基于FLUXNET数据集对陆面模式CoLM能量通量的单点评估

模式评估是模式发展中的重要一环。本文利用来自FLUXNET2015数据集的30个站点的涡动相关系统观测数据，重点关注能量通量，对通用陆面模式（Common Land Model version 2014，CoLM2014）在不同典型下垫面的模拟能力进行评估。结果表明，模式总体上能抓住感热、潜热和净辐射通量在日、季节和年平均等不同时间尺度上的变化特征，对感热、潜热和净辐射通量都有较好的模拟能力，净辐射的模拟效果最好，潜热通量次之。季节变化模拟中，感热、潜热通量在夏季不同植被型下站点的空间离散程度大于冬季，不同站点间模拟效果相差较大，净辐射多站点标准差变化幅度要小于感热、潜热，不同站点间模拟效果偏差较小。CoLM在常绿针叶林、稀树林地、草地、农田模拟感热、潜热通量的效果相对较好，在永久湿地、落叶阔叶林下模拟感热通量较差。本研究对CoLM2014在未来的改进和发展中提供了有用的参考。

Evaluating Energy Fluxes of the Common Land Model Based on FLUXNET Dataset

Model evaluation is an important part of model development. In this study, the eddy covariance measurements from 30 FLUXNET sites of FLUXNET2015 dataset were used to evaluate the performance of the Common Land Model, version 2014 (CoLM2014) over different land-cover types, focusing on energy fluxes. The results show that the model captures the variation characteristics of sensible heat flux, latent heat flux, and net radiation flux at different time scales, such as the daily, seasonal, and annual averages, and has a good simulation ability for these fluxes. The simulation effect of net radiation is the best, followed by that of latent heat flux. When simulating seasonal variation, the spatial dispersion degree of sensible and latent heat fluxes under different vegetation types is greater in summer than in winter, and the simulation effect varies greatly among different stations. The variation range of net radiation standard deviation is smaller than sensible and latent heat fluxes, and the deviation of the simulation effect between different stations is small. The evaluation performances for evergreen needleleaf forests, savannas, grasslands, and croplands were relatively good, but poor performances were obtained when simulating the sensible heat flux in wetlands and deciduous broadleaf forests. This study provides a useful reference for improving and developing CoLM2014.