The monsoon mission is planned with an overall objective to improve monsoon prediction over India on all time scales. Specific objectives of the monsoon mission are:

• To build a working partnership between the academic and R&D Organisations both national and international, and the MoES to improve the operational monsoon forecast skill over the country. 
• To setup a state of the art dynamical modelling frame work for improving prediction skill of ‘Seasonal and Extended range predictions’ and ‘Short and Medium range (up to two weeks) predictions’. 

At IITM, the principle focus of the Monsoon Mission is to develop Seasonal and Extended range prediction systems with the following objectives:

• To develop a fully coupled ocean-atmosphere-land modelling system for dynamical Prediction on Extended range and seasonal time scales and to improve the prediction skill.
• Development of Data assimilation system for Climate Forecast System (CFS)
• To improve parameterisation schemes in the coupled ocean-atmosphere models.
• To study and understand the monsoon variability over different spatio-temporal scales.
• To coordinate the working partnership amongst the ESSO-MoES organisations and various national and international R&D and academic institutions.

First phase of monsoon mission completed successfully in 2017 and handed over the seasonal and extended range prediction system developed to India meteorological Department (IMD) for operational use.

Second phase of monsoon mission started by October 2017. The second phase focuses on both model development and application development for agriculture, hydrology, energy etc sectors.

The Indian summer (southwest) monsoon is referred as lifeline of India, as variability in any of its aspects (onset, withdrawal and quantum of rainfall) greatly influences the agriculture yield, economy, water resources, power generation and ecosystem. Hence, if the variations in monsoon rainfall are known well in advance, it would be possible to reduce the adverse impacts related to excess or deficient rainfall, providing us prior information about droughts and floods. The accurate prediction of monsoon rainfall is a basic need for the nation but remained a challenge over the decades. The long range prediction of the seasonal mean monsoon rainfall depends on dynamics of its year-to-year variations. Recent improvements in dynamical numerical models with ocean-atmosphere coupling has been useful for improvement of the monsoon forecast skill through a collective effort.

Ministry of Earth Sciences (MoES), Government of India launched 'National Monsoon Mission' (NMM) in 2012 with a vision to develop a state-of-the-art dynamical prediction system for monsoon rainfall on different time scales. MoES bestowed the responsibility of execution and coordination of this mission to the Indian Institute of Tropical Meteorology (IITM), Pune. For this national mission, IITM is collaborating with NCEP (USA), MoES organisations and various academic institutions/organizations under NMM. Climate Forecast System (CFS) of NCEP, USA has been identified as the basic modelling system for the above purpose, as it is one of the best among the currently available coupled models. However, it had a moderate skill for retrospective forecast (hindcast) of seasonal monsoon rainfall and this skill needed to be improved to make the forecasts more useful. Thus, there was an urgent need to develop an Indian model based on CFS coupled model with an improved hindcast skill so that it can be transferred to the India Meteorological Department for operational forecasting. With this objective, and to accomplish this task, MoES/IITM invited proposals from national and international scientists/organizations. Through model development works at IITM, the skill of the model has improved and it has been transfered to IMD.

Base Models used

The Ministry of Earth Sciences (MoES) has considered to use the following numerical models : (i) The American model called “Climate Forecast System” (CFS) developed by National Centres for Environmental Prediction (NCEP), NOAA National Weather Service, USA. CFS is a coupled ocean-atmosphere modeling system that combine data from ocean, atmosphere and land for providing long range forecasting (seasonal prediction of Indian Monsoon); [ Model developments on CFS will be implemented by IITM, with atmospheric initial conditions from NCMRWF and Ocean initial conditions from INCOIS] and (ii) The Unified Model (UM), developed by the United Kingdom Meteorological Office (UKMO), UK. This model will be utilized for short to medium range prediction [and the Model developments on UKMO will be implemented by NCMRWF, in association with IMD.]

