Modeling Approaches to Investigating Distributions, Abundances, and Connectivity of Mosquito Vector Species

Abstract

Vector-borne and zoonotic diseases comprise a serious public health concern globally. Over the past 30 years, an increase in newly-emerging vector-borne pathogens, coupled with the broader dispersal of known pathogens, has resulted in substantial challenges for public health intervention and prevention programs. This burden highlights the need for continued improvement of modeling approaches and prediction methods to help identify areas vulnerable to infection, thereby contributing toward more efficient distributions of limited public health resources. The field of disease ecology emphasizes interactions between disease system components and the natural environment, recognizing that humans are not always the catalyst for pathogen dispersal and distributions. While incorporating environmental factors in assessing potential pathogen risk is a logical first step, complexities in this approach exist because pathogens are nested within the broader community ecology of host, vector, and reservoir species, and often, not all of these elements are known. Although this element poses challenges to understanding limiting factors of specific environmental pathogens, the multitude of components within individual disease systems offer several avenues from which to study patterns, providing insight into risk. Mosquito vectors are one such component. This knowledge, coupled with advances in geospatial technologies, provides excellent opportunities to model environmental factors contributing to potential pathogen distributions and to help predict disease risk in humans. Here, I present three ecological modeling approaches to quantify and predict suitable environments, abundances, and connectivity for three mosquito vector species important to human and domestic livestock health. The first chapter delivers a global model of suitable environments for Aedes aegypti and Ae. albopictus under present and future climate change calibrated on presence-only data. The second chapter outlines a new approach to predicting Ae. mcintoshi abundances in Kenya and western Somalia at an 8-day temporal resolution during October to January from 2002 – 2015. The third chapter demonstrates the potential to investigate Ae. mcintoshi population genetic structure and associations between environmental variables across eastern Kenya using gene sequence data. Each of these chapters address individual research questions using a disease ecology approach, while contributing more broadly to knowledge of mosquito vector ecologies and the potential for human disease risk.