Using regional systems based on ecology, not geopolitical boundaries, to determine the spread of West Nile virus

Scientists often use forecasting models to predict the spread of infectious diseases. The data informing these models are usually framed by geopolitical boundaries such as states or counties.

In a new paper published in GeoHealth, S. Kane Moser and colleagues argue that using such boundaries based on human activity and physical geography—rather than climatic or ecologic factors—may limit the accuracy of forecasts by obscuring relationships between disease and environment.

Mosquitoes carrying pathogens such as West Nile virus, for instance, pay no mind to the geopolitical boundaries along which health data are divided.

Regional U.S. disease forecasts often use the nine climate regions defined by NOAA. Though these regions consider climate factors, they are still grouped along state lines. In this study, the researchers instead used 10 level I ecoregions—or areas with similar ecosystems and comparable environmental resources—defined by the EPA.

The researchers studied Centers for Disease Control and Prevention (CDC) data on West Nile from 2005 to 2019 to examine correlations between temperature, precipitation, and West Nile cases in both NOAA climate regions and EPA ecoregions.

In some areas, the correlation between climate and virus incidence levels changed depending on which regional breakdowns were used in the analyses. For example, the Southern Semi-Arid Highlands EPA ecoregion is within the Southwest NOAA climate region, but the former showed a negative correlation between temperature and virus incidence levels, whereas the latter showed a positive correlation.

The correlations found within EPA ecoregions more often agreed with established knowledge about how environmental dynamics and mosquito biology affect the spread of disease.

For instance, virus rates tend to increase with anomalously high temperatures, but extremely high temperatures can have the reverse effect by causing mosquitoes to die off. (Because NOAA-defined regions tend to include more than one type of ecology, trends calculated for the entire region might be less accurate for specific locations within that region.)

The authors note that their results suggest that using regions partitioned by ecology (such as EPA ecoregions) instead of geopolitical regions could enhance future disease forecasts.

This story is republished courtesy of Eos, hosted by the American Geophysical Union. Read the original story here.