Need for NMM

El Nino and Southern Oscillation (ENSO) being a dominant mode of global inter-annual variability and due to its vast influence on other regional climates, in last few decades researchers have made large number of studies on the ENSO phenomena and its various impacts using atmospheric and ocean-atmosphere coupled general circulation models. In recent decades, dynamical numerical models have considerably improved and most of the global coupled models have shown good prediction skill of ENSO SST with six months lead time. The seasonal mean rainfall hindcast skill, at one season lead time, over the central Pacific is also very good. This has been possible due to a concerted effort by a group of devoted scientists. However in earlier decades, not much breakthrough has taken place in improving the prediction skill of Indian summer monsoon rainfall, even though it was expected as a prominent heat source over Indian region during summer monsoon period that drives the major atmospheric circulations. Recently, a lot of research on model development works are gradually improving the skills of the models.

Monsoon Mission Phase 2

After the suscessful completion of phase 1, phase 2 of the project is started giving emphasis to application development and model development both together. Under phase 2, project proposals are sanctioned both nationaly and internationally .

For facilitating the above objectives, the Monsoon Mission at IITM is divided into three sub-projects for better execution.

Sub Projects

• Seasonal Prediction
• Extended Range Prediction
• Parameterisation of Physical Processes and Analysis

Indian summer monsoon is the lifeline of the Indian subcontinent, especially India. Any variability in its onset, withdrawal or quantum of rainfall affects the country dearly. Knowledge of variability in advance would reduce the adverse impacts related to excess or deficient rainfall. Unfortunately, the prediction of the Indian Summer Monsoon rainfall has remained a challenge over the decades. Empirical models have not shown any improvement in prediction skill over the years. Dynamical models show some hope in improving skill in monsoon prediction. Many centres in the world are routinely using dynamical models to predict seasonal mean climate. Here, the NCEP CFS model is used as one of such reliable models.

However, there are some intrinsic problems in predicting the Indian monsoon rainfall through such dynamical models. Hence, to improve the assimilation and forecasting system especially for the monsoon region, the Ministry of Earth Sciences (MoES), Government of India has formulated a focused mission mode programme ‘Monsoon Mission’ with a vision to develop a state-of-the-art dynamical prediction system for monsoon rainfall on all the different time scales with concerted efforts by various research and academic institutes in India and abroad. The mission is being executed under the leadership of IITM for designing a prediction model to improve the Indian Monsoon weather and climate forecasts.

To make the mission a success, the MoES has signed a Memorandum of Understanding (MoU) with the National Oceanic and Atmospheric Administration (NOAA), USA under the agreement ‘Technical Cooperation for the Study of Dynamical Seasonal Prediction of Indian Monsoon Rainfall’. The Mission is working on developing an Indian model based on NCEP CFSv2 model by improvising on its strengths and weaknesses and by incorporating new physics/parameterization schemes for improving its simulations/prediction skill of the monsoon rainfall. Further, the Unified Model (UM) developed by the United Kingdom Meteorological Office (UKMO) will also be utilized for short to medium range prediction.

Seasonal prediction component of the mission is being coordinated by IITM, and short to medium range prediction by NCMRWF. While INCOIS is providing ocean observations, and IMD is implementing the research results in operational model and evaluating the verification of forecasts. Academic institutions from national and international organizations are participating through extramural funding to improve the modeling framework adopted by India.

To oversee the overall functioning of the mission, a Mission Directorate is set up at IITM. The overall execution of the mission is vested in the Mission Director, who is advised by a high level national Scientific Steering Committee. Further, a standing Scientific Review and Monitoring Committee reviews various research proposals from national and international partners and recommends the proposals that have direct relevance to the mission objectives.

The main objectives of the mission are to build a working partnership between the academic R&D organizations and the operational agency to improve the monsoon forecast skill, and to set up a state-of-the-art dynamical modeling frame work for improving prediction skill of (i) seasonal and extended range prediction system and (ii) short and medium range prediction system.

IITM has developed systems for seasonal and extended range prediction of Indian Summer Monsoon and these have been shared with IMD.

First phase of monsoon mission completed successfully in 2017 and handed over the seasonal and extended range prediction system developed to India meteorological Department (IMD) for operational use.

Second phase of monsoon mission started by October 2017. The second phase focuses on both model development and application development for agriculture, hydrology, energy etc sectors.

Why monsoon mission CFSv2 has better skill for ISMR compared to other present generation seasonal prediction models? [Reference: Pillai PA, Rao SA, Ramu DA, Pradhan M, George G (2018) . Seasonal prediction skill of Indian summer monsoon rainfall in NMME models and Monsoon Mission CFSv2. International journal of climatology (online)].

