti is the main vector of dengue and urban yellow fever in America, Ae. albopictus is vector of dengue in Asia, and members of the Cx. pipiens complex are responsible of the transmission of filarial worms and encephalitis viruses in different regions (see Service 1995; Forattini 2002 for a review).

Urbanization leads to an increase in the amount of artificial containers, including some useful stuff (e.g. water-storage tanks) and many that are urban trash (e.g. discharged tires and bottles). Despite their utility, artificial containers are common in urban settlements throughout the world because of human activities. Urban environments are heterogeneous mosaics of residential dwellings, commercial properties, parks, and other land-use types, providing an array of habitat types that can be used by arthropods (McIntyre 2000). In this fragmented landscape, parks, gardens, roadsides and cemeteries provide refuge to insects (Samways 1995), and each one of these ‘less urbanized’ areas could be a suitable habitat for different taxonomic groups.

Cemeteries are widely recognized as (i) areas of high mosquito productivity, (ii) priority sites for control of mosquito vectors, and (iii) strategic sites for monitoring the infestation or reinfestation by Aedes vectors (e.g. PAHO 1994; Natal et al. 1997; OPS 1997, 2000; WHO 1997, 1999; Anonymous 1998; Rodríguez Cruz 2002; Parks & Lloyd 2004). But, which characteristics do these anthropogenic environments meet to become highly suitable habitats for mosquito proliferation and in consequence to be used as study areas? The aim of this work was to answer these questions and synthesize the available information about mosquito studies performed in cemeteries.

Cemeteries as habitats: general concepts

Cemeteries are an obligatory component of the urban landscape in human settlements around the world, from megalopolises with millions of inhabitants to small villages with a few hundreds. They are specific urban habitats distinguished within the city boundaries. Usually, older cemeteries are immersed within densely populated neighborhoods as a consequence of an accelerated and unplanned urbanization, whereas newer ones are in the outskirts. As human population increases, so do the number and/or size of cemeteries; for example, there are 23 cemeteries in urban Hong Kong (Teather 1998), 52 in Middlesex County (New Jersey) (Schmidt 2003), and about 115 000 in the US (Stowe et al. 2001). Two types of administration could be distinguished, private and municipal. Private cemeteries are established as commercial business. Although some cemeteries are non-profit organizations, many modern cemeteries today are profit-generating corporations while municipal cemeteries are maintained and supported through taxes (Salisbury 2002). The type of religious affiliation (e.g. Christian, Jewish) may determine some customs of the visitors, e.g. leaving flowers is largely a Christian tradition (Schmidt 2003).

Cemeteries might be considered as environments to preserve due to their diversity of plants and animals (McPherson & Nilon 1987; Laske 1994; Prober 1996; Barrett & Barrett 2001). From the viewpoint of human recreative activities, Laske (1994) wrote: ‘cemeteries are not only burial grounds, but also, especially in the urban environment, they are places to relax and enjoy nature’, whereas Stowe et al. (2001) stated: ‘cemeteries are beautified with turf and invasive exotic species. The result is an artificial and toxic environment’. Perhaps, both extreme perspectives are right. Although a great number of plants are introduced to cemeteries as decorative vegetation (Sudnik-Wójcikowska & Galera 2005), a cemetery could act as a recreative site (with an appropriate behaviour) in big metropolis, if it is pleasant and well maintained. Many urban cemeteries were originally designed as a mixed-use open space in the city (Salisbury 2002), and joggers, bicyclists and walkers often use these sites as relax areas (McPherson & Nilon 1987). On the other hand, a polluted cemetery, full of trash or with vegetation out of control, is a harmful environment to visitors, where nuisance species must be controlled. Certainly, the soil, ground water and air in many cemeteries are contaminated due to the chemicals used in the embalming of the corpses, the materials of the coffins, and increasing concentrations of natural organic and inorganic substances (Üçisik & Rushbrook 1998; Stowe et al. 2001; Salisbury 2002; Creely 2004; Trick et al. 2005). Finally, some cultures as the Chinese regard cemeteries as landscapes to be feared because they link the material and non-material worlds of existence, thus resulting in a marked negative impact on nearby land and property prices (Teather 1998).

