Thursday, 10 January 2019: 2:00 PM
North 131AB (Phoenix Convention Center - West and North Buildings)
Technological advances have allowed for the operational use of large, convective-scale ensembles. However, these ensembles typically only launch new forecasts once or twice a day due to computational expense, resulting in idle observations that could improve predictability. Ensemble sensitivity is a useful and computationally inexpensive tool for analyzing how features in the flow at early forecast times affect different relevant forecast features later in the forecast window. Given the surplus of observations relative to data assimilation cycles, ensemble sensitivity may be used to increase predictability and forecast accuracy through an objective ensemble subsetting technique.
This technique identifies ensemble members with the smallest errors in regions of high sensitivity to produce a smaller, more representative ensemble subset. The procedure was subjectively evaluated in the 2018 NOAA Hazardous Weather Testbed and has since been objectively tested and optimized. Sensitivity-based subsets are generated from the 00 UTC TTU WRF 4-km ensemble runs from the spring and summer of 2016. Errors of the sensitivity-based subsets are compared to those of the full ensemble as well as errors from non-sensitivity-based subsets. A number of degrees of freedom associated with this technique including sensitivity variable, sensitivity time, sensitivity threshold, subset size, and analysis source, are tested to identify the optimal subsetting technique for convective scales. Results are analyzed and ideal parameters for convective-scale, sensitivity-based subsets are discussed.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner