What: This workshop was held as a follow-up to another workshop INTROSPECT 2017, to review the current status of forecasting extreme events and tropical cyclones particularly against the backdrop of climate change and to explore future pathways.
When: 25–28 November 2019
Where: Pune, India
Against the backdrop of increasingly frequent extreme precipitation events and the increasing severity of tropical cyclones, it is time to evaluate current forecast systems and develop pathways for improving the accuracy and reliability of predicting these high impact weather events in a changing climate. In India recurring recent extreme precipitation events in August 2018 and 2019 in the southern Indian state of Kerala severely disrupted lives and caused immense damage of more than $5 billion (U.S. dollars, 2018) to public properties. The Indian seas have experienced anomalously strong severe cyclones, such as Tropical Cyclone Ockhi (29 November–6 December 2017), at very low latitudes. There were other extreme rain events recently, such as flooding in the southern metropolis of Chennai during November and December 2015 as well as floods in the northern mountainous state of Uttarakhand in June 2013. For all of these events the existing forecast system provided useful guidance in the short-term (up to 10 days) and extended-range (up to 2–3 weeks) time scales. Though the short-range forecast succeeded in providing the guidance of ∼3 days, it could not provide the probability of extremes associated with those events, with longer lead time. Therefore, there remains a most urgent need to provide these probabilities with longer lead times (∼5–7 days).
Keeping in view the recent improvement in the forecast of extreme precipitation events and the forecast of tropical cyclones, the joint international workshop on “Prediction Skill of Extreme Precipitation Events and Tropical Cyclones: Present Status and Future Prospects (IP4), & Climate Change” was held in November 2019 at the Indian Institute of Tropical Meteorology, Pune, India (Fig. 1). The workshop included 31 invited talks covering different aspects of spatiotemporal ranges of forecasts, namely, subseasonal to seasonal, short-range weather forecasts, and climate projections. The details of the talks are mentioned in Table 1. The workshop was attended by experts from various operational forecasting centers.1 Sixty students/postdocs/early career scientists also attended (see Fig. 2 for a group photo), and there were 22 posters presented (mentioned in Table 2). The posters were displayed on all four days of the workshops to allow students/postdocs/scientists maximum time for interactions. Workshop talks/proceedings were live streamed and remote viewers texted questions as a way to participate in the discussions.
Speaker presentations have been uploaded and are available online (www.tropmet.res.in/ip4/presentations.php). Also, all talks have been uploaded to the IITM YouTube Channel (www.youtube.com/watch?v=i8KKsKsz8Q4&list=PLgQCKqNw6z_CZUThN5-dvZC3qaCQDfJzX).
In the workshop, experts deliberated on forecast techniques such as the ensemble prediction system and its utility for extreme event forecasts. There was extensive discussion of the strength and deficiencies of cloud and convection parameterizations in the context of increasing resolution in future climate models. Experts also mentioned the strength of machine learning (ML)/artificial intelligence (AI) and emphasized their use in the gray areas of parameterization.
The following emerged as consensus recommendations from the workshop:
Significant modeling development and improvement in forecast skill for extreme rain and tropical cyclones/hurricanes/typhoons have taken place in the last few decades. However, there is a need for further improvement in higher rainfall thresholds at longer lead times. Likewise, there is a need to improve tropical cyclone intensity forecasts.
The path ahead for further improvement could include increasing model resolution in a coupled ocean–atmosphere framework along with suitable physics for high resolution and further application of AI/ML in the representation of physical processes.
Use of improved and latest assimilation techniques involving data from remote sensing/satellites, particularly of low-level winds, was mentioned as one of the key aspects in improving model initial conditions vis-à-vis the skill of the forecast.
Apart from improving the parameterization of important physical processes such as clouds and convection, special effort should be made to improve the cloud-radiative feedback in the model; for climate change it needs to account for cloud–aerosol–climate feedback. The understanding of warm and cold rain processes and their representation in climate models is one of the key gap areas where further research is needed.
Against the backdrop of climate change, the increase of intense tropical cyclones (category 5), extreme rain events, rapid warming of the Indian Ocean, and sea level rise were highlighted and their consequences to society were discussed. It was recommended that models should be equipped for the representation of related processes so as to make the projections of these parameters more reliable.
Centers represented at the workshop: European Centre for Medium-Range Weather Forecasts (ECMWF), National Centers for Environmental Prediction (NCEP), the Met Office (UKMO), Korean Institute of Atmospheric Prediction System (KIAPS), National Centre for Medium Range Weather Forecasting (NCMRWF), and India Meteorological Department (IMD) and also from research institutes/universities such as the Massachusetts Institute of Technology, Colorado State University, California Institute of Technology, Jet Propulsion Laboratory, University of Oxford, University of Victoria, National Center for Atmospheric Research, Indian Institute of Science (Bengaluru, India), The University of Tokyo, Sant’Anna University, Goddard Space Flight Center, University of Kyoto, IIT, Bhubaneswar, Space Application Centre, Indian Space Research Organization, Indian National Centre for Ocean Information Services (INCOIS), National Centre for Polar and Ocean Research (NCPOR), National Centre for Coastal Research (NCCR), and Indian Institute of Tropical Meteorology (IITM).