The present study compares the Indian summer monsoon rainfall (ISMR) prediction skill of monsoon mission climate forecast system version 2 (CFSv2‐T382) with that of the seasonal prediction models participating in US National Multi‐Model Ensemble (NMME) project. It is noted that even though the prediction skill for SST boundary forcings like El Niño‐Southern Oscillation (ENSO) and Indian Ocean Dipole (IOD) is not at the best in CFSv2‐T382 compared to a few of the NMME models, it shows better skill for ISMR hindcasts initialized at 3‐month lead time (February IC). This may be attributed to the better teleconnection pattern of ENSO and IOD in CFSv2‐T382, which has minimum biases in equatorial Indo‐Pacific region. It also has a better ISMR–SST teleconnections in the Tropics with a pattern correlation of around 0.6. In many of the NMME models, the better prediction skill of the inter‐annual variability of SST indices is not transformed into the improvement of ISMR skill through teleconnections (Figure 5). It is therefore concluded that having good prediction skill for major SST boundary forcings is not sufficient, but capturing the appropriate teleconnections of these SST boundary forcings in the model is critical for the better prediction of ISMR. The study points out that the present‐day seasonal prediction systems need to be improved in their simulation of tropical SST–monsoon teleconnections, which can improve the seasonal prediction skill of Indian summer monsoon further. One area where the immediate focus is required is the Indian Ocean SST and ISMR teleconnection.

Figure 5: Spatial pattern of correlation between ISMR and seasonal SST anomalies from (a) observations, (b)–(p) individual model hindcasts and (o) MME of all models. Statistically significant (90% confidence level) correlations are stippled. The negative values are represented as dotted contours also with zero value as thick continuous line.

Prediction of Indian Summer-Monsoon Onset Variability: A Season in Advance

Indian Summer Monsoon is found to be influenced by atmospheric and oceanic conditions over various parts of the globe (for example, above or below normal rainfall years related to ENSO conditions over Pacific). Monsoon onset is a transient phenomenon and prior studies have shown that onset is teleconnected to large scale sea surface temperature (SST) changes over Pacific and Indian Ocean (also shown in the figure). Present day general circulation models (GCM) have shown good skill in predicting/capturing these large scale features. Therefore, by observing the state of large scale parameters, one can predict the status of monsoon onset. Despite of the tremendous skill achieved by the state-of-the-art models in predicting such large scale processes, the prediction of monsoon onset variability by the models is still limited to just 2–3 weeks in advance. Our study is based on the evidences that monsoon onset is governed by large scale processes to assess its predictability by general circulation models at a season in advance. Therefore primary motive of our study was to show the possibility of using a global seasonal forecasting model (Monsoon Mission Model; MMM) to predict early or delayed onset from the large scale atmosphere-ocean conditions from February itself. Using an objective definition (thermodynamic criteria) of onset in MMM simulations, it is shown that the skill-full prediction of onset variability is feasible under seasonal prediction framework. The better representations/simulations of not only the large scale processes but also the synoptic and intra-seasonal features during the evolution of monsoon onset are the comprehensions behind skill-full simulation (with an anomaly correlation of 0.47 between observed and simulated onset dates) of monsoon onset variability. The changes observed in convection, tropospheric circulation and moisture availability prior to and after the onset are also simulated in model simulations, which results in high hit rate (60-70 %) of early/delay in monsoon onset in the high resolution model.

[Pradhan M., Rao S.A., Srivastava A., Dakate A., Salunke K., Shameera K.S., Prediction of Indian Summer-Monsoon Onset Variability: A Season in Advance, Scientific Reports, 7:14229, October 2017, DOI:10.1038/s41598-017-12594-y]

