Andrew Hansen¹, Lisa Graumlich¹, Warren Cohen², Michael Lefsky^2
1Montana State University, Bozeman, MT
²Oregon State University, Corvallis, OR

Funding: NASA Land Cover Land Use Change Program

Understanding past global change provides an important context for designing monitoring protocols to detect future change. New data from the Greater Yellowstone Ecosystem indicate that vegetation change has been dramatic here over the last century. Conifer forests have both increased in density and expanded into previously unforested areas. Concurrently, hardwood, shrubland, and grassland habitats have declined. Fire exclusion by humans may explain these forest dynamics. However, pilot dendrochronological studies of tree growth rates suggest climate variability has also contributed to conifer expansion. These vegetation dynamics appear to have important implications for carbon sequestration, fire and risk to humans, and biodiversity. The responsiveness of vegetation to past land use and climate, and the climate changes predicted for the GYE in the future, suggest that the GYE is an important site for monitoring for early detection of global change. The objectives of the this study are:

1. Quantify change in forest cover, density, and composition across the GYE during 1950-2000.
2. Assess the consequences of this change for carbon sequestration, fire and risk to humans, and biodiversity.
3. Devise a monitoring strategy to detect future change in forests of the GYE that, based on results from Objectives 1 and 2, optimizes our ability to detect changes that are most critical to land management.
4. Ensure the results are useful and available to decision-makers by involving forest managers and regional stakeholders in the research design and implementation process.

We will quantify change in forest composition and structure over the GYE for 1975-2000 using Landsat imagery calibrated with reference data from aerial photographs and for focal watersheds for 1950-2000 using aerial photographs. The consequences of these changes for carbon accumulation will be quantified by estimating carbon storage for each cover type based on allometric relationships and field data. Fire spread and risk to rural homes and communities will be assessed with sensitivity analyses using the fire model EMBYR. Habitat functions will be used to estimate change in the abundances of several bird and shrub species. A monitoring strategy will be implemented for locations of rapid change and for the GYE as a whole. Analyses of past change will be used to locate monitoring transects through locations of rapid change. Lidar imagery, aerial photographs, and field samples will be used to accurately quantify forest composition and structure along the transects. These data will also be used to refine the GYE-wide classification done with Landsat 7 ETM+. The resulting monitoring protocol can be repeated to track change in the future. The protocol will also be applicable to other locations in the Rocky Mountains.

Figure 1. The Greater Yellowstone Ecosystem study area and example locations for the focal watersheds
and transects for rapid change detection. The actual locations will be selected as part of the study.

Figure 2. Paired photographs of Panther Creek and Bannok Peak, Yellowstone National Park from 1880 and 1990.
Note the loss of willow in the foreground and the expansion of the conifer forest in the midground.
From Meager and Huston (1998).

Figure 3. Vegetation cover in the Bridger and Bangtail mountains, Montana in 1985 and1994. Maps were
derived from Landsat imagery in our current NASA LCLUC study. Note that much of the mixed
conifer (<70% cover) in the valley between the two ridge north/south tending ridges in 1985
changed to conifer (>70% cover) by 1994. Inset shows that the percentage of conifer in
1994 that was mixed conifer in 1985 varied among aspects.

Figure 4. Since 1950, growth rates of limber pine growing at tree line near Big Sky, Montana, have exceeded
those observed in the previous 550 years. Preliminary analyses indicate that growth is strongly governed
by growing season temperature and depth of snow pack. Data based on >50 trees. (Graumlich, in prep.)

Research	Prospective Students	Personnel	Facilities	Course Offerings	Employment	Related Links
Return to Top