All Issue

2025 Vol.58, Issue 11S-1 Preview Page

Review Article

Quantitative assessment of spatial heterogeneity in extreme snowfall using frequency analysis and robust clustering techniques

극한 적설의 공간적 이질성 정량 평가를 위한 빈도해석 및 강건 군집 기법 적용

Hyeongjoo Leea
,
Gunhui Chungb*
이 형주a
,
정 건희b*
aResearch Professor, Fine Dust and Carbon Neutrality Institute (FNI), Hoseo University, Asan, Korea
bProfessor, Department of Architecture & Civil Engineering, Hoseo University, Asan, Korea
a호서대학교 미세먼지 및 탄소중립 연구소 연구교수
b호서대학교 건축토목공학부 교수
*Corresponding Author
30 November 2025. pp. 1167-1180
PDF XML Citation
Abstract

This study presents a comprehensive framework for quantifying the spatial heterogeneity of extreme snowfall in South Korea. Using maximum snow depth data for various return periods derived from nationwide meteorological stations, the analysis integrates frequency analysis, spatial interpolation, clustering, and robustness evaluation. Among five extreme value distributions (Gumbel, GEV, Weibull, Pearson-III, and Lognormal), the best-fitting distribution at each station was selected based on the Kolmogorov-Smirnov (K-S) test, with GEV (42%), Gumbel (31%), and Weibull (18%) being the most frequently optimal. The spatial distribution of maximum snow depth for return periods (2, 5, 10, 30, 50, 100, 200 years) was derived using Ordinary Kriging (OK), revealing that as the return period increases, the zones of extreme snowfall intensify and expand inland, particularly across the East Sea coast and mountainous regions. At the 200-year return period, the maximum snow depth reached approximately 350 cm. Clustering was conducted using K-means, Gaussian Mixture Models (GMM), and Hierarchical clustering, with robustness assessed via 200 bootstrap iterations using the Adjusted Rand Index (ARI). GMM (median ARI = 0.746) and K-means (0.574) demonstrated higher stability and consistency in defining cluster boundaries. These findings offer valuable insights for precise regional risk assessment of extreme snowfall and for the design of resilient infrastructure and disaster preparedness strategies.

Keywords
Extreme snowfall
Return period
Frequency analysis
Spatial interpolation
Cluster analysis
Robustness evaluation

본 연구는 전국 기상관측소에서 산정된 재현기간별 최대 적설심을 바탕으로 빈도해석-공간보간-군집화-강건성 평가를 통합하여 극한 적설의 공간적 이질성을 정량적으로 규명하였다. 5개 극치분포(Gumbel, GEV, Weibull, Pearson-III, Lognormal) 중 Kolmogorov-Smirnov 검정을 통해 지점별 최적 분포를 선정한 결과, GEV (42%), Gumbel (31%), Weibull (18%) 순으로 높은 적합도를 보였다. 재현기간(2, 5, 10, 30, 50, 100, 200년)별 최대 적설심의 공간 분포는 Ordinary Kriging (OK) 보간을 통해 도출되었으며, 재현기간 증가에 따라 적설 심도가 집중되는 지역이 동해안 및 산악지대를 중심으로 내륙까지 확장되고, 200년 재현기간에서는 최대 적설심이 약 350 cm 수준까지 도달하는 것으로 나타났다. 군집화(K-means, GMM, Hierarchical)와 200회 부트스트랩 기반 Adjusted Rand Index (ARI) 평가 결과, GMM (중앙값 0.746)과 K-means (0.574)가 상대적으로 높은 안정성을 보였으며, 이는 군집 경계의 신뢰성 확보에 기여할 수 있음을 시사한다. 이러한 결과는 극한 적설에 대한 공간적 위험도를 정밀하게 평가하고, 지역 특성을 반영한 재난 대응 및 인프라 설계에 활용될 수 있다.

키워드
극한 적설
재현기간
빈도해석
공간 보간
군집 분석
강건성 평가
References
1

Blanchet, J., Marty, C., and Lehning, M. (2009). “Extreme value statistics of snowfall in the Swiss Alpine region.” Water Resources Research, Vol. 45, No. 5, W05424. doi: 10.1029/2009WR007916.

10.1029/2009WR007916
2

Cho, E., McCrary, R.R., and Jacobs, J.M. (2021). “Future changes in snowpack, snowmelt, and runoff potential extremes over North America.” Geophysical Research Letters, Vol. 48, pp. e2021GL094985.

10.1029/2021GL094985
3

Eccel, E., Cau, P., and Ranzi, R. (2012). “Data reconstruction and homogenization for reducing uncertainties in high-resolution climate analysis in Alpine regions.” Theoretical and Applied Climatology, Vol. 110, pp. 345-358.

