The purpose of the Long-Range Forecast Transient Intercomparison Project (LRFTIP) is to develop an archive of hindcast climatologies and associated diagnostics that can inform investigations into the transient behavior of initialized subseasonal to decadal climate predictions, the development of model biases, and the relative merits of different initialization methods.

In support of the project, a “core” archive based on publically available hindcast datasets including the Subseasonal to Seasonal Prediction Project (S2S), the Climate-system Historical Forecast Project (CHFP) and the Coupled Model Intercomparison Project Phase 5 (CMIP5) is being developed. Additional contributions are sought from forecast systems not represented in these projects.


  • All data and diagnostics will be publicly accessible at
  • This location also contains a guide describing requested and archived data variables, frequencies, etc.
  • At present, decadal hindcast climatologies from 12 CMIP5 models are represented
  • Seasonal hindcast climatologies are from 19 models contributing to CHFP, including 6 models from the


  • Subseasonal hindcast climatologies based on daily fields are being developed from the S2S database

Expected outcome:

  • better understanding of the mechanisms and processes responsible for transient behavior, drifts and biases in coupled models
  • development of a standard array of diagnostics describing model drifts and biases


Mulholland, D. P., P. Laloyaux, K. Haines and M.A. Balmaseda (2015) Origin and impact of initialization shocks in coupled atmosphere-ocean forecasts, Mon. Wea. Rev., 143, 4631–4644, doi: 10.1175/MWR-D-15-0076.1

Sanchez-Gomez, E., C. Cassou, Y. Ruprich-Robert, E. Fernandez and L. Terray (2016), Drift dynamics in a coupled model initialized for decadal forecasts, Clim Dyn, 46, 1819, doi:10.1007/s00382-015-2678-y