The future of our forests
I recently got an email from a reader who was considering moving to Flagstaff. With its excellent bike trails, university, and a populace full of outdoor nuts, it sounded like a pretty nice spot. So he paid a visit, and while there, sought out the answer to a big question: “I needed to know when 100 percent mortality of the forests around Flag would occur,” he wrote to me. “Why would I want to move there if all the trees were going to die?”
He met someone who turned him on to the research of a guy named Park Williams, who works out of Los Alamos National Lab. Williams seemed to have an answer: 2050. "I decided not to move to Flagstaff," the reader wrote. "Too hot in the summer and surrounded by desert.”
Here at HCN, we’ve been keeping an eye on recent research about the climate change-induced decline of Southwestern forests. This reader’s email was a good nudge to give Park Williams a call. I chatted with him this week about a paper he published this fall in Nature Climate Change. While the paper doesn’t go so far as to predict the complete demise of Southwestern forests by 2050, it offers strong evidence that trees will be under extreme stress due to rising temperatures by then, making them much more vulnerable to death.
Here are some excerpts from our conversation:
High Country News: What was the major question you set out to answer?
Park Williams: The major question was really: As the globe warms, how should we expect forests in drought-limited places to respond – places where drought has meant trouble for forests? The reason that question was important is because it was unclear how important warming itself is to forests. It doesn’t necessarily affect how much water falls out of the sky, and the previous thought on drought-(troubled) forests was that they’re mainly sensitive to (precipitation).
We wanted to ask the trees directly and see what they said about the importance of warmth.
HCN: What did you find out?
PW: We did a tree ring study where we used tree ring data from hundreds of locations throughout the Southwest. The first interesting thing we saw was that all the trees in our data set were telling the same story. They all agreed on the years of good growth, and which years were years of bad growth. All the trees are dancing to the same tune, and that tune must be regional climate variability.
We created one regional tree ring record, and were able to compare the most recent 100 years to observed climate records. This is where we allowed the trees to tell us how important temperature is relative to the amount of water falling out of the sky. What we found was temperature during the warmest part of the year is at least as important as the amount of water that falls out of the sky.
HCN: How so?
PW: When we think of the effect of warmth on forests, we're really thinking about ... the sponginess of the atmosphere. It has more potential to suck water out of soil and plants the warmer it is.
Trees are scared of dying. When the air gets very dry, they close up their stomata – the holes in their leaves, which are like their mouths. While bringing in CO2, they accidentally let water vapor escape. They know they make that mistake. So when the air gets very dry, they close up their stomata and stop growing.
When you have very dry air, changes in (the sponginess of the atmosphere) is almost entirely driven by temperature. That makes ecosystems in these dry regions – forests on the border of desert – it makes those uniquely vulnerable to warming.
HCN: For a long time, people have said global climate models are too coarse to provide meaningful information to land managers about how warming will impact local or regional landscapes. But it seems like you’ve come up with such information.
PW: Using global climate models to derive information about specific regions can be dangerous -- especially when dealing with variables that are not temperature. Precipitation, for instance, is very unreliable in many parts of the globe.
One thing you can do to see how much confidence you should have in global projections is to take all the models, and say you’re interested in future rainfall in the Southwest, then ask each of the models what they say about rainfall in the Southwest. If they all agree, then you can have a little more confidence. In the Southwest, almost all agree that precipitation in the cold season should decrease slightly. But there’s still a lot of uncertainty.
The reason that doesn’t matter so much for our study is that the models all agree so strongly on increased temperatures in the Southwest. And in the Southwest, because (the atmosphere's sponginess, or) its "vapor pressure deficit" (VPD)-- (the difference between the moisture that's in the air and the total amount that could be in the air) -- is so strongly dictated by temperature, we have a lot of confidence that VPD is going to increase a lot. VPD is so important that even if the models have over-predicted our rate of warming to a factor of two, drought stress will still reach megadrought-type levels by the end of the century.
HCN: So it doesn’t take much?
PW: No. This is assuming a business as usual emissions trajectory. But we redid all these projections considering alternate scenarios. I considered a scenario where we begin curbing emissions significantly yesterday – a path we are definitely not on. What is interesting but also really grim about that analysis is even in that most optimistic scenario, we’re still looking at megadrought conditions by the 2070s. So it’s not really a question of will forests start experiencing these conditions this century, it’s when?
HCN: What are the implications of this for land management in the Southwest?
PW: In order to grow a tree ring, you need to be alive and growing. But is anybody really going to notice if forests just start growing slower? People are more concerned with death.
So we looked at death. Two popular ways to die in the Southwest if you’re a tree are by bark beetles and wildfires. One of the reasons we chose to focus on the Southwest in this study is that it has such a rich range of data sets on forests. We have great wildfire records, and since 1997, we’ve had decent data on bark beetle outbreaks.
We found an exceptionally strong relationship between (the drought stress on trees), and the areas of forest that were killed by wildfire and bark beetles in the past 20 to 30 years. Using climate variables to estimate drought stress in 2012, you can estimate how many acres burned almost exactly.
What’s important about the relationship between (drought stress on forests) and the area of forest killed is that the relationship is exponential. So a small increase in drought stress tends to lead to a very large increase in the vulnerability to death. Even if we think of a couple degrees of warming as relatively minor, forests notice a couple of degrees and they express it by dying.
When large fires get going, it seems like agencies are more likely to let the fires burn. It's recognized now that forests are artificially dense, and it's really increasing the probability of having monster wildfires. So in that way, they've been making decisions to try to bring the forests to more reasonable densities.
HCN: Is that likely to do any good?
PW: When I was doing this work, I was thinking, “Gee what’s the use of spending all this money trying to preserve our forests if they’re just going to go in flames no matter what?” But I don’t think there’s any reason to believe we should lose all of our forests in the next century. There are some parts of the Southwest that receive enough precipitation and are cool enough now, that even with a few degrees warming, the climate should still be suitable for forests.
If we decide as a society that we really like forests and want to have them around, I do believe it’s going to take some work to ensure that forests can still be in places where the climate is suitable. Maybe the correct species aren’t there right now. Maybe one of the things we want to consider is planting them in the climates that may be suitable.
Even in climates that are suitable to forests, the forests are still vulnerable to large wildfires. It may be worth thinking about what we can do to protect these forests. Maybe that means thinning in forests that we think we have a chance of preserving. Or building fire breaks to protect those forests. A strategic approach there will be good.
Of course, these ideas cost a lot of money to put into practice. But it costs a lot of money to fight huge wildfires as well. The current approach is to wait to spend most of our resources once a disaster is in progress. I think it would be worth looking into the idea of spending those resources beforehand.
Cally Carswell is HCN's assistant editor.
Photo: The 2012 Hewlett Gulch Fire near Fort Collins, Colorado. Courtesy USDA.