Monthly weather forecasts made by the NWP global model GLOBO of CNR-ISAC, Italy

Monthly ensemble forecasts are produced using the global circulation model GLOBO with horizontal grid spacing of 1.0 deg longitude x 0.75 deg latitude and 50 vertical (hybrid) levels. 32 forecast runs are initialized with GFS-NCEP analyses: 2 unperturbed analyses (00 and 12 UTC GFS control runs) and 30 perturbed analysis members (13 at 00 and 17 at 12 UTC). Sea-surface temperature (SST) is computed by overlaying and partially relaxing the SST anomaly for the initialization day to the SST climatological trend. The SST "climate" has been recomputed from a 20-year dataset. During the forecast integration, the observed sea ice included in GFS analysis evolves according to the climatological trend. Monthly ensemble reforecasts are also produced to determine the climate variability of the forecast month. 42 reforecast members are initialized using ECMWF ERA-Interim reanalyses at 00 and 12 UTC of the selected initialization day for each of the 21 years currently available from the ERA-Interim dataset.

Mastrangelo, D., Malguzzi, P., Rendina, C., Drofa, O., and Buzzi, A.: First outcomes from the CNR-ISAC monthly forecasting system, Adv. Sci. Res., 8, 77-82, 2012. [PDF]