skip to main content


Title: Prediction skill of tropical synoptic scale transients from ECMWF and NCEP ensemble prediction systems

The prediction skill of tropical synoptic scale transients (SSTR) such as monsoon low and depression during the boreal summer of 2007–2009 are assessed using high resolution ECMWF and NCEP TIGGE forecasts data. By analyzing 246 forecasts for lead times up to 10 days, it is found that the models have good skills in forecasting the planetary scale means but the skills of SSTR remain poor, with the latter showing no skill beyond 2 days for the global tropics and Indian region. Consistent forecast skills among precipitation, velocity potential, and vorticity provide evidence that convection is the primary process responsible for precipitation. The poor skills of SSTR can be attributed to the larger random error in the models as they fail to predict the locations and timings of SSTR. Strong correlation between the random error and synoptic precipitation suggests that the former starts to develop from regions of convection. As the NCEP model has larger biases of synoptic scale precipitation, it has a tendency to generate more random error that ultimately reduces the prediction skill of synoptic systems in that model. Finally, the larger biases in NCEP may be attributed to the model moist physics and/or coarser horizontal resolution compared tomore » ECMWF.« less
 [1] ;  [2] ;  [3] ;  [4]
  1. Pennsylvania State Univ., University Park, PA (United States); Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
  2. Indian Institute of Tropical Meteorology (IITM), Pune (India)
  3. Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
  4. Pacific Northwest National Lab. (PNNL), Richland, WA (United States); Indian Institute of Technology Bhubaneshwar, Bhubaneshwar (India)
Publication Date:
Report Number(s):
Journal ID: ISSN 2353-6438; KP1703010
Grant/Contract Number:
Accepted Manuscript
Journal Name:
Mathematics of Climate and Weather Forecasting
Additional Journal Information:
Journal Volume: 2; Journal Issue: 1; Journal ID: ISSN 2353-6438
de Gruyter
Research Org:
Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
Sponsoring Org:
Country of Publication:
United States
54 ENVIRONMENTAL SCIENCES; prediction skill; synoptic scale transients; random Error
OSTI Identifier: