Korean Journal of Agricultural and Forest Meteorology (한국농림기상학회지)

Korean Society of Agricultural and Forest Meteorology (한국농림기상학회)
권호 표지
DOI QR코드

DOI QR Code

A Statistical Correction of Point Time Series Data of the NCAM-LAMP Medium-range Prediction System Using Support Vector Machine

서포트 벡터 머신을 이용한 NCAM-LAMP 고해상도 중기예측시스템 지점 시계열 자료의 통계적 보정

  • Received : 2021.11.10
  • Accepted : 2021.12.14
  • Published : 2021.12.30
Download PDF
⟨ Previous Next ⟩

Abstract

Recently, an R-based point time series data validation system has been established for the statistical post processing and improvement of the National Center for AgroMeteorology-Land Atmosphere Modeling Package (NCAM-LAMP) medium-range prediction data. The time series verification system was used to compare the NCAM-LAMP with the AWS observations and GDAPS medium-range prediction model data operated by Korea Meteorological Administration. For this comparison, the model latitude and longitude data closest to the observation station were extracted and a total of nine points were selected. For each point, the characteristics of the model prediction error were obtained by comparing the daily average of the previous prediction data of air temperature, wind speed, and hourly precipitation, and then we tried to improve the next prediction data using Support Vector Machine( SVM) method. For three months from August to October 2017, the SVM method was used to calibrate the predicted time series data for each run. It was found that The SVM-based correction was promising and encouraging for wind speed and precipitation variables than for temperature variable. The correction effect was small in August but considerably increased in September and October. These results indicate that the SVM method can contribute to mitigate the gradual degradation of medium-range predictability as the model boundary data flows into the model interior.

NCAM-LAMP 중기예측 자료의 통계적 후처리와 개선을 위하여 R 기반의 지점 시계열 자료 검증 체계를 구축하였다. 이 시계열 검증체계를 이용하여 기상청 AWS 관측 자료와 NCAM-LAMP, KMA GDAPS 중기예측 모델 자료를 비교하였다. 이를 위해 관측 지점에 가장 근접한 모델 위도 및 경도 자료를 추출하여 총 9개 지점을 선정하였다. 각 지점에 대해 NCAM-LAMP, GDAPS 모델의 기온, 강수량, 풍속 일평균 예측 자료를 관측과 비교한 결과, 모델들은 풍속의 과대예측 경향을 뚜렷이 보였으며, 기온과 강수의 경우에는 두 모델의 예측력이 월별 및 변수별로 다르게 나타났다. 이를 바탕으로 본 연구에서는 통계적 기법을 개발하여 NCAM-LAMP가 가지고 있는 오차를 줄이고자 하였다. 모델 오차를 줄이기 위해 일반적으로 쓰이는 MOS(Model Output Statistics)기법 중에 인공지능 SVM(Support vector machine) 방식을 8~10월 기간에 적용한 결과, 8월에 비해서 10월이, 기온 변수에 비해서 바람과 강수 변수가 개선된 효과를 보여주었다. 이러한 결과는 풍속의 과대예측을 줄이고, 농림 가뭄지수와 산사태 예측 등을 개선시키며, 지역 수치예보 모델이 시간 적분됨에 따라 영역 내 예측가능성이 점점 저하되는 현상을 완화시키는데 SVM 방법이 일정 부분 기여할 수 있음을 가리키며, 현업 표출 중인 NCAM Agro-Meteogram 개선에도 도움을 줄 것으로 기대된다.

Keywords

Acknowledgement

본 연구는 산림청(한국임업진흥원) 산림과학기술연구개발사업(2021341B10-2123-CD01)의 지원으로 수행되었습니다.

