تاثیر انتخاب تابع درستنمایی در تخمین عدم قطعیت مدل شبیه سازی سیلاب HEC-HMS با استفاده از الگوریتم مونت کارلو زنجیر مارکوف

نوع مقاله : مقاله پژوهشی

نویسندگان

1 دانشجوی دکتری /آبیاری و زهکشی, پردیس بین الملل دانشگاه فردوسی مشهد

2 استاد/ گروه مهندسی آب، دانشکده کشاورزی، دانشگاه فردوسی مشهد

3 استادیار/گروه مهندسی آب، دانشکده کشاورزی، دانشگاه بیرجند

4 استاد/ گروه مهندسی آب، دانشکده کشاورزی، دانشگاه فردوسی مشهد.

چکیده

در تحقیق حاضر از الگوریتم DREAM(ZS) (از الگوریتم‌های مبتنی بر مونت کارلو زنجیره مارکوف) به‌منظور بررسی عدم قطعیت پارامترهای مدل-هیدرولوژیکی HEC-HMS در حوزه‌آبخیز تمر به مساحت 1530کیلومتر‌مربع واقع در استان گلستان استفاده شد. از سه رویداد برای واسنجی و یک رویداد در اعتباریابی استفاده گردید و تعداد 24 پارامتر واسنجی برای کل حوزه درنظر گرفته شد. همچنین تأثیر 5 تابع درستنمایی بر روی نتایج روش DREAM(ZS) ارزیابی گردید. توابع درستنمایی L1 تا L3 به‌عنوان توابع درستنمایی غیرصریح (informal) و توابع درستنمایی L4 و L5 به عنوان توابع درستنمایی صریح (formal) در نظر گرفته شدند. تابع درستنمایی L1، راندمان ناش ساتکلیف (NS) می‌باشد. L2، حداقل میانگین مربعات خطا است. تابع درستنماییL3، از واریانس خطای تخمین مدل استفاده می‌کند. تابع درستنمایی L4، ارتباط بین برازش حداقل مربعات استاندارد (SLS) و استنباط بیزی را مشخص می‌کند. در تابع درستنمایی L5، وابستگی پیاپی خطاهای باقی‌مانده با استفاده از مدل خودرگرسیون مرتبه اول باقی‌مانده‌های خطا (AR) محاسبه می‌شود. نتایج نشان داد که حساسیت پارامترها وابسته به انتخاب تابع درستنمایی بوده و حساسیت همه پارامترها در برابر توابع مختلف درستنمایی یکسان نیستند. بیشتر پارامترها توسط تابع درستنمایی L4 و L5 بهتر تعیین شده و حساسیت بالایی را به عملکرد مدل نشان دادند. مقدار فاکتور P عدم قطعیت کل نشان داد که 75 تا 100 درصد مشاهدات دربازه‌های عدم اطمینان 95% پیش‌بینی مدل قرار می گیرد. نتایج بررسی معیارهای ارزیابی عدم قطعیت شامل فاکتورP، فاکتور R، RMSE، KGEو NS نشان داد که عملکرد DREAM(ZS) با توابع درستنمایی L4 و L5 بهتر از توابع دیگر درستنمایی است.

کلیدواژه‌ها

موضوعات


عنوان مقاله [English]

Effect of Likelihood Function Choice for Estimating Uncertainty of HEC-HMS Flood Simulation Model Using Markov Chain Monte Carlo Algorithm

نویسندگان [English]

  • M Nourali 1
  • B Ghahraman 2
  • M Pourreza Bilondi 3
  • K Davary 4
1 Ph.D. Candidate of Irrigation and Drainage, International Campus, Ferdowsi University of Mashhad, Iran.
2 Professor, Department of Water Engineering, Faculty of Agriculture, Ferdowsi University of Mashhad, Iran
3 Assistant Professor, Department of Water Engineering, College of Agricultural, University of Birjand, Iran
4 Ferdowsi Professor, Department of Water Engineering, Faculty of Agriculture, Ferdowsi University of Mashhad, Iran.
چکیده [English]

