Landscape determinants and remote sensing of anopheline mosquito larval habitats in the western Kenya highlands

• Published in: Malaria Journal
• Main Authors: Beck Louisa, Schmidt Cindy, Kitron Uriel, Bian Ling, Feng Chen-chieh, Li Li, Minakawa Noboru, Munga Stephen, Mushinzimana Emmanuel, Zhou Guofa, Githeko Andrew K, Yan Guiyun
• Format: Article in Journal/Newspaper
• Language: English
• Published: BMC 2006
• Subjects: Arctic medicine. Tropical medicine; RC955-962; Infectious and parasitic diseases; RC109-216; Arctic
• Online Access: https://doi.org/10.1186/1475-2875-5-13; https://doaj.org/article/87a580807a404fb1b268233bc8904fec

Abstract Background In the past two decades the east African highlands have experienced several major malaria epidemics. Currently there is a renewed interest in exploring the possibility of anopheline larval control through environmental management or larvicide as an additional means of reducing malaria transmission in Africa. This study examined the landscape determinants of anopheline mosquito larval habitats and usefulness of remote sensing in identifying these habitats in western Kenya highlands. Methods Panchromatic aerial photos, Ikonos and Landsat Thematic Mapper 7 satellite images were acquired for a study area in Kakamega, western Kenya. Supervised classification of land-use and land-cover and visual identification of aquatic habitats were conducted. Ground survey of all aquatic habitats was conducted in the dry and rainy seasons in 2003. All habitats positive for anopheline larvae were identified. The retrieved data from the remote sensors were compared to the ground results on aquatic habitats and land-use. The probability of finding aquatic habitats and habitats with Anopheles larvae were modelled based on the digital elevation model and land-use types. Results The misclassification rate of land-cover types was 10.8% based on Ikonos imagery, 22.6% for panchromatic aerial photos and 39.2% for Landsat TM 7 imagery. The Ikonos image identified 40.6% of aquatic habitats, aerial photos identified 10.6%, and Landsate TM 7 image identified 0%. Computer models based on topographic features and land-cover information obtained from the Ikonos image yielded a misclassification rate of 20.3–22.7% for aquatic habitats, and 18.1–25.1% for anopheline-positive larval habitats. Conclusion One-metre spatial resolution Ikonos images combined with computer modelling based on topographic land-cover features are useful tools for identification of anopheline larval habitats, and they can be used to assist to malaria vector control in western Kenya highlands.