Comparison of global sensitivity analysis techniques based on a process-based model CROBAS.

doi:10.13287/j.1001-9332.202101.005

Abstract

Abstract: Constructions of process or mechanistic models are limited by physiological parameters, due to difficulty in direct and precise measurement. Global sensitivity analysis could evaluate the response of model outputs to changes in physiological parameters, and provide information for improving model structure, data collection, and parameter calibration. Based on a process model CROBAS, 10 parameters related to tree structure of Pinus armandii were selected to compare three widely used global sensitivity analysis methods (the Morris screening method, the variance-based Sobol indices, and the Extended Fourier Amplitude Sensitivity Test (EFAST)), with the objective function formulated by the Nash-Sutcliffe Efficiency (NSE) of tree height and biomass. The results showed that the sensitivity order of parameters slightly varied across different methods, which considerably changed with different objective functions. Both the Morris method and the EFAST method outperformed the Sobol method in terms of time consuming and convergence efficiency. All outputs were sensitive to the maximum rate of canopy photosynthesis per unit area, the specific leaf area, and the extinction coefficient. The light interception of tree canopy played a key role in the simulation of tree growth with CROBAS, suggesting that the module of photosynthetic carbon fixation took priority over any other modules for data collection and model validation during module calibration and tree growth simulation for CROBAS. The calculation and validation of foliage biomass module were crucial when applying carbon balance theory to biomass simulations. In conclusion, for the sensitivity analysis of a complex process-based model, the Morris method was suitable for qualitative studies, while the EFAST method was recommended for quantitative studies.

Key words: process-based model, global sensitivity, Morris, Sobol, EFAST

XUE Hai-lian, TIAN Xiang-lin, WANG Bin, SUN Shuai-chao, CAO Tian-jian. Comparison of global sensitivity analysis techniques based on a process-based model CROBAS.[J]. Chinese Journal of Applied Ecology, 2021, 32(1): 134-144.

