یک چارچوب چندمعیاره (Entropy–TOPSIS) برای انتخاب ترکیب‌های بهینه مدل–روشِ CMIP6 جهت تصحیح اریبی  بارش و دما در شمال شرق ایران

رازبان, آرش; موسوی بایگی, سید محمد; بابائیان, ایمان; کامکار, بهنام; صراف زاده, نسرین

doi:10.30499/ijg.2026.579122.1763

یک چارچوب چندمعیاره (Entropy–TOPSIS) برای انتخاب ترکیب‌های بهینه مدل–روشِ CMIP6 جهت تصحیح اریبی بارش و دما در شمال شرق ایران

نوع مقاله : مقاله پژوهشی‌

نویسندگان

آرش رازبان ¹

سید محمد موسوی بایگی ²

ایمان بابائیان ³

بهنام کامکار ⁴

نسرین صراف زاده ⁵

¹ دانشجوی دکتری، دانشکده کشاورزی، دانشگاه فردوسی مشهد، مشهد، ایران

² استاد، دانشکده کشاورزی، دانشگاه فردوسی مشهد، مشهد، ایران

³ استاد، پژوهشگاه هواشناسی و علوم جو، پژوهشکده اقلیم شناسی، مشهد، ایران

⁴ دانش آموخته کارشناسی ارشد، دانشکده کشاورزی، دانشگاه فردوسی مشهد، مشهد، ایران

⁵ مهندسی منابع آب، گروه علوم و مهندسی آب، دانشکده کشاورزی، دانشگاه فردوسی مشهد

10.30499/ijg.2026.579122.1763

چکیده

تغییر اقلیم و افزایش اثرات آن، نیاز به بازتولید دقیق بارش و دما در مقیاس ایستگاهی را تشدید کرده است؛ اما خروجی CMIP6 به‌علت تفکیک مکانی پایین و خطاهای سامانمند، نیازمند تصحیح اریبی آماری و انتخاب مدل–روش است. این پژوهش برای رفع خلأ ارزیابی‌های پراکنده و تک‌متغیره در شرق ایران، چارچوبی ایستگاه‌محور و چندمعیاره برای رتبه‌بندی هم‌زمان ترکیب‌های «مدل–روش» در بازتولید بارش، دمای حداقل و دمای حداکثرِ سه استان خراسان ارائه می‌کند (۱۹ ایستگاه، 1989–2014). در این چارچوب، خروجی‌های ۸ مدل از CMIP6(۵ مدل ESM و 3 مدل GCM) با ۸ روش تصحیح اریبی نرم‌افزار CMHyd (۵ روش بارش، ۴ روش دما) مقایسه شدند. سنجه‌ها: NRMSE، MAE، r و KGE؛ و به‌صورت متغیرمحور PBIAS برای بارش و MBE برای دما می‌باشد. معیارها پس از نرمال‌سازی با آنتروپی شانون (وزن ثابت در منطقه) وزن‌دهی و رتبه‌بندی ایستگاهی با TOPSIS انجام شد؛ سپس نتایج با میانه رتبه‌ها، IQR و تعداد ایستگاهِ برنده تجمیع و با آزمون‌های فریدمن و نمنـی اعتبارسنجی شد. اختلاف عملکرد گزینه‌ها برای هر سه متغیر معنادار بود (Tmin: χ²≈538, p≈10⁻⁹⁴؛ Tmax: χ²≈458, p≈10⁻⁷⁷؛Pr: χ²≈286, p≈10⁻⁴⁴). MPI|VarianceScaling (میانگین رتبه≈1/3) برای Tmin و MPI|DistributionMapping (میانگین رتبه≈2/3) برای Tmax بهترین گزینه‌های منطقه‌ای بودند؛ بااین‌حال، آزمون نمنـی نشان داد ۱۲ ترکیب (Tmin) و ۱۱ ترکیب (Tmax) از نظر آماری با گزینه برتر قابل تفکیک نیستند و «گروه ممتاز» با شاخص‌های همگرا و پایدار می‌سازند. برای بارش، HadGEM|DistributionMapping (میانگین رتبه≈1) برتر بود و نسبت به سایر اعضای گروه ممتاز شاخص‌های دقیق‌تری ارائه داد. در مقابل، بسیاری از ترکیب‌های مبتنی بر INM، CMCC و GFDL و نیز برخی روش‌های DeltaChange/LinearScaling و LocalIntensity/PowerTransformation فاصله معناداری با گروه ممتاز داشتند و در رتبه‌های انتهایی قرار گرفتند. بنابراین در خراسان، مدل MPI همراه Distribution Mapping و Variance Scaling و HadGEM همراه Distribution Mapping به‌ترتیب پایدارترین گزینه‌ها برای بازتولید دما و بارش‌ هستند. این چارچوب پیشنهادی مبنایی برای کاهش عدم‌قطعیت و انتخاب اتکاپذیر مدل–روش در پیش‌نگری اقلیم و مدیریت مخاطره منطقه‌ای است.

