Regional patterns of tree mortality in old-growth forests of the Pacific Northwest

Background/Question/Methods

There is increasing evidence that tree mortality rates in old-growth forests have increased in the western United States since the 1970’s.  Some of the highest rates have been documented in the Pacific Northwest, but there is little understanding of how mortality rates vary along major environmental and climatic gradients at the regional scale.  Although previous research has focused on mortality in unburned forests, the recent increase in the frequency and area burned by wildfires suggests that these disturbances must also be taken into account to fully understand contemporary regional forest dynamics.   We investigate stand level (1 ha) mortality rates of trees >33 cm dbh in old-growth (>150 years) forests during a period from the mid-1990’s to the mid-2000’s in more than 2,500 forest inventory plots across National Forests of Oregon and Washington.  We investigate the relationship between stand level mortality rates and tree density, topography, and climate and compare rates among major potential vegetation types in unburned and burned forests across the region. 

Results/Conclusions

Plot level mortality rates in unburned stands were highly variable but the mean rate (0.76%/yr) corresponded reasonably well with published rates for the Pacific Northwest.  Mortality rates in dry potential vegetation types increased from lower elevation ponderosa pine (0.43%/yr) and Douglas-fir forests (0.75%/yr) to higher elevation grand fir forests (0.88%/yr).  Mortality rates in moist potential vegetation types also generally increased with elevation from lower western hemlock forests (0.53%/yr) to higher Pacific silver fir (0.65%/yr) and mountain hemlock forests (0.71%/yr).  High elevation subalpine fir forests had the highest mortality rates (1.35%/yr).  As expected, the percent of trees that died during the sample period in burned stands (40%) was much greater than that of unburned forests (5%) despite the fact that almost twice as many trees in the sample died in unburned forests than in burned forests at the regional scale.