In the present study, DREAM(ZS), Differential Evolution Adaptive Metropolis combined is used to investigate uncertainty of parameters of the HEC-HMS model in Tamar watershed (1530 km2), Golestan province. In order to assess the uncertainty of 24 parameters used in HMS, three flood events were used to calibrate and one flood event was used to validate the posterior distributions. Moreover, performance of five different likelihood functions (L1–L5) was assessed by means of DREAM(ZS) approach. Three likelihood functions, L1‌‌‌–L3 is considered as informal, whereas remaining (L4 and L5) is represented in formal category. Likelihood function L1 is Nash–Sutcliffe (NS) efficiency. L2 is based on minimum mean square error. L3 uses estimation of model error variance and L4 focuses on the relationship between the traditional least squares fitting and the Bayesian inference. Finally, in likelihood function L5, serial dependence of residual errors is accounted using a first-order autoregressive (AR) model of the residuals. According to the results, sensitivities of the parameters strongly depend on the likelihood function, and vary for different likelihood functions. Most of the parameters were better defined by likelihood functions L4 and L5 and showed a high sensitivity to model performance. By calculating uncertainty assessment indicator (P-factor), 95% total prediction uncertainty ranges covers 75–100% of observed data.
Considering all the statistical indicators and criteria of uncertainty assessment, including RMSE, KGE, NS, P-factor and R-factor, results showed that DREAM(ZS) algorithm performed better under formal likelihood functions L4 and L5.

کلیدواژه‌ها [English]