کلیدواژه‌ها

CMIP6

Entropy-TOPSIS

بارش و دما

خراسان

تصحیح اریبی آماری

موضوعات

فیزیک فضا

عنوان مقاله English

A multi‑criteria framework (Entropy–TOPSIS) for selecting optimal CMIP6 model–method combinations for precipitation and temperature bias correction in northeastern Iran

نویسندگان English

Arash Razban ¹

Seyyed Mohammad Mosavi Baygi ²

Iman Babaeian ³

Behnam Kamkar ⁴

Nasrin Sarrafzadeh ⁵

¹ Ph.D. Student, Faculty of Agriculture, Ferdowsi University of Mashhad, Mashhad, Iran

² Professor, Faculty of Agriculture, Ferdowsi University of Mashhad, Mashhad, Iran

³ Professor, Meteorological Research Institute and Atmospheric Science, Mashhad, Iran

⁴ M.Sc., Faculty of Agriculture, Ferdowsi University of Mashhad, Mashhad, Iran

⁵ Department of Water Science and Engineering, Faculty of Agriculture, Ferdowsi University of Mashhad

چکیده English

Climate change and the intensification of its impacts have increased the need for accurate station-scale reproduction of precipitation and temperature. However, raw CMIP6 outputs, because of their coarse spatial resolution and systematic biases, require bias correction and a robust procedure for selecting appropriate model–method combinations. To address the gap of fragmented and single-variable evaluations in eastern Iran, this study proposes a station-based, multi-criteria framework for the simultaneous ranking of model–method combinations in reproducing precipitation, minimum temperature, and maximum temperature across the three Khorasan provinces (19 stations, 1989–2014). Within this framework, outputs from eight CMIP6 models (five ESMs and three GCMs) were evaluated using eight statistical bias-correction methods implemented in the CMHyd software (five methods for precipitation and four for temperature). The evaluation metrics included NRMSE, MAE, r, and KGE, together with variable-specific PBIAS for precipitation and MBE for temperature. After normalization, the criteria were weighted using Shannon entropy with regionally fixed weights, and station-level ranking was performed using TOPSIS. The results were then aggregated using median ranks, IQR, and the number of winning stations, and statistically assessed using the Friedman and Nemenyi tests. Performance differences among alternatives were significant for all three variables (Tmin: χ² ≈ 538, p ≈ 10⁻⁹⁴; Tmax: χ² ≈ 458, p ≈ 10⁻⁷⁷; Pr: χ² ≈ 286, p ≈ 10⁻⁴⁴). MPI|VarianceScaling and MPI|DistributionMapping were identified as the best regional options for Tmin and Tmax, respectively. Nevertheless, the Nemenyi test showed that 12 (Tmin) and 11 (Tmax) combinations were not statistically distinguishable from the top-ranked option, forming an excellent group characterized by convergent and stable performance indices. For precipitation, HadGEM|DistributionMapping ranked first and produced more accurate performance indices than other members of the excellent group. In contrast, many INM-, CMCC-, and GFDL-based combinations, together with some DeltaChange/LinearScaling and LocalIntensity/PowerTransformation methods, were significantly separated from the excellent group and occupied the lowest ranks. Accordingly, in the Khorasan region, MPI combined with Variance Scaling and Distribution Mapping, and HadGEM combined with Distribution Mapping, are the most stable options for reproducing temperature and precipitation, respectively. The proposed framework provides a practical basis for reducing structural uncertainty and supporting robust selection of climate model–bias correction combinations for climate change impact assessment and regional hazard management.

