MultiScheme Comparative Test and Ensemble of Cloud Top Height and Temperature Forecasting
- Article
- Figures
- Metrics
- Preview PDF
- Reference
- Related
- Cited by
- Materials
Abstract:
In this paper, three singlemoment microphysic schemes (Lin, WSM6, and Thompson) and three doublemoment microphysic schemes (Morrison 2mom, CAM51, and NSSL 2mom) are used to simulate 11 cases of cloud processes over Jiangsu Province based on the WRF model version 361 Using the cloud top height and temperature retrieved from satellite measurements to test the impact of different schemes on the two macro cloud parameters and select the best scheme by evaluating the performance of each scheme. Meanwhile, considering the uncertainty of individual schemes, the multischeme ensemble is used to reduce the systematic errors. Results show that all six schemes can roughly simulate the distribution of cloud bands, but there are still some differences in the range of cloud bands. The simulated cloud bands using CAM51 and NSSL 2mom scheme are larger than observations, while the others are smaller. The CAM51 forecasts of clouds are superior to those of the other schemes in terms of the anomaly correlation coefficients (ACC) and the rootmeansquare errors (RMSE). Although the NSSL scheme can reasonably simulate the location of the extreme center of cloud top height and temperature, it is still the worst scheme in terms of RMSE. The simulation performances of the Lin, WSM6 and Thompson schemes are almost the same，which perform better than that of the Morrison scheme. The multischeme ensemble mean and biasremoved ensemble mean reduce the RMSE of the cloud top height and temperature forecasts considerably. The biasremoved ensemble mean shows an improvement of forecast skill over the best single scheme forecast and the multischeme ensemble mean.
Keywords:
Project Supported:
Clc Number:
You are thevisitors
Supported by:Beijing E-Tiller Technology Development Co., Ltd.
Supported by:Beijing E-Tiller Technology Development Co., Ltd.