An Unexpected Shift in Treeline Vegetation

[caption id="attachment_3405" align="alignright" width="300"] Photo 1: A map of the GTREE observation network. Image credit: Carissa Brown.


Scientists usually aim to collect as much data as possible to help detect pervasive or repeated patterns often hidden by the high variability of individual observations. Drawing strong inference from individual, often anecdotal occurrences can be misleading because it is difficult to know whether the relationships we see can be generalized to other instances. On the other hand, as human beings we seem to learn well from stories of individual events, and these have the power to transform our understanding in ways that may have a bigger impact than reams of data. To me, this is one of the challenges of working as a field ecologist, where there is often a tension between developing strong inferential power by assembling datasets with many observations and developing a deeper understanding of the dynamics unfolding in any one place.

Important StoriesWhen I look at the state of treeline research today, I see a lot of evidence of this tension. There is a strong push to assemble data on treeline movement from alpine and latitudinal treelines around the world, in search of patterns that can tell us something about how treelines will respond to environmental change in the future. However, synthesis efforts to date have encountered a tremendous amount of variability in treeline responses to recent environmental change, and this variability encompasses opposing directions and magnitudes of responses across local to global scales. Finding common patterns in treeline dynamics amidst the growing number of treelines being studied around the world is not a trivial problem.

[caption id="attachment_3406" align="alignleft" width="300"] Photo 2: Finger Mountain in northern Alaska. Photo credit: Mélanie Jean.


Coordinated studies that use a common methodology are a powerful tool for monitoring treeline dynamics and investigating possible causes in a way that helps remove some of the noisy error that arises when comparing observations. For example, the GTREE (Global Treeline Range Expansion Experiment) study uses a common methodology for examining how seed availability and disturbances to intact plant communities may control the potential for trees to migrate into tundra environments around the world. However, ecological dynamics can be driven by unusual events that are unlikely to be captured within planned observation networks, no matter how well designed. As ecologists, we also need to take advantage of unexpected occurrences that may only happen rarely within our planned observations, but which tell us important stories about how ecological change may unfold.

 
On Finger MountainHere, I’d like to tell the story of one set of treeline observations in northern Alaska that I think has something to teach us about the potential for rapid and unexpected treeline responses to environmental change.

The location is near the gentle summit of Finger Mountain, about 25 km south of the Arctic Circle in the southern foothills of Alaska, USA.  The summit of Finger Mountain is covered with a mixture of tall and low shrub tundra, which gradually changes to continuous black spruce forest as you descend the shallow slopes. The road to Prudhoe Bay and the Trans-Alaska Pipeline go right past Finger Mountain and its iconic set of rocks that stick out like a finger from the rolling tundra at the top.

[caption id="attachment_3407" align="alignright" width="300"] Photo 3: Picture of a burned treeline site near Finger Mountain in 2005, one year after fire. Photo credit: Jill Johnstone.


2004 was a record year in Alaska, with warm and dry summer conditions leading to the greatest area burned in the state’s recorded fire history; a large complex of fires burned along the Dalton Highway from the Yukon River north to Finger Mountain. The burning reached all the way into the upper slopes, where sparse spruce trees transition to shrubby tundra. Although fighting these many fires caused a large economic deficit for state and federal land managers in Alaska, the fires also created an opportunity for ecologists to study the impacts of fire in road-accessible areas that permit repeated site visits.

Burning AnalysisI was involved in a project that received funding from the US Joint Fire Science Program to track ecological responses to fire in forests dominated by black spruce (Picea mariana) in interior Alaska. We established a network of 90 sites in order to capture a wide range of fire severity conditions, as well as pre-fire conditions related to vegetation structure and moisture drainage. We included sites that ranged from dense stands of black spruce up to treeline sites where there were only scattered individuals of black and white spruce (Picea glauca). Three of these treeline sites were located in the area around Finger Mountain (with five others located at burned treelines in the White Mountains to the south and east).

[caption id="attachment_3404" align="alignleft" width="300"] Photo 4: Picture of unburned site near treeline, where the vegetation is dominated by shrubs and sparse spruce trees. Photo credit: Jill Johnstone.


During the initial years after the 2004 burns, we collected detailed observations of fuel combustion (Boby et al. 2010), post-fire seed rain (Johnstone et al. 2009), and tracked the emergence of tree seedlings in seeded and unseeded plots (Brown et al. 2015). Our analyses of initial tree seedling establishment at the 90 sites found that plots with high severity burning, where much of the surface organic layer was consumed by fire, had altered seedbeds that promoted the recruitment of deciduous broadleaf trees such as aspen (Populus tremuloides) and birch (Betula neoalaskana). There were a number of sites at lower elevations that appeared to be shifting from expected patterns of black spruce self-replacement to deciduous or mixedwood dominance (Johnstone et al. 2010). However, at high elevations close to the treeline, recruitment of all tree species tended to be low and we predicted that these sites would recover to sparsely treed areas similar in composition to those that existed before the fire.

