Dynamical Downscaling of GCM Simulations: Toward the Improvement of Forecast Bias over California

Chin, H S

关键词: AGRICULTURE;    AIR QUALITY;    ATMOSPHERIC CIRCULATION;    CALIFORNIA;    CLIMATE MODELS;    CLIMATES;    DISTRIBUTION;    MANAGEMENT;    MOUNTAINS;    PHYSICS;    PRECIPITATION;    RESOLUTION;    TOPOGRAPHY;    WATER;    WATER SUPPLY;   

DOI  :  10.2172/945745
RP-ID  :  LLNL-TR-407576
PID  :  OSTI ID: 945745
Others  :  TRN: US200904%%139

学科分类：环境科学（综合）

美国|英语

来源: SciTech Connect

【 摘 要 】

The effects of climate change will mostly be felt on local to regional scales. However, global climate models (GCMs) are unable to produce reliable climate information on the scale needed to assess regional climate-change impacts and variability as a result of coarse grid resolution and inadequate model physics though their capability is improving. Therefore, dynamical and statistical downscaling (SD) methods have become popular methods for filling the gap between global and local-to-regional climate applications. Recent inter-comparison studies of these downscaling techniques show that both downscaling methods have similar skill in simulating the mean and variability of present climate conditions while they show significant differences for future climate conditions (Leung et al., 2003). One difficulty with the SD method is that it relies on predictor-predict and relationships, which may not hold in future climate conditions. In addition, it is now commonly accepted that the dynamical downscaling with the regional climate model (RCM) is more skillful at the resolving orographic climate effect than the driving coarser-grid GCM simulations. To assess the possible societal impacts of climate changes, many RCMs have been developed and used to provide a better projection of future regional-scale climates for guiding policies in economy, ecosystem, water supply, agriculture, human health, and air quality (Giorgi et al., 1994; Leung and Ghan, 1999; Leung et al., 2003; Liang et al., 2004; Kim, 2004; Duffy et al., 2006). Although many regional climate features, such as seasonal mean and extreme precipitation have been successfully captured in these RCMs, obvious biases of simulated precipitation remain, particularly the winter wet bias commonly seen in mountain regions of the Western United States. The importance of regional climate research over California is not only because California has the largest population in the nation, but California has one of the most sophisticated water collection and distribution systems in the world. Therefore, adapting California's water management system to climate change presents significant challenges. Besides, the strong scale interaction between atmospheric circulation and topography in this region provides a challenging testbed for RCMs. Thus, the success of California winter precipitation forecast over mountains would greatly help develop a reliable water management system to adapt to climate change.