Use of a remote sensing-based geographic information system in the characterizing spatial patterns for Anopheles minimus A and C breeding habitats in western Thailand.

A remote sensing (RS)-based Geographic Information System (GIS) was used to characterize the breeding habitats of Anopheles minimus species A and C in five different districts of Kanchanaburi Province in western Thailand. The GIS and RS were used to monitor the area for the presence and absence of An. minimus A and C in five major land areas, forest, agriculture, urban, water and bare land. The results show that An. minimus A survives both in dense canopy forest and in open fields where agriculture is dominant. A scatter plot of land-use/land-cover for An. minimus, considering proximities to the forest and proximities to agriculture, suggests that An. minimus A has a wider habitat preference, ranging from dense canopy forest to open agricultural fields. A scatter plot for An. minimus C, on the other hand, showed a narrow habitat preference. A scatter plot for proximities performed on separate populations of An. minimus species A, one in the north and the other in the south, showed that there was an association in the northern population with the forest and in the southern population with agricultural areas. There were no statistically significant differences in the scatter plot of proximities to urban areas and water bodies with the An. minimus A north, south, and An. minimus C. LANDSAT TM satellite data classification was used to identify larval habitats that produce An. minimus A and C and analyze proximities between land-use/land-cover classes and locations of larval habitats. An. minimus A has a wide habitat preference, from dense canopy forest to open agricultural fields, while An. minimus C has a narrow habitat preference.

Ikonos-derived malaria transmission risk in northwestern Thailand.

We mapped overall malaria cases and located each field observed major malaria vector breeding habitat using Global Positioning System (GPS) instruments from September 2000 to October 2003 around the three malaria-endemic villages of Ban Khun Huay, Ban Pa Dae, and Ban Tham Seau, Mae Sod district, Tak Province, Thailand. The land-use/land-cover classifications of the three villages and surrounding areas were performed on IKONOS satellite images acquired on 12 November 2001 with a spatial resolution of 1 x 1 m. Stream network was delineated and displayed. Proximity analysis was performed on the locations of the houses with and without malaria cases within a 1.5 km buffer from An. minimus immature mosquito breeding habitats, mainly stream margins. The 1.5 km used in our proximity analysis was arbitrarily estimated based on the An. minimus flight range. A statistical t-test at 5% significance level was performed to evaluate whether houses with malaria cases have higher proximities to streams than houses without malaria cases. The result shows no significant difference between proximity to streams between houses with malaria cases and houses without malaria cases. We suspect that the actual flight range of An. minimus may be greater than 1.5 km. The An. minimus larval habitat deserves more detailed investigation. Further studies on human behavior contrary to that required for adequate malaria control among these three villages are also recommended.