植物功能性状对动态全球植被模型改进研究进展

Research progresses in improving dynamic global vegetation models(DGVMs) with plant functional traits

动态全球植被模型(dynamic global vegetation models,DGVMs)在模拟和预测陆地生态系统对气候变化响应中表现出很大的不确定性,重要原因之一在于动态全球植被模型将定义植物功能型的性状值设置为常数,忽略了植物功能性状对环境变化的响应.动态全球植被模型现有的植物功能型框架已经严重地阻碍了其发展,因此迫切需要一种新的方法来克服这种局限性.植物功能性状不仅可以反映植物对环境连续变化的响应,而且与生态系统的结构和功能密切相关,可提升当前动态全球植被模型对生态系统过程的模拟和功能的预测.本文从动态全球植被模型发展和植物功能型局限性入手,详细介绍了植物功能性状发展现状及其对动态全球植被模型改进的重要价值,归纳总结了植物功能性状对动态全球植被模型改进的主要方法,并指明植物功能性状对动态全球植被模型改进的发展方向.以期通过凝练植物功能性状在构建下一代动态全球植被模型中发挥作用,推动动态全球植被模型在我国的发展和应用.

Dynamic global vegetation models（DGVMs） are key components of earth system models（ESMs）, which aim to understand ecosystem processes and their interactions with the atmosphere. DGVMs are designed to simulate the structural and functional responses of global vegetation to changes in climate and atmospheric CO2 concentration with the coexistence of plant functional types（PFTs）. PFTs are groups of plant species with similar functions based on morphological, physiological, biochemical, reproductive, and demographic characteristics. DGVMs typically assign each PFT with fixed parameters representing the above characteristics, consequently ignoring their variations within PFT and their responses to environmental changes. This parameterization scheme can simplify plant species represented in DGVMs but inevitably result in large uncertainties in model predictions on ecosystem processes. Therefore, a new generation of DGVMs are urgently needed to overcome the limitations of PFTs by replacing the PFTs parameterization scheme with continuous variation in plant functional traits. Plant functional traits（FTs） are defined as morphological, physiological, and phenological characteristics that affect plants via their effects on the growth, reproduction, and survival. Plant FTs that mediate the structure and function of ecosystems can be implemented into DGVMs as variables rather than PFT-specific parameters with fixed values. The plant FT scheme not only provides a widely applicable approach for forecasting ecosystem shifts and changes in ecosystem structure, but can also be linked with ecosystem functions under a changing climate. This holds great potentials for predicting possible responses of terrestrial ecosystems to environmental changes. This review first introduces the state-of-the-art DGVMs based on PFTs. PFTs represent most of the world＇s vegetation types and characteristics through their functional behaviors and attributes. Although PFT-based DGVMs play a pivotal role in simulating atmosphere-land interactions by quantifying the processes of global carbon, nitrogen, and water cycles, uncertainties arise from inadequate PFT parameters and incomplete PFT classification. For example, many traits vary more within PFTs than between PFTs. Moreover, the values assigned to different PFTs often do not differ much—so little is gained（and much uncertainty is added） compared to a generic model where all plants behave identically. This paper also reviews the recent progress in plant FTs for developing new generations of DGVMs, which includes：（1） descripting adaption strategies of plants to the changing environment,（2） revealing the mechanisms of coexistence between plant species,（3） highlighting the close relationship with the structures and processes of ecosystems, and（4） summarizing their application in the parameterization of vegetation models. We also summarize the main approaches to improve current or build new DGVMs based on FTs, such as building a transitional framework for a PFT-FT hybrid DGVM and building a completely FT-based DGVM. Based on those discussions, several future research directions are recommended for developing a new-generation DGVMs, such as improving the prediction power of traits within the PFTs, building the mechanism expressions between plant adaptations and environments, and standardizing of the data sharing for FTs. We argue that constructing next generation of DGVMs is not simply a matter of incorporating trait-climate relationships, but more importantly to combine optimality concepts and classical vegetation dynamic theories making vegetation modelling more reliable and robust. Botanists, geographers, vegetation modelers, and other relevant scientists should cooperate and share the trait data to elucidate the huge potential of plant FTs in constructing the next generation of more reliable, robust and realistic DGVMs. We hope this review will promote the developments and applications of new generation DGVMs in China.