Atmosphere (대기)

Korean Meteorological Society (한국기상학회)
권호 표지
DOI QR코드

DOI QR Code

Enhancing Medium-Range Forecast Accuracy of Temperature and Relative Humidity over South Korea using Minimum Continuous Ranked Probability Score (CRPS) Statistical Correction Technique

연속 순위 확률 점수를 활용한 통합 앙상블 모델에 대한 기온 및 습도 후처리 모델 개발

  • Hyejeong Bok (Numerical Data Application Division, Numerical Modeling Center, Korea Meteorological Administration) ;
  • Junsu Kim (Numerical Data Application Division, Numerical Modeling Center, Korea Meteorological Administration) ;
  • Yeon-Hee Kim (Numerical Data Application Division, Numerical Modeling Center, Korea Meteorological Administration) ;
  • Eunju Cho (Numerical Data Application Division, Numerical Modeling Center, Korea Meteorological Administration) ;
  • Seungbum Kim (Numerical Data Application Division, Numerical Modeling Center, Korea Meteorological Administration)
  • 복혜정 (기상청 수치모델링센터 수치자료응용과) ;
  • 김준수 (기상청 수치모델링센터 수치자료응용과) ;
  • 김연희 (기상청 수치모델링센터 수치자료응용과) ;
  • 조은주 (기상청 수치모델링센터 수치자료응용과) ;
  • 김승범 (기상청 수치모델링센터 수치자료응용과)
  • Received : 2023.10.13
  • Accepted : 2023.11.21
  • Published : 2024.02.29
Download PDF
⟨ Previous Next ⟩

Abstract

The Korea Meteorological Administration has improved medium-range weather forecasts by implementing post-processing methods to minimize numerical model errors. In this study, we employ a statistical correction technique known as the minimum continuous ranked probability score (CRPS) to refine medium-range forecast guidance. This technique quantifies the similarity between the predicted values and the observed cumulative distribution function of the Unified Model Ensemble Prediction System for Global (UM EPSG). We evaluated the performance of the medium-range forecast guidance for surface air temperature and relative humidity, noting significant enhancements in seasonal bias and root mean squared error compared to observations. Notably, compared to the existing the medium-range forecast guidance, temperature forecasts exhibit 17.5% improvement in summer and 21.5% improvement in winter. Humidity forecasts also show 12% improvement in summer and 23% improvement in winter. The results indicate that utilizing the minimum CRPS for medium-range forecast guidance provide more reliable and improved performance than UM EPSG.

Keywords

Acknowledgement

본 논문을 심사해주신 두 분 심사위원님들께 깊은 감사드립니다. 이 연구는 기상청 수치모델링센터 『수치예보 및 자료응용 기술개발(KMA2018-00721)』 과제의 일환으로 수행되었습니다.

