Eddy covariance measurements of whole-forest carbon fluxes

With average global temperature continually increasing over the past 100 years and seemingly no abatement in sight, there is particular focus on the potential affects of temperature on terrestrial ecosystems. Forested systems can be a carbon source with increases in temperature (Goulden et al., in press). Tropical systems may be particularly sensitive to changes in temperature and could potentially contribute more carbon from soils to the atmosphere than any other terrestrial ecosystem (Trumbore et al. 1996). In contrast, Grace et al. (1995) argue that tropical ecosystems may actually be accumulating carbon at a rate of 1.0 ton C ha/y. What physical environmental parameter(s) may be controlling the net energy and carbon fluxes in a primary moist tropical lowland forest (TLF)?

The primary objectives of our above-canopy net ecosystem exchange measurements are:

• To determine the sink and source strength of a TLF and attempt to quantify the effects of meteorological variables on carbon and energy fluxes.
• To develop a process level carbon balance model to provide long-term data for global climate change models, and examine the potential consequences of increases in average global temperature.

Our eddy covariance (EC) system, mounted atop a 42m tower (Upright Inc., Selma, CA.) estimates CO2, H2O, sensible heat and momentum fluxes above the canopy (Eq.1). Our system contains three basic components; 3-D sonic anemometer (K-probe, AT Inc., Boulder CO.), a closed-path infra-red gas analyzer (IRGA Model 6262, Li-Cor Inc., Lincoln, NE) and a computer to operate software that processes anemometer signals, computes lag times and fluxes, and stores data.

Because in times of rain (wet transducers), mid-day stalled windspeeds and nighttime laminar flow (U* < 0.2 m sec-1) violates assumptions of turbulent transfer, a profile system is also used to estimate fluxes. Inlets are mounted at six locations (0, 6.8, 11.5, 16, 20.8, and 27 m) and scalar fluxes are calculated assuming that turbulent transfer and molecular diffusivity and analogs (Eq. 2).

Eq. 1

Where,

Fs = flux of scalar (g m-3 hr-1),

r a = mean density of air (g m-3),

W' = instantaneously measured vertical windspeed

C' = instantaneously measured concentration of scalar, and

The bar represents the covariance.

Eq. 2

Where,

k = molecular diffusivity,

c = concentration of scalar, and

z = profile height.

Because these systems are running continuously for a minimum of three years, an automated program cycles calibration gases through the IRGA to check for calibration drift. Moreover, several corrections are made to the measured flux to calculate a 'true' flux from both systems i.e.; density corrections (Webb et al., 1980), power spectra and spectral corrections, and sensor separation.

Fluxes were measured from January 1998 to March of 2000, when funds for this portion of the project were exhausted. Above canopy microclimate continues to be measured to date.

Works cited:

Goulden, M.L., Wofsy, S.C., Harden, J.W., Trumbore, S.E., Crill, P.M., Gower, S.T., Fries, T., Duabe, B.C., Fan, S., Sutton, D.J., Bazzaz, A., Munger, J.W. Sensitivity of boreal forest carbon balance to soil thaw. Science.

Grace, J., Malhi, Y., Lloyd, J., McIntyre, J., Miranda, A.C., Mier, P., Miranda, H.S. (1996) The use of eddy covariance to infer the net carbon dioxide uptake of Brazilian rain forest. Global Change Biology 2:209-217.

Trumbore, S.E., (1996) Science 276: 393-396.

Webb, E.K., Pearman, G.I., Luening, R. (1980) Correction fo flux measurements of density effects due to heat and water vapour transfer. Quarterly Journal of the Royal Meteorological Society 106: 85-100.