Journal of Korea Water Resources Association (한국수자원학회논문집)

Korea Water Resources Association (한국수자원학회)
권호 표지
DOI QR코드

DOI QR Code

A Probabilistic Estimation of Changing Points of Seoul Rainfall Using BH Bayesian Analysis

BH 베이지안 분석을 통한 서울지점 강우자료의 확률적 변화시점 추정

  • 황석환 (한국건설기술연구원 수자원연구실) ;
  • 김중훈 (고려대학교 공과대학 건축.사회환경공학부) ;
  • 유철상 (고려대학교 공과대학 건축.사회환경공학부) ;
  • 정성원 (한국건설기술연구원 수자원연구실)
  • Received : 2009.04.01
  • Accepted : 2010.06.30
  • Published : 2010.07.31
Download PDF
⟨ Previous Next ⟩

Abstract

In this study, occurrences of relative probabilistic changing points between Chukwooki rainfall data (CWK) and modern rain gage data (MRG) were analyzed using Barry and Hartigan (BH) Bayesian changing points estimation method which estimated the changing points by calculation of change probabilities at each point. Since any natural phenomenon cannot be simulated identically and perfectly, a statistical method which can not consider the sequential order has its limitation on prediction of a specific time of occurrence. In this respect, Homogeneity analysis between CWK and MRG was performed through the occurrence investigation of relative probabilistic changing points for four rainfall characteristics of data sets using BH bayesian model which estimate the change point by calculating the relative probabilities in each data points. The results show that statistical characteristics of CWK are not different significantly from MRG, even though considered that there may be little quantitative difference CWK and MRG caused from limitation of measurement accuracy of CWK.

본 논문에서는 각각의 시점에서의 변화확률을 산정하여 변화시점을 추정하는 Barry and Hartigan (BH)의 베이지안 변화시점 추정방법(Bayesian changing points estimation method)을 이용하여 측우기 관측자료계열(CWK)과 근대우량계 관측자료계열(MRG)간의 변화에 대한 상대확률적 절점의 발생여부를 분석하였다. 어떠한 자연 현상도 완전히 동일하게 재현되지 않기때문에 시간적인 순서를 고려하지못하는 통계적 방법은 구체적인발생시간을 예측하기 어렵다는 점에서 자료의 변화량 파악은 가능하나 변화시점을 명확히 추정하는데 한계가 있다. 이러한 배경에서, 본 논문에서는 각 시점에서의 변화확률을 산정하여 변화시점을 추정하는 BH 베이지안 방법을 적용하여 CWK와 MRG의 각 강우특성별로 상대확률적인 변화시점 분석을 통하여 CWK와 MRG 간의 동질성 분석을 실시하였다. 분석 결과, CWK의 정성적인(본질적인) 통계적 특성은 MRG와 큰 차이가 없는 것으로 나타났다. 다만, 관측정밀도의 한계로 인한 정량적인 차이가 존재하는 것으로 분석되었다.

