P6 The NCAR Ensemble

Fossell, Kate, Glen Romine, Craig Schwartz, and Ryan Sobash, National Center for Atmospheric Research

The National Center for Atmospheric Research (NCAR) ensemble is an experimental convection-allowing ensemble prediction system run in real-time since the Spring of 2015, culminating in a two year dataset of high-resolution ensemble forecasts. The system employs a continuously cycled ensemble analysis system using the Data Assimilation Research Testbed (DART) to provide initial conditions for daily 10-member, 48-hour forecasts over the conterminous United States at 3-km horizontal grid spacing using the WRF-ARW model. The broader community has enthusiastically embraced this novel system in both operations and research. Forecasts are disseminated through a publicly available website that has been lauded as an effective demonstration of how to make ensemble forecast information more useable and useful. A subset of raw model output products has also been made available to community collaborators in real-time and an archive has been established at NCAR for public access. An overview of this long-term effort will be presented, including a description of the system design as well as highlights of system performance. Examples of current collaborations and research using the data will be offered along with data availability, access, and archive information. Challenges and lessons learned will be outlined and the fate of the ensemble and plans for future research will be briefly discussed.