During 1995/1997 Westgec provided funds to research the use of remotely sensed data for scaling up ecophysiological models. The study is being conducted at three study sites in Washington State: a clonal Populus stand near Wallula, old-growth and second-growth silver fir (Abies amabilis) stands in the Cedar River Watershed and Old-growth mixed Douglas Fir/Hemlock at the Wind River Canopy Crane site. A key question we are addressing is how plant physiological function varies depending on leaf position, canopy architecture and the spectral quality of vertically distributed light. To address this question we have measured the spectral quality of light as it varies vertically in architecturally diverse canopies. We have also measured spectral reflectance at leaf to canopy scales and the chemistry (pigment and water content) of leaves sampled from a diversity of lighting environments. Our objective is to use these measures to improve, or develop, new models of canopy photosynthesis that account for the effects of leaf position, light and architecture. Field and laboratory data to support the study were obtained in three field campaigns, one in late summer of 1995, a second in mid-summer, 1996 and a third in early summer 1997. In this report we focus on results obtained from the 1996 field campaign conducted at the Wind River Canopy Crane Site (Fig 1).

Field measures at the Crane Site included destructive leaf samples, above canopy reflectance and vertical profiles of diffuse visible/NIR light. Profiles were collected in 10 gaps at the Crane Site, spanning a range from closed Douglas fir to open hemlock. Spectra were collected between 400 and 2250 nm, providing the first measurements of the spectral quality of light across such a wide spectral range (Fig. 2). The most significant patterns observed moving down the profiles included a decrease in diffuse PAR, increase in diffuse NIR and increase in the depth of liquid water bands at 980 and 1200 nm. In order to investigate the role of canopy structure in modifying light quality, several spectral indices were developed. These included 550 nm diffuse downwelling light (analogous to PAR measurements), a ratio of NIR over red (sensitive to leaf area), a 1650/830 nm ratio (sensitive to the presence of trunks, branches and bark) and a measure of the depth of the 1200 nm water band (Fig. 3). These measures showed an exponential decrease in PAR, and increases in the depth of liquid water and NIR/red ratio with Douglas fir showing a relatively uniform increase in leaf area down the profile and hemlock showing a greater concentration of leaves towards the base. Significant differences were also observed in the 1650/830 nm ratio, which was highest in closed canopies suggesting a greater ratio of branches and trunks to leaves. Spectral measures were complemented by hemispherical photographs, a light bar and a LAI-2000 Plant Canopy Analyzer. When combined with leaf chemical measurements and field measures of photosynthesis we should be able to use this information to develop models relating light quality, vertical position and architecture to photosynthetic function across a gradient of forest types. These results were presented at the 1997 Annual meeting of the Ecological Society of America (Roberts et al., 1997a).

To extrapolate our field measures to larger scales we are using the Advanced Visible/Infrared Imaging Spectrometer (AVIRIS: Fig. 1). At each study site we are using AVIRIS to map canopy liquid water content, quantify canopy composition and texture and map the distribution of water vapor across the landscape (Roberts et al., 1997b). Early results were presented in Roberts et al., (1995) and Hinckley et al., (1996a/b). Above canopy spectral reflectance measured along transects at the Wind River Crane in 1996 were used to improve the reflectance retrieval and to develop a regionally specific spectral library for the Pacific Northwest (Roberts et al., 1997c: Fig. 4). Potential applications of such a library include mapping species dominants using multiple endmember spectral mixture analysis (Roberts et al., 1997d). For example, hemlock and Douglas fir could be distinguished along the transect on the basis of spectral differences in which hemlock had lower visible reflectance, higher NIR and lower SWIR reflectance than Douglas fir. These observations are consistent with a higher foliage to branch ratio in hemlock.

References:

Heilman, P.E., Hinckley, T.M., Roberts, D.A., Isebrands, J.G., and Ceulemans, R., Chapter in Production Physiology, in press.

Hinckley, T., Sprugel, D., Brooks, J.R., Brown, K.J., Martin, T.A., Roberts, D.A., Segura, G., Schaap, W., and Wang, D., 1996, Scaling and Integration in Trees, Ch. 15 in Ecological Scale: Theory and Applications, (Peterson D., and Parker, V.T., Eds.), in press.

Roberts, D.A., Brown, K.J., Hinckley, T.M., Green, R.O., and Ustin, S.L., 1995, Remote Estimates of Canopy Coupling and Architecture Using an Imaging Spectrometer: Conifer vs. Hardwood, 1995 Ecological Society of America, Snow Bird, Utah, July 30-Aug 3, 1995.

Roberts, D.A., Ustin, S., Waller, E., Brown, K., and Hinckley, T., 1997a, Spectral Quality of Vertically Distributed Visible and Near-Infrared Light in Old Growth Forest, in Bull. Ecol. Soc. Amer. Program & Abstracts, 82nd Annual ESA Meeting, Albuquerque, New Mexico, Aug. 10-14, 1997

Roberts, D.A., Green, R.O., and Adams, J.B., 1997b,Temporal and Spatial Patterns in Vegetation and Atmospheric Properties from AVIRIS, Remote Sens. Environ., in press.

Roberts, D.A., Gardner, M., Church, R., Ustin, S.L., and Green, R.O., 1997c, Optimum Strategies for Mapping Vegetation using Multiple Endmember Spectral Mixture Models, in SPIE Conf. Vol 3118, Imaging Spectrometry III, 12 p., San Diego, CA July 27-Aug 1, 1997.

Roberts, D.A., Gardner, M., Church, R., Ustin, S., Scheer, G.,and Green, R.O., 1997d, Mapping Chaparral in the Santa Monica Mountains using Multiple Endmember Spectral Mixture Models, Rem. Sens. Environ. 65:267-279.

Graduate Students Supported:

Eric Waller: Winter 1996 to Fall 1996
Keir Keightley: Winter 1997 to present.