Response of Ponderosa Pine to Variable Scale Climate Influences, Salmon River Canyon, Idaho

Growth of trees in sparse stands on low-productivity sites is often strongly controlled by climate variation. We examined tree rings in cores collected from 73 ponderosa pine trees (Pinus ponderosa) at a dry upland site near the confluence of French Creek and the main fork of the Salmon River. Cores were mounted, processed, and visually and statistically cross-dated following standard dendrochronological methods. Ultimately, 41 tree-ring measurement series with a continuous time span of 278 years were used to create ring-width indices of tree growth for the site. These indices were tested against annual and monthly climatic variables. Simultaneous dating of fires scars from trees and snags at the site enabled reconstruction of a 160-year fire history. There were strong (p<0.01) positive correlations between ring width indices and annual Palmer Drought Severity Index (PDSI) and precipitation, indicating trees grew best in wet years. Strongest correlations with monthly climate variables were for prior-year fall and winter temperature and precipitation, as well as November to April snow water equivalent (SWE). The seasonal variable found most strongly correlated with tree growth was September-January total precipitation, most of which falls as snow at this site, which explained 34 percent of the total variance in annual ring widths. The strong relationship with monthly SWE is corollary to the relationship observed in the fall-winter precipitation, but high correlation with SWE in April and May underscores the positive influence of late season snowpack on current year summer growth. The occurrence of fires was greater during years with low precipitation and high PDSI, but growth responses to climate variables were not affected by fires. Although no significant correlation existed between tree growth and Pacific Decadal Oscillation Index, teleconnections with oceanic climatic influences were present in a positive relationship with the Atlantic Multidecadal Oscillation Index. The AMO was also correlated to fire recurrence at the site; all fires occurred during the AMO negative (cool) phase. The warm phase of the AMO was also correlated warmer winter temperatures that could lead to greater winter precipitation. Variance in tree ring growth indices was lower during the 20th century than in earlier years, suggesting greater amplitude in the AMO before about 1850.