Temporal carbon dynamics of forests in Washington, US: Implications for ecological theory and carbon management

We quantified carbon (C) dynamics of forests in Washington, US using theoretical models of C dynamics as a function of forest age. We fit empirical models to chronosequences of forest inventory data at two scales: a coarse-scale ecosystem classification (ecosections) and forest types (potential vegetation) within ecosections. We hypothesized that analysis at the finer scale of forest types would reduce variability, yielding better fitting models. We fit models for three temporal dynamics: accumulation of live biomass, accumulation of dead biomass, and net primary productivity (NPP). We compared fitted model parameters among ecosections and among forest types to determine differences in potential C storage and uptake. Models of live biomass C accumulation and NPP fit the data better at the scale of forest types, suggesting this finer scale is important for reducing variability. Model fit for dead biomass C accumulation depended more on the region than on the scale of analysis. Dead biomass C was highly variable and a relationship with forest age was found only in some forest types of the eastern Cascades and Okanogan Highlands. Indicators of C storage potential differed between forest types and differences were consistent with expectations based on spatial variability in climate. Across the study area, maximum live biomass C varied from 6.5 to 38.6 kg C m(-2) and the range of ages at which 90% of maximum is reached varied from 57 to 838 years. Maximum NPP varied from 0.37 to 0.94 kg C m(-2) yr(-1) and the age of maximum NPP varied from 65 to 543 yrs. Forests with the greatest C storage potential are wet forests of the western Cascades. Forests with the greatest potential NPP are 65-100-year-old mesic western redcedar-western hemlock forests and riparian forests, although limited data suggest maximum NPP of coastal sitka spruce forests may be even greater. The observed relationship between the ages at which maximum NPP and maximum live biomass are reached for a given forest type suggests that there is a trade-off between managing for maximum live biomass (storage) vs. NPP (uptake) in some forest types but an optimal age for C management in others. The empirical models of C dynamics in this study can be used to quantify the effects of age-class distributions on C storage and NPP for large areas composed of different forest types. Also, the models can be used to test the effects of current or future natural and anthropogenic disturbance regimes on C sequestration, providing an alternative to biogeochemical process models and stand-scale methods. (C) 2013 Elsevier B.V. All rights reserved.