Cemeteries are usually similar to parks regarding vegetation cover, but different to them due to the presence of some edification in the formers. Compared with residential and commercial areas, cemeteries provide higher vegetation cover and lower edification level resulting in higher landscape connectivity, which facilitates insect dispersal. However, the main characteristic of the necropolis is the high density of artificial containers by the use of flower vases. Despite this generalized description (i.e. high vegetation cover, low-mid edification, high connectivity, and high container availability), remarkable differences are found among and within cemeteries because of the diverse anthropogenic structures related to different burial types. The larger cemetery of Buenos Aires City, namely Chacarita, occupies 72 ha and serves as an example of the heterogeneity of environments that can be found (Figure 1). The main structures are as follows: ground graves in open field (Figure 1a), mausoleums (Figure 1b), wall niches in open field (Figure 1c) or within galleries (Figure 1d), columbaria, administrative areas, and gardened areas and courtyards without tombs (Figure 1e,f). Mausoleums are small and closed buildings for a few coffins, and usually with a door with glass. Wall niches are arranged to form a single block. Similarly, columbaria are buildings where ashes (or more rarely, bones) are stored in sealed niches holding small urns. Wall niches and columbaria are important to save space in highly urbanized areas, as was described in Hong Kong by Teather (1998). Different burial structures could determine habitats of different suitability for mosquito proliferation, as discussed in the following section.

Figure 1.

 Burial structures. (a) Ground graves in open field, (b) mausoleums, (c) wall niches in open field, (d) wall niches within galleries, (e) gardened areas without tombs, and (f) courtyards without tombs.

Cemeteries as habitats for mosquitoes

To evaluate cemeteries as habitats for mosquitoes, first it is necessary to understand mosquito requirements. Adults of both sexes feed on sweet secretions, such as nectar, extrafloral nectars, honey-dew, damaged fruits, and damaged and intact vegetative tissues. Sugar-meals provide energy necessary for flight, dispersal and most other biological activities, though not usually for egg development. Females of most mosquitoes need a blood meal before they can develop their eggs, and host preference varies within a range that embraces mammals, birds, reptiles, amphibians and insects. Blood also could be an energy source for females. Adults of both sexes need shelter after emergence until ready for mating, and females seek refuge after feeding to digest their blood meal, develop their eggs and become gravid. Common places to rest are inside houses, rodent burrows, cracks in the ground, tree holes, amongst grass and other vegetation, on tree trunks, as well as on man-made structures such as fences and under bridges (see Clements 1992; Service 1995). Additionally, container-breeding species need water-filled containers for the development of immatures. In brief, there are four basic requirements to take into account when considering cemeteries as mosquito habitat: (i) sugar substances, (ii) blood, (iii) shelter, and (iv) water-filled containers.

Carbohydrates for mosquito adults are highly abundant within cemeteries. The main sources throughout the year are the numerous fresh-cut flowers brought in by visitors to commemorate their dead. These bouquets of flowers are placed in flower vases with water or wet sand, directly on the grave surface, or tied to the handle of the wall niche. Another source of sugar is the vegetation planted as ornament and/or to provide shadow, commonly found in these environments. For example, in a patch of 1 ha within a cemetery of Buenos Aires there were 35 plant species, with 15 of them flowering during summer (D. Vezzani, unpublished data).

Blood sources for mosquito females are also highly abundant. Permanent sources of human blood are visitors and caretakers during the day hours, and watchmen, furtive visitors and occasionally homeless people at night (e.g. Schultz 1989). Other common sources are represented by stray dogs and cats. Besides, birds, mammals, reptiles, amphibians, and insects usually are more abundant in cemeteries and parks than in other areas of the city.

A frequent landscape in urban cemeteries is a mixed environment composed of vegetation (trees, shrubs and grass) and small buildings. In general, this type of landscape is plenty of microhabitats suitable as shelters for adult mosquitoes. Specifically, each man-made structure may exhibit some favourable or unfavourable characteristics for mosquito life. Galleries provide an excellent buffer for extreme temperatures during the summer, but avoid the natural filling of containers with rainfall water. Open field areas usually are landscapes of high connectivity enhancing mosquito dispersal, contrary to built areas as mausoleums and multi-storey galleries. The burial structures could also determine the amount of vegetation; galleries usually have courtyards whereas open fields exhibit a matrix of grass. Also, the interior of containers are appropriate resting places for adult mosquitoes, which are enhanced by the presence of fresh or artificial flowers. In rural and park-cemeteries, vegetation probably plays a main role as mosquito shelter due to the scarcity of man-made structures.

