Our research questions
Is the nature of forest fires changing in a warming world? Are forests responding to disturbances differently than they have in the past? If so, how will these changes feedback to future disturbances such as fire and insect outbreaks? How are forests affected by the interaction of multiple disturbances (e.g., two successive fires, or insect outbreaks + fire)? How do these dynamics change over spatial and temporal scales? How can forest management shape forest trajectories and disturbance susceptibility?
These are just a few of the core research questions we ask in the Harvey Lab, where our core mission is to conduct cutting-edge scientific research in forest/landscape/disturbance ecology. We co-develop much of our research with partners and stakeholders, to further our understanding of the nature of forest ecosystems and connect our research to pressing issues in forest management across public and private lands.
Our current research projects
Forest Fires in western Cascadia: evaluating drivers & effects
Wet forests on the west side of the cascades are characterized by relatively infrequent but very severe fires. As the climate warms, fire potential is likely to increase, heightening the need for understanding the drivers of fires on the west side of the cascades and how they shape forests in the past, now and in the future. Along with our collaborators and partners, our research is studying recent wildfires that have burned in the west cascades of N OR and WA across a range of pre-fire stand ages (~50 years to ~500 years) in western hemlock / silver fir / Douglas-fir forests to ask the following questions:
1) What is the nature of the climate and weather events that drive fires on the west side and how are those conditions expected to change as the climate warms?
2) What are the drivers of burn severity patterns in recent west side fires and how do those patterns of stand-replacing fire compare to historical burn severity patterns?
3) How is post-fire forest succession and early-seral habitat affected by stand age at the time of fire, burn severity, and landscape position (topography and distance to burn patch edge)?
4) How are post fire vegetation responses affected by topoclimate and post-fire climate variability, and how can that inform climate adaptive management before and after fires?
Collaborators: Daniel C. Donato (WA DNR); Joshua Halofsky (WA DNR); Crystal Raymond (UW); Philip North (Tulalip Tribes); Nick Bond (UW); Karin Bumbaco (UW); Jerry F. Franklin (UW-SEFS); Dave W. Peterson (USFS)
Disturbance and restoration of dry forests in eastern Washington
Fire is a natural and integral disturbance that shapes forests in the Pacific Northwest. For example, many dry forests on the east side of the cascades were historically characterized by frequent fire and were dominated by thick bark fire “resisting” trees such as ponderosa pine. However, these forests are now at risk of high-severity fire from decades of fire exclusion and fuel buildup and a warming climate. To address these challenges, we are researching the fire ecology and management options in the dry forests on the east side of the Cascade Crest. We are studying dry forests where fuels treatments were implemented 15 years ago and that later burned in the 2012 Wenatchee Complex fire (Mission Creek Fire and Fire Surrogates site), and testing the longevity of fuels treatments from wildfire, mechanical treatment, and prescribed burning across eastern WA. In addition, we are assessing trends in forest condition and restoration need over the last few decades. Collectively, we are asking the following questions:
1) How do fuels treatments (e.g., mechanical thinning, prescribed burns) affect resistance and resilience to subsequent wildfire in low-elevation dry forests of the East Cascades?
2) What are the temporal trends in restoration need in dry forests of eastern WA and how are these trends affected by disturbance (e.g., wildfire)?
Serotinous forests in a warmer and more fire-prone world
Many forests are well adapted to severe, stand-replacing fire. Serotinous pine trees have evolved cones that are held sealed shut with resin until the heat of a fire opens the cones, releasing seeds to establish the next forest. Although well adapted to severe fires with predictable and long (relative to tree development) fire-free intervals, it is unknown how shortened intervals between severe fires may “squeeze” populations of serotinous conifers, such that a second fire may occur before young trees have produced cones. Further, warming temperatures may delay cone production during the time between fires, adding another factor that could threaten serotinous pines in a warmer and more fire-prone future.
Together with US and international collaborators, we are characterizing the race between “fuels vs. fruits (cones)” in forests dominated by serotinous pines in the western US, spanning systems in California that are water limited to systems in the interior Northwest that are energy (i.e., growing season) limited. We are asking the following questions:
1) How does cone production, and the relative proportion of open vs. closed cones, vary in young post-fire stands of serotinous conifers?
2) How does the trajectory of cone development compare to fuel development following one stand-replacing fire, and where are there critical points where fuels outpace cones?
3) How do dynamics differ across systems with different controls on tree demography (e.g., water- vs. energy-limited systems)?
