碩士 / 國立中央大學 / 大氣物理研究所 / 99 / Land use (LU) type is one of the important land surface data to describe the exchange of heat and momentum between land surface and atmosphere in the model. The U.S. Geological Survey (USGS) land use data used in Weather Research and Forecasting (WRF) model is roughly at 1-km resolution, this data set is outdated with the reference year of 1990s. The WRF version newer than 3.1, provides another LU datasets which is derived from 2001 MODIS satellites products and at 1-km resolution. However, none of these LU datasets correctly characterizes the land use and land cover classifications in Taiwan Island. Regarding this problem, a new land use data is derived using the 2007 SPOT satellite images (this data is named NCU). There are significant differences among these three datasets.
When synoptic flow was weak, the simulation results using different land use data showed big effect. Due to relative widely distribution of the type of urban in MODIS, the simulation result is unable to describe the real land use condition correctly. The WRF-MODIS simulation over-predicted maximum temperatures and induced stronger sea breeze flow. On the contrary, the main distribution of USGS land use data is irrigation in Taiwan area. The WRF-USGS simulation results to under-predicted minimum temperature and induced weaker sea breeze flow in the coast. The WRF-NCU simulation result is relatively near observed temperature. The primary land type in USGS is irrigation in Taiwan and the roughness length is set too small, so the WRF-USGS simulation over-predicted wind speed. The WRF-NCU and WRF-MODIS simulations are relatively near observed wind speed. Comparisons with various meteorological measurements show that the NCU land use description improve the simulation of meteorological field apparently.
Identifer | oai:union.ndltd.org:TW/099NCU05021030 |
Date | January 2011 |
Creators | Yu-ching Hsu, 許郁卿 |
Contributors | Fang-yi Cheng, Pay-liam Lin, 鄭芳怡, 林沛練 |
Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
Language | zh-TW |
Detected Language | English |
Type | 學位論文 ; thesis |
Format | 88 |
Page generated in 0.0137 seconds