Artificial containers are highly abundant in many cemeteries. For example, cemeteries have up to 2100 containers per ha in Buenos Aires (Vezzani et al. 2001), 2500 in Caracas (Barrera et al. 1979), and 6000 in Curitiba (Lozovei & Chahad 1997). The sources of these artificial containers are the administrative authorities of the necropolis and frequently the relatives and friends who bring in other containers to be used as flower vases. The most common types of containers provided by administrators are vases made of stone, plastic and metal, usually with an inner plastic or metallic container. The metals used are bronze, copper and aluminium. The external vase can have a drain hole, often plugged by leaves and other detritus (O'Meara et al. 1992b; Schmidt 2003; D. Vezzani, personal observation), turning it into a double breeding site, one external and other internal. Other containers brought in by visitors are glass or plastic bottles and flasks, toys, cans and vases made of every type of material. Cemetery containers were described in comparison with other urban breeding sites by Barrera et al. (1979) as follows: they possess more persistence, more homogeneity in form and capacity, higher density, and more heterogeneity in quality and quantity of organic matter. The organic matter in flower vases comes not only from fallen stems and leaves, but also from the fresh-cut flowers placed inside. Another particularity of these artificial containers is their single use as flower vases.

Overall, cemeteries could be perfect habitats for artificial container-breeding mosquitoes. Specifically for each cemetery, the combination of the factors discussed above determines its suitability as mosquito habitat. As it was previously stated, vegetation performs a main role due to its function as source of sugar, shelter and detritus. Obviously, some necropolis could exhibit characteristics unfavourable for mosquito breeding (e.g. the absence of flower vases and vegetation).

Mosquito species of the genera Aedes, Ochlerotatus, Culex, Toxorhynchites, Culiseta, Armigeres, Lutzia, Uranotaenia and Tripteroides were found breeding in artificial containers within cemeteries around the world; this information is summarized in Table 1. Among the 31 mosquito species reported in cemeteries from 16 countries, the invasive mosquitoes Ae. aegypti and Ae. albopictus were the most frequent ones. In addition, many species were captured in cemeteries as adults but not breeding in the containers; e.g. Schultz (1989) reported adults of 14 mosquito species not included in the table.

Table 1.   Mosquito species found breeding in artificial containers within cemeteries throughout the world
Mosquito speciesGeographic locationReferences
  1. *According to the most current study on generic-level names in Aedini (Reinert et al. 2004) the subgenus Stegomyia was elevated to generic rank. Therefore, the current names for Aedes aegypti, Ae. albopictus and Ae. riversi are Stegomyia aegypti, St. albopicta and St. riversi, respectively. Also, Ochlerotatus togoi is currently named Tanakaius togoi.

  2. †Referred by the authors as Cx. fatigans (syn. of Cx. quinquefasciatus).

