BEGIN:VCALENDAR VERSION:2.0 PRODID:-//wp-events-plugin.com//7.2.3.1//EN TZID:Asia/Kolkata X-WR-TIMEZONE:Asia/Kolkata BEGIN:VEVENT UID:114@cds.iisc.ac.in DTSTART;TZID=Asia/Kolkata:20250326T113000 DTEND;TZID=Asia/Kolkata:20250326T123000 DTSTAMP:20250313T035321Z URL:https://cds.iisc.ac.in/events/ph-d-thesis-colloquium-102-cds-26-march- 2025-quantifying-the-past-and-future-variability-in-the-bay-of-bengal-usin g-statistical-and-deep-learning-methods/ SUMMARY:Ph.D: Thesis Colloquium: 102 : CDS: 26\, March 2025 "Quantifying th e past and future variability in the Bay of Bengal using statistical and d eep learning methods" DESCRIPTION:DEPARTMENT OF COMPUTATIONAL AND DATA SCIENCES\nPh.D. Thesis Col loquium\n\n\n\nSpeaker : Mr. Abhishek.P\nS.R. Number : 05-03-00-10-12-19-1 -17560\nTitle : "Quantifying the past and future variability in the Bay of Bengal using statistical and deep learning methods"\nResearch Supervisor : Dr. Deepak Subramani\nDate &\; Time : March 26\, 2025 (Wednesday)\, 1 1:30 AM\nVenue : # 102 CDS Seminar Hall\n\n\n\nABSTRACT\n\nThe Bay of Beng al\, the world's largest bay\, along with the Andaman Sea\, a peripheral s ea situated in the southeastern part of the bay\, are crucial to the econo mic and maritime security of India. Understanding the dynamics and uncerta inties of the Bay of Bengal is particularly important with the increased i nvestment in the Deep Ocean Mission and Blue Economy by the Government of India. This thesis examines the variability in the features of the Bay of Bengal in the past and future using reanalysis and climate projections. In two parts\, this thesis makes the following specific contributions. In th e first part\, a new statistical analysis is completed on the reanalysis d ata to uncover the relationship of salinity dynamics in the Andaman Sea wi th the Southwest Monsoon Current\, and a data-assimilative forecast system is developed for synoptic forecasts of the Andaman Sea. In the second par t\, a novel deep learning model is developed to correct the future project ions of the climate models in the Bay of Bengal and uncover new dynamical insights.\n\nIn the first part of the thesis\, we begin by examining the s patiotemporal variability of salinity in the Andaman Sea to understand its dynamics. This research assesses seasonal and interannual salinity change s during the Boreal Summer (JJAS) in the area\, employing ocean reanalysis data from 1993 to 2018. River discharges and rainfall patterns were analy zed to relate them to the annual cycle of surface salinity. The subsurface dynamical patterns are identified\, and links are established between the interannual variability of salinity in the Andaman Sea and the intensity of the Southwest Monsoon Current (SMC). The spatial pattern of the subsurf ace salinity highlights a notable saltwater influx through the ten-degree channel\, which\, when juxtaposed with the primary model of variability in the Bay of Bengal\, shows a strong connection to the SMC. We conducted pa rticle trajectory experiments to further quantify the impact of the SMC's strength with the salinity variability of the Andaman Sea. This analysis m arks the first comprehensive attempt to decode the salinity dynamics of th e Andaman Sea using both Empirical Orthogonal Function (EOF) analysis and particle trajectories. In addition\, we propose a connection between the r eported decrease in shark catch and seasonal salinity change in the Andama n Sea.\n\nTo transition from analysis to forecasting\, we developed a data assimilation system that integrates satellite observations with numerical models\, producing data-driven synoptic forecasts for the Andaman Sea. Th is research marks the first application of a time-evolving\, high-resoluti on data assimilation ocean model in the Andaman Sea\, offering an accurate assessment of the ocean's state in this area. First\, we performed a high -resolution data-driven numerical Regional Ocean Modeling System (ROMS) co ntrol run. Second\, a data assimilative ROMS simulation is performed with MODIS SST observations and compared to the control run. The initial and bo undary conditions are taken from the Nucleus for European Modeling of the Ocean (NEMO) reanalysis model for the control and data assimilation runs. The incremental strong-constraint 4-dimensional variational data assimilat ion scheme (IS4D-Var) was used. The data assimilation run significantly re duces the difference between the Andaman Sea simulations and the in situ o bservations from a moored omni-bouy (BD12) over the control run. Significa ntly\, the data assimilation run reduces the surface temperature error by 0.5${^\\circ}$C compared to the control run. This data assimilation setup is crucial for improving our understanding of the synoptic dynamics of the Andaman Sea.