References

  1. Antolik, M. S., 2012: Model Output Statistics (MOS) - Objective Interpretation of NWP Model Output. Presented at the University of Maryland USA (2012).
  2. Asano, A., 2004: Support vector machine and kernel method Pattern information processing. Pattern information Processing (2004 Autumn Semester) Session 12. (05. 1. 21)
  3. Bae, K. Y., H. S. Jang, and D. K. Sung, 2017: Solar power prediction based on machine learning scheme and its error analysis. Proceedings of Symposium of the Korean Institute of communications and Information Sciences, 13-14.
  4. Hyeon, B., Y. Lee, and K. Seo, 2015: Evolutionary nonlinear regression based compensation technique for short-range prediction of wind speed using automatic weather station. The Transactions of the Korean Institute of Electrical Engineers 64(1), 107-112. https://doi.org/10.5370/KIEE.2015.64.1.107
  5. Hong, M.-K, S.-H. Lee, J.-Y. Choi, S.-H. Lee, and S.-J. Lee, 2015: Estimation of soil moisture and irrigation requirement of upland using soil moisture model applied WRF meteorological data. Journal of the Korean Society of Agricultural Engineers 57(6), 173-183. (in Korean with English Abstract) https://doi.org/10.5389/KSAE.2015.57.6.173
  6. Hong, M., S. Lee, S.-J. Lee, and J. Y. Choi, 2020: Application of high-resolution meteorological data from NCAM-WRF to characterize agricultural drought in small-scale farmlands based on soil moisture deficit. Agricultural Water Management 243, 106493. https://doi.org/10.1016/j.agwat.2020.106493
  7. Hwang, J., H.-O. Cho, Y. Lim, S.-W. Son, E.-J. Kim, J.-O. Lim, and K.-O. Boo, 2020: Extratropical prediction skill of KMA GDAPS in January 2019. Atmosphere 30(2), 115-124. (in Korean with English Abstract) https://doi.org/10.14191/ATMOS.2020.30.2.115
  8. Jacobson, T., J. James, and N. C. Schwertman, 2009: An example of using linear regression of seasonal weather patterns to enhance undergraduate learning. Journal of Statistics Education 17(2).
  9. Jo, N.-H., 2006: SVM load forecasting using cross-validation. The Korean Institute of Electrical Engineers A, 55(11), 485-491.
  10. Kim, D., and K. Seo, 2015: Comparison of linear and nonlinear regressions and elements analysis for wind speed prediction. Journal of Korean Institute of Intelligent Systems 25(5), 477-482. https://doi.org/10.5391/JKIIS.2015.25.5.477
  11. KMA (Korea Meteorological Administration), 2016: Ground Weather Observation Guidelines. (Available at https://book.kma.go.kr/viewer/MediaViewer.ax?cid=33393&rid=5&moi=5241 accessed on 13 December 2021)
  12. Kum, D., C. Jang, J. Lee, and K. J. Lim, 2014: Analysis of rainfall according to the bias correction of climate change scenarios. Proceedings of the Korea Water Resources Association Conference (2014-05), 470pp.
  13. Kwon, S., and S.-J. Lee, 2017: A statistical verification and improvement tool of point time series data of the NCAM-LAMP mid-term prediction system. Poster session presented at: 18th Conference on Agricultural and Forest Meteorology, 2017 November 10, Seoul National University.
  14. Lee, S., S.-J. Lee, J. H. Kang, and E. S. Jang, 2021: Spatial and temporal variations in atmospheric ventilation index coupled with particulate matter concentration in South Korea. Sustainability 13(16), 8954. https://doi.org/10.3390/su13168954
  15. Lee, S.-J., J. Kim, M. Kang, and B. Malla-Thakuri, 2014: Numerical simulation of local atmospheric circulations in the valley of Gwangneung KoFlux sites. Korean Journal of Agricultural and Forest Meteorology 16, 244-258.
  16. Lee, S.-J., J. S. Kang, and H. Yoo, 2015: Atmospheric Modeling, Data Assimilation, and Predictability (Korean Version). SigmaPress.
  17. Lee, S.-J., J. Song, and Y.-J. Kim, 2016: The NCAM Land-Atmosphere Modeling Package (LAMP) Version 1: Implementation and evaluation. Korean Journal of Agricultural and Forest Meteorology 18(4), 307-319. https://doi.org/10.5532/KJAFM.2016.18.4.307
  18. Lee, S.-J., Y. Kim, J. Song, and J. Kim, 2017: An Agrometeorological application of the meteogram: "Agrometeogram". American Meteorological Society 97th Annual Meeting, 22-26 January 2017, Seattle, WA, U.S.A.
  19. Lee, S., S.-J. Lee, and J. S. Koo, 2020: Database construction of high-resolution daily meteorological and climatological data using NCAM-LAMP: Sunshine hour data. Korean Journal of Agricultural Forest Meteorology 22(3), 135-143. (in Korean with English Abstract) https://doi.org/10.5532/KJAFM.2020.22.3.135
  20. Mokhtarzad, M., F. Eskandari, N. J. Vanjani, and A. Arabasadi, 2017: Drought forecasting by ANN, ANFIS, and SVM and comparison of the models. Environmental Earth Science 76, 729. https://doi.org/10.1007/s12665-017-7064-0
  21. Moon, H., J. Baik, S. Hwang, and M. Choi, 2014: Spatial downscaling of grid precipitation using support vector machine regression. Journal of Korea Water Resources Association 47(11), 1095-1105. https://doi.org/10.3741/JKWRA.2014.47.11.1095
  22. Park, J., H.-S. Kim, S.-J. Lee, and T. Ha, 2018: Numerical evaluation of JULES surface tiling scheme with high-resolution atmospheric forcing and land cover data. SOLA 14, 19-24. https://doi.org/10.2151/sola.2018-004
  23. Shin, Y. H., J. Y. Choi, S.-J. Lee, and S. H. Lee, 2017: Estimation of irrigation requirements for red pepper using soil moisture model with high resolution meteorological data. Journal of the Korean Society of Agricultural Engineers 59(5), 31-40. https://doi.org/10.5389/KSAE.2017.59.5.031
  24. So, Y. Y., S. J. Lee, S. W. Choi, and S.-J. Lee, 2020: Construction of NCAM-LAMP precipitation and soil moisture database to support landslide prediction. Korean Journal of Agricultural and Forest Meteorology 22(3), 152-163. (in Korean with English Abstract) https://doi.org/10.5532/KJAFM.2020.22.3.152
  25. Statnikov, A., C. F. Aliferis, D. P. Hardin, and I. Guyon, 2013: A Gentle Introduction to Support Vector Machines in Biomedicine: Case Studies. World Scientific Publishing Co., Inc., River Edge, NJ, USA, 1st edition, 2011.
  26. Wu, C. L., K. W. Chau, and Y. S. Li, 2008: River stage prediction based on a distributed support vector regression. Journal of Hydrology 358, 96-111. https://doi.org/10.1016/j.jhydrol.2008.05.028
  27. Wilks, D. S., 2009: Extending logistic regression to provide full-probability-distribution MOS forecasts. Meteorological Applications 16(3), 361-368. https://doi.org/10.1002/met.134
  28. Yoon, H., P. Yoon, E. Lee, and G.-B. Kim, and S.-H. Moon, 2016: Application of machine learning technique-based time series models for prediction of groundwater level fluctuation to national groundwater monitoring network data. Journal of the Geological Society of Korea 52(3), 187-199. https://doi.org/10.14770/jgsk.2016.52.3.187