Keywords

References

  1. 김기욱, 유철상, 박민규, 김현준(2007). “강우빈도해석에 서의측우기자료의유용성평가.” 한국수자원학회논문집, 한국수자원학회, 제40권, 제11호, pp. 851-859.
  2. 김상욱, 이길성(2008a). “Bayesian MCMC를 이용한 저수량 점 빈도해석: I. 이론적 배경과 사전분포의 구축.” 한국수자원학회논문집, 한국수자원학회, 제41권, 제1호, pp. 35-47. https://doi.org/10.3741/JKWRA.2008.41.1.035
  3. 김상욱, 이길성(2008b). “Bayesian MCMC를 이용한 저수량 점 빈도해석: II. 적용과 비교분석.” 한국수자원학회논문집, 한국수자원학회, 제41권, 제1호, pp. 49-63. https://doi.org/10.3741/JKWRA.2008.41.1.049
  4. 김상욱, 이길성(2008c). “Bayesian 다중회귀분석을 이용한 저수량(Low flow) 지역 빈도분석.” 한국수자원학회논문집, 한국수자원학회, 제41권, 제3호, pp. 325-340. https://doi.org/10.3741/JKWRA.2008.41.3.325
  5. 김상욱, 이길성(2008d). “베이지안 회귀분석을 이용한 수위-유량관계곡선의 불확실성 분석.” 한국수자원학회논문집, 한국수자원학회, 제41권, 제9호, pp. 943-958. https://doi.org/10.3741/JKWRA.2008.41.9.943
  6. 이병설(1970). “서울의 연강수량 및 하기강수량의 Normality에 관한 연구.” 한국기상학회지, 한국기상학회, 제5권, pp. 11-14.
  7. 전종갑, 문병권(1997). “측우기 강우량 자료의 복원과 분석.” 한국기상학회지, 한국기상학회, 제33권, 제4호, pp. 691-707.
  8. 정현숙, 임규호(1994). “서울 지역 월강수량 강수일의 관계, 1770-1907.” 한국기상학회지, 한국기상학회, 제30권, 제4호, pp. 487-505.
  9. 조희구, 나일성(1979). “18세기한국의기후변동: 강우량을 중심으로.” 동방학지, 제22권, pp. 83-103.
  10. 황석환, 김중훈, 유철상, 유도근(2009). “서울지점 강우자료의 정량적 동질성 분석.” 한국방재학회논문집, 한국방재학회, 제9권, 제4호, pp. 29-35.
  11. Barry, D., and Hartigan, J.A. (1992). “Product partition models for change point problems.” The annals of Statistics, Vol. 20, No. 1, pp. 260-279. https://doi.org/10.1214/aos/1176348521
  12. Barry, D., and Hartigan, J.A. (1993). “A bayesian analysis for change point problems.” Journal of the American Statistical Association, Vol. 88, No. 421, pp. 309-319. https://doi.org/10.2307/2290726
  13. Erdman, C., and Emerson, J.W. (2007b). “bcp: A Package for Performing a Bayesian Analysis of Change Point Problems.” R package version 1.8.4, URL http://CRAN.Rproject.org/.
  14. Kuczera, G. (1999). “Comprehensive at-site flood frequency analysis using Monte Carlo Bayesian inference.” Water Resources Research, Vol. 35, No. 5, pp. 1551-1557. https://doi.org/10.1029/1999WR900012
  15. Madsen, H., and Rojsberg, H.D. (1997). “Generalized least squares and empirical Bayes estimation in regional partial duration series index flood modeling.” Water Resources Research, Vol. 33, No. 4, pp. 771-781. https://doi.org/10.1029/96WR03850
  16. O'Connel, D.R.H., Ostenaa, D.A., Levish, D.R., and Klinger, R.E. (2002). “Bayesian flood frequency analysis with paleohydrologic bound data.” Water Resources Research, Vol. 38, Issue 5, p. 1058. https://doi.org/10.1029/2000WR000028
  17. Perreault, L., Bernier, J., Bobee, B., and Parent, E. (2000a). “Bayesian change-point analysis in hydrometeorological time series. Part 1. The normal model revisited.” Journal of Hydrology, Vol. 235, pp. 221-241. https://doi.org/10.1016/S0022-1694(00)00270-5
  18. Perreault, L., Bernier, J., Bobee, B., and Parent, E. (2000b). “Bayesian change-point analysis in hydrometeorological time series. Part 2. Comparison of change-point models and forecasting.” Journal of Hydrology, Vol. 235, pp. 242-263. https://doi.org/10.1016/S0022-1694(00)00271-7
  19. Reis Jr., D.S., and Stedinger, J.R. (2005). “Bayesian MCMC flood frequency analysis with historical information.” Journal of Hydrology, Vol. 313, pp. 97-116. https://doi.org/10.1016/j.jhydrol.2005.02.028
  20. Reis Jr., D.S., Stedinger, J.R., and Martins, E.S. (2005). “Bayesian generalized least squares regression with application to long Pearson type III regional skew estimation.” Water Resources Research, Vol. 41, W10419. https://doi.org/10.1029/2004WR003445
  21. Seidou, O., Ouarda, T.B.M.J., Barbet, M., Bruneau, P., and Bobee, B. (2006). “A parametric Bayesian combination of local and regional information in flood frequency analysis.” Wate Resources Research, Vol. 42, W11408. https://doi.org/10.1029/2005WR004397
  22. Vicens, G.J., Rodriguez-Itrube, I., and Schaake Jr., J.C. (1975). “A Bayesian Framework for the Use of Regional Information in Hydrology.”Water Reources Research, Vol. 11, No. 3, pp. 405-414. https://doi.org/10.1029/WR011i003p00405
  23. Wood, E.F., and Rodriguez-Itrube, I. (1975a). “Bayesian Inference and Decision Making for Extreme Hydrologic Events.” Water Resources Research, Vol. 11, No. 4, pp. 533-542. https://doi.org/10.1029/WR011i004p00533
  24. Wood, E.F., and Rodriguez-Itrube, I. (1975b). “Bayesian Approach to Analyze Uncertainty Among Flood Frequency Models.”Water Resources Research, Vol. 11, No. 6, pp. 839-843. https://doi.org/10.1029/WR011i006p00839
  25. Yao, Y.C. (1984). “Estimation of a noisy discrete-time step function: bayes and empirical bayes approaches.” Annals of Statistics, Vol. 12, pp. 1434-1447. https://doi.org/10.1214/aos/1176346802

Cited by

  1. Estimation of the Regional Future Sea Level Rise Using Long-term Tidal Data in the Korean Peninsula vol.47, pp.9, 2014, https://doi.org/10.3741/JKWRA.2014.47.9.753
  2. Investigating Changes over Time of Precipitation Indicators vol.15, pp.2, 2013, https://doi.org/10.17663/JWR.2013.15.2.233
  3. Intercomparison of Change Point Analysis Methods for Identification of Inhomogeneity in Rainfall Series and Applications vol.47, pp.8, 2014, https://doi.org/10.3741/JKWRA.2014.47.8.671