Journal of Korea Water Resources Association (한국수자원학회논문집)

Korea Water Resources Association (한국수자원학회)
권호 표지
DOI QR코드

DOI QR Code

Application of Artificial Neural Network to Improve Quantitative Precipitation Forecasts of Meso-scale Numerical Weather Prediction

중규모수치예보자료의 정량적 강수추정량 개선을 위한 인공신경망기법

  • 강부식 (단국대학교 공과대학 토목환경공학과) ;
  • 이봉기 (도화종합기술공사)
  • Received : 2010.12.22
  • Accepted : 2011.01.18
  • Published : 2011.02.28
Download PDF
⟨ Previous Next ⟩

Abstract

For the purpose of enhancing usability of NWP (Numerical Weather Prediction), the quantitative precipitation prediction scheme was suggested. In this research, precipitation by leading time was predicted using 3-hour rainfall accumulation by meso-scale numerical weather model and AWS (Automatic Weather Station), precipitation water and relative humidity observed by atmospheric sounding station, probability of rainfall occurrence by leading time in June and July, 2001 and August, 2002. Considering the nonlinear process of ranfall producing mechanism, the ANN (Artificial Neural Network) that is useful in nonlinear fitting between rainfall and the other atmospheric variables. The feedforward multi-layer perceptron was used for neural network structure, and the nonlinear bipolaractivation function was used for neural network training for converting negative rainfall into no rain value. The ANN simulated rainfall was validated by leading time using Nash-Sutcliffe Coefficient of Efficiency (COE) and Coefficient of Correlation (CORR). As a result, the 3 hour rainfall accumulation basis shows that the COE of the areal mean of the Korean peninsula was improved from -0.04 to 0.31 for the 12 hr leading time, -0.04 to 0.38 for the 24 hr leading time, -0.03 to 0.33 for the 36 hr leading time, and -0.05 to 0.27 for the 48 hr leading time.

수문학적 예측에 있어서 강우수치예보의 활용성을 제고하기 위하여 인공신경망을 이용한 정량강수예측기법을 제시하였다. 본 연구에서는 2001년 6월과 7월, 2002년 8월의 중규모수치예보자료와 AWS의 3시간 누적강수, 상층기상관측소에서의 가강수량과 상대습도, 각 선행시간별 강수발생확률을 이용하여 각 선행시간에 따른 강수량을 예측하였다. 강수는 대기변수의 물리적 비선형조합으로 발생하기 때문에 강수에 영향을 미치는 대기변수와 관측강수사이의 비선형관계를 고려하는데 유용한 인공신경망기법을 이용하였다. 인공신경망의 구조는 전방향 다층퍼셉트론(feedforward multi-layer perceptron)을선택하였으며, 신경망의 학습 시 음의 강수모의값을 고려하여 무강수로전환하기 위하여 비선형 양극활성화함수를 사용하였다. 중규모수치예보모형과 인공신경망에서 예측된 강수량은 Nash-Sutcliffe Coefficient of Efficiency (NS-COE)와 Coefficient of Correlation (CORR)로 선행시간별로 통계분석을 실시하였다. 3시간 누적강수를 기준으로 NS는 한반도영역에서 평균적으로 선행시간이 12 hr인 경우 -0.04에서 0.31로, 선행시간이 24 hr인 경우 -0.04에서 0.38로, 선행시간이 36 hr인 경우 -0.03에서 0.33으로, 선행시간이 48 hr인 경우 -0.05에서 0.27로 증가하여, 강수예측의 정확도가 향상됨을 확인할 수 있었다.

