Estimating gross photosynthesis using satellite and ancillary data: Approach and preliminary results [Review]

摘要

Observations by space-borne sensors can provide photosynthetically active radiation (PAR) intercepted by a canopy (IPAR), which, together with radiation zee efficiency (RUE), can be used to estimate carbon assimilation by terrestrial plant communities. Evaluation of biological and environmental factors determining variability of RUE for daily total gross photosynthesis within and among species suggests a biophysically based approach for estimating daily total grass photosynthesis using satellite and ancillary data. Calculations have been done using leaf photosynthesis and canopy characteristics derived from field measure ments for six cultivated herbaceous and 12 uncultivated (herbaceous and needle leaf and broadleaf woody) species, and mixed species stands of temperate deciduous forest and tropical moist and dry forests. These data are for 65 canopies, divided about equally between cultivated and uncultivated species, and the leaf area index varies between 0.6 and 9.5. A model for instantaneous canopy photosynthesis, considering sunlit and shaded leaves, variation of maximum rate of leaf photosynthesis within the canopy, and a solution of the radiative transfer equation for propagation of direct and diffuse PAR, is numerically integrated in space (angular variation of radiances and depth within the canopy) and time (diurnal variation) to obtain daily total gross photosynthesis and IPAR for varied incident irradiances. The results show that the maximum rate of leaf photosynthesis and magnitudes of direct and diffuse irradiance incident on the canopy are the major factors determining RUE for unstressed conditions. Also, while woody species on the average have lower RUE, they are found to utilize diffuse radiation more efficiently them cultivated species. Comparison of calculated RUE with observations for 10 cultivated species in varied environment gave differences of -23% to +11% for individual species, but aggregately they have almost equal coefficient of variation (similar to 25%) and their averages differ by -6% (underestimation), Monthly and annual gross photosynthesis determined using satellite and ancillary data are obtained for rainforest (in Brazil), boreal forest (in Canada), and cropland (in the United States). Annual photosynthesis calculated using satellite data for the three biomes differed from, that determined from field measurements by -29% to +15%, averaging to +6%. The above error estimates for RUE and photosynthesis are obtained without calibrating any model parameter. (C)2001 Published by Elsevier Science Ireland Ltd. [References: 142]