Aedes aegypti*USAEbeling 1975; O'Meara et al. 1992a,b, 1995; Juliano 1998; Comiskey et al. 1999; Lounibos et al. 2001; Juliano et al. 2002; Reyes-Villanueva et al. 2003; Juliano et al. 2004
MexicoGorrochotegui-Escalante et al. 1998; Torres-Estrada et al. 2001
VenezuelaAnduze 1973; Barrera et al. 1979; Kazana et al. 1983; Abe et al. 2005
PhillippinesSchultz 1989, 1993
ArgentinaAvilés et al. 1997; Almirón & Ludueña Almeida 1998; Vezzani et al. 2001, 2004a,b,c, 2005; García et al. 2002; Vezzani & Schweigmann 2002
Puerto RicoNovak et al. 1985
BarbadosBelkin & Heinemann 1976
MartiniqueMoutenda & Yebakima 1999
Costa RicaSoto et al. 1999
PerúJamanca et al. 2005
Chagos Archipelago (Indian Ocean)Lambrecht & Van Someren 1971
Ae. albopictus*USASmith et al. 1990; O'Meara et al. 1992a,b, 1995; Walker et al. 1996; Fukuda et al. 1997; Juliano 1998; Kitron et al. 1998; Comiskey et al. 1999; Swanson et al. 2000; Lounibos et al. 2001; Xue et al. 2001; Juliano et al. 2002, 2004; Reyes-Villanueva et al. 2003
Hawaii (USA)LaPointe 2002
JapanSota et al. 1994
PhillippinesSchultz 1989, 1993
ItalyRaineri et al. 1991, 1993; Romi et al. 2000
MalaysiaSulaiman et al. 1996a,b
BrazilBrito et al. 1986
Chagos Archipelago (Indian Ocean)Lambrecht & Van Someren 1971
Ae. riversi*JapanSota et al. 1994
Ochlerotatus triseriatusUSAO'Meara et al. 1992a,b; Fukuda et al. 1997; Swanson et al. 2000; Lounibos et al. 2001; Xue et al. 2001; Schmidt 2003
Oc. fluviatilisBrazilLozovei & Luz 1976; Silva & Lopes 1985; Lopes et al. 1993; Chahad & Lozovei 1994; Lozovei & Chahad 1997; Silva 2002
VenezuelaAnduze 1973
Oc. japonicusJapanSota et al. 1994
USASchmidt 2003
Oc. atropalpusCosta RicaKumm et al. 1940
Oc. sierrensisUSARusmisel et al. 1999
Oc. homoeopusGuatemalaHeinemann & Belkin 1977b
Oc. togoi*JapanSota et al. 1994
Culex pipiens-quinquefasciatus complexBrazilLozovei & Luz 1976; Silva & Lopes 1985; Lopes et al. 1993; Chahad & Lozovei 1994; Lozovei & Chahad 1997
USAShanafelt 1969; Ebeling 1975; Dhillon et al. 1980; Dhillon & Mulla 1982; Rusmisel et al. 1999; Schmidt 2003
Hawaii (USA)LaPointe 2002
PhillippinesSchultz 1989
Chagos Archipelago (Indian Ocean)Lambrecht & Van Someren 1971
MexicoPérez Pacheco et al. 2004
ArgentinaVezzani et al. 2001; García et al. 2002
GuatemalaHeinemann & Belkin 1977b
Venezuela†Anduze 1973; Barrera et al. 1979; Kazana et al. 1983
Cx. mollisBrazilSilva & Lopes 1985; Lopes et al. 1993; Chahad & Lozovei 1994
ArgentinaVezzani et al. 2001
Cx. coronatorBrazilSilva & Lopes 1985
Costa RicaHeinemann & Belkin 1977a
GuatemalaHeinemann & Belkin 1977b
Cx. cornigerVenezuelaBarrera et al. 1979
Costa RicaHeinemann & Belkin 1977a
MexicoStrickman & Pratt 1989
Cx. nigripalpusVenezuelaBarrera et al. 1979
Cx. maxiArgentinaOria et al. 2000
Cx. restuansUSAFoss & Dearborn 2002; Schmidt 2003
Cx. eduardoiArgentinaVezzani et al. 2001
Cx. peusUSADhillon & Mulla 1982
Cx. sasaiJapanSota et al. 1994
Cx. stigmatasomaUSARusmisel et al. 1999
Cx. tarsalisUSARusmisel et al. 1999
Cx. pinarocampaGuatemalaHeinemann & Belkin 1977b
Toxorhynchites theobaldiVenezuelaAnduze 1973; Barrera et al. 1979; Kazana et al. 1983
Tx. rutilusUSAO'Meara et al. 1992a
Culiseta incidensUSAShanafelt 1969; Ebeling 1975; Dhillon et al. 1980; Dhillon & Mulla 1982; Rusmisel et al. 1999
Cs. inornataUSARusmisel et al. 1999
Armigeres subalbatusJapanSota et al. 1994
Lutzia bigotiGuatemalaHeinemann & Belkin 1977b
Uranotaenia coatzacoalcosGuatemalaHeinemann & Belkin 1977b
Tripteroides bambusaJapanSota et al. 1994

Cemeteries as study areas: general concepts

‘In an increasingly fragmented and urbanized landscape, it is imperative to select model patches and systems where urban and natural influences and relationships can be quantified (i.e. where artificial and natural factors converge on a specific landscape element). We suggest that cemeteries provide such an integrative unit of investigation’. This statement by Barrett and Barrett (2001) clearly expresses the potential usefulness of urban cemeteries for researchers from different fields. Cemeteries were used as study areas for different biological groups, e.g. plants (Loneragan 1975; Hopper 1977; Prober & Thiele 1995; Casler 2004; Sudnik-Wójcikowska & Galera 2005), birds (Lussenhop 1977; Hopper & Burbidge 1978; Zalewski 1994; Kocian et al. 2003), mammals (McPherson & Nilon 1987; Arenz & Leger 1999; Baker et al. 2000, 2003), fungi, lichen and bacteria (Walker et al. 1996; Hyerczyk 1997; Gorbushina et al. 2002), reptiles (Barbault & Mou 1988; Thompson 1992, 1994), amphibians (Rice & Jung 2004; Severtsova & Severtsov 2005), and insects other than mosquitoes (Harris 1998; Luyt & Johnson 2001; Hughes et al. 2003; Bourel et al. 2004). But indubitably, the most frequent animals studied in cemeteries were the mosquitoes. The results of more than 30 researches concerning mosquitoes in cemeteries were published over the last decade.