Spatial patterns of burn severity under changing fire regimes
Even large forest fires are extraordinarily heterogeneous in how they burn; that is, they produce a complex mosaic of severely burned patches intermixed with unburned islands and lightly burned areas. This heterogeneity is critical to post-fire forest response (e.g., ability to disperse seed from unburned patches), yet the mechanisms that produce these patterns are not fully understood. Further, patterns may be changing as fire regimes are changing (e.g., increased frequency, severity, or size of fires). We are using hundreds of field plots across the Rocky Mountains and the Pacific Northwest where we have intensive measures of burn severity (e.g., tree mortality, soil charring) to calibrate satellite indices of burn severity. We are then using these calibrated satellite maps of burn severity to ask the following questions:
1) How well do satellite indices of burn severity perform across gradients of latitude, topography, stand structure, and pre-fire insect outbreaks? Further, can satellite indices accurately measure burn severity in short interval “reburns” that are becoming common throughout the western US?
2) What are the drivers of landscape patterns of burn severity (e.g., patch size and shape of different forests burned at different severity)? Do those drivers vary within and among regions?
3) Are the landscape patterns of burn severity different in short-interval reburns than in long-interval fires?
Collaborators: Monica G. Turner (University of Wisconsin); Craig Baker (USFS-GTAC)
Old growth, disturbance, and legacies in long-term plots
Bark beetle outbreaks are naturally occurring disturbances throughout upper-montane and subalpine forests in the Rocky Mountains, yet few long-term datasets exist to track forest trajectories pre-, during-, and post-outbreak, and how these outbreaks drive structure and function across spatial scales. Further, little is known about how the legacy of past thinning treatments may influence resistance or resilience to beetle outbreaks and potential for interactions with other disturbances such as fire. Building on permanent plots that were established in the late 1930s at Fraser Experimental Forest in Colorado, we are conducting stem mapping in four replicate 2-ha plots, re-surveying tree mortality, survival, growth, and reproduction, and measuring fuel profiles across a range of disturbance X treatment interactions. We are asking the following questions:
1) How do insect outbreaks affect the spatial pattern of dead and live trees within a stand, and what do these patterns mean for old-growth forest structure?
2) How do past thinning treatments affect resistance to beetle outbreaks occurring 60 years post-treatment?
3) How does the legacy of past thinning treatments interact with contemporary beetle outbreaks to affect stand structure, C storage, and fire hazard?
Causes and consequences of biotic forest disturbances
The future of subalpine forests in the Rocky Mountains is uncertain following widespread and severe bark beetle outbreaks [primarily mountain pine beetle (Dendroctonus ponderosae) and spruce beetle (Dendroctonus rufipennis) in lodgepole pine (Pinus contorta) and Engelmann spruce (Picea engelmannii) trees, respectively] in recent decades. Many subalpine forest trajectories will depend largely on survival of subalpine fir (Abies lasiocarpa), the primary tree species not attacked in recent outbreaks. Yet, subalpine fir has experienced escalating mortality since the mid-1990s across the US Rocky Mountains. The causes of this mortality and consequences for subalpine forest integrity are largely unknown, but such information is critical for designing and successfully achieving conservation goals in the face of novel climate and disturbance combinations.
We are using a long-term dataset on tree demography (e.g., establishment, growth, and mortality on >6,000 individual trees) located at the Niwot Ridge NSF-Long Term Ecological Research (LTER) site and combining these data with with satellite/aerial imagery, climate data, and forest simulation modeling to better understand the causes and consequences of accelerating rates of tree mortality in the subalpine forest zone of the US Rocky Mountains. Specifically, we are asking the following questions:
1) What are the spatio-temporal patterns of recent subalpine fir mortality and relationships with potential causal mechanisms?
2) What are the effects of recent tree mortality on future forest trajectories and susceptibility to disturbance?
3) How can different management treatments foster ecological resistance and resilience?
In our research we use a variety of tools and methods, letting the interesting and important questions guide our approach. We strive to connect insights across spatial scales (e.g., trees –> tree-neighborhoods –> forest stands –> watersheds –> regions), temporal scales (e.g., hours –> days –> weeks –> years –> decades –> centuries), and gradients (e.g., elevation, land use, management context). Doing so allows us to link a deep understanding of natural history born in the field with “big data” streams available from networks of satellites and sensors.
Our study areas
Our research is primarily focused in conifer forests of western North America (with particular focus on forests in the Pacific Northwest and the Rocky Mountains), where forests span gradients in: elevation ranging from sea level to ~3500 m; human influence from heavily-managed plantations to protected wilderness; biophysical settings from rainforests to dry woodlands; and disturbance regimes from frequent (e.g., decadal) low-severity fires to infrequent (e.g., many centuries) high-severity fires. We utilize long-term research sites (e.g., NSF Long-term Ecological Research stations and USFS Experimental Forests), extensive networks of field plots across the western US, and publicly available geospatial data (e.g., satellite burn severity maps).
We acknowledge that our research in the Pacific Northwest and beyond is located on homelands of Indigenous peoples, and we have a responsibility to critically reflect on the histories of dispossession and forced removal of the original inhabitants, as well as the resilience and vibrancy of their cultures today.