• Dutta S., Narkhedkar S.G., Mukhopadhyay P., Yadav M., Sunitha Devi, Use of large‑scale atmospheric energetics for understanding the dynamics of contrasting Indian summer monsoon rainfall in different years, Meteorology and Atmospheric Physics, online, June 2018, DOI:10.1007/s00703-018-0610-9, 1-12 (Impact Factor 1.159)
• Arora A., Siddharth Kumar, What makes protracted El Niño to last longer than canonical El Niño?, Theoretical and Applied Climatology, online, May 2018, DOI:10.1007/s00704-018-2503-8, 1-17 (Impact Factor 2.640)
• Chaudhari H.S., Hazra A., Pokhrel S., Chakrabarty Chandrima, Saha Subodh K., Sreenivas P., SST and OLR relationship during Indian summer monsoon: a coupled climate modelling perspective, Meteorology and Atmospheric Physics, 130, April 2018, DOI:10.1007/s00703-017-0514-0, 211-225 (Impact Factor 1.159)
• De S., Agarwal N.K., Hazra A., Chaudhari H.S., Sahai A.K., On unravelling mechanism of interplay between cloud and large scale circulation: a grey area in climate science, Climate Dynamics, online, April 2018, DOI:10.1007/s00382-018-4211-6, 1-22 (Impact Factor 4.146)
• Dandi Ramu A., Chowdary J.S., Ramakrishna S.S.V.S., Kumar O.S.R.U.B., Diversity in the representation of large-scale circulation associated with ENSO-Indian summer monsoon teleconnections in CMIP5 models, Theoretical and Applied Climatology, 132, April 2018, DOI:10.1007/s00704-017-2092-y, 465-478 (Impact Factor 2.640)
• Seelanki V., Pentakota S., Prasad K.V.S.R., Impact of aquarius sea-surface salinity assimilation in improving the ocean analysis over Indian Ocean, Marine Geodesy, 41, April 2018, DOI:10.1080/01490419.2017.1422817, 144-158 (Impact Factor 1.000)
• Shrivastava S., Kar S. C., Sahai A. K., Sharma A.R., Identification of drought occurrences using Ensemble predictions up to 20-Days in advance , Water Resources Management, 32, April 2018, DOI: 10.1007/s11269-018-1921-9, 2113-2130 (Impact Factor 2.848)
• Chinthalu G.R., Dharmaraj T., Patil M.N., Dhakate A.R., Pawar S.D., Siingh D., Severe cyclonic storm JAL, air-sea interaction perspectives and floods along Andhra Pradesh-Tamilnadu coast, Journal of Indian Geophysical Union, 22, May 2018, 341-348 (Impact Factor 0.000)
• Dwivedi S., Uma R., Lakshmi Kumar T.V., Narayanan M.S., Pokhrel S., Kripalani R.H., New spatial and temporal indices of Indian summer monsoon rainfall, Theoretical and Applied Climatology, online, February 2018, DOI:10.1007/s00704-018-2428-2, 1-12 (Impact Factor 2.640)
• Pokhrel S., Hazra A., Saha Subodh K., Chaudhari H.S., Metya A., Ghude S.D., Konwar M., Contrast in monsoon precipitation over oceanic region of north Bay of Bengal and east equatorial Indian Ocean, International Journal of Climatology, online, February 2018, DOI:10.1002/joc.5433, 1-15 (Impact Factor 3.760)
• Abhilash S., Mandal R., Dey A., Phani R., Joseph S., Chattopadhyay R., De S., Agarwal N.K., Sahai A.K., Devi S.S., Rajeevan M., Role of enhanced synoptic activity and its interaction with intraseasonal oscillations on the lower extended range prediction skill during 2015 monsoon season, Climate Dynamics, online, January 2018, DOI:10.1007/s00382-018-4089-3, 1-12 (Impact Factor 4.146)
• Arora A., Rao Suryachandra A., Pillai P., Dhakate A., Salunke K., Srivastava A., Assessment of prediction skill in equatorial Pacific Ocean in high resolution model of CFS, Climate Dynamics, online, January 2018, DOI:10.1007/s00382-018-4084-8, 1-15 (Impact Factor 4.146)
• Chattopadhyay N., Rao K.V., Sahai A.K., Balasubramanian R., Pai D.S., Pattanaik D.R., Chandras S.V., Khedikar S., Usability of extended range and seasonal weather forecast in Indian agriculture, Mausam, 69, January 2018, 29-44 (Impact Factor 0.467)
• Rai A., Saha Subodh K., Evaluation of energy fluxes in the NCEP climate forecast system version 2.0 (CFSv2) , Climate Dynamics, 50, January 2018, DOI:10.1007/s00382-017-3587-z, 101-114 (Impact Factor 4.146)

