ASPEN BARK PHOTOSYNTHESIS AND ITS SIGNIFICANCE TO REMOTE-SENSING AND CARBON BUDGET ESTIMATES IN THE BOREAL ECOSYSTEM

Aspen bark was investigated for photosynthetic function, pigment content, and spectral characteristics during the 1993-1994 Boreal Ecosystem-Atmosphere Study (BOREAS) summer field campaigns in the boreal zone of Saskatchewan, Canada. Parameters related to photosynthetic function were similar for bark and leaves: chlorophyll (Chl) concentration; fluorescence responses; and spectral reflectance. Similar increases along a vertical gradient from base to tree top were observed for incident photosynthetically active radiation (PAR), photosynthetic pigment content, photosynthetic capacity, and spectral reflectance variables. Since transmittance of aspen bark periderm was 20-30% in the blue, and 50-60% in the red Chl absorption bands, the PAR available to the photosynthetic cortical layer in the natural, canopy environment (<1000 mu mol m(-2) s(-1)) was sufficient to support positive net assimilation (<8-10 mu mol CO2 m(-2) s(-1)) under ideal conditions (e.g., light, temperature, saturating CO2), a rate approximately 30-50% that of leaves. However, the respiring tissues comprising the greater fraction of bark tissue bias the balance of CO2 exchange in favour of respiration for the whole bark. Therefore, net photosynthesis under ambient conditions on the whole bark was, in general, negative. The total bark surface area was estimated to contain 17-40% of the whole tree Chl. The contribution of the bark surface area fraction of the full canopy (leaves plus bark) increased with age (<60 years), with a similar trend expected for bark in total tree (and stand) photosynthesis. A spectral reflectance variable, the red edge inflection point (REIP), was related to total bark Chl content (r(2)=0.74). A better predictive relationship (r(2)=0.82) for total bark Chl was observed using a spectral index calculated from the reflectance ratio of two narrow wavebands (R3/R2: R2 and R3 are between 0.715-0.726 mu m and 0.734-0.747 mu m, respectively), which may have greater utility in landscape remote sensing. The bark spectra for Chl-containing bark should improve understanding of carbon balance in aspen forests, based on landscape-level radiative transfer simulations.