کلیدواژه‌ها English

CMIP6

entropy-TOPSIS

Khorasan

precipitation and temperature

statistical bias correction

Abdolalizadeh, F., Khorshiddoust, A. M., & Jahanbakhsh Asl, S. (2023). Projection of the future outlook of temperature and precipitation in Urmia Lake basin by the CMIP6 models. Physical Geography Research Quarterly, 55(1), 95–112. https://doi.org/10.22059/JPHGR.2023.352727.1007737

Agrawal, N., & Mujumdar, S. S. (2025). A multimodel ensemble using the entropy-TOPSIS method for projecting temperature for the Lower Tapi River Basin. Water Practice & Technology. https://doi.org/10.2166/wpt.2025.166

Ansari Mahabadi, S., Dehban, H., Zareian, M. J., & Farokhnia, A. (2022). Trend analysis of temperature and precipitation changes in Iran’s river basins over a 20-year horizon based on CMIP6 model outputs. Iran Water Research Journal, 16(1), 11–24. https://doi.org/10.22034/iwrj.2022.11204

Azad, N., & Ahmadi, A. (2024). Assessment of CMIP6 models and multi-model averaging for temperature and precipitation over Iran. Scientific Reports, 14, 24165. https://doi.org/10.1038/s41598-024-74789-4

Babaeian, I., Modirian, R., Khazanedari, L., Karimian, M., Kouzegaran, S., Kouhi, M., Falamarzi, Y., & Malbusi, S. (2023). Projection of Iran’s precipitation in 21st century using downscaling of selected CMIP6 models by CMHyd. Journal of the Earth and Space Physics, 49(2), 431–449. https://doi.org/10.22059/jesphys.2023.332410.1007436

Babaeian, I., Rahmatinia, A. E., Entezari, A., Baaghideh, M., Aval, M. B., & Habibi, M. (2021). Future projection of drought vulnerability over northeast provinces of Iran during 2021–2100. Atmosphere, 12, 1704. https://doi.org/10.3390/atmos12121704

Brumatti, L. M., Sant’Anna Commar, L. F., de Oliveira Neumann, N., Ferreira Pires, G., & Avila-Diaz, A. (2024). Bias correction in CMIP6 model simulations and projections for Brazil’s climate assessment. Earth Systems and Environment, 8, 121–134. https://doi.org/10.1007/s41748-023-00368-8

Eshaghi, A., & Kamkar, B. (2025). A remote sensing-based framework for agricultural drought risk monitoring and assessment: Introducing SADFI for disaster risk assessment in Northeastern Iran. Theoretical and Applied Climatology, 156, Article 684. https://doi.org/10.1007/s00704-025-05888-z

Eyring, V., Bony, S., Meehl, G. A., Senior, C. A., Stevens, B., Stouffer, R. J., & Taylor, K. E. (2016). Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization. Geoscientific Model Development, 9(5), 1937–1958.

Ghorbani-Minaei, L., Mosaedi, A., Zakerinia, M., Kalbali, E., & Ghabaei Sough, M. (2024). Study of future climate change on temperature and precipitation trends in Qarasu basin based on the CMIP6 models. Iranian Journal of Soil and Water Research, 55(2), 245–268. https://doi.org/10.22059/ijswr.2024.369146.669613

Hamidianpour, M., Baaghideh, M., & Abbasnia, M. (2016). Assessment of precipitation and temperature changes over southeast Iran using downscaling of general circulation model outputs. Physical Geography Research, 48(1), 107–123. https://doi.org/10.22059/jphgr.2016.57030

Han, R., Li, Z. L., Han, Y. Y., Huo, P. Y., & Li, Z. J. (2023). A comparative study of TOPSIS-based GCM selection and multi-model ensemble. International Journal of Climatology, 43(12), 5348–5368. https://doi.org/10.1002/joc.8150

IPCC. (2021). Climate Change 2021: The Physical Science Basis: Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, United Kingdom and New York, NY: Cambridge University Press. https://doi.org/10.1017/9781009157896

Krishnan, R., Swapna, P., Dey Choudhury, A., Narayansetti, S., Prajeesh, A. G., Singh, M., Modi, A., Mathew, R., Vellore, R., Jyoti, J., Sabin, T. P., Sanjay, J., & Ingle, S. (2021). The IITM Earth System Model (IITM ESM). arXiv preprint arXiv:2101.03410. https://arxiv.org/abs/2101.03410

Knutti, R., Furrer, R., Tebaldi, C., Cermak, J., & Meehl, G. A. (2010). Challenges in combining projections from multiple climate models. Journal of Climate, 23(10), 2739–2758.

