The researcher's findings about the impact of wildfires raise questions about whether fire suppression or controlled periodic burning are the best strategies in areas being gradually inundated by rising seas resulting from global warming, said Benjamin Poulter. He is a research scientist at Duke's Nicholas School of the Environment and Earth Sciences as well as a visiting lecturer at the University of North Carolina at Chapel Hill.
Poulter will describe his findings during a 5:45 p.m. talk on Thursday, Oct. 20, 2005, during an Estuarine Research Federation conference at the Marriott Waterside Hotel and Convention Center in Norfolk, Va. His research was mostly funded by NASA.
In analyses for his doctoral degree, Poulter mapped how plant life is responding to one of the world's highest rates of sea-level rise -- about 30 to 40 centimeters per century -- along the western shoreline of two interconnected North Carolina estuaries, the Pamlico and Albemarle Sounds.
As a starting point, Poulter took advantage of new higher-resolution digitized data that he said enable the first sensitive analysis of elevations lower than 1 meter above sea level. Ocean heights are predicted to rise by 1 meter by the year 2100.
Overlaying those maps with aerial photographs made in the years 1932, 1969 and 1998, Poulter could then assess visually how far marshlands had penetrated into forested areas at various points within this critical inundation zone over a 66-year period.
"From looking at these photographs we found that there are some areas that haven't changed at all," Poulter said in an interview. "There were other areas where there have been gradual changes of just several meters. But there were also areas where large expanses of forest have been converted into marsh."
Scientists initially hypothesized that since the sea level is rising gradually all changes would also be gradual. But the photographs showed that, in places, "the forest is retreating in large patches," he said. "Those retreats have been about 200 meters over 70 years."
Visiting some of those sites of large retreat, Poulter noticed dead trees with fire scars amid the invading marsh grasses. That led him to consult U.S. Fish and Wildlife Service fire history records for the area, which since 1970 have identified "ignition points" where each wildfire originated.
When he digitally superimposed those ignition points on the digitized landscape he found a correlation. "It appeared that the more frequent the fires were the more likely the forest was to retreat in this patchy way," he said.
At a site where a forest of loblolly and pond pines had been burned in May 2004, he found only marsh grasses growing and no signs of regenerating pine seedlings. He hypothesized that, in the absence of trees to release water vapor into the air, the water table had moved upward.
Attempting to explain the lack of forest regeneration, Poulter initially hypothesized that when marsh grass species migrate in following a fire, plant competition and the higher water table prevent the pines from returning.
But after further experiments showed that pine seedlings were able to grow prolifically under the marsh plants at high water tables, Poulter developed a new hypothesis.
He suggests that after a fire both marsh grass and pine seedling can coexist and proliferate in burned areas. Then, if land managers prevent new wildfires for longer than 10 to 15 years, the forest may re-establish itself. "In some of those areas where fire has been suppressed, you do see forests moving back out into the marsh," he said.
On the other hand, if a new fire occurs over a shorter interval, "you would have to start the process over," he said. And the pine seedlings could lose out in the second round of competition.
"I think this is a great example of how there is some kind of built-in tolerance to environmental change that gets disrupted when a disturbance such as fire passes through the system," he said.
"The changes that you end up having are unpredictable, punctuated changes rather than gradual change."
According to Poulter, land managers are rethinking their old extended fire suppression policies in his study area, a pristine region of protected land especially rich in plant and animal species.
Plans call for restoring the environment to more natural and frequent cycles of periodic burning and regeneration. "But under certain situations there is the potential to increase the rates of forest retreat in this region," he said.
Poulter noted that both marsh grasses and pine forests actually depend on fire for periodic renewal. "These are such highly fire-dependent systems that fire suppression as a strategy to reduce marsh invasion does not make much sense," he said.
"The reintroduction of controlled burning to minimize marsh plant invasion of forests should be done under optimal weather conditions and during the appropriate season of the year," he said.