The March 2024 issue of IEEE Spectrum is here!

Close bar

Satellite-Based Early Warnings Help Contain Cholera Outbreaks

Better analysis of geophysical data predicts impending cholera epidemics in time to intervene

3 min read
Satellite-Based Early Warnings Help Contain Cholera Outbreaks

To most people living in the industrialized West, cholera is a relic of the 19th Century. But it is very much a 21st-Century problem, and researchers are using modern satellite imaging techniques to anticipate and short circuit new outbreaks.

The current worldwide cholera surge, the Seventh Pandemic, began in the 1960s and brought new cases to areas that hadn’t seen the disease in a hundred years. According to a 2012 World Health Organization report, it now kills between 100 000 and 120 000 of the 1.2 million to 4.3 million people who contract the disease each year. Almost all of these cases occur in tropical regions of South Asia, Africa, and South America.

London physician John Snow’s medical sleuthing helped end the London cholera epidemic of 1854 and lay the groundwork for modern scientific epidemiology. (For a highly readable account of the hunt the source of the London epidemic, which turned out to be a public water pump, see Steven Johnson’s The Ghost Map.) That was part of the Third Pandemic, which carried off millions. The Fifth Pandemic—the last major cholera outbreak to affect Europe—ended  in 1897 after claiming some 250 000 lives there and tens of thousands more in Asia and the Americas. In 1910-11, the last U.S. outbreak killed 11 people in New York City.

Since then, though, two more pandemics have claimed hundreds of thousands of lives in some of the world’s most heavily populated and poorest regions.

In Remote Sensing Letters, lead author Antarpreet Jutla (now at West Virginia University), team leader  Shafiqul Islam (Tufts University), and collaborators from the University of Maryland and the International Center for Diarrhoeal Disease Research in Dhaka, Bangladesh, describe refined satellite measurement techniques that greatly improve public health workers’ ability to forecast outbreaks of the disease in time to intervene and reduce their severity.  

Researchers doubt that an organism as widespread as V. cholerae can be eliminated, but they say the spread of the disease can be inhibited before big seasonal outbreaks with a bit of warning…and that’s what improved satellite and weather monitoring can provide.

Plankton Early Warning. For years, researchers have used satellite-borne optical sensors to detect the green chlorophyll signatures of phytoplankton in the Bay of Bengal and other inshore waters. The flourishing plants are a good initial indication that V. cholerae may be breeding off-shore, but the correlation is not exact. To build a better early warning system, the Islam group developed a “Satellite Water Marker” (SWM)—a fairly simple index of the contrast between the reflected blue signature (at a wavelength of 412 nanometers) of clear water and the green, 555-nm signature of turbid, plankton-rich water. If Blue and Green are the intensities of the respective signals picked up by the satellite, then:

SWM = [(Green-Blue) / (Green + Blue)] x 100

SWM can range from +100 (very turbid, plankton-rich, and presumably teeming with V. cholerae) to -100 (open water, clear epidemiological sailing).

Applied to the NASA/Goddard “Sea-viewing Wide Field of view Sensor” (SeaWiFS) data set—which includes monthly radiance readings at 412 and 555 nm, as well as at 443-, 490-, and 510 nm)—the composite index improves markedly on the predictions based on chlorophyll readings alone. Comparisons of SWM-based models and actual spring cholera data for 1998-2009 showed models could predict the actual cholera incidence rates within a few percent, and accurately forecast low-incidence years (like 2002) and major outbreaks (like 2005). As an additional test, the researchers analyzed data for the winter cholera season in Mozambique, which is driven by similar environmental factors. Here, the SWM-based models forecast the number of cholera cases to within a few hundred cases out of about 5500 in the peak year of 1998, and predicted the 2008 total of just under 1000 to within 100 or so.

As the authors readily admit, these space-borne solutions are not final, but they are a big step in the right direction: increasing lead time to strengthen interventions before an outbreak can start.

For a good overview of the interplay between environmental factors and epidemics of human disease, see the “Oceans, Climate, and Global Infectious Disease: The Cholera Paradigm,” Rita R. Colwell’s October 2012 IEEE Committee on Earth Observation “Blue Marvel-Ocean Mysteries” webinar. Colwell, a co-author of the Remote Sensing Letters paper, is perhaps the world’s leading investigator of Vibrio cholerae, the cholera bacterium.

The Conversation (0)