Sub-Seasonal Predictability of Extreme Precipitation in CCSM4 Climate Forecasts

Accurate prediction of extreme rainfall events is particularly important due to the associated risk of flooding in many regions.  As the extreme rainfall associated with flooding is often short lived, skillful sub-seasonal climate predictions on time-scales of weeks to months can be of great benefit.  This timescale allows for predictions of rainfall variations that have the potential to lead to flooding, among many other things.  While estimates of the skill of sub-seasonal climate predictions is very important, estimates of the predictability of extreme rainfall events can aid determining the extent variability is predictable given ideal conditions.  We examine the predictability of sub-seasonal extreme rainfall in Community Climate System Model version 4 (CCSM4) May initialized hindcasts, targeting predictability in week 3/4 for the central US.  To compare the CCSM4 sub-seasonal predictions to a real-life example, we also include case study assessment of predictions during the May 2015 central US flooding, which saw record-breaking precipitation over Texas and Oklahoma.