Interpolation of Point Prevalence Rate of the Highly Pathogenic Avian Influenza Subtype H5N8 Second Phase Epidemic in South Korea

Abstract

Humans and animals are both susceptible to highly pathogenic avian influenza (HPAI) viruses. In the future, HPAI has the potential to be a source of zoonoses and pandemic disease drivers. It is necessary to identify areas of high risk that are more vulnerable to HPAI infections. In this study, we applied unbiased predictions based on known information to find points of localities with a high probability of point prevalence rate. To carry out such predictions, we utilized the inverse distance weighting (IDW) and kriging method, with the help of the R statistical computing program. The provinces of Jeollanam-do, Gyeonggi-do, Chungcheongbuk-do and Ulsan have high anticipated risk. This research might aid in the management of avian influenza threats associated with various potential risks.

Keywords: highly pathogenic avian influenza; inverse distance weighting; kriging; predictions; prevalence.

Figure 1

The overall scheme for the IDW and kriging analyses in this study.

Figure 2

The known point prevalence rate of HPAI H5N8 epidemic phase 2. Brown dots denote high prevalence while white circles denote low prevalence.

Figure 3

IDW interpolation with different powers. Power indicates the power function.

Figure 4

Depicts the comparison of observed prevalence rate with cross validated predicted prevalence rate. (A) Mean squared error against different powers. (B) “Predicted prevalence” vs. “observed prevalence”. (C) “Cross-validated predicted prevalence” vs. “observed prevalence”.

Figure 5

Showing a measure of variability between pairs of points at various distances. (A) A variogram displayed as a “vario cloud”. (B) Bin points by distance using 2nd order trend, depicting relationship between semi-variance and distance class. (C) Simplified variogram depicting relationship between semi-variance and bin points at distances using decimal degrees. (D) Variogram binning at 0.2 decimal degrees. (E) The blue line indicates spherical model fit, while red line indicates exponential model fit.

Figure 6

Showing the intensity of the point prevalence rate on xy coordinate plane. (A) Kriged predictions of HPAI H5N8 epidemic second phase in the coordinate plane indicating longitude on X-axis and latitude on Y-axis. (B) Prediction raster with colored scale from gray to yellow and green on the right panel indicating the intensity of the point prevalence rate. The green color shows the highest predicted point prevalence rate followed by yellow, gray, and then white for the lowest point prevalence rate.

Figure 7

Showing the point prevalence rate of HPAI H5N8 during 2nd phase, also depicting risk predictions based on kriging and IDW methods. (A) Known point prevalence rate during second phase of HPAI H5N8 epidemic. (B) Kriged prediction from the points where data were collected. (C) IDW predictions from the points of collected data. (D) Kriged prediction for the whole prediction grid window. (E) IDW prediction for the whole prediction grid window. (F) Comparison of known prevalence and kriged predicted prevalence with scale bar on the upper right corner. (G) The difference between using IDW versus kriging to interpolate our point prevalence data. Regions of greatest difference are indicated in green.

Figure 8

Maps showing domestic poultry density (number per square kilometer, colors in key) in ROK according to Gridded Livestock of the World 2.0 (Robinson et al., 2014). (a) Domestic chicken density. (b) Domestic duck density. (c) Map of provinces. Colors correspond to the branch color scheme used in Figure 3. Province abbreviations are as follows; CB: Chungbuk, CN: Chungnam, DG: Daegu, GB: Gyeongbuk, GG: Gyeonggi, GN: Gyeongnam, GW: Gangwon, IC: Incheon, JB: Jeonbuk, JJ: Jeju, JN: Jeonnam, US: Ulsan. (For interpretation of the references to colors in this figure legend, the reader is referred to the web version of this article.).