Li, X., & Li, Z. (2025). Assessment of bias correction methods for high-resolution daily precipitation projections with CMIP6 models: A Canadian case study. Journal of Hydrology: Regional Studies, 58, 102223. https://doi.org/10.1016/j.ejrh.2025.102223

Maraun, D., & Widmann, M. (2018). Statistical downscaling and bias correction for climate research. Cambridge: Cambridge University Press. https://doi.org/10.1017/9781107588783

Mianabadi, A., Mohammadi, S., & Bateni, M. M. (2023). Projection of changes in precipitation and temperature distributions using bias-corrected simulations of CMIP6 climate models (Case study: Kerman synoptic station). Climate Change Research, 4(14), 65–84.

Nguyen, P. L., Alexander, L. V., Thatcher, M. J., Truong, S. C. H., Isphording, R. N., & McGregor, J. L. (2024). Selecting CMIP6 global climate models for CORDEX downscaling. Geoscientific Model Development, 17, 7285–7315.

Ozbuldu, M., & Irvem, A. (2025). Projecting and downscaling future temperature and precipitation based on CMIP6 models using machine learning in Hatay Province, Türkiye. Pure and Applied Geophysics, 182, 1825–1842. https://doi.org/10.1007/s00024-024-03656-0

Raeesi, M., Zolfaghari, A. A., Kaboli, S. H., Rahimi, M., de Vente, J., & Eekhout, J. P. C. (2024). Using quantile mapping and random forest for bias correction of high-resolution reanalysis precipitation data and CMIP6 climate projections over Iran. International Journal of Climatology, 44(12), 4495–4514.

Rashidi Ghane, M., Motavalli, S., Janbaz Ghobadi, G., & Kouhi, M. (2023). Evaluating the capability of three statistical downscaling methods for temperature and precipitation outputs of CMIP6 models in the Kashafrood watershed. Journal of Climatological Research, 14(53), 117–132.

Seraj Ebrahimi, R., Zareian, M. J., & Dehban, H. (2024). Evaluation of the performance of CMIP6 models in estimating temperature and precipitation in the Sefidrood Basin. Journal of Water and Irrigation Management, 14(2), 277–289.

Thakur, R., & Manekar, V. L. (2023). Ranking of CMIP6-based high-resolution global climate models for India using TOPSIS. ISH Journal of Hydraulic Engineering, 29(2), 175–188. https://doi.org/10.1080/09715010.2021.2015462

Wei, S., Wang, X., Liu, L., Qie, L., Li, Y., Wang, Q., Wang, T., Wang, J., Gou, X., & Yang, M. (2025). Bias correction of CMIP6 GCMs for historical and future air temperatures across China. Atmospheric Research, 323, 108193. https://doi.org/10.1016/j.atmosres.2025.108193

Yazdandoost, F., Moradian, S., Izadi, A., & Aghakouchak, A. (2020). Evaluation of CMIP6 precipitation simulations across different climatic zones: Uncertainty and model intercomparison. Atmospheric Research, 250, 105369. https://doi.org/10.1016/j.atmosres.2020.105369

Yazdani, D., Zarrin, A., & Roudbari-Dadashi, A. A. (2024). Statistical downscaling of general circulation models (GCMs): History, principles and methods. Journal of Water and Sustainable Development, 11(2), 15–26. https://doi.org/10.22067/jwsd.v11i2.2401-1305

Zabihi, O., & Ahmadi, A. (2024). Multi-criteria evaluation of CMIP6 precipitation and temperature simulations over Iran. Journal of Hydrology: Regional Studies, 52, 101707. https://doi.org/10.1016/j.ejrh.2024.101707

Zareian, M. J. (2022). Effects of climate change on temperature and precipitation in Yazd Province based on combined output of CMIP6 models. Journal of Water and Soil Science, 26(2), 91–105.

Zareian, M. J., Dehban, H., & Gohari, A. (2022). Evaluation of the accuracy of CMIP6 models in estimating the temperature and precipitation of Iran based on a network analysis. Journal of Water and Irrigation Management, 12(4), 783–797. https://doi.org/10.22059/jwim.2022.345975.1006

Zhu, Y., Tian, D., & Yan, F. (2020). Effectiveness of entropy weight method in decision-making. Mathematical Problems in Engineering, 2020, 3564835. https://doi.org/10.1155/2020/3564835.