Among the features described for these habitats, some of them present clear advantages or disadvantages for their use as field sampling or experimental areas. There are logistical problems incurred from working in urban settings, such as difficulty in obtaining permission to conduct large-scale experiments on private property, as well as vandalism to field equipment (McIntyre et al. 2000). In this sense, it is reasonably easy to obtain authorization from local authorities to access and work within cemeteries, and it could be possible to avoid vandalism to field equipment by an arrangement with cemetery employees. In particular, the homogeneity in the use of containers offers a unique chance to study the suitability of some container types for mosquito breeding, independently of the human use. On the other hand, the main problem about the use of cemeteries to study certain issues related with mosquitoes is the validity of the results, i.e. are results extrapolatable to other urban environments? In some particular cases, the answer is probably not. For example, Ae. aegypti dispersal is driven by oviposition behaviour and depends on container availability (Reiter et al. 1995; Edman et al. 1998), and therefore, cemeteries with high container densities would not be appropriate to study the typical dispersal range of this species.

In view of these advantages and disadvantages, the use of cemeteries as study areas has to be evaluated considering the objectives of each particular study and the specific characteristics of local cemeteries. In some cases, cemeteries were useful to collect mosquitoes for laboratory bioassays (e.g. Reyes-Villanueva et al. 2003), or to know species composition (e.g. Rusmisel et al. 1999). In contrast, some complex ecological studies about competence among sympatric mosquito species were performed within these environments (e.g. Juliano et al. 2004).

Mosquito studies performed in cemeteries

This section deals with available information regarding mosquito studies carried out in cemeteries, summarized in Figure 2. Although many of these studies could be related simultaneously to different subjects, they were grouped into the following categories to facilitate reading: early detection, ecological approaches, and mosquito control studies.

Figure 2.

 Mosquito studies carried out in cemeteries grouped by subject.

Early detection and monitoring of Aedes mosquitoes

Monitoring the appearance of an insect pest is essential to avoid its establishment, and it is reasonable to use highly suitable habitats to detect the early presence of invasive species. The guidelines provided by the WHO through its regional offices for monitoring the infestation or reinfestation of an area by Aedes vectors usually include cemeteries as strategic sites, together with seaports, airports and tire facilities (e.g. PAHO 1994; OPS 1997, 2000; Anonymous 1998). Probably, cemeteries and tire dumps are ideal sites for monitoring the spreading of Aedes vectors over a region, whereas seaports and airports are useful to detect the arrival of these vectors at country or continental scales, e.g. the appearance and establishment of Ae. albopictus and Ochlerotatus japonicus in the US was mainly through the arrival of shipments of used tires (Lounibos 2002).

Despite the guidelines provided by the WHO, reports about invasive mosquitoes detected early in cemeteries are scarce. The most notorious example is the case of Ae. albopictus in the State of Florida (USA). Until 1990, the detection of Ae. albopictus exclusively in a cemetery was registered only in one county (Smith et al. 1990), and a few years later the initial discovery of this species was made at cemeteries in 17 counties (O'Meara et al. 1993). Moreover, the establishment of Ae. albopictus in some cemeteries before appearing in nearby accumulations of waste tires suggested that its spread could have been due to the passive transportation of eggs attached to flower vases from one cemetery to another (O'Meara et al. 1992a, 1993), being an alternative spreading route of container-breeding mosquitoes. Aedes albopictus was also detected earlier in a cemetery than in surrounding areas in Memphis, TN, USA (Reiter & Darsie 1984), and in São Paulo State, Brazil (Brito et al. 1986).

Another example is the discovery of Ae. aegypti in two localities from Argentina, Buenos Aires and Villa María, using larvitraps placed within cemeteries (Junín et al. 1995; Avilés et al. 1997). In addition, larvae and pupae of this species were found in a cemetery of Cordoba City as part of a vector monitoring study within the city (Almirón & Ludueña Almeida 1998).

Ecological approaches

Independent of the study complexity, each mosquito survey performed within the cemetery boundaries contributes to the knowledge of the mosquito community present in the area. In these sense, Anduze (1973) and Barrera et al. (1979) were pioneers in cemeteries in Venezuela, Shanafelt (1969) in USA, and Lozovei and Luz (1976) in Brazil. From the 1980s onwards, many studies designed with the aim of registering species composition included a cemetery as part of the study area, e.g. Lopes et al. (1993), Oria et al. (2000), Foss and Dearborn (2002) and Silva (2002).

