Carbon storage in large woody debris in low-order streams at the Hubbard Brook Experimental Forest, NH: Implications for carbon budget assessments

Background/Question/Methods

Streams in forested landscapes are heavily influenced by the surrounding forest. Riparian trees directly impact light availability, nutrient inputs, and are sources of large woody debris (LWD). While many functions of in-stream LWD are well studied, there is limited research on their carbon storage potential. Carbon stored in terrestrial LWD is well researched, but as wood accumulation dynamics, residence times, and decay rates in streams are less well understood, obtaining stream specific LWD carbon data is key to developing accurate landscape carbon assessments. In New England, USA, due to continued forest recovery and maturation following historic clearing, LWD is expected to increase as young forests enter late-successional stages. Understanding how this will alter stream LWD carbon is key to long-term predictions of landscape carbon storage. To better understand the carbon stored in in-stream LWD, we conducted a census of all LWD in sixteen 300m reaches in low order streams at the Hubbard Brook Experimental Forest (HBEF) in NH. These mid-elevation, northern hardwood-conifer forests have a history of partial harvesting. To understand how in-stream carbon storage will change as these second-growth forests develop, we also collected LWD data at eleven comparison streams in old-growth stands in the Adirondacks of NY.

Results/Conclusions

Streams at HBEF contained 4.19 Mg C/ha in LWD, exhibiting substantial spatial variation within and between sites (range 1.23-10.54). This variation is linked to heterogeneity in riparian forest structure, with large, legacy trees explaining some of these differences. Despite high variability, the mean carbon pool is comparable to region-wide assessments of forest floor LWD. Our data also suggest the storage potential of stream LWD will increase as riparian forests age, with comparison old-growth stands having a significantly larger in-stream LWD pool of 19.84 Mg C/ha. This aligns closely to assessments of terrestrial LWD in old-growth forests, indicating that despite different decay and export dynamics, low-order stream LWD carbon stocks may not differ significantly from the forest floor in the long-term. Our results argue that while stream LWD does not play an outsized role in carbon storage, at least in mature stands, it can be as important to carbon accounting as forest floor LWD on a per area basis. Additionally, the considerable spatial variability in carbon storage indicates that stream carbon dynamics are highly sensitive to changing ecosystem structure. Understanding the specific structural drivers of this variability will be important for developing robust models of carbon storage in forested streams.