Rare events can determine biomass dynamics and carbon status of old-growth forest

Background/Question/Methods

Are very late-successional (old-growth) mesic forests carbon neutral or, as suggested by some recent studies, carbon sinks?  How do interactions between growth of living biomass pool and inputs to dead biomass (coarse woody debris) pools affect overall carbon status?  Very large trees contribute large portions of total biomass, but do they contribute disproportionately to carbon sink strength either through biomass gains or higher mortality?  Most attempts to address these questions rely on short-term measurements of ecosystem-scale carbon fluxes and on biometric estimation of standing pools.  I address these questions using repeated (up to 8x) measurements of all trees over >75 yrs on 250 0.2 -acre permanent plots in old-growth hemlock-hardwood and swamp conifer forests (dominant species Acer saccharum, Acer rubrum, Betula alleghaniensis, Fagus grandifolia, Tsuga canadensis, Thuja occidentalis) in northern MI, USA.   Tracking of >10,000 individual trees since 1989 in upland plots allows analysis of individual contributions to biomass dynamics.  

Results/Conclusions

Above-ground living biomass densities were high for temperate forests, ranging from 250-400 Mg/ha in upland stands.  For most of the 75-yr study period, across all stand-types, average plot-scale living biomass increased at about 0.5%/yr.  However, biomass losses due to a blow-down in 2002 approximately equaled increases over the previous 65 yr, resulting in no net change in biomass density (but increased variation among plots).  CWD inputs through mortality averaged 0.5-2% of standing biomass per year, but the 2002 blow-down caused a one-time influx about an order of magnitude greater.  Biomass pools were increasingly dominated by large trees from 1935 to 2002, but biased mortality in the 2002 blow-down reversed this trend.   Individual-tree biomass increments showed strong species- and size-dependence.  Both average and variability of biomass increment increased linearly with diameter for all species; large trees showed extreme variability. Species differences suggest life-history-related patterns in size-growth rate relationships and in biomass-growth response to disturbance.  These results show that rare events may play a major role in biomass dynamics of old-growth forests, suggesting caution in inferring long-term carbon status from short-term measurements. The role of large trees in forest carbon cycling is likely important but complicated.