Detecting Tree Migration with Forest Inventory Analysis

Preparers

Christopher Woodall, Northern Research Station; Gretchen Moisen, Rocky Mountain Research Station; Louis Iverson, Northern Research Station; Nicholas Crookston, Rocky Mountain Research Station

Issues

The spatial contraction, expansion, and persistence of tree ranges in response to climate change is species-specific and can be collectively referred to as tree migration. Past migration of tree species across long time scales is well-documented, especially since the last ice age (1). However, given recent changes in climate and projected future conditions, the ability of tree species to migrate at the pace of rapid climate change is unlikely (2). Since tree species will respond individually to climate change, changes in the spatial co-occurrence of some species are expected (3, 4, 5), potentially resulting in future North American biomes for which there is no contemporary analog (6). The rapid rate of climate change combined with the response of individual tree species could lead to extirpation, or regional loss of certain tree species. This in turn could lead to a loss in biodiversity (7), loss or gain of forest area, or all of these. For example, forests may be converted to a different ecosystem such as grassland, or alpine tundra could be converted to forest.

Scientists are actively engaged in efforts to both model potential future tree ranges (for example see 8) and monitor current tree species distributions. Monitoring of current shifts in tree ranges is often limited by the lack of consistent data over the 1900’s with the first national atlas of tree distributions not developed until the 1970’s (9). Due to this relatively short temporal scale, evaluating range boundaries across the multitude of tree species in US forests is problematic. Novel indicators of tree range dynamics have been developed to assist, for example comparing seedlings to adult tree locations (10).

Likely Changes

To date, contemporary tree migration has been documented in a portion of the eastern US along elevational gradients, (e.g., 11) and observed globally (12). Tree migration up elevation gradients usually affects relatively small geographic areas, and therefore may occur more readily than tree migration latitudinally, which concerns broad regions especially in flat terrain (13, 14). For example, a 1˚C increase in mean annual temperature may correspond to 100 to 200 m of elevation in contrast with 150 km of latitude (Fig. 1 in 13).

Tree range studies have been initiated across latitudinal scales for much of the eastern US forested area as more inventory data have become available. Current research suggests that tree ranges in the eastern US may be fairly static along their margins, but with tree seedlings for some species present at higher densities in the northern part of their range causing a slight northward shift in their mean latitudinal location compared to their adult counterparts (10). If the margins of tree ranges remain static while regeneration fails within the range such as along southern range boundaries, range contraction will occur (15). Emerging work has suggested such a dynamic may be currently occurring for certain species (2). The vast presence of invasive plant species across US forests (16) coupled with the advanced stage of stand development and stocking of forests (17) suggests there will be difficulty in recruiting tree regeneration in the future, especially for native tree species subjected to deer herbivory. Even over the past decade, the abundance of tree regeneration in eastern US forests has decreased, potentially as a result of these factors (18).

If climate change continues or accelerates, one would expect to see more rapid loss of tree species with narrow ecological niches along their southern range boundaries. One would also expect to see tree regeneration failure and the loss of small trees in these zones. Some tree species have genetically distinct subpopulations that have become adapted to local conditions over time, and climate change could lead to their maladaptation. Maladaptation and potential mortality will impact forest stand composition in species and tree size. Therefore the predicted sequestration of carbon, yields of wood volume, and provision of other ecosystem services from forests will differ greatly from those forest managers would expect without climate change (19). For example, in the western US, climate change is expected to cause locally adapted Douglas-fir to become maladapted (20, 21) leading to growth loss (22). On the other hand, some populations may benefit; certain subpopulations of western larch may be more genetically suited to future climates than others and could serve as seed sources for management interventions (23). Emerging research suggests that range contraction may be expected in the more montane areas of the west where regeneration may fail to match adult tree distributions (24).

Beyond climate change itself, regional land use legacies can compound the complexity of future tree range dynamics. For example, populations of piñons and junipers across the interior western US have been highly dynamic over the last two centuries, undergoing an overall expansion but punctuated with regional mortality. These species have areas of long-term persistence relative to their centurial (piñons) and millennial (junipers) life spans which in turn reflect differing land management legacies.

Management Options

It is important to recognize that under future climates there will be species-specific zones of contraction, expansion, and persistence across the landscape. Management practices that are developed for certain species under certain conditions cannot necessarily be extrapolated to different species under disparate conditions.

Expected shifts in tree ranges mean that the healthy forests of the future may not look exactly like the healthy forests of today. The following considerations may help managers retain forests and the benefits they provide even as the areas they manage become more or less suitable for specific species.

• Consider rotation periods that match the time trees growing on a site will likely remain genetically attuned to the environment. This will often require the use of shorter rotations.
• For currently growing trees of a given species, introduce trees of the same species that are more climatically-adapted to projected future climate (e.g., from a different subpopulation).
• Establish species and seed imported from a wide array of locations to increase the genetic diversity at a location, with the expectation that some portion of the trees will be viable at the site as the climate changes.

Finally, it is important to remember that tree regeneration is a complex dynamic, mediated by numerous processes in addition to climate. Tree regeneration and mortality are affected by site factors (e.g., soils and elevation), climate (e.g., precipitation and temperature regimes), herbivory (i.e., deer browse), mast periodicity, management actions, stochastic disturbances, forest succession, and competing vegetation (e.g., invasives). This complicates both tree range monitoring and the development of adaptive management responses to climate. Managing tree ranges, whether through assisted migration or species selection during silvicultural operations, will need to take into account the fact that more than climate will be affecting tree ranges in any given region of the US.