Satellite imaging is being used to locate potential mosquito breeding sites in southern Zambia, in a bid to reduce malaria transmission in the area.
Researchers use the data, containing information such as soil moisture and water drainage patterns, to identify areas where the mosquitoes live and breed.
They can use the data to gain a more accurate picture of 'at risk' areas, and interventions such as bednets and insecticides can be targeted more effectively, says lead researcher Gregory E. Glass, director of the Environmental Surveillance Core at the Johns Hopkins Malaria Research Institute, United States.
Locating breeding sites is particularly difficult in Sub-Saharan Africa, where multiple malaria vectors breed in specific areas.
"In Macha [Zambia] we see some villages where most households are infected with malaria, but six miles away, no one is," says Glass. "Not all people, even in endemic areas, are at equal risk of acquiring infection."
Mosquitoes in the region breed "exclusively in very specific habitats that represent a small fraction of the landscape", said Glass. "These aspects of the environment can be measured with satellite imagery."
The researchers are trying to predict where breeding sites should be, rather than merely using geographical information systems to map where they have found breeding sites previously. "Both approaches are important," he said.
As well as satellite data, the researchers use environmental and hydrologic models, as well as field surveys, to determine mosquito breeding sites.
Glass said the method can be used in other areas and is being expanded to other parts of southern Africa.
Jonathan Cox, a senior lecturer at the London School of Hygiene and Tropical Medicine, told SciDev.Net: "This type of research is a useful way to get a better understanding of the factors driving spatial variations in malaria in a particular setting."
"However, they only become useful tools for malaria control if their predictive skill is rigorously validated in a variety of settings and they are accessible to malaria control programmes and district-level teams," said Cox.
"Strengthening routine malaria surveillance activities and introducing basic tools to analyse and map case data would be a more appropriate and sustainable approach to mapping variations in malaria cases," he added.
The research was presented this month (12 November) at a web summit highlighting the institute's latest research.