Presentation on theme: "INITIAL CONDITION: 31 May 2014 Extended range prediction during 2014 season Extended range prediction during 2014 season Using CFS based Grand Ensemble."— Presentation transcript:
INITIAL CONDITION: 31 May 2014 Extended range prediction during 2014 season Extended range prediction during 2014 season Using CFS based Grand Ensemble Prediction System (CGEPS)
Ocean and Atmospheric Initial conditions (IC) are obtained from NCEP. Technique was developed to perturb ICs to generate 11 members. (Abhilash et al., 2013) Model has been integrated at low resolution (T126, ~100km, termed as CFS126 hereafter; Abhilash et al. 2014) and high resolution (T382, ~38km, termed as CFS382 hereafter; Sahai et al. 2014) for 45 days for each 11 ICs at five day intervals starting from 16th May (Eg: 16 May, 21 May, 26 May, 31 May, 05 Jun........etc). Climatology has been calculated from 10-year hindcast. Bias correction in daily forecasted SST from CFS126 for each lead time has been done by removing the daily mean bias for corresponding lead time (model climatology-observed climatology) from forecasted daily SST (Sahai et al., 2013; Abhilash et al., 2013). This model is hereafter termed as GFSbc. Strategy: Ensemble Prediction using CFSv2
Generation of CGEPS In the formulation of MME from CGEPS, 21 ensembles of GFSbc, 11 ensembles of CFS126 and 11 ensembles of CFS382 have been used (Abhilash et al. 2014, submitted ). Hence, total 43 ensemble members were produced independently from 3 variants of CFS model to generate the CGEPS and forecast consensus is done by making simple average among the members.
Extended EOF analysis is carried out similar to Wheeler and Hendon 2004 using standardized rainfall anomalies up to lag -15 days, averaged between 60-95E for the latitudes -12 to 30. The rainfall anomalies for the lag days are appended side by side to create the extended data matrix. The EOF analysis is carried out using IMD-TRMM merged data from 1998-2011. The real time data for 2013 is projected onto the EOFs created from the 14 years of past data. The amplitude of EOF1 and EOF2 (PC1 and PC2 ) are plotted in a PC1/PC2 phase space similar to Wheeler Hendon 2004 to get an idea of the evolution of ISO and its strength. Computation of MISO (Ref: Suhas et al. 2012, Sahai et al. 2013) Phase1: Peninsular India; Phase2: Central India; Phase3: Central India; Phase4: North India Phase5: Foothills Phase6: South Indian Ocean; Phase7: Indian Ocean; Phase8: Southern tip
Real-time forecast based on 31 May 2014 initial condition
The rainfall would be confined to west coast and NE India up to 15 June. The surface pressure gradient pattern and low level circulation indicates that (a) monsoon trough will be developing around 12 June over Indo-Gangetic plains (b) presence of ridge east of western Ghats (anticyclonic circulation over central India) will hamper the establishment of monsoon over Indian land. Thus, the strengthening and progression of the monsoon seems to be slackened till 18 June and monsoon would reach central India by 20 June as a feeble current. Large scale MISO forecast also suggests that it will be over peninsular India up to 25 June and may advance to Central India afterwards. Overall, monsoon activity in June will be mainly confined to west coast, NE India and southern peninsula. Key points from the present forecast