References

  1. Ashkboos, S., L. Huang, N. Dryden, T. Ben-Nun, P. Dueben, L. Gianinazzi, L. Kummer, and T. Hoefler, 2022: ENS-10: A Dataset For Post-Processing Ensemble Weather Forecasts. Adv. Neural Inf. Processing Syst., 35, 21974-21987, doi:10.48550/arXiv.2206.14786.
  2. Buizza, R., and D. Richardson, 2017: years of ensemble forecasting at ECMWF. ECMWF Newsletter, 153, 20-31.
  3. Dallavalle, J. P., 1996: A perspective on the use of model output statistics in objective weather forecasting. Conference on Weather Analysis and Forecasting, Amer. Meteor. Society, 15, 479-482.
  4. Freebairn, J. W., and J. W. Zillman, 2002: Economic benefits of meteorological services. Meteor. Appl., 9, 33-44, doi:10.1017/S1350482702001044.
  5. Frich, P., L. V. Alexander, P. Della-Marta, B. Gleason, M. Haylock, A. K. Tank, and T. Peterson, 2002: Observed coherent changes in climatic extremes during the second half of the twentieth century. Climate Res., 19, 193-212, doi:10.3354/cr019193.
  6. Geman, S., E. Bienenstock, and R. Doursat, 1992: Neural networks and the bias/variance dilemma. Neural Comput., 4, 1-58, doi:10.1162/neco.1992.4.1.1.
  7. Gneiting, T., A. E. Raftery, A. H. Westveld, and T. Goldman, 2005: Calibrated probabilistic forecasting using ensemble model output statistics and minimum CRPS estimation. Mon. Wea. Rev., 133, 1098-1118, doi:10.1175/MWR2904.1.
  8. Gneiting, T., and A. E. Raftery, 2007: Strictly proper scoring rules, prediction, and estimation. J. Amer. Stat. Assoc., 102, 359-378, doi:10.1198/016214506000001437.
  9. Gronquist, P., C. Yao, T. Ben-Nun, N. Dryden, P. Dueben, S. Li, and T. Hoefler, 2021: Deep learning for post-processing ensemble weather forecasts. Philos. Trans. Roy. Soc. A, 379, 20200092, doi:10.1098/rsta.2020.0092.
  10. Haiden, T., M. Janousek, F. Vitart, Z. Ben-Bouallegue, L. Ferranti, and F. Prates, 2021: Evaluation of ECMWF forecasts, including the 2021 upgrade, ECMWF technical memoranda 884. European Centre for Medium Range Weather Forecasts, 56 pp.
  11. Han, C.-H., J.-W. Lee, and K.-K. Lee, 2009: Analyzing information value of temperature forecast for the electricity demand forecasts. Korean Manage. Sci. Rev., 26, 77-91.
  12. Han, K., C. Kim, and C. Kim, 2016: Probabilistic forecast of temperature in Pyeongchang using homogeneous and nonhomogeneous regression models. J. Climate Res., 11, 87-103, doi:10.14383/cri.2016.11.1.87.
  13. Hong, H., W. Kim, J. Kim, and B.-J. Kim, 2013: Analysis of demand characteristics for long-term forecasts. J. Climate Res., 8, 117-126, doi:10.14383/cri.2013.8.2.117.
  14. Hwang, Y., and C. Kim, 2018: A combination and calibration of multi-model ensemble of PyeongChang area using ensemble model output statistics. Atmosphere, 28, 247-261, doi:10.14191/Atmos.2018.28.3.247.
  15. Hyun, Y.-K., J. Park, J. Lee, S. Lim, S.-I. Heo, H. Ham, S.-M. Lee, H.-S. Ji, and Y. Kim, 2020: Reliability assessment of temperature and precipitation seasonal probability in current climate prediction systems. Atmosphere, 30, 141-154, doi:10.14191/Atmos.2020.30.2.141.
  16. Jang, D. H., and C. Kim, 2017: Probabilistic forecast of temperature using station-specific nonhomogeneous regression model. J. Climate Res., 12, 277-287, doi:10.14383/CRI.2017.12.3.277.
  17. Kang, S.-C., M.-S. Won, and S.-H. Yoon, 2016: Large fire forecasting depending on the changing wind speed and effective humidity in Korean red pine forests through a case study. J. Korean Assoc. Geographic Inf. Stud., 19, 146-156, doi:10.11108/kagis.2016.19.4.146.
  18. Kim, K., J. Shin, and S.-I. Lee, 2012: An overview of methodologies for socio-economic value evaluation of long-range weather forecasts in energy industry. J. Climate Res., 7, 98-118.
  19. Kioutsioukis, I., and S. Galmarini, 2014: De praeceptis ferendis: good practice in multi-model ensembles. Atmos. Chem. Phys., 14, 11791-11815, doi:10.5194/acp-14-11791-2014.