Several authors studied the occurrence of mosquitoes in flower vases of different materials with dissimilar results. Among the studies focused on Ae. aegypti, one of them did not find differences among materials (Abe et al. 2005), another one reported higher infestations for some materials (García et al. 2002), while a third one documented both findings in contrasting patches within the same cemetery (Vezzani & Schweigmann 2002). Differences were also observed for Cx. quinquefasciatus and Oc. fluviatilis (Chahad & Lozovei 1994; Lozovei & Chahad 1997), for Aedes spp. and Culex spp. (Silva & Lopes 1985), and for mosquito fauna in general (Schmidt 2003). But the preferred materials observed in these studies were usually different, e.g. plastic, concrete, glass or aluminium. In this sense, Vezzani and Schweigmann (2002) suggested that microenvironmental conditions, as solar exposition level, affect the suitability of each type of container. In Florida cemeteries, O'Meara et al. (1992a,b) found that Aedes larvae were less frequent and less abundant in bronze and copper containers than in other material containers. These authors hypothesized that cooper ions released into the water may be toxic to the larvae or to microorganisms serving as larval food. Later, also in a Florida cemetery, Walker et al. (1996) determined that bacteria (one component of larval food) were more abundant in bronze containers than in non-bronze containers, thus rejecting the second hypothesis.

In regard to container capacity, Vezzani et al. (2004b) and Abe et al. (2005) reported that Ae. aegypti productivity was higher in containers of 1–5 l than in those up to 1 l. Sota et al. (1994) studied the relationship of container capacity with water level and with the mosquito community present along the seasons in natural and artificial containers, including vases from a cemetery.

Besides the features of flower vases, different approaches were used to assess the habitat suitability for mosquitoes at different spatial scales. Mosquito fauna was compared among urban habitats such as residential areas, junkyards, parks and cemeteries (Lozovei & Luz 1976; Lopes et al. 1993; Silva 2002). Sota et al. (1994) compared mosquito community between artificial containers from cemeteries and natural containers (bamboo stumps and treeholes) from wooded areas. Particularly for Ae. aegypti and Ae. albopictus, Schultz (1989) found in Philippines that the former predominated in cemeteries with scarce vegetation whereas the second was highly abundant in vegetated cemeteries. In Florida, O'Meara et al. (1992a) observed that well-shaded cemeteries are likely to harbour more Aedes mosquitoes in general. In the absence of Ae. albopictus in temperate Argentina, Ae. aegypti infestation level was correlated with the vegetation cover when five cemeteries were studied (Vezzani et al. 2001). Within a cemetery, vegetation cover was positively associated with the number of pupae (Abe et al. 2005), and at a microhabitat scale the presence of Ae. aegypti immatures was enhanced by sites less exposed to sunlight, with taller and closer vegetation, and shaded and vegetated surroundings (Vezzani et al. 2005). Silva and Lopes (1985) reported dissimilar mosquito frequencies for Aedes spp. and Culex spp. in shaded and exposed containers through the seasons. Barrera et al. (1979) also reported some differences among mosquito species; Cx. corniger preferred containers exposed to sunlight and Cx. fatigans (syn. Cx. quinquefasciatus) and Ae. aegypti were more frequently found in shaded vases. They also detected a temporal succession from Cx. corniger to Cx. fatigans to Ae. aegypti following the sequence fresh flowers, wilted flowers, and organic matter without flowers. Partially similar results were observed for Ae. aegypti by García et al. (2002).

Mosquito population seasonality was studied in cemeteries worldwide. As expected, in temperate areas mosquito abundance was associated with temperature (Sota et al. 1994; Swanson et al. 2000; García et al. 2002; Vezzani et al. 2004a), and the occurrence of mosquitoes during the unfavourable seasons depends on winter severity. In tropical areas with a marked rainy season, mosquito fauna proliferates all year round but higher abundances were recorded during the wet months (Barrera et al. 1979; Schultz 1989, 1993). In the dry season, immature mosquitoes can be found because of human-filled vases. Some studies in Brazil revealed the presence of mosquitoes during the four seasons and abundance patterns were correlated with both temperature and rain, with certain variations among species (Lozovei & Luz 1976; Silva & Lopes 1985; Chahad & Lozovei 1994; Lozovei & Chahad 1997).

A few attempts were made to identify natural enemies of mosquitoes in cemeteries. Parasites of the genus Ascogregarina were found infecting larvae of Ae. albopictus (by A. taiwanensis), Oc. triseriatus (by A. barretti) and Ae. aegypti (by A. culicis) (Fukuda et al. 1997; Comiskey et al. 1999). Negative effects of green algae, Chlorella ellipsoidea, on mosquito populations of Cx. quinquefasciatus and Culiseta incidens from cemetery vases were reported by Dhillon and Mulla (1982). Field surveys of aquatic insects from flower vases suggested that the predator mosquito Toxorhynchites theobaldi would have a low impact on Ae. aegypti and Cx. fatigans populations, although predation on the immature instars was demonstrated (Kazana et al. 1983). On the contrary, spiders seem to play a main role in suppressing mosquito adult populations; Sulaiman et al. (1996a) identified four spider species (Heteropoda venatoria, Heteropoda sp., Pardosa sp., Leucauge decorata) feeding on Ae. albopictus adults in a cemetery.

