8.4 Forecasting experiments of 2011-Korea 100-year flood weather using NCAR WRF-RTFDDA and Ensemble-RTFDDA

Thursday, 10 January 2013: 11:45 AM
Room 18B (Austin Convention Center)
Yubao Liu, NCAR, Boulder, CO; and W. Y. Y. Cheng, Y. Liu, S. W. Lee, J. Y. Byon, Y. J. Choi, Y. S. Park, and S. L. Kang

Heavy rainfall around the national capital Seoul and nearby Siheung regions in South Korea in late July 2011 triggered flash floods and landslides that killed nearly 70 people by July 28. At least 86 power outages followed the landslides, affecting 125,000 people by July 27. Over 11,000 South Koreans were forced to evacuate. The NCAR-ATEC WRF-based high-resolution Real-Time Four-Dimensional Data Assimilation and forecasting system (RTFDDA) and its ensemble model E-RTFDDA were used to conduct retrospective forecasting experiments aiming to test and demonstrate the WRF-RTFDDA modeling capability of forecasting such high impact weather event. The result of the modeling experiments shows that 1) the WRF-RTFDDA system forecast over 650mm rainfall during July 26 and 27 in the Seoul regions, which is very close to what was measured; 2) the spun-up cloud and precipitation, and dynamical processes with the WRF-RTFDDA 4D continuous data assimilation scheme significantly improve the prediction of the major precipitation episodes; 3) the rapid-update FDDA and forecasting cycles are able to continuously update the longer-range forecast with more accurate shorter-range forecasts; and 4) properly setting the model domain locations can have a great impact on the simulation of the convective systems. Analysis of the output of the E-RTFDDA forecasts for the event is under way and the result will also be reported at the meeting.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner