PS 69-50 - An assessment of tree encroachment into a serpentine pine savanna using remote sensing and dendroecological techniques

Thursday, August 9, 2012
Exhibit Hall, Oregon Convention Center
E. April Sahara1, Daniel A. Sarr2 and Erik S. Jules1, (1)Department of Biological Sciences, Humboldt State University, Arcata, CA, (2)Klamath Network-National Park Service, Ashland, OR
Background/Question/Methods

The dominant vegetation type in certain low-elevation serpentine habitats in the Klamath Mountains of California and Oregon is Jeffrey pine (Pinus jeffreyi) savanna. This vegetation type is characterized by scattered trees in the overstory and a more or less continuous layer of forbs and grasses in the understory, often including rare and serpentine endemics. The Little Bald Hills area of Redwood National Park encompasses the only known serpentine Jeffrey pine savanna within the park boundaries. Park scientists have noted encroachment by both Jeffrey pine and Douglas-fir (Pseudotsuga menziesii) for at least the past 20 years, and have become increasingly concerned about the impacts of encroachment on rare species and the continuing existence of this rare park vegetation type.

We analyzed vertical aerial photographs from 1942, 1960, 1975, 1980, 1993, 1998, and 2009 to create a predictive model of encroachment. We also collected tree cores and measured environmental variables for 27 0.2 ha circular plots in Little Bald Hills in order to determine historical tree basal area, tree establishment dates, and whether seedling establishment or change in tree basal area was correlated with slope, aspect, elevation, or annual precipitation.

Results/Conclusions

Our analysis of the historical aerial photographs indicates that in 1942 grassland covered 40% of the Little Bald Hills, but was less than half of that in 2009. A logistic regression model predicts that, if current encroachment rates continue, by the year 2109 grassland will represent less than 1% of the total historical Jeffrey pine savanna.

The earliest tree establishment detected by our field sampling was 1855, indicating there were few trees present before the 1850s. Pulses of establishment occurred in the 1860s and 1870s, and then again in the 1940s and 1980s. We were unable to correlate establishment or change in tree basal area to any of the environmental variables considered or to annual precipitation, indicating that the factors governing encroachment are more complex than topographic position or annual climatic variability.