Remote sensing approach to estimating net primary productivity of temperate deciduous forest in northeast China

The application of remote sensing in the study of terrestrial ecosystem opens up an effective way for simulating terrestrial productivity. In this paper, a RUE model is established by parameterized plant photosynthesis and respiration. Photosynthetically active radiation absorbed by all surface materials (APAR_SFC), obtained by correcting photosynthetically active radiation observed by MODIS for the effects of surface albedo, atmospheric total transmittance, and periodical change of solar incident angle and solar hour angle during daytime, together with the ratio of PAR absorbed by the canopy to APAR_SFC (RPAR) and radiation use efficiency (RUE) based on a relationship between RUE and maximum rate of leaf photosynthesis (A_max) governed by its dependence on seasonal dynamics of foliage nitrogen content, are used to estimate gross photosynthesis. Plant respiration is divided into growing respiration as constant fraction of GPP, and maintenance respiration from leaf, stem, and root. The model is adapted to calculate the net primary productivity (NPP) of temperate deciduous forest in northeast China, supported by auxiliary data composed of daily MODIS data, atmosphere profile, ice-snow cover and landform of the study area in June 2003. In June 2003, daily mean APAR_SFC of temperate deciduous forest in the study area reaches 8.1 MJ m^-2day^-1, RUE varies from 0.45 to 1.09 gC MJ^-1, and monthly NPP is 0.53 MgC ha^-1, 0.56 MgC ha^-1 for temperate deciduous coniferous forest, 0.51 MgC ha^-1 for temperate deciduous broadleaved forest. Compared with the process-based model, both of them are in agreement with simulated NPP