Sensitivity of topographic resolution to snow over the complex mountainous areas in Japan
Abstract
The reproducibility of snow cover is greatly dependent on the horizontal resolution of the numerical models. It is well known that the numerical simulation with high resolution improves the precipitation process and topography, resulting in better reproducibility in the distribution of snow cover. Although the snow cover is influenced by both the surface air temperature and precipitation, the contribution of these two effects has not been separately evaluated. The aim of this study is to investigate the resolution dependency of the snowfall and snow cover using the regional climate model. We focus on Hokuriku region, which is a heavy snowfall area in Japan. We used the Advanced Research Weather Research and Forecasting (WRF) modeling system Version 3.4. The initial and lateral boundary conditions are interpolated from the ERA-interim data set. We used the Noah-LSM including a one-layer snow model as a land surface model. We performed two numerical experiments with different horizontal resolutions; one is 4.5km (EXP4.5) and another is 1.5km (EXP1.5). Moreover, we performed the numerical experiment with 1.5 km horizontal resolution using the topography with 4.5 km resolution (EXP1.5_TOPO4.5). The EXP4.5 and EXP1.5 show that the difference in snow depth (Dif_snow) has a positive correlation with the difference in altitude (Dif_alt). The trend is 0.14 cm/m and determination coefficient is 078 (Figure). The comparison of EXP4.5 with EXP1.5 indicates that snow depth relates to not only surface temperature but precipitation. The surface temperature in the EXP1.5_TOPO4.5 is higher than the EXP1.5 in mountainous areas, which corresponds to the different elevations in the experiments. On the other hand, a difference in precipitation does not correspond to the elevation between the EXP1.5 and the EXP1.5_TOPO4.5. Precipitation decreases at the high altitude, i.e., around the top of mountain, while increase at the middle altitude i.e., around mountain slope. Our results indicate that the resolution of topography in the model contributes to snow depth through temperature, which corresponds to the model elevation, and precipitation, which corresponds to the shape and elevation of the mountains. Figure, Scatterplot of the altitude difference in the model versus the estimation difference of snow depth. The difference describes 4.5km-1.5km. The solid line indicates regression line.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2012
- Bibcode:
- 2012AGUFM.A41H0058U
- Keywords:
-
- 0550 COMPUTATIONAL GEOPHYSICS / Model verification and validation;
- 0736 CRYOSPHERE / Snow;
- 3354 ATMOSPHERIC PROCESSES / Precipitation;
- 9320 GEOGRAPHIC LOCATION / Asia