• After the successful completion of Phase-I (2012-2017), the Ministry of Earth Sciences (MoES) launched the Monsoon Mission Phase II (2017-2020) in September 2017 with emphasis on predicting extremes and development of climatic applications based on monsoon forecasts, especially in the field of agriculture and hydrology, while continuing model development activities. IITM has initiated the work for climatic applications in agriculture and hydrology in collaboration with IMD, ICRISAT, etc. Project scientists have been recruited to address these developmental issues. IMD has issued LRF for 2018 SW monsoon season using MMCFS model (developed at IITM).

• Efforts are being made to link the IITM forecast products with the crop simulation models for major crops grown under variable agro-climatic zones of India to find out the model performance as well as their sensitivity and usability for growth and yield prediction for the selected crops. It requires fine tuning of the model for the major crops and varieties grown in a particular region as well as to identify the major climatic risks of the particular region. It can also be helpful to decide the most appropriate crop and variety to be grown as well as to finalize various management practices in a particular area as per the forecast outlook.

• IITM developed world’s highest resolution Global Ensemble Forecast System (GEFS) for short range prediction at 12 km using 21 members of the model. The GEFS prediction system provided probabilistic rainfall for next 10 days.  The very high (~12 km) resolution Ensemble Prediction System (EPS) with 21 ensemble members for short range forecast system based on GEFS (T1534) has been put in place by IITM and has been handed over to IMD for operational implementation since 1 June 2018.  For the benefit of IMD forecasters, various new model diagnostics and products have been developed and added based on GEFS T1534 forecast Presently the GEFS (12km) forecast is utilised to develop the block level forecast of rainfall probability for IMD’s agrimet application.

• A coupled Ocean-Atmospheric data assimilation system using Local Ensemble Transform Kalman Filter (LETKF) technique (weakly coupled) has been developed and implemented for CFSv2 at Aaditya HPC, IITM. The LETKF uses time-dependent covariance from an ensemble of a forecast.

• The extended range prediction products for research/scientific use (based on weekly initial conditions) are being uploaded at http://www.tropmet.res.in/erpas/. These forecast products are based on the real time weekly operational forecast generated by IMD using the Multi Model Extended Range Prediction System developed at IITM.  The MME forecasts are prepared using CFS (T126 & T382) and GFS (T126 & T382). Each resolution of CFS and GFS is having 4 ensemble members.

i) Monsoon 2018 has started with deficient June rainfall. Experimental extended range prediction of 2018 monsoon season predicted a revival in July (positive anomaly) when monsoon was in subdued condition during the end of June.

 ii) Application of extended range forecast to Gramin Krishi Mausam Seva (GKMS) is discussed in a recent meeting with IMD Kolkata. Under this GKMS project the block level outlook would be prepared for next 2-3 weeks based on extended range forecasts for crop management and other agromet advisories.

iii) Efforts are being made to incorporate the extended range forecast in the health sector.

• Various modifications in the CFS source code as suggested by different projects under Monsoon Mission phase-I are being tested for possible improvement in skill and associated processes. Sensitivity runs using ocean initial conditions from INCOIS and atmospheric initial conditions from NCMRWF, various simulations, hindcast runs, etc. were carried out. Some bugs were identified and appropriate corrections were made in the model. Replacing the old land albedo used in CFS, a new albedo product (CFS-ALBEDO) have been made using the Moderate Resolution Imaging Spectro-radiometer (MODIS) in CFSv2T126 runs.

• As per the operational forecast requirements of IMD, updated hindcasts at T382 resolution for the period 2003-2017 are being made available to them (INCOIS-T382). INCOIS-NCMRWF initial conditions are employed for this purpose. Hindcasts for all calendar months have been completed and handed over to IMD for operational forecast purpose.

• Efforts are being made to test the ability of weakly coupled data assimilation (WCDA) to further enhance model skilland for coupling a hydrological model to the CFSv2 in order to close the hydrological cycle in CFS.  Various other simulation works are being carried out for further improving the model output. Also, a 21 ensemble member GEFS model at T1534 resolution has been ported on Pratyush with the help of HPC team, and hindcast experiments are being planned.