植物生态学报 ›› 2013, Vol. 37 ›› Issue (6): 492-502.DOI: 10.3724/SP.J.1258.2013.00051

• 研究论文 • 上一篇    下一篇

流域水碳过程耦合模拟——WaSSI-C模型的率定与检验

刘宁1,孙鹏森1,*(),刘世荣1,孙阁2   

  1. 1中国林业科学研究院森林环境与保护研究所, 国家林业局森林生态环境重点实验室 北京 100091
    2Eastern Forest Environmental Threat Assessment Center, USDA Forest Service, Raleigh, NC 27606, USA
  • 收稿日期:2013-01-09 接受日期:2013-03-27 出版日期:2013-01-09 发布日期:2013-06-05
  • 通讯作者: 孙鹏森
  • 基金资助:
    国家科技支撑项目(2012BAD22B01)

Coupling simulation of water-carbon processes for catchment―calibration and validation of the WaSSI-C model

LIU Ning1,SUN Peng-Sen1,*(),LIU Shi-Rong1,SUN Ge2   

  1. 1Key Laboratory of Forest Ecology and Environment of the State Forestry Administration, Institute of Forest Ecology, Environment and Protection, Chinese Academy of Forestry, Beijing 100091, China
    2Eastern Forest Environmental Threat Assessment Center, USDA Forest Service, Raleigh, NC 27606, USA
  • Received:2013-01-09 Accepted:2013-03-27 Online:2013-01-09 Published:2013-06-05
  • Contact: SUN Peng-Sen

摘要:

在复杂的气候变化条件下, 利用水碳耦合模型进行生态水文学研究成为主要的研究手段和途径。该文以杂谷脑河上游流域为例, 在确定生态水文模型WaSSI-C模拟尺度的基础上, 探讨水碳耦合模型在中国西南湿润地区的适用性。杂古脑河上游流域位于岷江上游, 隶属于长江流域。在分析和讨论了模型结构和机理的基础上, 分别对模型蒸散和融雪计算进行了补充改进, 以提高模型的适用性。将1988-1996年作为模型的率定期, 1997-2006年作为模型的验证期, 分别在率定期和验证期利用实测的径流数据和中分辨率成像光谱仪数据的总初级生产力、蒸散(ET)数据, 对模拟结果进行对比验证。并利用决定系数(R2)和Nash-Sutcliffe效率系数(NS)两个指标对模拟效果进行评价。流域总径流率定期和验证期对比验证的R2分别为0.86和0.78; NS分别为0.82和0.67。总生态系统生产力和ET验证期的R2分别为0.89和0.78。可见模型模拟结果的两个评价指标都处于较为理想的区间内, 说明WaSSI-C模型在研究区内具有较好的适用性。并对模型的蒸散计算方法进行了讨论, 在此基础上提出了模型中存在的问题和改进的方向。

关键词: 生态水文模型, 模型率定, 模型验证, WaSSI-C模型

Abstract:

Aims The water-carbon coupling model has become a major approach of eco-hydrology study under the complicated situation of climate changes. Our objective was to explore the applicability of the WaSSI-C model for the humid region in southwestern China.
Methods We focused on a large watershed, the upper Zagunao River watershed, situated in the upper reach of the Minjiang River, Yangtze River Basin. Applicability of the WaSSI-C model was assessed by evaluating the modeling results with two indexes: determination coefficient (R2) and Nash-Sutcliffe efficiency coefficient (NS). In order to make it better suited for the study area, two main improvements were introduced regarding to the snowmelt and evapotranspiration (ET) calculation. The calibration period was set as 1988-1996, and the validation period was set as 1997-2006. We validated the model using observed runoff and gross ecosystem productivity (GEP) and ET production of moderate resolution imaging spectroradiometer (MODIS) in the calibration and validation periods, respectively.
Important findings There was a good performance of our application of the WaSSI-C model in the study area. The R2of calibration period and NS for the total runoff of the study watershed were 0.86 and 0.82, respectively, and 0.78 and 0.67, respectively, for the validation period. Moreover, the R2 of the GEP and ET were 0.89 and 0.78, respectively. We discussed the modeling process and results, and the future direction of improvements of the ET calculation in the end.

Key words: eco-hydrological model, model calibration, model validation, WaSSI-C model