Background

During 2017 November/December and during monsoon season of 2018, the southern state of India: Kerala experienced unusually extreme weather systems causing unprecedented damage of public property and loss of life. The event that struck during 29 November – 6 December 2017, was the tropical cyclone (TC) “Ockhi”. TC “Ockhi” was a very severe cyclonic storm that devastated the state of Kerala. Although climatologically, tropical cyclone over lower latitude near Kerala coast are relatively rare, this event posed a challenge to the numerical forecaster, as it required enhancing the model fidelity to predict such rare extreme event with sufficient lead time. The other event that shook the whole globe was the exceptionally heavy rainfall over Kerala during the month of August 2018. The state received around 164% more rainfall than its climatology during 1-19 August. Such unprecedented and exceptionally heavy rain caused wide spread flooding over the whole state of Kerala and caused loss of precious life and immense damage to public property. Apart from these events, there were other extreme events over Mega cities Mumbai, Chennai and many other places over India such as Uttarakhand heavy rain and flood etc., which posed the challenge of accurate forecasting of extremes with longer lead time.

While the numerical prediction did provide guidance of heavy rainfall during the period, there appears to be room for improving the forecasts. Keeping these high impact and exceptionally unusual weather events in background, the present workshop emphasizes and focuses on following eleven topics:

• Present skills of Global operational model in predicting extreme rainfall events
• Challenges in enhancing the skill of forecasting extreme events.
• Role of Physical parameterization, dynamical core, assimilation, resolution, land use land cover in numerical model in improving the skill of extreme rain events.
• What are the future roadmaps in improving the lead time of forecast of extreme precipitation events?
• Probabilistic prediction and stochastic physics on improving the forecast skill.
• Current status of forecast skill of Tropical Cyclone genesis, track and intensity.
• Whether propagating MJO prediction with longer lead time can help in predicting TC genesis.
• Roadmaps and future prospect of improving next generation TC forecast model
• Latest observational platform and their role in improved monitoring and better data assimilation for extreme precipitation events and tropical cyclone forecast.
• Lessons learnt in understanding extreme precipitation events and tropical cyclones in the backdrop of climate change
• New approach of Deep learning/machine learning for breaking the parameterization deadlock

Experts will deliberate on these and related topics during 25-28 November 2019 and come out with recommendation at the end of the workshop for possible improvement and future development of model for enhancing the skill of forecast. Students/postdocs will be given opportunities to display their posters to the National and International experts.
This workshop is a follow up of an earlier International workshop (INTROSPECT) held during Feb 2017 on improving the physical process parameterization of the model.