It is widely recognized that transmission of malaria and other vector-borne diseases (VBDs) is greatly influenced by environmental factors, especially temperature, due to the ectothermic nature of insect and arthropod vectors. However, the response of transmission to temperature and other drivers is complex, as thermal traits of ectotherms are typically non-linear, and they interact to determine transmission constraints. In this study, we assess and compare the effect of temperature on the transmission of the two most common malaria parasites Plasmodium falciparum and Plasmodium vivax by two of the most common malaria mosquito vector species, Anopheles gambiae and Anopheles stephensi. We model the non-linear responses of temperature dependent mosquito and parasite traits (mosquito development rate, bite rate, fecundity, egg to adult survival, vector competence, mortality rate, and parasite development rate) and incorporate these traits into a model for the basic reproductive number, R0(T), across temperatures.
Results/Conclusions
Our model predicts that transmission of both, P. falciparum and P. vivax by An. gambiae peak at 25 °C with confidence intervals (CI) of 23.8 - 26.1 °C and 23.9 - 26 °C respectively; and that the transmission of P. falciparum and P. vivax by An. stephensi peak at 24.8 °C (CI: 23.1 - 26.2 °C) and at 24.6 °C (CI: 23.2 - 25.6 °C) respectively. Optimum temperature for malaria transmission is very similar for the four mosquito/parasite combinations assessed in this study; the main differences are at the thermal limits. At the maximum temperature limit, we found a significant difference between An. stephensi/P. falciparum versus An. stephensi/P. vivax (5 °C) and An. stephensi/P. falciparum versus An. gambiae/P. falciparum (7.1 °C). At the minimum temperature limit, it was a significant difference between An. stephensi/P. falciparum and An. gambaie/P. falciparum (3.9 °C) and between An. stephensi/P. vivax versus An. gambiae/P. vivax (3.7 °C). Using prevalence data from Africa and Asia, we show that the reproductive number R0(T) calculated by our mechanistic model is consistent with observed data. We also use our model to construct visual representations of relative risk of transmission. Our risk maps show the areas in Africa and Asia that are suitable for malaria transmission year-round based on R0(T)>0. This study could expand the understanding of the effects of current and future temperatures on malaria transmission and could be useful in planning preventive strategies for areas where malaria outbreaks might occur in the future due to global warming.