Soil Moisture Data Used to Predict Severity of Malaria Outbreaks
Embargoed for Release: Monday, October 12, 1998
Contact: Lisbeth Pettengill (410)-955-6878 or [email protected]
A large multidisciplinary study has shown that the amount of water held in the soil of a region, along with such factors as the local vegetation and soil type, can more accurately predict the incidence of malaria outbreaks than more conventional variables such as temperature and rainfall. The study appeared in the October 1998 issue of Tropical Medicine and International Health.
Since mosquitoes breed in water, rainfall data have long been used to predict the seasonality of malaria. However, the timing and severity of malaria outbreaks, which in 1995 killed over two million people worldwide, remain difficult to predict. The soil-moisture model may be especially useful under the more extreme weather conditions that climatologists predict will accompany long-term global warming and climate change.
Lead author Jonathan Patz, MD, MPH, assistant scientist, Environmental Health Sciences, the Johns Hopkins School of Public Health, said, "Even if steep rocky terrain and flat silty areas receive identical amounts of rain, for example, the effects of the water on those two spots will differ greatly. Soil-moisture models can reflect those differences, and can also account for changes in land use, such as draining of swamps or water development projects--changes that could substantially influence mosquito ecology and subsequent malaria transmission."
The researchers analyzed data from Kenya of the biting rates and infectiousness of two local species of mosquitoes, Anopheles gambiae and Anopheles funestus, and combined this information with data on local temperature, precipitation, relative humidity, and "evapotranspiration" (a function of temperature, wind speed, humidity, and solar radiation). The researchers then used information on vegetation and soil features to construct a picture of surface-water availability that allowed them not only to estimate weekly levels of soil moisture and river runoff, but also to predict Anopheles biting rates and the proportion of female mosquitoes carrying the malaria parasite. These results from the soil-moisture model were then compared with results derived from both raw weather data and from satellite pictures of vegetation.
The soil-moisture model predicted biting rates much more accurately than did the raw rainfall data. Moreover, when the soil-moisture model's data were lagged by two weeks, accuracy was further improved, allowing the scientists to explain up to 45 percent of An. gambiae's biting variability, compared to just eight percent when raw rainfall data were used. The soil-moisture model accounted for up to 56 percent of the variation in An. gambiae inoculation rates and up to 32 percent for An. funestus.
Although satellite images yielded predictions of mosquito biting rates that were nearly as accurate as those from the researchers' model, the authors noted that soil-moisture modeling had several advantages over satellite technology. Soil-moisture estimates could be calculated weekly or even daily, whereas the predictions derived from satellite data were only robust over periods of a month or more. The soil-moisture model was also relatively inexpensive and, unlike the satellite method whose data gathering can be hampered by clouds, soil moisture could be used even during the rainy season, which is the most critical period for assessing surface water. Finally, the model may be more practical for use by resident public health scientists who have access to local streamflow and weather data.
-end-
This study was funded in part by the Climate and Policy Assessment Division, Office of Policy, of the U.S. Environmental Protection Agency, and the National Institutes of Health.
RSS