  20. Klein, W. H., F. Lewis, and G. P. Casely, 1967: Automated nationwide forecasts of maximum and minimum temperature. J. Appl. Meteor. Climatol., 6, 216-228, doi:10.1175/1520-0450(1967)006<0216:ANFOMA>2.0.CO;2.
  21. KMA, 2022: Weather yearbook, Korea Meteorological Administration, 416 pp (in Korean).
  22. Lazo, J. K., R. E. Morss, and J. L. Demuth, 2009: 300 billion served: Sources, perceptions, uses, and values of weather forecasts. Bull. Amer. Meteor. Soc., 90, 785-798, doi:10.1175/2008BAMS2604.1.
  23. Lorenz, E. N., 1969: Atmospheric predictability as revealed by naturally occurring analogues. J. Atmos. Sci., 26, 636-646, doi:10.1175/1520-0469(1969)26<636:APARBN>2.0.CO;2.
  24. Lyu, Y., X. Zhi, H. Wu, H. Zhou, D. Kong, S. Zhu, Y. Zhang, and C. Hao, 2022: Analyses on the multimodel wind forecasts and error decompositions over North China. Atmosphere, 13, 1652, doi:10.3390/atmos13101652.
  25. Mailier, P. J., 2010: Can we trust long-range weather forecasts? Manage. Wea. Climate Risk in the Energy Industry, 227-239, doi:10.1007/978-90-481-3692-6_15.
  26. Mazrooei, A., T. Sinha, A. Sankarasubramanian, S. Kumar, and C. D. Peters-Lidard, 2015: Decomposition of sources of errors in seasonal streamflow forecasting over the US Sunbelt. J. Geophys. Res. Atmos., 120, 11809-11825, doi:10.1002/2015JD023687.
  27. Meehl, G. A., and C. Tebaldi, 2004: More intense, more frequent, and longer lasting heat waves in the 21st century. Science, 305, 994-997, doi:10.1126/science.1098704.
  28. Murphy, K. P., 2012: Machine learning: a probabilistic perspective. MIT press, 1067 pp.
  29. Neal, B., S. Mittal, A. Baratin, V. Tantia, M. Scicluna, S. Lacoste-Julien, and I. Mitliagkas, 2018: A modern take on the bias-variance tradeoff in neural networks. arXiv preprint arXiv:1810.08591, doi:10.48550/arXiv.1810.08591.
  30. Raftery, A. E., T. Gneiting, F. Balabdaoui, and M. Polakowski, 2005: Using Bayesian model averaging to calibrate forecast ensembles. Mon. Wea. Rev., 133, 1155-1174, doi:10.1175/MWR2906.1.
  31. Rasp, S., and S. Lerch, 2018: Neural networks for postprocessing ensemble weather forecasts. Mon. Wea. Rev., 146, 3885-3900, doi:10.1175/MWR-D-18-0187.1.
  32. Schultz, M. G., C. Betancourt, B. Gong, F. Kleinert, M. Langguth, L. H. Leufen, A. Mozaffari, and S. Stadtler, 2021: Can deep learning beat numerical weather prediction? Philos. Trans. Roy. Soc. A, 379, 20200097, doi:10.1098/rsta.2020.0097.
  33. Seong, M.-G., C. Kim, and M.-S. Suh, 2015: Inter-comparison of prediction skills of multiple linear regression methods using monthly temperature simulated by multi-regional climate models. Atmosphere, 25, 669-683, doi:10.14191/Atmos.2015.25.4.669.
  34. Sinha, T., A. Sankarasubramanian, and A. Mazrooei, 2014: Decomposition of sources of errors in monthly to seasonal streamflow forecasts in a rainfall-runoff regime. J. Hydrometeor., 15, 2470-2483, doi:10.1175/JHM-D13-0155.1.
  35. Taylor, K. E., 2001: Summarizing multiple aspects of model performance in a single diagram. J. Geophys. Res. Atmos., 106, 7183-7192, doi:10.1029/2000JD900719.
  36. Teisberg, T. J., R. F. Weiher, and A. Khotanzad, 2005: The economic value of temperature forecasts in electricity generation. Bull. Amer. Meteor. Soc., 86, 1765-1772, doi:10.1175/BAMS-86-12-1765.
  37. Yang, Y., I. Kang, J. Yoo, and K. An, 2004: Analysis of economical & social impact of meteorological information. J. Korean Meteor. Soc., 40, 159-175.
  38. Yun, J., Y.-H. Kim, and H.-W. Choi, 2021: Analyses of the meteorological characteristics over South Korea for wind power applications using KMAPP. Atmosphere, 30, 1-15, doi:10.14191/Atmos.2021.31.1.001.
  39. Zhang, H., M. Chen, and S. Fan, 2019: Study on the construction of initial condition perturbations for the regional ensemble prediction system of North China. Atmosphere, 10, 87, doi:10.3390/atmos10020087.