[A-HW28] Hydrology and Water Environment

[AHW28-P05] Historical and future simulation of snow fraction and snowmelt by applying a temperature index snow model in the wider Japan Sea region

*Al- Shakil¹, Andrew Charles Whitaker² (1.Graduate School of Science and Technology, Niigata University, Japan, 2.Institute of Science and Technology, Niigata University, Japan)

Keywords:Snow fraction, Snowmelt, Snow model, Japan Sea region, Climate change

The Japan Sea region is one of the heaviest snowfall regions in the world. Here we investigate whether warming temperatures due to climate change are already causing the fraction of precipitation falling as snow (snow fraction) to decrease in recent decades in this region, which is having an impact on the seasonal timing of runoff. There are two mechanisms having an influence on seasonal timing of runoff; (1) the temperature dependence of precipitation phase (snow fraction), and (2) the temperature (energy) dependence of the snowmelt process. In this study, six AMeDAS stations (Automated Meteorological Data Acquisition System) were selected from five different prefectures; Oisawa (440m elevation, Yamagata), Tadami (377m elevation, Fukushima), Hirosaki (30m elevation, Aomori), Sumon (222m elevation, Niigata), and Kazuno and Yunotai (123m and 335m elevation, respectively, Akita prefecture). We apply a temperature index snow model using critical temperature (1.5℃) to distinguish rainfall and snowfall under historical and future conditions. For the period 1980 to 2024, four out of six stations show decreasing snow fractions, while five stations show increasing temperature trends (except Kazuno), but not significant. The average DJF temperature of these six stations range between 0.35℃ to -1.88℃. Yunotai station shows a slightly increasing snow fraction trend while Kazuno shows significant increasing trend at 10% level (Sen’s slope is 0.0033). The average DJF snow fractions in these six stations range between 0.75 to 0.86. The snow fractions of these six stations are predicted to range between 0.53 to 0.73 (T+1.5℃ scenario). Among the six stations, Tadami station shows the strongest relationship between the mean DJF temperature and snow fraction and Kazuno station shows the weakest relationship (R² values range between 0.38 to 0.81).
We looked at the differences between historical and future (T+1.5ºC) for the simulated DJFM rainfall and simulated DJFM snowmelt to try and determine which of the two mechanisms introduced above is dominant in influencing the seasonal timing of runoff. The differences (future minus historical) for average DJFM rainfall in Tadami, Sumon, Hirosaki, Kazuno, and Yunotai stations were much greater than the differences for average DJFM snowmelt. For Oisawa station the average increase in DJFM rainfall and the average increase in DJFM snowmelt is approximately the same. In the case of every station, the largest increases in rainfall amount occurred during December and January, being the months with the highest precipitation. Results are still tentative, but we can conclude that the mechanism (1) is dominant over the mechanism (2) in this region.