References

[1] Petritsch R, Hasenauer H, Pietsch SA. Incorporating forest growth response to thinning within Biome-BGC. Forest Ecology and Management, 2007, 242: 324-336
[2] Korol RL, Running SW, Milner KS. Incorporating inter tree competition into an ecosystem model. Canadian Journal of Forest Research, 1995, 25: 413-424
[3] Saltelli A, Tarantola S, Campolongo F, et al. Sensitivity Analysis in Practice: A Guide to Assessment Scientific Models. Chichester, UK: John Wiley and Sons Ltd. 2004
[4] Mahamah DS. Simplified sensitivity analysis applied to a nutrient-biomass model. Ecology Modelling, 1988, 42: 103-109
[5] Morris MD. Factorial sampling plans for preliminary computational experiments. Technometrics, 1991, 33: 161-174
[6] Sobol’I M. Sensitivity estimates for nonlinear mathema-tical models. Matematicheskoe Modelirovanie, 1990, 2: 112-118
[7] Cukier RI, Fortuin CM, Shuler KE, et al. Study of the sensitivity of coupled reaction systems to uncertainties in rate coefficients. Ⅰ. Theory. Journal of Chemical Physics, 1973, 59: 3873-3878
[8] Cukier RI, Levine HB, Shuler KE. Nonlinear sensitivity analysis of multi parameter model systems. Journal of Computational Physics, 1978, 26: 1-42
[9] Cukier RI, Schaibly JH, Shuler KE. Study of the sensitivity of coupled reaction systems to uncertainties in rate coefficients. Ⅲ. Analysis of the approximations. Journal of Chemical Physics, 1975, 63: 1140-1149
[10] Campolongo F, Cariboni J, Saltelli A. An effective screening design for sensitivity analysis of large models. Environmental Modelling and Software, 2007, 22: 1509-1518
[11] Saltelli A. Making best use of model evaluations to compute sensitivity indices. Computer Physics Communications, 2002, 145: 280-297
[12] Saltelli A, Tarantola S, Chan KPS. A quantitative model-independent method for global sensitivity analysis of model output. Technometrics, 1999, 41: 39-56
[13] 何丽鸿, 王海燕, 雷相东. 基于BIOME-BGC模型的长白落叶松林净初级生产力模拟参数敏感性. 应用生态学报, 2016, 27(2): 412-420 [He L-H, Wang H-Y, Lei X-D. Parameter sensitivity of simulating net primary productivity of Larix olgensis forest based on BIOME-BGC model. Chinese Journal of Applied Ecology, 2016, 27(2): 412-420]
[14] Minunno F, van Oijen M, Cameron DR, et al. Selecting parameters for Bayesian calibration of a process-based model: A methodology based on canonical correlation analysis. SIAM/ASA Journal on Uncertainty Quantification, 2013, 1: 370-385
[15] Song X, Bryan BA, Almeida AC, et al. Time-dependent sensitivity of a process-based ecological model. Ecological Modelling, 2013, 265: 114-123
[16] Salvador R, Pinol J, Tarantola S, et al. Global sensitivity analysis and scale effects of a fire propagation model used over Mediterranean shrublands. Ecological Modelling, 2001, 136: 175-189
[17] Ewen S. Evaluation and Calibration of the CroBas-PipeQual Model for Jack Pine (Pinus banksiana Lamb) Using Bayesian Melding Hybridization of a Process-based Forest Growth Model with Empirical Yield Curves. Master Thesis. Québec, Canada: Laval University, 2013
[18] Mäkelä A. A carbon balance model of growth and self-pruning in trees based on structural relationships. Forest Science, 1997, 43: 7-24
[19] 李悦黎, 杜纪山. 火地塘教学实验林场森林资源的数据分析及经营对策. 西北林学院学报, 1993, 8(3): 53-58 [Li Y-L, Du J-S. Analysis and management strategy of forest resources of Huoditang teaching and experimental forest farm. Journal of Northwest Forestry College, 1993, 8(3): 53-58]
[20] 陈存根, 彭鸿. 华山松林木的生长与分化. 西北林学院学报, 1994, 9(2): 1-8 [Chen C-G, Peng H. Growth and differentiation of Amandi pine trees. Journal of Northwest Forestry College, 1994, 9(2): 1-8]
[21] Shinozaki K, Yoda K, Hozumi K, et al. A quantitative analysis of plant form-the pipe model theory: I. Basic analyses. Japanese Journal of Ecology, 1964, 14: 97-105
[22] Reineke PJ. Perfecting a stand-density for even-aged forests. Agricultural Research, 1933, 46: 627-638
[23] 陈存根, 彭鸿. 秦岭火地塘林区主要森林类型的现存量和生产力. 西北林学院学报, 1996, 11(suppl.): 92-102 [Chen C-G, Peng H. Standing crops and productivity of the major forest-types at the Houditang forest region of the Qinling Mountains. Journal of Northwest Forestry College, 1996, 11(suppl.): 92-102]
[24] 梁军生, 陈晓鸣, 杨子祥, 等. 云南松与华山松人工混交林针叶光合速率对光及CO₂浓度的响应特征. 林业科学研究, 2009, 22(1): 22-25 [Liang J-S, Chen X-M, Yang Z-X, et al. Photosynthesis rate in response to light intensity and CO₂ concentration in the mixed plantation of Pinus yunnanensis and Pinus armandii. Forest Research, 2009, 22(1): 22-25]
[25] Pappenberger F, Iorgulescu I, Beven KJ. Sensitivity analysis based on regional splits and regression tree (SARA-RT). Environmental Modeling and Software, 2006, 21: 976-990
[26] 梁保松, 朱景乐, 王军辉, 等. Pilodyn在华山松活立木木材材性估测中的应用. 南京林业大学学报: 自然科学版, 2008, 32(6): 97-101 [Liang B-S, Zhu J-L, Wang J-H, et al. The application of the Pilodyn to assess wood traits of living trees in Pinus armandii. Journal of Nanjing Forestry University: Natural Science, 2008, 32(6): 97-101]
[27] Mohren MJ. Simulation of Forest Growth, Applied to Douglas Fir Stands in the Netherlands. PhD Thesis. Wageningen: Wageningen Agriculture University, 1987
[28] 张小全. 森林细根生产和周转研究. 林业科学, 2001, 37(3): 126-138 [Zhang X-Q. Fine-root production and turnover for forest ecosystems. Scientia Silvae Sinicae, 2001, 37(3): 126-138]
[29] 刘华, 侯琳, 雷德瑞. 秦岭火地塘林区油松和华山松林的空间分布格局及碳储量与碳密度研究. 中国生态农业学报, 2007, 15(1): 5-8 [Liu H, Hou L, Lei D-R. Carbon storage and carbon density of Pinus tabuliformis and Pinus armandii forests at Huoditang forest region in Qinling Mountain. Chinese Journal of Eco-Agriculture, 2007, 15(1): 5-8]
[30] 时忠杰, 王彦辉, 徐丽宏, 等. 六盘山华山松(Pinus armandii)林降雨再分配及其空间变异特征. 生态学报, 2009, 29(1): 76-85 [Shi Z-J, Wang Y-H, Xu L-H, et al. Rainfall redistribution and its spatial variation in the stand of Pinus armandii in the Liupan Mountains, China. Acta Ecologica Sinica, 2009, 29(1): 76-85]
[31] 宋子炜, 郭小平, 赵廷宁, 等. 北京山区油松林光辐射特征及冠层结构参数. 浙江林学院学报, 2009, 26(1): 38-43 [Song Z-W, Guo X-P, Zhao T-N, et al. Light environment and canopy structure of a Pinus tabuliformis community in the mountainous area of Beijing. Journal of Zhejiang Forestry College, 2009, 26(1): 38-43]
[32] Zeide B, Pfeifer P. A method for estimation of fractal dimension of tree crowns. Forest Science, 1991, 37: 1253-1265
[33] 徐崇刚, 胡远满, 常禹, 等. 生态模型的灵敏度分析. 应用生态学报, 2004, 15(6): 1056-1062 [Xu C-G, Hu Y-M, Chang Y, et al. Sensitivity analysis in ecological modeling. Chinese Journal of Applied Ecology, 2004, 15(6): 1056-1062]
[34] Nash JE, Sutcliffe JV. River flow forecasting through conceptual models part Ⅰ. A discussion of principles. Journal of Hydrology, 1970, 10: 282-290
[35] Yang J. Convergence and uncertainty analyses in Monte-Carlo based sensitivity analysis. Environmental Modelling and Software, 2011, 26: 444-457
[36] Shin MJ, Guillaume JHA, Croke BFW, et al. Addressing ten questions about conceptual rainfall-runoff models with global sensitivity analyses in R. Journal of Hydrology, 2013, 503: 135-152
[37] Poorter H, Nagel O. The role of biomass allocation in the growth response of plants to different levels of light, CO₂, nutrients and water: A quantitative review. Australian Journal of Plant Physiology, 2000, 27: 595-607
[38] Le Roux X, Lacointe A, Escobar-Gutierrez A, et al. Carbon-based models of individual tree growth: A critical appraisal. Annals of Forest Science, 2001, 58: 469-506
[39] Mäkelä A, Hari P. Stand growth model based on carbon uptake and allocation in individual trees. Ecological Modelling, 1986, 33: 205-229
[40] Dietze MC. Ecological Forecasting. Princeton, NJ, USA: Princeton University Press, 2017