中文摘要：

物候和叶面积指数的季节动态在落叶林中是决定生态系统净生产力的关键因素。尽管物候对能量和CO2通量的影响可以简单地通过描述发芽和落叶的时间以及叶面积指数的季节动态来表示．但是由于对驱动物候的物理过程缺乏全面正确的理解．在陆地生态系统模型中物候就成为最难以参数化的一个过程。目前，在陆地生态系统模型中描述物候主要有两种不同的方法：一种是基于气候变量（主要是温度或积温）的经验方法，即是通过建立物候不同阶段与气候变量的经验关系来预测关键物候事件发生的时间。另一种方法是基于碳吸收的物候参数化方案，物候的任何阶段都和当前的碳平衡相联系。在生态系统模型中，基于碳吸收的物候参数化方法可以大大降低物候模拟的经验性。提高模型的适用性和模拟精度。比基于气候变量的经验模型更适于模拟未来气候变化影响。未来随着生理和分子水平上，对控制物候和LAI动态过程机理的揭示．建立基于过程的物候参数化方案和LAI动态模拟模型就成为生态系统模型或气候模型的发展方向。

英文摘要：

The seasonal phenology of the leaf area index （LAI） is a major determinant of net ecosystem production in deciduous forest ecosystems. In the simplest case, the nology on energy and CO2 fluxes is represented by prescribing leaf onset and effect of leaf pheoffset times or the seasonal evolution of LAI. Leaf phenology remains one of the most difficult processes to be parameterized in terrestrial ecosystem models because the understanding of the physical processes that initiate leaf onset and senescence is incomplete. At present, to describe phenology, the terrestrial models generally use one of two different approaches. One approach uses empirical formulations to estimate the timing of crucial phonological events like leg-on/off dates only based on abiotic variables, especially temperature or growing degree-days （GDDs）. The other approach is a carbon-gain-based scheme. In this approach, the phonological stages are directly determined from the current carbon balance. A carbon-gain-based parameterization of phenology in ecosystem models offers the possibility of reduced empiricism, general applicability, and increased robustness; and it is more suitable for being used in climate change models. In the future, while the mechanisms of processes controlling leaf phenology are clarified with progress being made at the physiological and molecular level, ecosystem models have to set up processbased model of leaf phenology and LAI to promote the robust of simulation results.