The results of some studies performed in cemeteries have direct ecoepidemiological implications. Annual mosquito productivity in a cemetery was estimated to be 12 million larvae (Lozovei & Chahad 1997), and at a given date of the year it was between 2.5 and 60 million larvae (Barrera et al. 1979; Silva & Lopes 1985; Schmidt 2003). For Ae. aegypti, the output estimated was 3000 females per day from a small cemetery of 3.5 ha from Venezuela (Abe et al. 2005). Survival rates of Ae. albopictus immatures were estimated in a Malaysian cemetery (Sulaiman et al. 1996b), and adult survival rates were estimated in some areas of China from populations of several environments including cemeteries (Almeida et al. 2005). The latter study also included other aspects of the bioecology of Ae. albopictus related to the transmission of dengue virus, e.g. feeding preference, daily activity cycle, human biting rate and vectorial capacity. Biting activities of Ae. albopictus and Ae. aegypti were also investigated in five cemeteries from Philippines (Schultz 1989). The dispersal range of Ae. albopictus was studied in a rural cemetery of Singapore by rubidium-marked eggs, previous elimination of removable breeding habitats to increase the proportion of oviposition in ovitraps (Liew & Curtis 2004). Kitron et al. (1998) performed an interesting survey in which a chipmunk seropositive for La Crosse virus was trapped within a cemetery and near to breeding sites of Ae. albopictus, a potential vector of this virus.

Studies about interspecific competition between sympatric mosquito species were performed in cemeteries only in USA and mainly in Florida. O'Meara et al. (1992a, 1993, 1995) observed declines in Ae. aegypti and Oc. triseriatus populations associated with the invasion and expansion of Ae. albopictus populations in Florida, providing indirect evidence of interspecific competition. A similar assumption was inferred from results concerning the oviposition activity of Oc. triseriatus and Ae. albopictus during a longitudinal survey in Illinois (Swanson et al. 2000). However, Lounibos et al. (2001) found only weak evidence to support displacement of Oc. triseriatus by Ae. albopictus in cemetery vases. A few studies have tested for competitive effects among Aedes species in the field. Juliano (1998) estimated the intensity of competition between Ae. albopictus and Ae. aegypti by measuring the numbers and sizes of adults emerged from cemeteries with known histories of invasion by the former and decline of the second one. Similarly, Juliano et al. (2004) demonstrated interspecific larvae competition by field experiments in Florida cemeteries. Finally, data obtained from six cemeteries by Juliano et al. (2002) supported their laboratory observations that suggest differential egg mortalities for Ae. aegypti and Ae. albopictus, facilitating the coexistence of both species in areas with warm and dry climate.

Mosquito control studies

Field assays to evaluate biological, environmental and chemical mosquito control were performed within some necropolis. Among biological control agents, the nematode Romanomermis culicivorax was not effective to control larval populations of Cx. quinquefasciatus and Cs. incidens in a cemetery from California, contrary to laboratory observations (Dhillon et al. 1980). In Mexico, Gorrochotegui-Escalante et al. (1998) achieved a reduction of 67% in the number of vases positive for Ae. aegypti after the inoculation of the copepod Mesocyclops longisetus, although they had to refill the containers periodically to avoid desiccation. Another copepod, M. thermocyclopoides, has also been successfully used for Ae. aegypti control involving community participation in a cemetery from Costa Rica, but Culex spp. larvae were not predated (Soto et al. 1999). Community compromise to replace water by wet sand was evaluated by Moutenda and Yebakima (1999) in a cemetery from Martinique during All Saint's Day. They observed an excellent response from the visitors, but unfortunately the mosquito infestation levels before and after the treatment were not measured.

