Breeding site preferences, prevalence, and predictive modelling of Aedes aegypti and Aedes albopictus in Matara District, Sri Lanka

Aedes aegypti and Aedes albopictus are major vectors of arboviral diseases and pose significant public health concerns. This study evaluated their breeding site preferences and prevalence across sub-regions of the Matara District, Sri Lanka. Monthly entomological surveillance was conducted in all Medical Officer of Health (MOH) areas from 2022 to 2024 using standard sampling methods. Chi-square tests were used to identify preferred breeding container types for each species. Cluster analysis using Ward’s linkage grouped MOH areas into three clusters based on vector prevalence, representing a gradient of vector receptivity. Monthly Premise Index (PI) values were analyzed alongside average rainfall to assess seasonal trends. Stepwise multiple regression was used to develop predictive PI models incorporating rainfall variables with different time lags. A total of 58,745 water-holding containers were recorded, of which 824 and 8,015 were positive for Ae. aegypti and Ae. albopictus, respectively. Oviposition preferences differed markedly between species: Ae. aegypti was associated exclusively with man-made containers, whereas Ae. albopictus showed a strong preference for containers linked to the natural environment. Spatially, Ae. aegypti predominated in densely populated coastal areas, while Ae. albopictus exhibited a broader inland distribution. Regression analysis revealed significant linear relationships between rainfall and PI, with a one-month lag for Ae. albopictus and no lag for Ae. aegypti (P < 0.005). The predictive models for both species and pooled PI demonstrated low root mean square error (RMSE) and high explanatory power. Notably, the pooled PI model showed a high R² value, indicating strong predictability of district-level vector prevalence using rainfall data, supporting its use as an early warning tool for proactive vector management. These findings highlight the importance of integrating ecological, spatial, and meteorological factors into routine surveillance systems and strengthening community-based interventions to enhance dengue prevention and control in endemic settings such as Sri Lanka.