Warmer temperatures play big role in droughty tree die-offs

What does it take to kill a tree?

"That's a little bit of an embarrassing question," acknowledged David Bershears, an ecologist at the University of Arizona, during a chat in his office earlier this year.

Oh, sure, he knows about bulldozers, axes, fires, bugs, and diseases. But when it comes to understanding how plants respond to drought, "I don't think as a community we really know: What does it take to kill a tree?" he concedes.

Now, scientists are a lot closer to an answer. All other things being equal, temperature appears to play the dominant role, at least for piñon pines, according to a new study. It's authors, who include Dr. Breshears, estimate that as global warming tightens its grip in the US Southwest, large-scale loss of trees from drought is likely to occur five times more frequently than they have in the past century. You can find a description of the study here.

Why? The experiment conducted for this study indicates that in an extreme drought, a significant temperature increase -- in this case 4.3 degrees Celsius above current levels -- will kill off a tree far faster than a lack of water under cooler temps. (That temperature increase corresponds to atmospheric concentrations of carbon dioxide at just over twice preindustrial levels. If current emission trends continue, concentrations will overshoot that significantly.) So as temperatures warm, shorter droughts, which over the course of a century tend to occur more often than long ones, have the potential to trigger a level of damage long drought used to inflict.

And the five-fold rise in the frequency of large die-backs understates the effect -- for a couple of reasons.

"If you think about drought frequency, nobody expects it to stay the same as it was during the last 100 years," says Henry Adams during a phone chat. Mr. Adams is a graduate student Dr. Bershears is guiding toward his PhD in ecology and evolutionary biology. And he ran the experiment for the international research team and wrote up the results.

Nor do the results take into account other climate-related threats forests face. Bark beetles, for instance, are ravishing western forests. Global warming is a key driver, bug ecologists say, because the West is experiencing fewer winters with temperatures cold enough to kill off the larvae. More bugs emerge the following year to overpower the trees' natural defenses.

Following the trail of droughts 'n trees

For his part, Breshears traces his own work on the subject to the 1990s.

The challenge, he says, is that for a long time scientists have tried to see how various ecosystems might respond to long-term trends, such as rising CO2 levels or warming temperatures on average.

But organisms adapt to their environment not on the basis of how well they survive averages or gradual trends, but on the basis of how well they survive the climate extremes in their area.

Breshears says his work tends to focus these "threshold events." These don't just rock the ecological canoe. They can capsize it. They can force significant, fast changes in a region's status as habitat for wildlife, it water resources, its ability to act as a storage depot for atmospheric carbon that trees take up, as well as is ability to provide other so-called ecosystem services.

A decade ago, he and Craig Allen, a researcher with the US Geological Survey's field station in the Bandelier National Monument near Los Alamos, N.M., teamed up to explore the effects of a drought that struck northern New Mexico in the early 1950s. Up to that time, it was the most severe drought the area had experienced in 500 years. You can read more about some of the work Dr. Allen does here.

During that drought, the Ponderosa pine forests died at lower elevations, leaving the field largely to juniper and piñon woodlands. In effect, the boundary between the two moved upslope. The change occurred over five years. And it has persisted -- certainly long enough for the scientists to see its effects when they conducted their study 40 years later. One byproduct of the change: higher rates of soil erosion. The team published its results in the Proceedings of the National Academy of Sciences in 1998.

In the end, the change "was not a surprise," Breshears says. "But it hadn't been well documented."

That study illustrated how rapidly an ecosystem can shift -- for the long haul. Fast-forward to the drought of 2000-2003. This time, bark beetles had entered the picture. And instead of looking at one location, the team looked at the entire Four Corners area, in addition to the site they studied in their earlier work. Now they could document a sudden drought-driven piñon die-off covering some 4,600 square miles.

The kicker: In general the drought of 2000-2003 "was not necessarily drier than the 1950s drought," Breshears says. But it was measurably warmer. Yet it's hard to peg the tree die-off under drought conditions to higher temperatures. Those pesky bark beetles clouded the picture. The team published its results in 2005.

So, says Mr. Adams, it was time to take the notion of temperature as the dominant villain governing tree die-offs during a drought and see if it could be shifted from "it might be" or "it looks like it is" to "yes, it well could have been."

On to the Biosphere 2 lab

So, it was out onto a ranch to dig up 20 reproductively mature piñons of like stature and from similar ground conditions. They went into large nursery pots and thence to the controlled environment of the University of Arizona's Biosphere 2 lab, near Oracle,Ariz. Ten trees were kept at temperatures that tracked those the trees would, on average, experience in the wild. The other 10 were kept in temperatures that were an average of 4.3 degrees Celsius above the first group's. Once the trees adjusted to their new digs, the team stopped watering five trees in each group. No bark beetles, no wildfires, only differences in temperature for the "drought-stressed" trees.

Water-stressed trees in the "global warming" group lasted an average of 18 weeks before dying. Water-stressed trees in the "ambient" environment survived for 25.1 weeks.

Based on a suite of measurements the team took, carbon starvation killed the high-temp piñons. In essence, the tree closed the pores, or stomata, on its needles to conserve moisture it otherwise would have lost through evapotranspiration. The closed stomata prevented the plant from taking up CO2 for photosynthesis. This carbon starvation also undercuts a tree's ability to mount its natural defenses against pests. Ergo, dead trees.

The results appear in today's edition of the Proceedings of the National Academy of Sciences.