Abstract
The role of earlier forecast errors on subsequent convection forecasts is evaluated for a northern Great Plains severe convective event on 11-12 June 2013 during the Mesoscale Predictability Experiment (MPEX) by applying the ensemble-based sensitivity technique to Weather Research and Forecasting (WRF) Model ensemble forecasts with explicit convection. This case was characterized by two distinct modes of convection located 150 km apart in western Nebraska and South Dakota, which formed on either side of an axis of high, lower-tropospheric equivalent potential temperature. Convection forecasts over both regions are found to be sensitive to the position of this θe axis. The convection in Nebraska is sensitive to the position of the western edge of the θe axis near an upstream dryline, which modulates the preconvective θe prior to the diurnal maximum. In contrast, the convection in South Dakota is sensitive to the position of the eastern edge of the θe axis near a cold front, which also modulates the preconvective θe in that location. The position of the θe axis is modulated by the positions of both upstream and downstream mid- to upper-tropospheric potential vorticity anomalies, and can be traced backward in time to the initial conditions. Dropsondes sampling the region prior to convective initiation indicate that ensemble members with better representations of upstream conditions in sensitive regions are associated with better convective forecasts over Nebraska.
| Original language | English |
|---|---|
| Pages (from-to) | 2343-2361 |
| Number of pages | 19 |
| Journal | Monthly Weather Review |
| Volume | 145 |
| Issue number | 6 |
| DOIs | |
| State | Published - Jun 1 2017 |
Keywords
- Convective storms
- Ensembles
- Model errors
- Numerical weather prediction/forecasting
- Statistical techniques
- Synoptic-scale processes