Due to climate differences, an extreme range in productivity occurs along a 250—km, west—east transect at °44° north latitude in western Oregon, USA, where coniferous evergreen forests dominate. As part of the Oregon Transect Ecosystem Research (OTTER) project, our objective was to evaluate how climate constrains net primary production (NPP) by limiting the utilization of intercepted photosynthetically active radiation (IPAR). The forests measured along the transect intercepted from 22% to 99.5% of the incident PAR. With data collected from recording meteorological stations installed near each site, we defined the hourly conditions when photosynthesis was partly or completely limited by drought, extreme humidity deficits, or frost. From this analysis we calculated that the fraction of incident PAR that could be utilized throughout the year ranged from 92% in the coastal rainforests to <25% in the juniper woodland. NPP varied from 3 to 26 Mg°ha—1°yr—1 with the fraction of belowground NPP, estimated from litterfall, increasing from 20% to 60% of the total as the environment becomes harsher. Light—use efficiency (°u) calculated under conditions when the environment did not constrain photosynthesis, averaged 0.8 g/MJ for aboveground NPP and 1.3 g/MJ for total NPP.

Remotely sensed data acquired from four remote—sensing instruments on three different aircraft platforms over a transect of coniferous forest stands in Oregon were analyzed with respect to seasonal leaf area index (LAI). Data from the four instruments were corrected for the varying seasonal and geographic atmospheric conditions present along the transect. Strong logarithmic relationships were observed between seasonal maximum and minimum LAI and the simple ratio (SR) (near infrared/red reflectance) calculated from the broad—spectral—band Thematic Mapper Simulator (TMS), as well as from the narrow—spectral—band Airborne Visible/Infrared Imaging Spectrometer (AVIRIS), the Compact Airborne Spectrographic Imager (CASI), and the Spectrom SE590 spectro—radiometer (R2 = 0.82—0.97). The TMS SR reached an asymptote at an LAI of °7—8. However, the SE590 and the CASI SR continued to increase up to the maximum LAI of 10.6. The variability of the relationship between the AVIRIS SR and LAI increased at stands with LAIs >7, making a trend in the AVIRIS SR—LAI relationship at LAIs >7 difficult to discern. The SRs of the coniferous forest stands measured by the narrow—spectral—band instruments were higher than they were from the broad—spectral—band TMS. This is attributed partially to the integration of the TMS over a broad wavelength region in the red and more strongly to calibration differences between the sensors. Seasonal TMS SR trends for four time periods for some of the stands deviated from the expected seasonal LAI trends, possibly because of smoke and very low sun angles during some of the acquisition periods. However, the expected SR differences for the seasonal minimum and maximum LAI were observed for all of the sensors for nearly all of the forest stands. This study demonstrates that remotely sensed data from both broad— and narrow—spectral—band instruments can provide estimates of LAI for use in forest ecosystem simulation models to estimate evapotranspiration, photosynthesis, canopy turnover, and net primary production over large areas.