[A-TT30] Machine Learning Techniques in Weather, Climate, Ocean, Hydrology and Disease Predictions

[ATT30-04] Utilizing a Causal Discovery to Identify Robust Tropical Cyclone Predictors across the North Indian Ocean

*Akshay Kumar Sagar¹、Arun Chakraborty²、Pankaj Kumar³、Swadhin Behera⁴ (1.Indian Institute of Technology Kharagpur, INDIA、2.Centre for Ocean, River, Atmosphere and Land Sciences (CORAL), Indian Institute of Technology Kharagpur, Kharagpur, West Bengal, India、3.Institute of Meteorology and Climate Research - Department Troposphere Research, Karlsruhe Institute of Technology (KIT), Germany、4.Application Laboratory, JAMSTEC)

キーワード：Tropical Cyclone, Causal Discovery, Predictors, North Indian Ocean

The intensity and destruction caused by tropical cyclones (TCs) varies significantly from year to year. Longer-term tropical cyclone (TC) forecasting is critical to minimizing the damage and fatalities caused by TCs. Using a Causal-network-based methodology, this study determined the main development regions and antecedents accountable for the origin and intensification of TC. All of the extreme events are, nonetheless, connected by a multitude of global linkages. Therefore, it is examined how tropical cyclone teleconnection and correlation with El Niño Southern Oscillation (ENSO), Indian Ocean Dipole (IOD), and North Atlantic Oscillation (NAO) occurred over the North Indian Ocean (NIO) basins during the satellite era using this Causal Effect Network (CEN)-based algorithm (1980-2020). Accumulated Cyclone Energy (ACE) is the most appropriate measure for cyclone energy; its link with the various components is explored. We investigated the variance in TC activity during all three periods (positive, negative, and neutral phases).
The results show that, there was an increasing tendency in ACE over the NIO region. At this time, cyclones of the highest intensity linger longer, but they are generally less common. Following 2005, ACE started to shift and kept rising quickly. The favorable changes and fluctuation of ACE are indicated by the vertical wind shear (VWS) analysis between 850 and 250 hPa & sea surface temperature (SST). Causal graphs, which influence the ACE value, display the contributing areas with direct or indirect relationships between the various regions at a good significance level. The teleconnections between events in the atmosphere and the ocean may become clearer as a result of these results, and improved tropical cyclone forecasting may reduce the damage caused by TCs and demonstrate how other phenomena influence the various parameters.