Environmental Conditions and Forecast Error Evolution Associated with Synoptic Progressiveness Errors in GFS Forecasts

Details

• Journal Title:
  Weather and Forecasting
• Personal Author:
  Lupo, Kevin M. ; Schwartz, Craig S.
• NOAA Program & Office:
  OAR (Oceanic and Atmospheric Research) ; WPO (Weather Program Office)
• Description:
  Operational forecasters often note that medium-range forecasts of midtropospheric cutoff lows using the Global Forecast System (GFS) are typically characterized by excessive synoptic progressiveness, i.e., a tendency to advance features too quickly downstream. Recent research examining the displacement error characteristics of cutoff lows in 0–10-day GFS forecasts objectively corroborates anecdotal impressions of a progressive bias. To better understand the processes associated with progressiveness errors in the GFS, this research examines composite 500-hPa environments and forecast error evolutions that favor erroneously progressive cutoff lows. Time-lagged, feature-relative composite analyses of erroneously progressive cutoff lows indicate that these displacement errors are preceded by the onset of negative 500-hPa geopotential height errors in the environment surrounding the composite feature, most notably through an underamplification of an upstream ridge and heights that are too low poleward of the forecast feature. Concurrently, midtropospheric winds and cyclonic vorticity advection are too strong on the downstream and equatorward side of the forecast feature, consistent with greater forcing for height falls and faster downstream progression than verifying analyses. Furthermore, a systematic negative 500-hPa height bias during GFSv15 coincides with a worsening of this progressive bias over North America, while a systematic positive 500-hPa height bias during GFSv16 coincides with a reduction of synoptic progressiveness over North America. While GFSv16 partially remedies the progressive bias over North America, it is unclear if the underlying processes reducing these progressiveness errors have been resolved or if they are masked by the positive height bias in this model version.
• Keywords:
  Numerical weather forecasting Cutoff Lows Model Errors Model Evaluation Numerical Weather Prediction Operational Forecasting
• Source:
  Wea. Forecasting, 40, 1103–1124
• DOI:
  https://doi.org/10.1175/WAF-D-24-0137.1
• Format:
  PDF
• Document Type:
  Journal Article
• Funding:
  Grant no. NA21OAR4590184
• Rights Information:
  Other
• Compliance:
  Submitted
• Main Document Checksum:
  urn:sha-512:04977742e8380d43f59bd28f87f3f8020df9ea367bd321e3894af0fed7daab7cf1d91ed4c792c9041c5cce62aabcd273fa2d794855dff5ac1fb25e43f3bbdf75
• File Type: