The relationship between mosquito abundance and rice field density in the Republic of Korea

• Main Authors: Klein Terry A, Smith Matthew, Brett-Major David, Masuoka Penny, Richards Erin E, Kim Heung, Anyamba Assaf, Grieco John
• Format: Article
• Language: English
• Published: BMC 2010-06-01
• Series: International Journal of Health Geographics
• Online Access: http://www.ij-healthgeographics.com/content/9/1/32

<p>Abstract</p> <p>Background</p> <p>Japanese encephalitis virus (JEV), the causative agent of Japanese encephalitis (JE), is endemic to the Republic of Korea (ROK) where unvaccinated United States (U.S.) military Service members, civilians and family members are stationed. The primary vector of the JEV in the ROK is <it>Culex tritaeniorhynchus</it>. The ecological relationship between <it>Culex </it>spp. and rice fields has been studied extensively; rice fields have been shown to increase the prevalence of <it>Cx. tritaeniorhynchus</it>. This research was conducted to determine if the quantification of rice field land cover surrounding U.S. military installations in the ROK should be used as a parameter in a larger risk model that predicts the abundance of <it>Cx. tritaeniorhynchus </it>populations.</p> <p>Mosquito data from the U.S. Forces Korea (USFK) mosquito surveillance program were used in this project. The average number of female <it>Cx. tritaeniorhynchus </it>collected per trap night for the months of August and September, 2002-2008, was calculated. Rice fields were manually digitized inside 1.5 km buffer zones surrounding U.S. military installations on high-resolution satellite images, and the proportion of rice fields was calculated for each buffer zone.</p> <p>Results</p> <p>Mosquito data collected from seventeen sample sites were analyzed for an association with the proportion of rice field land cover. Results demonstrated that the linear relationship between the proportion of rice fields and mosquito abundance was statistically significant (R²= 0.62, r = .79, F = 22.72, p < 0.001).</p> <p>Conclusions</p> <p>The analysis presented shows a statistically significant linear relationship between the two parameters, proportion of rice field land cover and log₁₀of the average number of <it>Cx. tritaeniorhynchus </it>collected per trap night. The findings confirm that agricultural land cover should be included in future studies to develop JE risk prediction models for non-indigenous personnel living at military installations in the ROK.</p>