The lowest recorded rainfall for over 40 years characterised the period between 1992 and 1997 in the northern Queensland centres of Townsville and Charters Towers. Despite seemingly unfavourable conditions for Aedes aegypti (L.) during the drought years of 1992-1993, northern Queensland was affected by one of the largest Australian dengue fever epidemics recorded in recent history.
Preliminary surveys of Ae. aegypti of breeding sites in Townsville and Charters Towers (1994—1997) revealed that subterranean water containers such as wells and service manholes provided mosquitoes with a stable aquatic microhabitat with optimal physicochemical parameters for larval development, irrespective of the cold and dry conditions experienced at ground level during the northern Queensland dry season.
The main objective of this thesis was to investigate the importance of subterranean larval habitats to mosquito-borne disease. To do this, I investigated both the ecological and epidemiological significance of subterranean mosquito habitats in the dry tropics of northern Queensland, with a special focus on the dengue vector, Ae. aegypti. Subsidiary aims were to develop better tools and techniques for the sampling and control of subterranean mosquito populations.
The calibrated funnel trap was found to be a sensitive and an accurate tool for both the detection and quantification of larval populations of Ae. aegypti in subterranean habitats such as wells. A specially modified version of the funnel trap called the Vietrap, sampled on average 20.5% of the 3rd and 4th instars when set overnight for 12 hours. The sticky entry-exit trap for sampling adult mosquitoes also proved to be a simple and cost effective method for detecting mosquitoes that utilise subterranean habitats.
In north Queensland, eight species of mosquitoes were found breeding in subterranean habitats; Ae. aegypti, Aedes notoscriptus (Skuse), Aedes tremulus (Theobald), Culex annulirostris Skuse, Culex pullus Theobald, Culex quinquefasciatus Say, Tripteriodes atripes (Skuse) and Toxorhynchities specious (Skuse). Ae. aegypti and Ae. tremulus were the most frequently found in wells and manholes. In both the wet and dry season, the larval productivity of subterranean containers supported 97% of the total subterranean and surface larval population in the dry season and 59% in the wet season. This demonstrated the importance of subterranean breeding sites as dry season refuges.
The mean numbers of ova carried by gravid Ae. aegypti and Ae. tremulus entering subterranean habitats during the wet and dry seasons were compared. The mean egg load of Ae. aegypti increased from 39.8 in the dry season to 63.7 eggs per female in the wet season (t=2.6, df=30, P=0.014). Eggs oviposited in subterranean habitats which did not hatch quickly were destroyed by cockroaches or mycotoxins produced by opportunistic species of fungi such as Penicillium citrinum Thom.
Despite poor subterranean egg survival, cohort survival analysis revealed that, significantly more adult mosquitoes per 1000 eggs emerged from subterranean containers versus surface breeding sites during the dry season. However there was no significant difference between development times of Ae. aegypti in subterranean and surface containers, despite differences in temperature. Mean wing length of females emerging in the cooler surface sites, were larger than those mosquitoes that developed in subterranean containers. As there is a relationship between longer wing length and increased fitness, it can be argued that, although subterranean habitats produce greater numbers of Ae. aegypti in relation to surface sites during the dry season, the quality (fitness) of adults produced was significantly less.
The epidemiological significance of subterranean mosquito breeding sites to the 1993 outbreak of dengue fever (type 2) in the northern Queensland town of Charters Towers, Australia was investigated. This study was the first to demonstrate a direct epidemiological association between subterranean breeding sites and dengue virus infection. The mean distance between residents seropositive for dengue 2 and the nearest subterranean container (113 metres) was significantly less than for randomly selected controls (191 metres), (F=81.9, d.f=l,478, P<0.001). Residents living within 160 metres of a well or service manhole, were 2.5 (95% confidence interval 1.88- 3.24) times more likely to have dengue 2 antibodies, compared to those residing further away.
In an effort to find a low cost, sustainable and environmentaly friendly method to control subterranean populations of Ae. aegypti, three biological control agents were trialled, the copepod Mesocyclops aspericornis (Daday) and native freshwater fish, Melanotaenia splendida splendida (Peters) and Craterocephalus stercusmuscarum stercusmuscarum (Günther). The copepod was an excellent biological control agent for subterranean populations of Ae. aegypti. The fish were shown to be predators of Ae. aegypti. However, M s. splendida was also an efficient predator of Litoria caerulea tadpoles, a species of native frog found in wells during the dry season. This may limit the usefulness of M.s.splendida as a biological control agent of well breeding Ae. aegypti. Despite early promise C. s. stercusmuscarum was found to be a poor biological control agent due to its high mortality rates during even limited periods of road transport.