Temporal and Spatial Synchronicity in West Nile Virus Cases Along the Central Flyway, USA

Abstract

This study of West Nile virus (WNV) examined the possibility of avian transmission to explain synchronicity in the year-to-year variability of WNV case numbers from Texas northward to the Dakotas, and reasons for the large case numbers on the northern Great Plains. We determined correlation coefficients between annual disease incidence per 100,000 people among states within the Great Plains Region, as well as the Central Flyway. There was spatial and temporal synchronicity, as evidenced by Pearson "r," with values along the core of the Central Flyway (Oklahoma, Kansas, Nebraska, and South Dakota) varying between 0.69 and 0.79. Correlations for North Dakota (r = 0.6), however, were affected by local conditions. The concept of relative amplification is helpful in explaining why northerly states along the Central Flyway have larger annual case numbers per 100,000 than Texas but preserve the temporal signal. States differed in their capacity for amplifying the temporal signal in case numbers. For example, Nebraska, South Dakota, and North Dakota case numbers were commonly amplified relative to Texas, with Oklahoma and Kansas deamplified. Relative amplification factors for all states increased as a function of increasing case numbers in Texas. Thus, increased numbers of initially infected birds in Texas likely led to the rapid intensification of the zoonotic cycle as compared to more typical years. The study also confirmed the importance of winter weather in locally modulating disease cases. North Dakota appeared most impacted by these factors to the extent of reducing WNV case numbers in colder years and years with deep snow.

Keywords: Amplification Factor; Central Flyway; Temporal and Spatial Analysis; Vector Competence; West Nile virus.

Figure 1

Annual West Nile virus incidence per 100,000 population by state, 1999–2021 (CDC, 2022).

Figure 2

Map showing the spatial distribution of dominant Culex species in the continental U.S. based on Darsie and Ward (2005) (reprinted permission from University Press of Florida), and Data‐driven identification of potential Zika virus vectors by Evans et al. (2017), licensed under CC BY 4.0, Modified from original.

Figure 3

Migratory bird flyways in the U.S (Differential changes in the onset of spring across US National Wildlife Refuges and North American migratory bird flyways by Waller et al. (2018), licensed under CC0 1.0).

Figure 4

(a) Pearson correlation coefficients created by linear correlations of annual total West Nile virus cases between Texas and each state in the continental U.S., (b) heat map illustrating r and p values for similar correlations for states in the Great Plains Region, and (c), an example of general migratory bird activity on flyways in the United States (reprinted with permission from BirdCast, Van Doren & Horton, 2018).

Figure 5

Statewide correlation analysis. Panels (a, b) show linear regressions between North Dakota (ND) and Texas (TX) and residual deviates from the linear line for each year, respectively. Panels (c, d) present the linear regressions between ND and South Dakota (SD), and residual deviates from the linear line for each year, respectively.

Figure 6

The amplification factor between Texas and each state along the Central Flyway changed as a function of disease cases per 100,000 population in Texas. The two lines for North Dakota/Texas reflected variability in winter weather in North Dakota. Data for all years are plotted for the indicated states. Trend lines were generated using logarithmic and power fittings.

Figure 7

Panel (a) shows a weak correlation between case numbers and heating degree days (HDD). The HDD are divided into a warm‐weather set (red) and a cold set (blue) for reference purposes. Panel (b) shows the correlation between total snowfall and annual disease cases described by two lines. The top line characterized by relatively low HDD represents warmer conditions. The lower line represents colder conditions. Panel (c) shows the correlation between the annual West Nile virus cases in Texas and North Dakota with smaller case numbers in colder years and larger case numbers in warmer years.