\n\nIn the second part of the thesis\, we study the variabili ty of the features of the Bay of Bengal in the future. Climate change impa cts the ocean state\, including temperature\, salinity\, and sea level\, a ffecting monsoons and ocean productivity. Future projections based on shar ed socioeconomic pathways (SSP) from the Coupled Model Intercomparison Pro ject (CMIP) are widely used to understand the effects of climate change. H owever\, CMIP models exhibit considerable errors compared to observations in the Bay of Bengal for the time period when both projections and observa tions are available. In this region\, there is a 1.5$^{\\circ}$C root mean square error (RMSE) in the sea surface temperature of CMIP6 compared to t he Ocean Reanalysis System (ORAS5). We introduce data-driven deep learning models to correct for this error in three important variables: sea surfac e temperature\, sea surface salinity\, and dynamic sea level. The deep neu ral model for each variable is trained using pairs of monthly CMIP6 projec tions and the corresponding month's ORAS5 as input and output. This model is trained and validated with historical data (1950-2014) and future proje ction data (2015-2020) and tested with future projections from 2021 to 202 4. Ablation studies are conducted to identify the best neural architecture . The final developed model has a UNet architecture and uses a climatology -removed CMIP6 projection as input and predicts the climatology-removed co rrected fields. The trained model is then used to correct the future proje ctions from 2025 to 2100. Our new deep learning-based CMIP6 correction app roach has 15% lower RMSE compared to the traditional statistical correctio n method called the Equidistant Cumulative Distribution Function (EDCDF). Further goodness of correction metrics such as pattern correlation coeffic ient and image-based similarity metrics all indicate the superiority of ou r correction model.\n\nTo uncover the dynamical implication of the correct ed projections\, a detailed analysis of the monthly\, seasonal\, and decad al mean and variability of the projections is performed. Compared to the r aw projections\, our corrected projections indicate increased warming in t he bay\, altered patterns of salinity\, and dynamic sea level that affects monsoons and vital oceanographic features of the BoB. The new findings fr om the corrected projections are as follows. In winter\, the north-south t emperature gradient weakens more than the raw projections\, potentially in fluencing the northeast monsoon dynamics. The intensified warming in the c orrected projections during the pre-monsoon season in the central bay has implications for cyclogenesis and may potentially delay or weaken the mons oon onset. The corrected SST shows a stronger coastal warm front and a cha nged thermal structure in central and southern bays during the monsoon\, p otentially impacting monsoon rainfall and SMC propagation. The corrected s alinity shows a stronger variability in the propagation of SMC in the Bay of Bengal\, potentially impacting the influx of high salinity to the regio n and subsequently affecting productivity in the region. The increased pos t-monsoon bay warming in the corrected projections may aid cyclogenesis an d affect the EICC and its eddies. In general\, the impact of climate chang e on the mean and variability of the characteristics in the Bay of Bengal\ , revealed by our new corrected model\, would be substantially different f rom the projections of the uncorrected climate models.\n\n\n\nALL ARE WELC OME CATEGORIES:Events,Ph.D. Thesis Colloquium END:VEVENT BEGIN:VTIMEZONE TZID:Asia/Kolkata X-LIC-LOCATION:Asia/Kolkata BEGIN:STANDARD DTSTART:20240326T113000 TZOFFSETFROM:+0530 TZOFFSETTO:+0530 TZNAME:IST END:STANDARD END:VTIMEZONE END:VCALENDAR