June, 2004 Journal of Vector Ecology 94 The invasion of urban forest by dengue vectors in Rio de Janeiro Ricardo Lourenço-de-Oliveira, Márcia G. Castro, Marieta A.H. Braks 1 , and L. Philip Lounibos 1 Laboratórios de Transmissores de Hematozoários, Instituto Oswaldo Cruz - Fiocruz, Av. Brasil 4365, 21045-900 Rio de Janeiro, RJ, Brazil 1 Florida Medical Entomology Laboratory, University of Florida, 200 9th St. SE, Vero Beach FL 32962, U.S.A. Received 8 July 2003; Accepted 22 September 2003 ABSTRACT: The invasion of a secondary forest within the city of Rio de Janeiro by Aedes aegypti and Ae. albopictus was evaluated from July 1997 to June 1998 through collections of immature stages in ovitraps set at 1m, 10 m, 100 m, 500 m, and 1,000 m into the forest from houses on the periphery. Both mosquito species were much more abundant close to houses (1-10 m). Aedes aegypti was not collected beyond 100 m, while Ae. albopictus was the most abundant species overall and in ovitraps at all distances from houses. Abundances of Ae. albopictus were significantly correlated with time-lagged rainfall and with abundances of Ae. aegypti. Co-occurrences of Ae. albopictus in traps with Ae. aegypti and Limatus durhami, but not with Culex dolosus, were more likely close to houses. The results suggest that the urban forest is a refuge for both Aedes species, but especially for Ae. albopictus, whose abundance both near houses and in the forest raises concern that this invader may transmit arboviruses to humans that are presently restricted to the sylvan environment. Journal of Vector Ecology 29 (1): 94-100. 2004. Keyword Index: Aedes aegypti, Aedes albopictus, secondary forest invasion, Rio de Janeiro. INTRODUCTION Aedes aegypti, eradicated from Brazil in the 1950s, reinfested the country during the late 1970s when it was detected in the cities of Salvador and Rio de Janeiro (Vasconcelos et al. 1999, Schatzmayr 2000). Dengue has been endemic in Rio de Janeiro since the introduction of dengue virus serotype 1 in 1986, and this city as well as other Brazilian urban centers have experienced severe outbreaks with the spread of serotype 1 and subsequent invasion of serotypes 2 and 3 (Schatzmayr, 2000, Nogueira et al. 2001, PAHO 2002). The control of Ae. aegypti, the main vector of dengue and urban yellow fever, has chiefly been based on chemical insecticides, mainly organophosphates, applied to artificial containers and through ULV spraying (Lima et al. 2003). The control of Aedes albopictus, which invaded the country in the 1980s (Consoli and Lourenço-de-Oliveira 1994), is not a priority for the national campaign, since this species has never been incriminated as a natural dengue vector in Brazil. Aedes albopictus is considered an important dengue vector in Asia (Shroyer 1986) and was the primary transmitter of this virus to man during a recent outbreak in Hawaii (MMWR 2002). Dengue virus was once isolated from its immature stages in Brazil (Serufo et al. 1993), and laboratory assays have shown that Brazilian populations of this mosquito species are susceptible to both dengue and yellow fever viruses (Miller and Ballinger 1988, Lourenço-de-Oliveira et al. 2003, Johnson et al. 2002). The application of insecticides to artificial containers is limited to indoors and in the vicinity of houses because Ae. aegypti is believed to be strictly peridomestic in Brazil. Phytotelmata, such as bromeliads growing close to houses, are rarely used by this species that prefers artificial containers (Cunha et al. 2002). By contrast, Ae. albopictus uses various kinds of human-made and natural larval containers and is not restricted to habitats associated with humans (Lounibos 2002). Such ecological adaptability may favor linkages of Ae. albopictus between sylvatic cycles of pathogens such as yellow fever, and the urban area (Marques et al. 1998, Vasconcelos et al. 1999), just as this species may serve as a bridge vector for West Nile virus in the U.S.A. (Turell et al. 2001, Holick et al. 2002). In the city of Rio de Janeiro, Ae. aegypti and Ae. albopictus occupy container habitats differently depending on the level of urbanization, the former species favoring more urban zones and the latter more rural areas (Braks et al. 2003a). In shady containers, larvae of Ae. albopictus outcompete and may exclude Ae. aegypti (Juliano 1998, Braks et al. 2003b). Larval interactions