中文摘要：

生态系统光能利用率（LUE）反映了植被通过光合作用利用光能吸收和固定大气中CO2的能力,是表征生态系统生产力的重要指标。选取长白山温带阔叶红松（Pinus koraiensis）林生态系统为研究对象,利用涡度相关通量观测数据,采用直角双曲线方程获取了生态系统光合作用的表观量子效率（ε）;基于总生态系统初级生产力（GEP）与下垫面入射光合有效辐射（Q）的比值得到生态光能利用率（LUEeco）。研究表明：在季节尺度上,ε与LUEeco均表现出显著的单峰变化特征,并主要受到土壤温度和归一化植被指数（NDVI）的调控,同时,ε和LUEeco都受到GEP的显著影响,而与Q的相关性较弱或无显著相关关系,但散射辐射的增加在一定程度上有助于提高生态系统的LUE。ε与LUEeco存在显著的线性正相关关系,但ε明显高于LUEeco。2003–2005年,ε与LUEeco每年最大值的平均值分别为（0.087±0.003）和（0.040±0.002）μmol CO2·μmol photon–1,年际间变异度分别为4.17%和4.25%,而不同年份之间最大差异均达到8%或8%以上,从而对模型模拟结果产生明显影响。因此,在基于光能利用率模型的模拟研究中,最大LUE的年际变异需要在参数反演和优化中给予重要考虑。

英文摘要：

Aims Ecosystem light use efficiency（LUE） reflects the ability of CO2 uptake and light utilization via photosynthesis, which is a key parameter in ecosystem models to evaluate ecosystem productivity. The objectives of this study were to：（1） compare the differences of LUE derived from different methods;（2） elucidate the seasonal dynamics of LUE and its regulatory factors; and（3） evaluate the maximum LUE（LUEmax） and its variability based on eddy-covariance（EC） flux. Methods Using the flux data from an EC tower during 2003–2005 at a broad-leaved Korean pine（Pinus koraiensis） mixed forest, Changbai Mountain, two types of LUE indicators were generated from： 1） the apparent quantum yield（ε） estimated with rectangular hyperbolic curve, and 2） the ecological light use efficiency（LUEeco） calculated as the ratio between gross ecosystem productivity（GEP） and photosynthetically-active radiation（Q）. Important findings The seasonal variation of ε and LUEeco appeared a unimodal pattern within a year, with the variations significantly dominated by soil surface temperature and Normalized Difference Vegetation Index（NDVI）. A positive correlation between GEP and LUE was found for both ε and LUEeco, with the effect of Q on LUE relatively weak. The increase in diffusion radiation appeared favorable for enhanced LUE. Generally, there was a significant positive relationship between ε and LUEeco, while ε was higher than LUEeco, especially duringthe mid-season. The annual maximum value of ε and LUEeco was（0.087 ± 0.003） and（0.040 ± 0.002） μmol CO2·μmol photon–1 over the three years, respectively. The interannual variability of LUEmax for ε and LUEeco was 4.17% and 4.25%, respectively, with a maximum difference of 8%, likely resulted from considerable uncertainty in model simulations. Our results indicated that the inversion and optimization of maximum LUE should be taken seriously in the application of LUE models.