YukioroSignal was developed to mitigate house damage and accidents during snow removal, which is the most common cause of casualties in snow and ice disasters. It provides the spatial distribution of snow weight over a wide area to help people make decisions about roof snow removal. However, the current YukioroSignal cannot estimate the danger level of roof snow falling. SNOWPACK, a numerical snowpack model used in the YukioroSignal system, calculates the detailed layer structure of snow cover, and therefore has the potential to estimate this hazard. In this study, we attempted to develop a machine learning method for estimating roof-snow-falling hazard using the snow stratigraphy simulated by SNOWPACK and information on roof-snow-falling accidents from the snow and ice disaster database. Individual roof-snow-falling cases were classified into wet type and dry type based on the SNOWPACK simulation results. The hazard model was constructed using the past roof-snow-falling accident data as training data. As a result, the model showed a certain level of accuracy for both wet type and dry type during the fitting stage. In contrast, the evaluation conducted separately on training and test data showed some accuracy only for wet-type events and indicated the need for improvement for practical use. Based on the constructed machine learning model, a prototype of a roof-snow-falling hazard distribution estimation system was developed in conjunction with YukioroSignal and used to provide hazard distributions during the winter of fiscal year 2024.

1. [1] Cabinet Office, Government of Japan, “White paper disaster management in Japan 2022,” 2022. https://www.bousai.go.jp/en/documentation/white_paper/pdf/2022/R4_hakusho_english.pdf [Accessed April 10, 2026]
2. [2] H. Hirashima, T. Iyobe, K. Kawashima, and H. Sano, “Development of a snow load alert system, ‘YukioroSignal’ for aiding roof snow removal decisions in snowy areas in Japan,” J. Disaster Res., Vol.15, No.6, pp. 688-697, 2020. https://doi.org/10.20965/jdr.2020.p0688
3. [3] H. Hirashima, K. Kawashima, K. Motoya, and H. Sano, “Operation, expansion, and improvement of the snow load alert system ‘YukioroSignal’,” J. Disaster Res., Vol.19, No.5, pp. 741-749, 2024. https://doi.org/10.20965/jdr.2024.p0741
4. [4] N. Wever et al., “Verification of the multi-layer SNOWPACK model with different water transport schemes,” Cryosphere, Vol.9, No.6, pp. 2271-2293, 2015. https://doi.org/10.5194/tc-9-2271-2015
5. [5] H. Hirashima, “Numerical snowpack model simulation schemes for avalanche prediction in Japan,” Bull. Glaciol. Res., Vol.37S, pp. 31-41, 2019. https://doi.org/10.5331/bgr.18SW02
6. [6] National Research Institute for Earth Science and Disaster Resilience, https://yukibousai.bosai.go.jp/obs/news/index.php [Accessed April 10, 2026]
7. [7] J. Paine and L. Bruch, “Avalanches of snow from roofs of buildings,” Proc. Int. Snow Sci. Workshop (ISSW 1986), pp. 138-142, 1986.
8. [8] X. Cao, H. Mo, G. Zhang, Q. Zhang, and F. Fan, “Experimental study on shear strength of roof–snow interfaces for prediction of roof snow sliding,” Buildings, Vol.14, No.4, Article No.1036, 2024. https://doi.org/10.3390/buildings14041036
9. [9] T. Tomabechi, T. Ito, M. Takakura, and H. Yamaguchi, “Fundamental study on snow sliding on the roof,” J. Snow Eng. Jpn., Vol.11, No.2, pp. 88-95, 1995 (in Japanese with English abstract). https://doi.org/10.4106/jsse.11.88
10. [10] G. Zhang et al., “Numerical investigations of snow load interference effects on multiple arched roofs under wind–snow coupled actions,” Appl. Sci., Vol.15, No.23, Article No.12414, 2025. https://doi.org/10.3390/app152312414