Keywords

References

  1. 김호준, 백희정, 권원태, 최병철(2001). “구간 연산 신경망을 이용한 강수량 장기예측 기법.” 한국기상학회지, 한국기상학회, 제37권, 제5호, pp. 443-452.
  2. 안중배, 박정규, 임은순, 차유미(2003). “인공신경망 모형을 이용한 기온과 강수량 규모축소 연구.” 한국기상학회지, 한국기상학회, 제13권, 제1호, pp. 476-477.
  3. 정슬(2004). “인공지능시스템 I (신경회로망의 구조 및 사용법).” 충남대학교 출판부. pp. 76-106.
  4. 차유미, 안중배(2005). “역학적으로 규모축소된 남한의 여름철 강수에 대한 인공신경망 보정 능력평가.” 한국기상학회지, 한국기상학회, 제41권, 제6호, pp. 1125-1135.
  5. Bras, R.L. (1990). Hydrology: An Introduction to Hydrologic Science, Addison-Wesley Publishing company, pp. 82-92.
  6. Glahn, H.R., and Lowry, D.A. (1972). “The use of model output statistics(mos) in objective weater forecasting.” J. Appl. Meteor. Vol. 11, No. 8, pp. 1203-1211. https://doi.org/10.1175/1520-0450(1972)011<1203:TUOMOS>2.0.CO;2
  7. Gupta, M.M. (2003). Staticand Dynamic Neural Networksfrom Fundamentalsto Advanced Theory, Wiley-Interscience, IEEEPress. pp. 80-118.
  8. Hall, T., Brooks, H.E., and Doswell, III C.A. (1999). “Precipitation forecasting using a neural network.” Wea. Forecasting, Vol. 14, No. 3, pp. 338-345. https://doi.org/10.1175/1520-0434(1999)014<0338:PFUANN>2.0.CO;2
  9. Klein, W.H. (1971). “Computer prediction of precipitation probability in the united states.” J. Appl. Meteor. Vol. 10, No. 5, pp. 903-915. https://doi.org/10.1175/1520-0450(1971)010<0903:CPOPPI>2.0.CO;2
  10. Kuligowski, R.J., and Barros, A.P. (1998). “Localized precipitation forecasts from a numerical weathrer prediction model using artificial neural networks.” Wea. Forecasting, Vol. 13, No. 4, pp. 1194-1204. https://doi.org/10.1175/1520-0434(1998)013<1194:LPFFAN>2.0.CO;2
  11. Kuligowski, R.J., Barros, A.P., and Ferreira, N.J. (1998). “Experiments in short-term precipitation using artificial neural networks.” Mon. Wea. Rev., Vol. 126, No. 2, pp. 470-482. https://doi.org/10.1175/1520-0493(1998)126<0470:EISTPF>2.0.CO;2
  12. Nash, J.E., and Sutcliffe, J.V. (1970). “River flow forecasting through conceptual models part I-A discussion of principles.” Journal of Hydrology, Vol. 10, No. 3, pp. 282-290. https://doi.org/10.1016/0022-1694(70)90255-6
  13. Schoof, J.T., and Pryor, S.C. (2001). “Downscaling Temperature and Precipitation: A Comparison of Regression-Based Methods and Artificial Neural Networks.” Int. J. Climatol. Vol. 21, No. 7, pp. 773-790. https://doi.org/10.1002/joc.655
  14. Valverde Ramirez, M.C., and de Campos Velho, H.F. (2005). “Artificial neural network technique for rainfall forecasting applied to the São Paulo region.” J. Hydrol. Vol. 301, No. 1-4, pp. 146-162. https://doi.org/10.1016/j.jhydrol.2004.06.028
  15. Wilks, D.S. (1989). “Probabilistic quantitative precipitation forecasts derived from PoPs and conditional precipitaiton amount climatologies.” Mon. Wea. Rev., Vol. 118, No. 4, pp. 874-882.
  16. Wilks, D.S. (1995). Statistical methods in the Atmospheric Sciences. Academic Press, pp. 233-250.

Cited by

  1. Drought Forecasting Using the Multi Layer Perceptron (MLP) Artificial Neural Network Model vol.46, pp.12, 2013, https://doi.org/10.3741/JKWRA.2013.46.12.1249
  2. Terrestrial Sediment Yield Projection under the Bias-Corrected Nonstationary Scenarios with Hydrologic Extremes vol.8, pp.10, 2016, https://doi.org/10.3390/w8100433
  3. ROC evaluation for MLP ANN drought forecasting model vol.49, pp.10, 2016, https://doi.org/10.3741/JKWRA.2016.49.10.877