High-resolution hydrometeorological forecast in Southwest China based on a multi-layer nested WRF model. Issue 1 (December 2020)

• Record Type: Journal Article
• Authors: Chen, Z N; Li, J; Zhu, Y; Hu, L C; Wen, X; Lei, X H
• Is Part Of: IOP conference series. Volume 612:Issue 1(2020)
• Publication Date: 2020-12
• Subjects: Earth sciences -- Periodicals; Environmental sciences -- Congresses; Environmental sciences -- Periodicals; 550.5
• DOI: 10.1088/1755-1315/612/1/012062
• Languages: English
• ISSNs: 1755-1307

Abstract

Abstract: In this study, a high-resolution (5km:1km) regional hydrometeorological simulation (Weather Research and Forecasting, WRF) in Southwest China was evaluated by comparisons with the multiple General Circulation Model (multi-GCM) ensemble mean from Coupled Model Intercomparison Project phase 5 (CMIP5) and in-situ observation data, to prove its advantage to precisely delineate the regional complex topographical and climatic conditions. The temperature and precipitation were selected to evaluate the model performance skills. Simulations of the spatiotemporal rainfall and near-surface air temperature distribution across the entire research area and at four specific sites (Ganzi, Daofu, Jiulong Huili) were analyzed based on observational data from 2007–2010. Overall, both the WRF and multi-GCM demonstrated satisfactory capabilities in representing seasonal variation, but systematic biases remained. The regional average near-surface air temperature of WRF outputs had cold biases of −4.91, −1.96, −3.92 and −8.17°C in spring, summer, autumn and winter, respectively, and wet biases of 40.5 – 428.5 mm in cumulative precipitation over the four seasons. Overall, the multi-GCM means had consistent bias, but were closer to regional averages derived from in-situ data. At the four validation stations, the WRF outputs consistently performed better for temperature and precipitation according to the correlation coefficient, root-mean-square error, and index of agreement. The simulation capabilities identified herein can serve as a foundation for addressing WRF model biases and improving projection accuracy in the future.