With long-term support from the Bonanza Creek LTER and more targeted projects funded by the US Department of Defense Environmental Research Programs, we have continued to track patterns of ecological recovery into the second decade after the fire. Although much of the ecological dynamics have played out as we predicted, there have been a few surprises. In particular, the treeline sites near Finger Mountain have provided us with what I consider to be some of the most interesting and unexpected results.

[caption id="attachment_3401" align="alignright" width="300"] Photo 5: Low establishment of tree seedlings leads to shrub dominance at a treeline site that burned with low fire severity. Photo credit: Jill Johnstone.


Several SurprisesWe had three burned sites representing sparsely treed stands near the treeline at Finger Mountain. The fire severity at these sites ranged from very low combustion of the organic layer to more extensive combustion that exposed about 50 percent of the mineral soil surface. At the low severity site, limited seed availability for spruce and poor quality seedbeds on organic soils has resulted in very little tree seedling establishment. Growth of tree seedlings transplanted into the site was slow with substantial mortality, suggesting very limited potential for tree expansion. This site appears to be on a trajectory to maintain a stable ecological condition as sparsely treed tundra.

Ecological conditions have been more conducive to tree growth at the site with intermediate fire severity, although seedling establishment has still been relatively sparse. This site is likely to recover to a relatively productive, but still very open, treeline stand that has black spruce as the most common tree species.

[caption id="attachment_3402" align="alignleft" width="241"] Photo 6: Tiny aspen establishing from seed on the ground where most of the soil has been burned away. Seedlings were marked with colored toothpicks to track mortality. Photo credit: Jill Johnstone.


Our third treeline site at Finger Mountain burned with higher severity than the other two, and this site has shown a very different pattern of recovery after the fire. Our first surprise at this site was the appearance of tiny seedlings of trembling aspen that established from seed in the year after the fire. There are very few aspen trees in the local area that could have acted as a local seed source for post-fire colonization. However, aspen seeds are lightweight and attached to a buoyant, hairy plume that allows them to be transported very long distances. In 2005, the year after the fires, aspen seed production in interior Alaska was the highest I’d seen in that decade, and the air in mid-June was full of aspen seeds being blown about by the wind. High seed production combined with a mechanism for long-distance transport allowed abundant aspen seed to arrive at the upper slopes of Finger Mountain.

Our second surprise did not really become apparent until we re-surveyed these sites a decade after the fire. Not only did the aspen seedlings establish at relatively high rates at this site, but they also grew rapidly to develop a tall deciduous canopy that dwarfs the tundra vegetation and sparse spruce seedlings below. The unexpectedly high recruitment and growth of aspen seedlings is transforming the site from open spruce treeline to something that resembles the aspen woodlands I am used to seeing at the southern margins of the boreal forest. This type of dramatic ecological transition is not something I would have expected to see at this northern treeline site. Currently, we don’t know how widespread or common such a transition may be in the area burned by the 2004 fires. It will probably require another decade of growth before patches dominated by deciduous trees can be distinguished from shrubby areas and mapped with satellite imagery. However, this single observation of unexpected change in treeline vegetation triggered by fire is pointing a finger at the potential for treeline change to be both rapid and dramatic when the right conditions come together. Although it may be a be just one data point among many, it tells a story that I think we should pay attention to.

[caption id="attachment_3403" align="alignright" width="300"] Photo 7: Aspen woodland developing at a treeline site on the slopes of Finger Mountain that was characterized by sparse, slow-growing spruce prior to the 2004 fire. Photo credit: Jill Johnstone.


References & Further Reading: Boby, L. A., E. A. G. Schuur, M. Mack, D. L. Verbyla, and J. F. Johnstone. (2010) Quantifying fire severity, carbon, and nitrogen emissions in Alaska’s boreal forest: The adventitious root method. Ecological Applications 20:1633–1647.
Brown, C. D., J. Liu, G. Yan, and J. F. Johnstone. (2015) Disentangling legacy effects from environmental filters of post-fire assembly of boreal tree assemblages. Ecology 96:3023–3032.
Johnstone, J. F., L. A. Boby, E. Tissier, M. Mack, D. L. Verbyla, and X. Walker. (2009) Post-fire seed rain of black spruce, a semi-serotinous conifer, in forests of interior Alaska. Canadian Journal of Forest Research 39:1575–1588.
Johnstone, J. F., T. N. Hollingsworth, F. S. Chapin III, and M. C. Mack. (2010) Changes in fire regime break the legacy lock on successional trajectories in Alaskan boreal forest. Global Change Biology 16:1281–1295.
This article was written by Dr. Jill Johnstone, Adjunct Professor of the Department of Biology at the University of Saskatchewan, Canada.