Journal of Integrative Agriculture ›› 2022, Vol. 21 ›› Issue (2): 375-388.DOI: 10.1016/S2095-3119(20)63437-2

• • 上一篇    下一篇

基于不同方法的ORYZA(v3) 模型参数估计值的差异性 基于不同方法的ORYZA(v3) 模型参数估计值的差异性

  

  • 收稿日期:2020-04-16 接受日期:2020-09-29 出版日期:2022-01-02 发布日期:2022-01-02

Differences in parameter estimates derived from various methods for the ORYZA (v3) Model

TAN Jun-wei1, DUAN Qing-yun2, GONG Wei3, DI Zhen-hua3    

  1. 1 College of Water Resources and Civil Engineering, China Agricultural University, Beijing 100083, P.R.China
    2 College of Water Resources and Hydrology, Hohai University, Nanjing 210098, P.R.China
    3 Institute of Land Surface System and Sustainable Development, Faculty of Geographical Science, Beijing Normal University, Beijing 100875, P.R.China
  • Received:2020-04-16 Accepted:2020-09-29 Online:2022-01-02 Published:2022-01-02
  • About author:Correspondence TAN Jun-wei, E-mail: tanjw@cau.edu.cn
  • Supported by:
    This research was supported by the National Natural Science Foundation of China (NSFC 51909004).  

PDF

摘要：

参数估计对于模型用户来说一直是个棘手的话题。不准确的参数值将会导致模型预测结果不准确。由于“异参同效”的影响和参数估计过程的差异，基于不同参数估计方法得到的模型参数值可能存在一定的差异。因此，分析参数估计的影响因素、比较当前常用参数估计方法的有效性对于更好地进行模型参数率定具有重要意义。本研究采用了3种常规优化方法(SCE-UA、GA、PEST)和2种基于贝叶斯原理的优化方法(GLUE、MCMC-AM)对ORYZA(v3)模型中的9个作物参数进行估计。结果表明，基于不同参数估计方法得到的参数值存在较大的差异且对模型模拟结果具有较大的影响。基于SCE-UA、GA、PEST方法得到的参数值对参数初始值比较敏感，但敏感程度随算法和目标函数的不同而存在明显的差异。 在常规的优化方法中，SCE-UA方法兼备收敛的稳定性和较高的效率，适用性较好。 基于所有优化方法的参数值都显著提高了模型拟合的精度，其中贝叶斯方法的有效性总体上弱于常规方法，但基于MCMC-AM方法得到的参数后验分布对应概率密度最大的参数值(MCMC_Pmax)与常规法方法得出的结果具有相当的有效性，甚至效果更好。此外，模型中物候参数值的准确性在验证期对模拟结果的精度具有较大的影响。

Abstract: Parameter estimation is always a difficult issue for crop model users, and inaccurate parameter values will result in deceptive model predictions.  Parameter values may vary with different inversion methods due to equifinality and differences in the estimating processes.  Therefore, it is of great importance to evaluate the factors which may influence parameter estimates and to make a comparison of the current widely-used methods.  In this study, three popular frequentist methods (SCE-UA, GA and PEST) and two Bayesian-based methods (GLUE and MCMC-AM) were applied to estimate nine cultivar parameters using the ORYZA (v3) Model.  The results showed that there were substantial differences between the parameter estimates derived by the different methods, and they had strong effects on model predictions.  The parameter estimates given by the frequentist methods were obviously sensitive to initial values, and the extent of the sensitivity varied with algorithms and objective functions.  Among the frequentist methods, the SCE-UA was recommended due to the balance between stable convergence and high efficiency.  All the parameter estimates remarkably improved the goodness of model-fit, and the parameter estimates derived from the Bayesian-based methods had relatively worse performance compared to the frequentist methods.  In particular, the parameter estimates with the highest probability density of posterior distributions derived from the MCMC-AM method (MCMC_Pmax) led to results equivalent to those derived from the frequentist methods, and even better in some situations.  Additionally, model accuracy was greatly influenced by the values of phenology parameters in validation.

Key words: parameter estimation , frequentist method , Bayesian method , crop model , calibration