The efficiency of chemical control of mosquitoes was exhaustively assessed during the last century, but only a few studies were performed within cemeteries. Shanafelt (1969) tested five insecticides applied by hand method to individual containers in a cemetery from California. Although the author observed 100% mortality among the immature mosquito population, he concluded that hand methods would be impractical in large cemeteries. Also in California, temephos from a boom sprayer was found to be ineffective, chlorpyrifos provided excellent control for only 1–2 months, and slow-release charcoal briquettes containing methoprene resulted in excellent inhibition of adult emergence for over 5 months (Mulla et al. 1977). Aedes larvae were eliminated for 4–6 weeks post-treatment with deet and two piperidines in stone vases without flowers in Florida (Xue et al. 2001). In a small cemetery in Buenos Aires the percentage of Ae. aegypti breeding sites decreased from 18.4% to 2.2% with only two applications of temephos during the reproductive season, and to 0.05% with five applications (Vezzani et al. 2004c). Based on the observations of O'Meara et al. (1992a,b), Romi et al. (2000) performed field assays to evaluate the effect of cooper ions on Ae. albopictus development within a cemetery from Genoa, Italy. A multi-wire electric cable was used as cooper source in ovitraps and a reduction of 90% in the number of immatures with several months of persistence of the toxic effects was observed.

The control of mosquitoes within cemeteries

Larval populations developing in containers are intrinsically difficult targets to control (Washburn 1995). In particular, the control of artificial container-breeding mosquitoes within cemeteries is a challenging task due to the high density of flower vases holding water. Several authors agree that human habit of leaving flowers in water is hard to break, and cultural and religious customs must be considered to prevent mosquito proliferation (Barrera et al. 1982; Natal et al. 1997; Romi et al. 2000; Schmidt 2003).

In the compiled bibliography, only about 10 papers proposed some specific recommendations to control mosquitoes in cemeteries. The most obvious action is to remove the vases (Schultz 1989; Lozovei & Chahad 1997). Definitely, it would be largely effective to avoid mosquito breeding, but it is in direct conflict with population customs. Furthermore, in many cemeteries caretakers obtain their remuneration from people who commemorate their loved ones with fresh-cut flowers. The second most obvious action is the frequent replace of water in vases (Schultz 1993; García et al. 2002; Schmidt 2003), but probably it is almost impracticable in large cemeteries or for long periods. Specific recommendations to limit mosquito production are as follows: fill the vases with wet soil or sand, use only artificial flowers on the ground, turn the containers upside down, use vases with a drain hole and keep it clean from obstruction, and use bronze vases (Shanafelt 1969; Barrera et al. 1982; Schultz 1989, 1993; O'Meara et al. 1992a,b). In regard to vases filled with wet sand, containers in Buenos Aires cemeteries that were partially empty after a few weeks by the effect of rain and wind, harboured mosquito larvae and pupae (D. Vezzani, personal observation). Natal et al. (1997) summarized some issues about biological and chemical control and highlighted that the elimination of water in vases along with education campaigns are the main activities in any control programme. Public awareness was also pointed by Schmidt (2003) and Lozovei and Chahad (1997). Mosquito control strategies in cemeteries should also consider the increase of water-filled vases around special dates, such as Easter and Memorial Day (Shanafelt 1969), Mother's Day (Barrera et al. 1979), Dead's Day (Chahad & Lozovei 1994), and All Saint's Day (Moutenda & Yebakima 1999). Considering mosquito resistance to insecticides, it is recommendable to avoid chemicals for routine control and keep them as an efficient tool for preventing potential epidemics of mosquito-borne diseases. In brief, environmental management and educational campaigns are the key factors, and cemeteries caretakers and administrators should play a central role.

At last, I believe that in many cemeteries the regulations are essentially appropriate to reduce mosquito abundance. Unfortunately, these sets of laws are poorly enforced and final responsibility belongs to local authorities. For example, in Buenos Aires cemeteries it is forbidden for visitors to bring in containers to be used as flower vases and only those provided by local authorities are permitted. However, about half of the flower vases surveyed (n = 20 032) were flasks, bottles and tins introduced by visitors (Vezzani & Velázquez 2003).

Final comments

Overall, cemeteries are highly suitable environments for artificial container-breeding mosquitoes, and the basic reason is the great availability of the different resources that they need (i.e. sugar substances, blood, shelter and water-filled containers). Also, these places are mostly ideal settings to perform mosquito studies in urbanized areas because of the high abundance of mosquitoes, the heterogeneity of macro- and microhabitats, and the easier access in comparison with private premises. However, the feasibility of a cemetery as a study area must be evaluated in each case considering the objectives of the study and cemetery characteristics.

The leading position of the USA with respect to mosquito studies performed in cemeteries is clearly demonstrated in Table 1, not only in number but also in quality. It is highly probable that many studies concerning mosquitoes and cemeteries in other countries were published in journals and proceedings of local scope and thus remain unnoticed. Similarly, to my knowledge there are only 31 mosquito species that have been recorded breeding in artificial containers from cemeteries. This number is surely underestimated. Finally, I hope that the present review encourages researchers from different disciplines to use cemeteries as great field laboratories within the cities.

Ancillary