Co-Authors: Xiao-Wei Quan; Jonathan Beverley, OAR/PSL & CIRES University of Colorado
We are in the first year of developing the Seasonal Forecast Application of the UFS. The early development is centered around a low resolution (1-degree) version of the UFS. In order to establish a baseline we carried out a 50-year free coupled simulation with the Prototype 8 version of the UFS, and an additional 50-year run with stochastic physics active. In addition, we also carried out a 30-year of seasonal re-forecasts for one initialization time as a test to see if the 1-degree model is capable of being a configuration that we can use to develop the Seasonal Forecast System (SFS).
Initial results show that the 1-degree version of the UFS lacks El Niño variability unless stochastic physics is active. Additionally, the re-forecasts show superior skill over the current operational seasonal forecast model (CFSv2). We will present an analysis of the climate simulations and seasonal forecasts runs.
Mr. Philip Pegion is a physical scientist and deputy division chief for the Modeling and Data Assimilation Division within the Physical Sciences Laboratory of NOAA. He was previously with the University of Colorado, Cooperative Institute for Research in Environmental Sciences (CIRES). Mr. Pegion received his Master of Science in Meteorology from Florida State University in 1999 then worked in the NASA Seasonal-to-Interannual Prediction Project/Global Modeling and Assimilation office for over eight years, then one year at the National Weather Service’s Climate Prediction Center before joining the University of Colorado in 2009. His current research focus is on improving numerical weather prediction through advances in atmospheric modeling, data assimilation, and better use of ensembles by better accounting of model uncertainty.
He has previously studied the role of SST forcing and atmospheric internal variability in limiting the predictability of seasonal climate, statistical post-processing of seasonal forecasts, initialization of the coupled model for ensemble forecasts, and supported the development of MERRA.