10.1007/s00704-012-0624-z
4

Hanel, M., Vizina, A., and Máca, P. (2009). “Hydrologic modeling and trend detection in time series of hydrologic variables.” Hydrology and Earth System Sciences, Vol. 13, No. 7, pp. 1291-1301.

5

Hossain, A., Lee, S., Kim, H., and Han, D. (2021). “Comparison of estimation techniques for snow depth analysis in cold regions.” Natural Hazards, Vol. 107, No. 1, pp. 239-258.

6

Kim, H.-U., Sohn, C., and Han, S.-G. (2012). “Identifying the optimal number of homogeneous regions for regional frequency analysis using self-organizing map.” Journal of Korea Spatial Information Society, Vol. 20, No. 6, pp. 13-21.

10.12672/ksis.2012.20.6.013
7

Kudo, K., Shimizu, K., and Ishikawa, Y. (2017). “The influence of temperature and precipitation on snowfall and snowpack in Japan.” Journal of Hydrology, Vol. 552, pp. 123-135.

8

Li, M., Liu, J., Zhang, X., and Wang, W. (2023). “Comparison of estimation techniques in snow frequency analysis.” Journal of Hydrology, Vol. 615, 128673.

9

Marty, C., and Blanchet, J. (2012). “Long-term changes in annual maximum snow depth and snowfall in Switzerland based on extreme value statistics.” Climatic Change, Vol. 111, No. 3, pp. 705-721.

10.1007/s10584-011-0159-9
10

Moller, A., Isaksen, K., Dyrrdal, A.V., and Benestad, R. (2017). “Modeling snow depth extremes in a changing climate.” Natural Hazards, Vol. 88, No. 3, pp. 1403-1425.

11

Nam, W.-S., Kim, T.-S., Shin, J.-Y., and Heo, J.-H. (2008). “Regional rainfall frequency analysis by multivariate techniques.” Journal of Korea Water Resources Association, Vol. 41, No. 5, pp. 517-525.

10.3741/JKWRA.2008.41.5.517
12

Oh, T.-S., Moon, Y.-I., and Oh, K.-T. (2008). “Estimation of probability precipitation by regional frequency analysis using cluster analysis and variable kernel density function.” Journal of the Korean Society of Civil Engineers, Vol. 28, No. 2B, pp. 225-236.

13

Park, H.S., and Chung, G.H. (2019). “Frequency analysis for annual maximum of daily snow accumulations using conditional joint probability distribution.” Journal of Korea Water Resources Association, Vol. 52, No. 9, pp. 627-635.

10.3741/JKWRA.2019.52.9.627
14

Shin, J.-Y., Kim, T.-W., Kim, S.-C., and Yoon, S.-B. (2013). “Statistical frequency analysis of earthquake data at East Sea using mixed distribution functions.” The Magazine of The Korean Society of Hazard Mitigation, Vol. 13, No. 5, pp. 347-354.

10.9798/KOSHAM.2013.13.5.347
15

Smyth, E., Pomeroy, J.W., and Janowicz, J.R. (2020). “Improving SWE estimation with data assimilation: The influence of snow depth.” Water Resources Research, Vol. 56, No. 4, e2019WR026853. doi: 10.1029/2019WR026853.

10.1029/2019WR026853
16

Troin, M., Caya, D., and Brissette, F. (2016). “Comparison of projected changes in snow hydrology in Canada using a suite of regional climate models.” Journal of Hydrology, Vol. 541, pp. 353-370.

17

Yoo, J.-Y., Kim, T.-W., and Kim, S.-D. (2010). “Drought frequency analysis using cluster analysis and bivariate probability distribution.” Journal of the Korean Society of Civil Engineers, Vol. 30, No. 6B, pp. 599-606.

18

Zarzycki, C.M. (2018). “Projecting changes in societally impactful northeastern U.S. snowstorms.” Geophysical Research Letters, Vol. 45, pp. 12067-12075.

10.1029/2018GL079820
Information
  • Publisher :KOREA WATER RESOURECES ASSOCIATION
  • Publisher(Ko) :한국수자원학회
  • Journal Title :Journal of Korea Water Resources Association
  • Journal Title(Ko) :한국수자원학회 논문집
  • Volume : 58
  • No :11
  • Pages :1167-1180
  • Received Date : 2025-09-22
  • Accepted Date : 2025-09-29
  • DOI :https://doi.org/10.3741/JKWRA.2025.58.S-1.1167
Journal Informaiton Journal of Korea Water Resources Association Journal of Korea Water Resources Association
Online Submission
Archives
  • scopus
  • KCI-수자원
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close