Wednesday, March 17, 2010

Forest Die-Offs

Climate Progress alerts me to a new paper out which reviews the literature on forest die-off associated with drought and/or heat stress.  The paper is Allen et al, A global overview of drought and heat-induced tree mortality reveals emerging climate change risks for forests. The paper is very well worth reading, though unfortunately it's rather expensive at $31.50 (unless you have access to a university library of course).

The abstract says:
Greenhouse gas emissions have significantly altered global climate, and will continue to do so in the future. Increases in the frequency, duration, and/or severity of drought and heat stress associated with climate change could fundamentally alter the composition, structure, and biogeography of forests in many regions. Of particular concern are potential increases in tree mortality associated with climate-induced physiological stress and interactions with other climate-mediated processes such as insect outbreaks and wildfire. Despite this risk, existing projections of tree mortality are based on models that lack functionally realistic mortality mechanisms, and there has been no attempt to track observations of climate-driven tree mortality globally. Here we present the first global assessment of recent tree mortality attributed to drought and heat stress. Although episodic mortality occurs in the absence of climate change, studies compiled here suggest that at least some of the world's forested ecosystems already may be responding to climate change and raise concern that forests may become increasingly vulnerable to higher background tree mortality rates and die-off in response to future warming and drought, even in environments that are not normally considered water-limited. This further suggests risks to ecosystem services, including the loss of sequestered forest carbon and associated atmospheric feedbacks. Our review also identifies key information gaps and scientific uncertainties that currently hinder our ability to predict tree mortality in response to climate change and emphasizes the need for a globally coordinated observation system. Overall, our review reveals the potential for amplified tree mortality due to drought and heat in forests worldwide.
The methodology of the paper is basically to do a literature search on various combinations of the search terms "tree," "forest," "mortality," "die-off," "dieback," "decline," and "drought". They assessed the resulting literature based on whether a) there was documentation of the extensiveness of the die-off, and b) there was clear attribution of the problem to some combination of higher temperatures or less precipitation. They then present a big table of all the resulting incidents they found around the globe, along with maps.

The number of papers meeting their criteria has been increasing over time:

This is their Figure 9, which shows the number of search results for "forest AND mortality AND drought", divided by the number of results for "forest".

Of course, this cannot distinguish between "forest die-offs because of drought are increasing", and "forestry scientists are becoming more interested in drought induced die-offs". Intuitively, to me, it seems most likely that there is some of both going on.  But that's only a guess - their data cannot distinguish (and the paper is careful not to say that they have documented a trend of increasing forest die-offs.

The most striking thing is the spatial pattern of incidents. In North America, the bulk of incidents are in the Western US, especially the interior:

which of course is generally much drier and more mountainous than the Eastern half of the country and the immediate coastal environment.

And in Europe, they are mainly in the drier Mediterranean region and the Alps:

(There are maps of all the other continents in the paper, but I think I'd be exceeding my fair-use limit to post them all and you'll have to go pay to look at them, as well as all the other interesting figures and tables).

If there were a generalized increased in forest die-offs associated with rising global temperatures, we might expect it to hit first and hardest in places that are already closer to the edge of viability for trees due to not very much water. So the geographical pattern is consistent with that idea.  Of course, one might argue that, since the paper was specifically looking for "drought" related die-offs, rather than all-causes tree mortality, even if the problem was due to natural variability it might be expected to show up in this paper mainly in more arid places.

In western North America, in particular, the problems are mainly disease/insect mediated, and there is now a fair amount of attributional literature suggesting they are climate change related:

The Bentz reference is a short book, Bark Beetle Outbreaks in Western North America: Causes and Consequences, that you can order from Amazon, so I did.

It's tough to know how to talk about this stuff.  On the one hand, it's impossible to read this paper without the obvious apocalyptic potential on one's mind - if half a degree Celsius of warming can cause a fairly substantial beginning to broad-scale die-off of forests, at least in drier regions, then what will a few degrees do?  In particular, the potential for nasty positive feedbacks is obvious as the carbon in dead and rotting trees is released into the air.  There are some references in the paper suggesting this is happening regionally in some of the worst affected regions of bark-beetle infestation, with the forest going from a net carbon sink to a net carbon source.  However, that is not a globally significant phenomenon yet (the fraction of carbon emissions taken up by the land does not have a significant trend).

On the other hand, there's no question, as an issue of scientific logic, that the case is unproven on a global scale by this paper (and the authors freely admit that).  They just don't have a time series of incidents that doesn't have obvious biasses and so would allow a firm conclusion to be drawn.  Apparently no such thing exists at present.  So really it's a call for forestry scientists to study the issue as intensively as possible (while the rest of us wait to discover whether we are already screwed or not).


kjmclark said...

Two notes. First, I don't see how the bark beetle problems could be anything other than climate change. That's an indigenous beetle, suddenly exploding in numbers, in part due to trees that are weakened from recent climate conditions. I suppose one could argue that the climate fluctuations are part of some natural cycle, but why haven't we seen that kind of die off before?

Related to that, though these trees are dying off, it's possible that the forest could recover, though maybe with different trees. Depending on the conditions, it isn't uncommon for deciduous trees to follow conifers in succession.

Second note is that the '2' over eastern Michigan is probably due to emerald ash borer, which is certainly an insect-related die off of a large percentage of our forest trees here, but isn't really related to climate change in any way. That's an insect introduced from asia that happens to have stumbled on a very popular urban tree population. The ashes would still have been viable here for several decades at least.

