Comparison of Two Detailed Models of Aedes aegypti Population Dynamics

LEGROS M; OTERO, M. J.; ROMEO AZNAR, V; H G SOLARI; GOULD F; LLOYD A

Ecosphere

Wiley

Año: 2016

The success of control programs for mosquito borne diseases can beenhanced by crucial information provided by models of the mosquitopopulations. Models, however, can differ in their structure, complexity andbiological assumptions, and these differences impact their predictions.Unfortunately, it is typically difficult to determine why two complex modelsmake different predictions because we lack structured side-by-sidecomparisons of models using similar parameter values. Here we present adetailed comparison of two complex, spatially-explicit, stochastic models ofthe population dynamics of Aedes aegypti, the main vector of dengue,yellow fever, and chikungunya viruses. Both models describe themosquito´s biological and ecological characteristics, but differ in complexityand specific assumptions. We compare the predictions of these models intwo selected climatic settings, a tropical and weakly seasonal climate inIquitos, Peru, and a temperate and strongly seasonal climate in BuenosAires, Argentina. Both models were calibrated to operate at identicalaverage densities in unperturbed conditions in both settings, by adjustingparameters regulating densities in each model (number of larvaldevelopment sites and amount of nutritional resources). We show that themodels differ in their sensitivity to environmental conditions (temperatureand rainfall), and trace differences to specific model assumptions.Temporal dynamics of the Ae. aegypti populations predicted by the twomodels differ more markedly under strongly seasonal Buenos Airesconditions. We use both models to simulate killing of larvae and/or adultswith insecticides in selected areas. We show that predictions of populationrecovery by the models differ substantially, an effect likely related to modelassumptions regarding larval development and (direct or delayed) densitydependence. Our methodical comparison provides important guidance formodel improvement by identifying key areas of Ae. aegypti ecology thatsubstantially affect model predictions, and revealing the impact of modelassumptions on population dynamics predictions in unperturbed andperturbed conditions.