NON-STATIONARITY IN MOUNTAIN HEMLOCK GROWTH ALONG THE GULF OF ALASKA AND DENDROCLIMATIC RECONSTRUCTION OF PDV

Mountain hemlock at treeline sites have been widely employed in reconstructions of Pacific Decadal Variability (PDV) due to their high sensitivity to temperature. We examine the climate response at near-coastal sites along the Gulf of Alaska with respect to elevation. While most high elevation treeline sites appear to be faithfully tracking contemporary warming, many low elevation sites show a marked decrease in ring-width after about 1950. This decrease in tree growth is reminiscent of the well-studied Alaska yellow cedar decline attributed to a decrease in insulating snowpack and a resulting increase in their susceptibility to frost damage. The cedars are in the midst of a century-long decline, coinciding with the end of the Little Ice Age (LIA), circa A.D. 1880. To determine if mountain hemlock at low elevations could be susceptible to cedar-like decline, we compare monthly temperature and precipitation records from Sitka, Alaska that begin in 1832 and includes data from the LIA through contemporary warming. Temperature correlations throughout the record show no significant change in response; tree growth favors warm temperatures in most months. Comparisons with precipitation records, however, suggest a shift in response at both low and high elevation sites; trees that were negatively correlated with precipitation during the Little Ice Age began in the mid 1900s to favor increased spring precipitation (snowfall) that may serve as protection against damaging spring frosts. These observations are consistent with the loss of snowpack, possibly due to earlier meltout or decreased snowfall over the last few decades. Low elevation sites experience loss of snow cover first and thus are more vulnerable, as in the case of the yellow cedar. Unlike the yellow cedar, however. mountain hemlock are responding to these changing conditions with an upward treeline migration, indicating that this “decline” at low elevation sites could be an ongoing adjustment of the species. A major research question remains: will the migration keep pace with the changing conditions? A closer examination of this question can lead to a better understanding of the climate controls on tree growth, valuable information for PDV reconstruction and for anticipating future changes in forests along the Gulf of Alaska.