Genomic vulnerability and socio-economic threats under climate change in an African rainforest bird.

Preserving biodiversity under rapidly changing climate conditions is challenging. One approach for estimating impacts and their magnitude is to model current relationships between genomic and environmental data and then to forecast those relationships under future climate scenarios. In this way, understanding future genomic and environmental relationships can help guide management decisions, such as where to establish new protected areas where populations might be buffered from high temperatures or major changes in rainfall. However, climate warming is only one of many anthropogenic threats one must consider in rapidly developing parts of the world. In Central Africa, deforestation, mining, and infrastructure development are accelerating population declines of rainforest species. Here we investigate multiple anthropogenic threats in a Central African rainforest songbird, the little greenbul (Andropadus virens). We examine current climate and genomic variation in order to explore the association between genome and environment under future climate conditions. Specifically, we estimate Genomic Vulnerability, defined as the mismatch between current and predicted future genomic variation based on genotype-environment relationships modeled across contemporary populations. We do so while considering other anthropogenic impacts. We find that coastal and central Cameroon populations will require the greatest shifts in adaptive genomic variation, because both climate and land use in these areas are predicted to change dramatically. In contrast, in the more northern forest-savanna ecotones, genomic shifts required to keep pace with climate will be more moderate, and other anthropogenic impacts are expected to be comparatively low in magnitude. While an analysis of diverse taxa will be necessary for making comprehensive conservation decisions, the species-specific results presented illustrate how evolutionary genomics and other anthropogenic threats may be mapped and used to inform mitigation efforts. To this end, we present an integrated conceptual model demonstrating how the approach for a single species can be expanded to many taxonomically diverse species.

A continental risk assessment of West Nile virus under climate change.

Since first introduced to North America in 1999, West Nile virus (WNV) has spread rapidly across the continent, threatening wildlife populations and posing serious health risks to humans. While WNV incidence has been linked to environmental factors, particularly temperature and rainfall, little is known about how future climate change may affect the spread of the disease. Using available data on WNV infections in vectors and hosts collected from 2003-2011 and using a suite of 10 species distribution models, weighted according to their predictive performance, we modeled the incidence of WNV under current climate conditions at a continental scale. Models were found to accurately predict spatial patterns of WNV that were then used to examine how future climate may affect the spread of the disease. Predictions were accurate for cases of human WNV infection in the following year (2012), with areas reporting infections having significantly higher probability of presence as predicted by our models. Projected geographic distributions of WNV in North America under future climate for 2050 and 2080 show an expansion of suitable climate for the disease, driven by warmer temperatures and lower annual precipitation that will result in the exposure of new and naïve host populations to the virus with potentially serious consequences. Our risk assessment identifies current and future hotspots of West Nile virus where mitigation efforts should be focused and presents an important new approach for monitoring vector-borne disease under climate change.

Economic conditions predict prevalence of West Nile virus.

Understanding the conditions underlying the proliferation of infectious diseases is crucial for mitigating future outbreaks. Since its arrival in North America in 1999, West Nile virus (WNV) has led to population-wide declines of bird species, morbidity and mortality of humans, and expenditures of millions of dollars on treatment and control. To understand the environmental conditions that best explain and predict WNV prevalence, we employed recently developed spatial modeling techniques in a recognized WNV hotspot, Orange County, California. Our models explained 85-95% of the variation of WNV prevalence in mosquito vectors, and WNV presence in secondary human hosts. Prevalence in both vectors and humans was best explained by economic variables, specifically per capita income, and by anthropogenic characteristics of the environment, particularly human population and neglected swimming pool density. While previous studies have shown associations between anthropogenic change and pathogen presence, results show that poorer economic conditions may act as a direct surrogate for environmental characteristics related to WNV prevalence. Low-income areas may be associated with higher prevalence for a number of reasons, including variations in property upkeep, microhabitat conditions conducive to viral amplification in both vectors and hosts, host community composition, and human behavioral responses related to differences in education or political participation. Results emphasize the importance and utility of including economic variables in mapping spatial risk assessments of disease.