There's a nice Forest Service tool for the eastern forests that allows you to look at individual tree species (and apparently some bird species) viability in the future (2100) as projected by three climate models. The page is here: The models are the Hadley, PCM, and GFDL.

My favorite example (though not my favorite situation) is sugar maple. Currently it's viable far south of Michigan, all the way down to Kentucky and Tennessee. The average of the 'high' scenario for the three climate models shows sugar maple mostly unviable south of northern Michigan, Wisconsin, and Minnesota. Pennsylvania loses most of its sugar maples, and they become a less common tree in New York, Vermont, and the other northern new england states. The average of the low scenarios isn't much better.

It really makes you wonder what will get people to see that we have a problem.

KLR said...

I suppose one could argue that the climate fluctuations are part of some natural cycle, but why haven't we seen that kind of die off before?

Oh, no doubt migrations of insects like this have happened in the past. After all, there are about 220 genera of Bark beetles. Don't know how fond God is of them as regards other species of beetle, but that's still a lot. Records of succession of trees post-LGM show how ecosystems can evolve rather rapidly.

I don't doubt the AGW factor in the current spate of hardiness zones moving around, but I wonder about the methodology in this study. Perhaps it's been observed on lower levels in the past but authors used terms other than "mortality," "die-off," "dieback," "decline," and "drought". This might skew their results, not that I disagree with the overall finding.

It's fun to trawl through the Google News Archive to search for terms like, oh, "peak oil." You get all kinds of curious results, including plenty of hits predating Colin Campbell's coining of the term, all of them unintended of course. It's a quick and dirty way of conducting a similar analysis as this one. Also you find how old everything new is. One piece from 1965 promised global oil RFs of 65% within 10 years from tertiary recovery.

Stuart Staniford said...

Yeah, I think the thing that can't be ruled out is this happened just as bad back in 17XX, but the only record of it is in some obscure Spanish priest's diary buried in a museum somewhere. Probably not, but their methodology is fundamentally limited (as they are very quick to acknowledge). But it's certainly points to the urgency of doing more research.

One thing I wondered about is the existence of natural clearings in forests. I wonder if those are evidence of past epidemics, or they have some other cause.

porsena said...

The mountain pine beetle (MPB) outbreak in British Columbia, Washington and Alberta is an outstanding example of forest change attributable to a warming climate.

The back story is that MPB outbreaks have occurred in British Columbia for hundreds of years, but forest analysis indicates nothing as severe as the present one. Past outbreaks were generally limited to small areas and were typically brought under control naturally by winter-kill of the larvae, which are sensitive to persistent low temperature early or late in the season.

Warmer winters in the last two decades, insufficient to kill the beetle larvae, have contributed to a sustained outbreak. BC Forestry predicts the present outbreak will kill 80% of the merchantable pine in the Province’s central and southern interior by 2013. Huge areas in the centre of B.C. have close to 100% kill of lodgepole pine. About 2002, the MPB crossed the Rockies into Alberta, into forests where the pest had never been recorded before. The Alberta government says that, left unmanaged, MPB could devastate Alberta's pine forests and spread eastward across Canada's boreal region.

The MPB outbreak is changing huge swaths of forest from a small carbon sink to a large carbon source. Here’s the abstract (minus the references) of a 2008 paper in Nature by Kurz et al, Mountain pine beetle and forest carbon feedback to climate change. Again, access to a library is helpful if you want to read the paper.

The mountain pine beetle (Dendroctonus ponderosae Hopkins, Coleoptera: Curculionidae, Scolytinae) is a native insect of the pine forests of western North America, and its populations periodically erupt into large-scale outbreaks. During outbreaks, the resulting widespread tree mortality reduces forest carbon uptake and increases future emissions from the decay of killed trees. The impacts of insects on forest carbon dynamics, however, are generally ignored in large-scale modelling analyses. The current outbreak in British Columbia, Canada, is an order of magnitude larger in area and severity than all previous recorded outbreaks. Here we estimate that the cumulative impact of the beetle outbreak in the affected region during 2000–2020 will be 270 megatonnes (Mt) carbon (or 36 g carbon m-2 yr-1 on average over 374,000 km2 of forest). This impact converted the forest from a small net carbon sink to a large net carbon source both during and immediately after the outbreak. In the worst year, the impacts resulting from the beetle outbreak in British Columbia were equivalent to 75% of the average annual direct forest fire emissions from all of Canada during 1959–1999. The resulting reduction in net primary production was of similar magnitude to increases observed during the 1980s and 1990s as a result of global change. Climate change has contributed to the unprecedented extent and severity of this outbreak. Insect outbreaks such as this represent an important mechanism by which climate change may undermine the ability of northern forests to take up and store atmospheric carbon, and such impacts should be accounted for in large-scale modelling analyses.

Mike Aucott said...


The fact that the fraction of fossil fuel emissions taken up by the land does not show a significant trend indicates that there has actually been a long-term increasing trend in the absolute value of the terrestrial sink. There's a good discussion on this in James Hansen's new book, Storms of My Grandchildren. See figure 16 (p.119). This figure shows that the airborne fraction of fossil fuel carbon emissions has been nearly constant for the last 50 years, even though emissions from combustion of fossil fuels have increased from about 2 Gt/y in 1950 to 8 Gt/y today. While risks to forests are worrisome, so far plants appear to be sequestering more carbon, not less, each year. This provides a glimmer of hope that atmospheric carbon levels can be stabilized if emissions can be limited.