  • Uncertainty
  • DREAM(ZS) Algorithm
  • Likelihood function
  • HEC-HMS
  • First-order autoregressive
Abbaspour KC (2011) User manual for SWAT-CUP4. SWAT calibration and uncertainty Programs, Swiss Federal Institute of Aquatic Science and Technology, Eawag, Duebendorf, Switzerland, 103p
Abebe NA, Ogden FL, Pradhan NR (2010) Sensitivity and uncertainty analysis of the conceptual HBV rainfall–runoff model:Implications for parameter estimation. Journal of Hydrology 389:301–310
Alazzy AA, Lü H, Zhu Y (2015) Assessing the uncertainty of the Xinanjiang rainfall-runoff model: effect of the likelihood function choice on the GLUE method. Journal of Hydrologic Engineering 20(10):04015016
Alizadeh, A (2000) Principles of Applied Hydrology. Imam Reza University Press, Mashhad, Iran, 622p (In Persian)
Baltas EA, Dervos NA, Mimikou MA (2007) Technical note: Determination of SCS initial abstraction ratio in an experimental watershed in Greece. Hydrology and Earth System Sciences 11(6):1825-1829
Bates BC, Campbell EP (2001) A Markov Chain Monte Carlo scheme for parameter estimation and inference in conceptual rainfall-runoff modeling. Water Resources Research 37(4):937–947
Beven KJ, Binley AM (1992) The future of distributed models: Model calibration and uncertainty prediction. Hydrological Processes 6(3):279– 298
Beven KJ (2006) A manifesto for the enquiringly thesis. Journal of Hydrology 320:18-36
Blasone RS (2007) Parameter estimation and uncertainty assessment in hydrological modeling. Ph.D. Thesis, Institute of Environment & Resources, Technical University of Denmark, 55p
Box GEP, Tiao GC (1992) Bayesian inference in statistical analysis. Wiley-Interscience, New York, USA, 608 p
DeChant CM, Moradkhani H (2012) Examining the effectiveness and robustness of sequential data assimilation methods for quantification of uncertainty in hydrologic forecasting. Water Resources Research 48:W04518
Dotto CBS, Kleidorfer M, Deletic A, Rauch W,  McCarthy DT, Fletcher TD (2011) Performance and sensitivity analysis of stormwater models using a Bayesian approach and long-term high resolution data. Environmental Modelling & Software 26(10):1225– 1239
Freer J, Beven KJ, Ambroise B (1996) Bayesian estimation of uncertainty in runoff prediction and the value of data: An application of the GLUE approach. Water Resources Research 32:2161–2173
Freni G, Mannina G, Viviani G (2009) Uncertainty in urban stormwater quality modelling: The influence of likelihood measure formulation in the GLUE methodology. The Science of the Total Environment 408(1):138–145
Gao GY, Fu BJ, Lu YH, Liu Y, Wang S, Zhou J (2012) Coupling the modified SCS-CN and RUSLE models to simulate hydrological effects of restoring vegetation in the Loess Plateau of China. Hydrology and Earth System Sciences 16(7):2347–2364
Gelman A, Rubin DB (1992) Inference from iterative simulation using multiple sequences. Statistical Science 7(4):457 – 472
Gupta HV, Kling H, Yilmaz KK, Martinez GF (2009) Decomposition of the mean squared error and NSE performance criteria: Implications for improving hydrological modeling. Journal of Hydrology 377(1):80–91
He J, Jones JW, Graham WD, Dukes MD (2010) Influence of likelihood function choice for estimating crop model parameters using the generalized likelihood uncertainty estimation method. Agricultural Systems 103:256–264
Heidari A, Saghafian B, Maknoon R (2006) Assessment of flood forecasting lead time based on generalized likelihood uncertainty estimation. Stochastic Environmental Research and Risk Assessment 20(5):363–380
Iran Water Research Institute, Water Resources Department (IWRI) (2008) Report on hydrologic model calibration: Gorganroud flood warning system project. Tehran, Iran (In Persian)
Jin XL, Xu CY, Zhang Q, Singh VP (2010) Parameter and modeling uncertainty simulated by GLUE and a formal Bayesian method for a conceptual hydrological model. Journal of Hydrology 383(3–4):147–155
Kamali B, Mousavi SJ, Abbaspour KC (2013) Automatic calibration of HEC-HMS using single-objective and multi-objective PSO algorithms. Hydrological Processes 27(26):4028–4042
Kavetski D, Kuczera G, Franks SW (2006) Bayesian analysis of input uncertainty in hydrological modeling: 1. Theory. Water Resources Research, 42:W03407
Koskela JJ, Croke BWF, Koivusalo H, Jakeman AJ, Kokkonen T (2012) Bayesian inference of uncertainties in precipitation-streamflow modeling in a snow affected catchment. Water Resources Research 48(11):W11513
Kuczera G (1983) Improved parameter inference in catchment models: 1. Evaluating parameter uncertainty. Water Resources Research 19(5):1151–1162
Kuczera G, Parent E (1998) Monte Carlo assessment of parameter uncertainty in conceptual catchment models: The Metropolis algorithm. Journal of Hydrology 211:69-85
Kuczera G, Kavetski D, Renard B, Thyer M (2010) A limited-memory acceleration strategy for MCMC sampling in hierarchical Bayesian calibration of hydrological models. Water Resources Research 46(7):W07602
Laloy E, Fasbender D, Bielder CL (2010) Parameter optimization and uncertainty analysis for plot-scale continuous modeling of runoff using a formal Bayesian approach. Journal of Hydrology 380:82–93
Laloy E, Vrugt JA (2012) High-dimensional posterior exploration of hydrologic models using multiple-try DREAM(ZS) and high-performance computing. Water Resources Research 48(1):W01526
Mahdavi M (1999) Applied hydrology. Vol. 2, 2nd edn., Tehran University Press, Tehran, Iran, 401p (In Persian)
Makowski D, Wallach D, Tremblay M (2002) Using a Bayesian approach to parameter estimation: comparison of the GLUE and MCMC methods. Agronomie 22(2):191–203
