Figure 1.
Areas highlighted represent the subsets of high resolution imagery analyzed; subsets in red indicate land under traditional management (unprotected), and subsets in yellow designate land that is managed for conservation (protected).
Figure 2.
Identification of timberline using remotely sensed imagery.
Digitized 1963 timberline (yellow line) draped over the ASTER GDEM2. A 2005 Quickbird multispectral true color composite serves as the background layer. The blue polygon highlights the timberline subset displayed in Fig. 3; red segments indicate change in timberline elevation between 1963 and 2005.
Figure 3.
Calculation of timberline movement.
An example of a 1963 aerial photograph (left) and 2005 Quickbird panchromatic imagery (right) of the same location in the study area. The yellow line delimits 1963 timberline and the red line delimits 2005 timberline. Differences between mean segment elevations (m) are indicated in red squares. Grid cells correspond to the 30-m GDEM cell boundaries.
Table 1.
Summary of timberline change between 1963 and 2005.
Figure 4.
Migration of timberline in protected and unprotected areas.
A graph showing the distribution of mean elevation change for timberline segments that migrated between 1963 and 2005 in protected areas (n = 32) and unprotected areas (n = 24). Mean elevation change for segments inside protected areas was significantly greater than zero (one-sided t-test, t = 5.01, df = 31, p<0.0001). Outside protected areas, a given change segment was equally likely to migrate up as to migrate down in elevation; the mean of elevation change for all segments in unprotected areas was not significantly different than zero (two-sided t-test, t = −0.13, df = 23, p = 0.90). Migration was more likely to occur in an upslope direction in protected areas compared to non-protected areas (Fisher’s Exact Test, df = 1, p<0.0003).
Table 2.
Summary of timberline segments that migrated between 1963 and 2005.
Figure 5.
Bootstrapping of timberline migration rates.
Bootstrapped sample probability of mean yearly migration rate for the entire length of measured timberline in protected and unprotected areas (∼40 km in each area), and differences between these treatments. Total samples equaled 20,000.
Table 3.
Annualized migration rates (in vertical m yr−1) and number of years required to reach equilibrium with 2100 climate projections.
Figure 6.
Influence of elevation upon upslope timberline migration rate.
A graph showing the rate of mean elevation change versus starting elevation for segments that migrated during the study period. Filled circles represent samples in protected areas; unfilled circles represent those samples in unprotected areas. Mean elevation was predicted as a function of segment starting (1963) elevation, protection status, and the interaction between the two effects.