Projected effects of climate change on old-growth forests carbon dynamics in the southern Sierra Nevada mountains

Background/Question/Methods

Climate change is projected to cause increased warming and inter-annual variability in precipitation in the Sierra Nevada of California.  With the potential to influence forest capacity for sequestering and storing carbon, understanding the impacts projected climate change may have on forest growth is important for long-term climate change adaptation planning and management.  We conducted a series of landscape scale simulations under historical and projected climate using outputs from the GFDL global climate model, driven by the A2 emission scenario.  One hundred fifty year simulations (1950-2099) were conducted for the Teakettle Experimental Forest in the southern Sierra Nevada using the LANDIS-II forest landscape model and Century Succession extension. Teakettle is an old-growth forest that is occupied by mixed-conifer forest at lower elevations and red fir at higher elevations.  We partitioned the forest by elevation, soil type, and forest type using a 400m grid.  We parameterized the grid-cell forest cohorts using US Forest Service Forest Inventory and Analysis data.  We sought to determine the effect of increasing temperature and altered precipitation projected by the GFDL-A2 simulation on forest carbon dynamics as compared to historical climate. 

Results/Conclusions

We found that assuming no disturbance forest growth and carbon accumulation continued to increase throughout the simulation period.  However, the trajectories began to diverge during the second half of the 21^st century, with the GFDL-A2 scenario having reduced total and soil carbon as compared to historical climate.  The mean difference in total and soil carbon over the 2050-2099 period was 12.3 and 5.9 Mg C ha^-1, respectively.  Over the 21^st century, the GFDL-A2 scenario has substantial projected warming coupled with precipitation deficits, as compared to historical climate, suggesting increasing drought stress may result in decreased carbon sequestration potential in this water-limited forest. Within the study area, the pattern of divergence in ecosystem carbon varied as a function of elevation, soil, and forest type.  Higher elevation red fir and mixed-conifer both exhibited similar patterns of end-of-century divergence in ecosystem carbon.  However, in the red fir dominated portion of the red fir/mixed-conifer ecotone ecosystem carbon was consistently lower under the GFDL-A2 scenario.  These results suggest that lower elevation red fir may experience reduced growth under warming and drying climate and maintaining the forest carbon sink strength may require managing for the more drought tolerant species that comprise mixed-conifer forests at lower elevation.