Improvement of Snow Depth Reproduction in Japanese Urban Areas by the Inclusion of a Snowpack Scheme in the SPUC Model

Journal of the Meteorological Society of Japan

Naoto HORI¹  Rui ITO²  Toshinori AOYAGI³  Hiroaki KAWASE⁴  Koji DAIRAKU⁵  Mitsuo OH'IZUMI⁶ 

[1] Japan Meteorological Agency, Tokyo, Japan;Japan Meteorological Business Support Center, Tsukuba, Japan;Meteorological Research Institute, Japan Meteorological Agency, Tsukuba, Japan;Meteorological college, Japan Meteorological Agency, Kashiwa, Japan;National Research Institute for Earth Science and Disaster Resilience, Tsukuba, Japan;Yamagata Meteorological office, Japan Meteorological Agency, Yamagata, Japan

关键词: snowpack;    urban canopy;    model development;    regional model;   

DOI  :  10.2151/jmsj.2018-053

学科分类：大气科学

来源: Meteorological Society of Japan

【 摘 要 】

 Accurate simulation of urban snow accumulation/melting processes is important to provide reliable information about climate change in snowy urban areas. The Japan Meteorological Agency operates a square prism urban canopy (SPUC) model within their regional model to simulate the urban atmosphere. However, presently, this model takes no account of snow processes. Therefore, in this study, we enhanced the SPUC by introducing a snowpack scheme, and assessed the simulated snow over Japanese urban areas by comparing the snow depths from the enhanced SPUC and those from a simple biosphere (iSiB) model with the observations. Snowpack schemes based on two approaches were implemented. The diagnostic approach (sSPUCdgn) uses empirical factors for snow temperature and melting/freezing amounts and the Penman equation for heat fluxes, whereas the prognostic approach (sSPUCprg) calculates snow temperatures using heat fluxes estimated from bulk equations. Both snowpack schemes enabled the model to accurately reproduce the seasonal variations and peaks in snow depth, but it is necessary to use sSPUCprg if we wish to consider the physical processes in the snow layer. Compared to iSiB, sSPUCprg resulted in a good performance for the seasonal variations in snow depth and the error fell to 20 %. While iSiB overestimated the snow depth, a cold bias of over 1°C appeared in the daily mean temperature, which can be attributed to excessive decreases in the snow surface temperature. sSPUCprg reduces the bias by a different calculation method for the snow surface temperature and by including heated building walls without snow; consequently, the simulated snow depth is improved. With an increase in the correlation coefficient, sSPUCprg generated a relationship between the seasonal variations in snowfall and snow depth close to the observed relationship. Therefore, the simulation accuracy of snowfall becomes more crucial for simulating the surface snow processes precisely by using the enhanced SPUC.

【 授权许可】

Unknown