Carbon Sequestration

Since 1996, the forest around the eddy covariance tower has been storing roughly 1.7 megagrams of carbon per year – about 1.5% of the total live forest biomass.   One of the critical questions surrounding carbon storage in forests is how long it will stay there, sometimes referred to as ‘permanence.’   This depends greatly on where the carbon is being stored.   If it is stored in trees, it is susceptible to being lost when/if a disturbance kills a tree.   In contrast, if it is stored in the mineral soil, it is more ‘protected’ and less likely to be lost.   As part of our ongoing work at Howland Forest, we are measuring changes in the amount of carbon stored in different components of the ecosystem, including live and dead trees, soil, and down-dead wood.   These results not only tell us how ‘stable’ the carbon sink is, but also provide data for comparison against the eddy covariance estimates of carbon storage.

Carbon storage in live and dead standing biomass

Picture 1: Measuring tree diameter for biomass measurements.

To measure changes in carbon stored in standing biomass, and also the transfer of standing biomass into the down-dead wood pool, we measure the diameter at breast height (Picture 1) of all trees in permanent plots surrounding the eddy covariance tower. In 2001, groups of 12 plots were located on concentric circles located 50, 100, 200,and 400m from the tower for a total of 48 plots.   Each plot is 14.6 meters in diameter.   These plots were established in 2001, and are being used to measure tree growth from that point onwards.   Prior to 2001, we depended on historical measurements made in a large (200 by 150 m) plot, where every stem has been mapped (Figure 1).

Figure 1: Plot design, and   trees measured in 1988, for the stem-map plot at Howland Forest.

We divided this plot into 30 by 30 m plots, 12 of which were measured in 1988, 1998, and 2002.   Using allometric equations to calculate individual tree biomass from the DBH measurement, we calculated that the forest stored about 1.7 megagrams of carbon per year between 1988 to 1998, and about 1.2 megagrams of carbon per year between 1998 and 2002, in live biomass (Figure 2).   The lower rate of storage in the live trees between 1998 and 2002 may have been caused by significant mortality from an ice storm in 1999.   The increment of standing dead material was very small.

Carbon storage in soils

While carbon storage in forest soils is generally low unless the forest is recovering from a disturbance, carbon stored in the soil is potentially more ‘permanent’ as a result of the way that soils protect carbon.   Based largely on work by colleagues Julia Gaudinski (UC Santa Cruz) and Sue Trumbore (UC Irvine), we measured the rate of carbon storage in the mineral soil and forest floor using bomb radiocarbon techniques.   This technique is based on the fact that atmospheric testing of nuclear weapons in the 1950’s left a discernable 14 C signature in the atmosphere.   This signal can be followed in terrestrial ecosystems as this carbon is removed from the atmosphere by the vegetation, then cycled through the ecosystem as a result of litterfall, mortality, and other carbon fluxes to the soil.

We found that most of the carbon stored in soils over the past 40 years is near the surface of the soil in either forest floor material or so-called ‘light fraction’ material in the surface mineral soils (Figure 3).   Using these results, estimates of carbon inputs to the soil, and a model, we predict that the rates of carbon storage in soils are very low, ranging from 0 to 0.3 megagrams of carbon per hectare per year.   We cannot rule out he possibility that the soil is5 accumulating no carbon at all.   If we take an average value of 0.15 megagrams of carbon per hectare per year, this is only about 10% of the observed carbon uptake and storage in the live vegetation.

Carbon storage in down-dead wood

Carbon can also be stored in dead wood, both standing and live.   This is particularly important in forests that are actively being managed, as management often enhances the production of dead wood (photo 2).   Our inventory measurements suggest little change in carbon stored in standing dead trees over time, but does not include carbon stored in down-dead wood (DDW).   We examine changes in carbon stored in DDW by estimating the decay rate of DDW and input rates of DDW to the soil.

Picture 2: Slash pile from shelterwood harvest at Howland Forest

We use two approaches to quantify the decay rate of DDW.   First, we estimated the rate of respiration from DDW by collecting pieces of dead wood at various stages of decay and measuring the rate of CO 2 production using a portable infra-red gas analyzer (see section on Dead wood respiration).   Based on a strong positive relationship between temperature and respiration rate (see section on Deadwood respiration), annual temperature data, and estimates of the total amount of dead wood, we estimate that annual CO 2 production from DDW is 0.10 megagrams of carbon per year.   Second, we estimated decay rate using radiocarbon measurements to quantify the date of mortality, then combined that with measurements of wood density to quantify changes in density over time.   These results suggest an annual loss of 0.08 megagrams of carbon, which is very similar to our estimate of 0.10 megagrams of carbon based on measurements of DDW respiration.   These are likely slight underestimates of decay as they do not include physical breakdown of the wood.

To determine the net carbon balance of DDW, we also have to estimate the input rates.   We used forest inventory measurements (see previous section on Carbon storage in live and dead standing biomass ) over time to quantify the amount of wood that went from standing to down and dead.   On average, over a 15-year period, the input of DDW was 0.30 megagrams of carbon per year.   When combined with estimates of carbon losses from DDW (0.10 megagrams of carbon per year), our results suggest that DDW may be accumulating at Howland Forest at a rate of about 0.20 megagrams of carbon per year, a number that is similar to that obtained for soils.

Net carbon sequestration

Figure 2: Comparison of net ecosystem C accumulation measured by eddy covariance (blue bars, pink is mean), and biometric estimates of C accumulation in vegetation, dead wood, and soil.

©Woods Hole Research Center, 2007