People who live in the Amazon basin are not likely to forget the great storm of January, 2005. Over the course of two days, a squall line measuring 620 miles (1,000 km) long and 124 miles (200 km) wide raged across the region from southwest to northeast, with buzzsaw-like winds of 90 mph (146 km/hr) causing widespread damage to property and a handful of deaths in the cities of Manaus, Manacaparu and Santarem.
The loss of human lives was, of course, the most tragic part of the historic storm. But a new study published in Geophysical Research Letters shows that our species wasn’t the only one that suffered—and that in terms of the actual number of deaths, we got off comparatively easy. During those same two days, a breathtaking half a billion trees might have been killed, taking a deep gouge out of a region that’s already been burned and clearcut to within an inch of its life.
The storm of 2005 was unusual for reasons beyond its mere severity. It also occurred during a severe drought—the worst in 103 years, as measured by the depth of the Rio Negro River. Dead fish and mud-mired boats were scattered across the countryside while vultures circled overhead. No surprise that trees adapted to an extremely wet region died by the hundreds of thousands when the rain quit falling. And while a complete botanical body count was never taken, it was the drought that was always considered responsible for most of the deaths. But the new study says no.
Researchers from both Tulane University and NASA conducted the work, relying on a combination of Landsat satellite photos, measurements in the field and computer modeling. Satellite images are always a good place to start for this kind of survey, but if you don’t conduct your studies fast, the pictures are no good to you. That’s because when trees are felled by a storm, the hole that’s left in the jungle canopy gapes for only a year, before rapid growth of ground cover and spreading of existing trees close it back up again. For that reason, the old Landsat photos had to be supplemented by new on-foot field surveys. Not only did that allow researchers to take a census of all the trees that died in that period, but also to help them distinguish the ones that had been felled by the storm from the ones that had died standing up—which is what happens when a drought is responsible.
Finally, since it was not practical to survey the entire 200,000 sq. mi. area that was affected by the storm, the investigators limited their survey to the area around Manaus, where they estimated that 300,000 t0 500,000 trees had died storm-related deaths. Expanded over the entire affected area, that factored out to 441 to 663 million trees.
Jeffrey Chambers, Tulane forest ecologist and the study’s lead author, admits that his methods were imperfect. “To quantify the potential basin-wide impact,” he said in a statement, “we assumed that the whole area impacted by the storm had a similar level of tree mortality as the mortality suffered in Manaus.” Nature is never that tidy of course, which is why Chambers and his collaborators hedged so much, with a range of 222 million trees between their high and low figures.
No matter the exact number, the deaths were staggering—and it’s more important than ever that we learn how such devastation can happen. “Under a changing climate, some forecasts say storms will increase in intensity,” say Chambers. “When we collect data in the field, we [must thus] do forensics on tree mortality.” As with other kinds of forensic science, battling the killer requires identifying it first.