Port Moresby, Papua New Guinea ? In one New Guinea hilltop village the message was rooted deep in lore: If you hunt in the valley below and sleep there overnight, evil spirits will possess you, you’ll become sick, and you’ll die.

It was a homespun kind of malaria control in the highlands of this western Pacific island, long free of the disease-bearing mosquitoes that plague the hot and humid nights of its lowlands, said Dr. Ivo Mueller, a lead scientist at the Papua New Guinea Institute of Medical Research.

As the Earth warms, however, “malaria epidemics in the highlands are now basically happening every year,” Mueller said.

The threat of collapsing ice sheets and super-hurricanes dominates many discussions at the annual U.N. climate conference now under way in Bali, Indonesia. On the litany of ills linked to climate change, the slow spread of warm-weather diseases is more a quiet scourge, one whose ultimate cost remains incalculable.

“What is going to be the burden on the health care infrastructure of poor, developing countries?” asked Hannah Reid, of London’s International Institute for Environment and Development, opening a panel session Wednesday in Bali on the health impacts of climate change.

Forecasting those impacts can be controversial, both politically and scientifically.

In Washington in October, for example, The Associated Press reported that the Bush administration, which opposes mandatory international action to rein in warming, expurgated pages discussing such negative health effects from a U.S. official’s congressional testimony.

At the technical level, researchers in poorer nations like Papua New Guinea often cannot find the reliable health statistics – or, sometimes, historical temperature readings – they need to reach scientific conclusions.

“Not having quality health data that spans many decades makes the long-term assessment of climate change impact on health rather difficult,” Dr. Jonathan Patz, an international expert on health and climate, said in a telephone interview from his office at the University of Wisconsin.

Mueller’s team, based in the highlands town of Goroka, faces that problem.

“Whether this is already climate change – it’s difficult to say because we don’t have time-series data,” Mueller said. “There’s no reliable malaria data from the late 1970s to 2000. But we do know that in the last 20 years temperatures have risen 1 degree in the highlands.”

And they know they’re seeing more malaria at higher altitudes. One statistical glimpse: In 2005 the World Health Organization said reported malaria cases in Papua New Guinea’s Western Highlands province rose to 4,986 in 2003 from 638 in 2000 – considered minimum figures in view of reporting deficiencies.

Two out of five Papua New Guineans live in the lush, densely populated highlands of the equatorial country, most between the altitudes of 5,000 and 6,700 feet, “where there’s no malaria or low epidemic outbreaks,” Mueller said.

“There’s talk of a 2- or 3-degree temperature rise (3 to 5 degrees Fahrenheit) in the future,” he said. If so, “perhaps 2 million people would go from a low- or no-risk area to considerable risk.”

International health authorities say more than 1 million people, mostly African children, die each year of malaria, caused by a parasite transmitted by the bite of the female anopheles mosquito. Tens of millions more suffer chronically from the debilitating disease.

The parasite needs temperatures above 64 degrees Fahrenheit to develop. As for the “vector,” the mosquito, scientists have found that even small temperature increases can produce disproportionately large increases in mosquito populations.

The Intergovernmental Panel on Climate Change, a U.N. network of climate scientists, has long projected that mosquito-borne tropical diseases would spread to new areas that grew warmer. But in its latest reports, issued this year, the IPCC panel was cautious about more specific projections.