McMillan H, Clark M (2009) Rainfall‐runoff model calibration using informal likelihood measures within a Markov chain Monte Carlo sampling scheme. Water Resources Research 45:W04418
Mousavi SJ, Abbaspour KC, Kamali B, Amini M, Yang H (2012) Uncertainty-based automatic calibration of HEC-HMS model using sequential uncertainty fitting approach. Journal of Hydroinformatics 14(2):286 –309
Nash JE, Sutcliff JV (1970) River flow forecasting through the conceptual model. Part 1: A discussion of principles. Journal of Hydrology 10(3):282–290
Natural Resourses and Watershed Management Administration of Golestan (2007) Report on: Gorganroud Watershed. Gorgan, Iran (In Persian)
Pourreza-Bilondi M, Akhond Ali AM, Ghahraman B (2012) Parameters uncertainty analysis in distributed single- event rainfall-runoff model with MCMC approach. Iranian Water Research Journal 6(11):165-173 (In Persian)
Pourreza-Bilondi M, Abbaspour KC, Ghahraman B (2013) Application of three different calibration-uncertainty analysis methods in a semi-distributed rainfall-runoff model application. Middle-East Journal of Scientific Research 15(9):1255-1263
Rafiei Sardoii E, Rostami N, Khalighi Sigaroudi S, Taheri S (2012) Calibration of loss estimation methods in HEC-HMS for simulation of surface runoff (Case Study: Amirkabir Dam Watershed, Iran). Advances in Environmental Biology 6(1):343-348
Renard B, Kavetski D, Leblois E, Thyer M, Kuczera G, Franks SW (2011) Toward a reliable decomposition of predictive uncertainty in hydrological modeling: Characterizing rainfall errors using conditional simulation. Water Resources Research 47:W11516
Schoups G, Vrugt JA (2010) A formal likelihood function for parameter and predictive inference of hydrologic models with correlated, heteroscedastic and non-Gaussian errors. Water Resources Research 46:W10531
Shafiei M, Ghahraman B, Saghafian B, Davary K, Pande S, Vazifedous M (2014) Uncertainty assessment of the agro-hydrological SWAP model application at field scale: A case study in a dry region. Agricultural Water Management 146(1):324-334
Smith PJ, Beven K, Tawn JA (2008) Informal likelihood measures in model assessment: Theoretic development and investigation. Advances in Water Resources 31(8):1087–1100
Sorooshian S, Dracup JA (1980) Stochastic parameter estimation procedures for hydrologic rainfall-runoff models: Corrolated and heteroscedastic error cases. Water Resources Research 16(2):430-442
Stedinger JR, Vogel RM, Lee SU, Batchelder R (2008) Appraisal of the generalized likelihood uncertainty estimation (GLUE) method. Water Resources Research 44:W00B06
Straub TD, Melching CS, Kocher KE (2000) Equations for estimating Clark unit-hydrograph parameters for small rural watersheds in Illinois. U.S. Geological Survey, Water-Resources Investigations Report 00-4184, 36p
ter Braak CJF (2006) A Markov chain Monte Carlo version of the genetic algorithm differential evolution: Easy Bayesian computing for real parameter spaces. Statistics and Computing 16(3):239 – 249
ter Braak CJF, Vrugt JA (2008) Differential evolution Markov chain with snooker updater and fewer chains. Statistics and Computing 18(4):435–446
Thyer M, Renard B, Kavetski D, Kuczera G, Franks SW, Srikanthan S (2009) Critical evaluation of parameter consistency and predictive uncertainty in hydrological modeling: A case study using Bayesian total error analysis. Water Resources Research, 45:W00B14
USACE (2000) HEC-HMS Technical  Reference Manual. US Army Corps of Engineers, Hydrologic Engineering Center, Davis, California,158p
USACE (2013) HEC-HMS User's Manual. US Army Corps of Engineers, Hydrologic Engineering Center, Davis, California, 442p
Vrugt JA, Gupta HV, Bouten W, Sorooshian S (2003) A Shuffled Complex Evolution Metropolis algorithm for optimization and uncertainty assessment of hydrologic parameter estimation. Water Resources Research 39(8):1201
Vrugt JA, Gupta HV, Dekker SC, Sorooshian S, Wagener T, Bouten W (2006) Application of stochastic parameter optimization to the Sacramento soil moisture accounting model. Journal of Hydrology 325(1–4):288–307
Vrugt JA, Robinson BA (2007) Treatment of uncertainty using ensemble methods: Comparison of sequential data assimilation and Bayesian model averaging. Water Resources Research 43:W01411
Vrugt JA, ter Braak CJF, Clark MP, Hyman JM, Robinson BA (2008) Treatment of input uncertainty in hydrologic modeling: Doing hydrology backward with Markov chain Monte Carlo simulation. Water Resources Research, 44(12):W00B09
Vrugt JA, ter Braak CJF, Diks CGH, Robinson BA, Hyman JM, Higdon D (2009a) Accelerating Markov Chain Monte Carlo simulation using self-adaptative differential evolution with randomized subspace sampling. International Journal of Nonlinear Sciences and Numerical Simulation 10(3):273-290
Vrugt JA, ter Braak CJF, Gupta HV, Robinson BA (2009b) Equifinality of formal (DREAM) and informal (GLUE) Bayesian approaches in hydrologic modeling?. Stochastic  Environmental  Research and Risk Assessment 23(7):1011–1026
Vrugt JA, Sadegh M (2013) Toward diagnostic model calibration and evaluation: Approximate Bayesian computation. Water Resources Research 49:4335–4345
Wang X, He X, Williams JR, Izaurralde RC, Atwood JD (2005) Sensitivity and uncertainty analyses of crop yields and soil organic carbon simulated with EPIC. Transaction of ASAE 48(3):1041–1054
Xiao B, Wang QH, Fan J, Han FP, Dai QH (2011) Application of the SCS-CN model to runoff estimation in a small watershed with high spatial heterogeneity. Pedosphere 21(6):738-749
Zhange X (2008) Evaluating and developing parameter optimization and uncertainty analysis methods for a computationally intensive distributed hydrological model. Ph.D. dissertation, Texas A&M University, 181p