11. [11] T. Tsutsumi et al., “Housing arrangement considering snow cover in heavy snow region—Experiments and numerical analysis on relationship between housing arrangement and snow distribution—,” J. Housing Res. Found. “JUSOKEN,” Vol.38, pp. 187-197, 2012 (in Japanese). https://doi.org/10.20803/jusokenronbun.38.0_187
12. [12] R. E. Pawluk, M. Rezvanpour, Y. Chen, and Y. She, “A sensitivity analysis on effective parameters for sliding/melting prediction of snow cover on solar photovoltaic panels,” Cold Reg. Sci. Technol., Vol.185, Article No.103262, 2021. https://doi.org/10.1016/j.coldregions.2021.103262
13. [13] M. Shimizu and T. Kimura, “Snow adhesion to roof sheets at sub-freezing temperatures,” J. Jpn. Soc. Snow Ice, Vol.54, No.3, pp. 269-275, 1992 (in Japanese with English abstract). https://doi.org/10.5331/seppyo.54.269
14. [14] T. Chiba and T. Tomabechi, “Study on the evaluation of snow load on pitched roof,” J. Struct. Constr. Eng. (Trans. AIJ), Vol.66, No.539, pp. 37-42, 2001 (in Japanese with English abstract). https://doi.org/10.3130/aijs.66.37_1
15. [15] T. Tsutsumi, “Snow and ice control on and around buildings—Summary of Japan Society for Snow Engineering Research Award 2010—,” Summ. JSSI JSSE Jt. Conf. Snow Ice Res. 2011, Session No.A1-11, 2011 (in Japanese). https://doi.org/10.14851/jcsir.2011.0.11.0
16. [16] T. Tomabechi and T. Kobayashi, “A study on countermeasures against cornice on roof,” J. Snow Eng. Jpn., Vol.5, No.1, pp. 3-9, 1989 (in Japanese with English abstract). https://doi.org/10.4106/jsse.5.3
17. [17] T. Chiba and M. Otsuki, “Study on snow cornice on eaves of flat roofs—Estimation method for forming location of the snow cornice based on roof shape coefficient considering wind speed dependence on the snow cornice—,” J. Struct. Constr. Eng. (Trans. AIJ), Vol.88, No.808, pp. 898-908, 2023 (in Japanese with English abstract). https://doi.org/10.3130/aijs.88.898
18. [18] T. Sato, “Criterion for warning of accidents related to roof snow based on analysis of their occurrence rate in the inland of Yamagata Prefecture,” J. Snow Eng. Jpn., Vol.26, No.4, pp. 211-219, 2010 (in Japanese with English abstract).
19. [19] Akita Prefecture (in Japanese). https://www.bousai-akita.jp/pages/index.html?article_id=99 [Accessed April 10, 2026]
20. [20] Yamagata Prefecture (in Japanese). https://www.pref.yamagata.jp/020072/bosai/kochibou/bousaijouhou/setsugai/yukioroshi.html [Accessed April 10, 2026]
21. [21] S. Kikuchi, K. Sugiura, and M. Hori, “Relationship between roof snow accidents and meteorological factors in Toyama Prefecture,” J. Jpn. Soc. Snow Ice, Vol.84, No.6, pp. 489-501, 2022 (in Japanese with English abstract). https://doi.org/10.5331/seppyo.84.6_489
22. [22] M. Lehning, C. Fierz, B. Brown, and B. Jamieson, “Modeling snow instability with the snow-cover model SNOWPACK,” Ann. Glaciol., Vol.38, pp. 331-338, 2004. https://doi.org/10.3189/172756404781815220
23. [23] C. Mitterer, F. Techel, C. Fierz, and J. Schweizer, “An operational supporting tool for assessing wet-snow avalanche danger,” Proc. Int. Snow Sci. Workshop (ISSW 2013), pp. 334-338, 2013.
24. [24] C. Pérez-Guillén et al., “Data-driven automated predictions of the avalanche danger level for dry-snow conditions in Switzerland,” Nat. Hazards Earth Syst. Sci., Vol.22, No.6, pp. 2031-2056, 2022. https://doi.org/10.5194/nhess-22-2031-2022
25. [25] M. Hendrick et al., “Automated prediction of wet-snow avalanche activity in the Swiss Alps,” J. Glaciol., Vol.69, No.277, pp. 1365-1378, 2023. https://doi.org/10.1017/jog.2023.24
26. [26] T. Iyobe and K. Kawashima, “Development of quasi-real-time monitoring system for regional snow cover by utilizing multiple snow data,” J. Snow Eng. Jpn., Vol.36, No.1, pp. 1-13, 2020 (in Japanese with English abstract). https://doi.org/10.32120/jsej.36.1_1
27. [27] “SNOWPACK: MET meteorological time series.” https://snowpack.slf.ch/doc-release/html/met_format.html [Accessed August 10, 2026]