Tropical cyclones (TCs), devastating weather events occurring over warm tropical waters, induce entrainment and upwelling in the ocean. Their formation depends on factors like atmospheric disturbances, warm sea surface temperatures (SSTs) exceeding 27°C, atmospheric instability, and low vertical wind shear. The North Indian Ocean (NIO), encompassing the Bay of Bengal (BoB) and the Arabian Sea (AS), is a region prone to TC activity, with differing characteristics between the two basins. The BoB, characterized by high precipitation, river runoff, and strong stratification, exhibits lower biological productivity than the AS, which is comparatively more productive. While riverine nutrient input in the BoB is often lost to deeper depths due to the narrow continental shelf, TCs can upwell subsurface waters, bringing nutrients to the sunlit zone and boosting phytoplankton growth. The BoB experiences higher cyclone frequency during the post-monsoon season due to the absence of upper-level jet streams and low vertical wind shear; however, the strong stratification during the post-monsoon, caused by heavy rainfall and freshwater influx, creates a barrier layer (BL) that inhibits mixing and upwelling. Previous research has examined Chl-a enhancement during individual cyclone events, but a comprehensive analysis across the entire NIO and various cyclone categories during the satellite era is needed to understand the dynamics and biological production linked to TCs, especially concerning the influence of global warming, rising SST, and altered wind patterns on cyclogenesis and biological productivity. This study aims to provide a comprehensive assessment of the link between tropical cyclones and phytoplankton blooms in the NIO, using satellite measurements from 1997 to 2019.

Several studies have individually analyzed the impact of specific cyclones on Chl-a concentrations in the BoB and AS. Nayak et al. (2001) examined the Orissa Super Cyclone (1999), while Vinayachandran and Mathew (2003) analyzed Chl-a blooms associated with northeast monsoon and selected cyclones. Patra et al. (2007) investigated contrasting Chl-a abundance during different cyclones using SeaWiFS data. Sarangi et al. (2008) highlighted cyclone-induced upwelling as a primary cause of elevated Chl-a. Rao et al. (2006) and Vidya et al. (2017) examined Chl-a enhancements following specific cyclone events using OCM and MODIS data, respectively. Chacko (2017, 2019) analyzed Chl-a changes associated with cyclones Hudhud and other NIO cyclones, respectively, while Subramanyam et al. (2002) and Chakraborty et al. (2018) reported significant Chl-a increases after cyclones ARB01 (2001), Gonu (2007), and Phyan (2009) in the AS. Pan et al. (2018) investigated the role of wind forcing and other parameters in phytoplankton blooms in the northwest Pacific and South China Sea. These studies highlight the dependence of bloom intensity on several factors, but mostly focused on individual events, necessitating a comprehensive analysis of all cyclones in both BoB and AS basins during the satellite era to gain a complete understanding of cyclone-induced primary production and its key drivers.

This study utilizes satellite-derived Ocean Colour Climate Change Initiative (OC-CCI) version 4.2 data for chlorophyll-a (Chl-a) concentrations (1997-2019), merging MERIS, MODIS, and SeaWiFS measurements. Cyclone track, category, and duration information were obtained from the Joint Typhoon Warning Centre (JTWC) best track data and the India Meteorological Department (IMD) for 2019. The Saffir-Simpson scale was used for cyclone categorization. Ekman Pumping Velocity (EPV) was calculated using European Centre for Medium Range Weather Forecast (ECMWF) Reanalyses (ERA5) 10 m wind data. Sea surface height anomaly (SSHA) data from Copernicus Marine Environment Monitoring Services (CMEMS) were used to identify oceanic eddies. Argo float measurements (WMO IDs 2902086, 2902114, and 2902120) provided subsurface Chl-a and physical parameters (ILD, MLD, BLT, D23). The study area included the entire NIO (3°-25° N, 50°-98° E), with separate analyses for BoB and AS. BoB analysis was further divided into four sub-regions based on cyclone landfall. Chl-a data were averaged for five days before, during, immediately after, and five days after the cyclone passage. A Chl-a threshold of 0.2 mg/m³ was used for bloom detection in BoB/AS, with blooms defined as Chl-a concentrations above 0.5 mg/m³. The peak Chl-a bloom value was calculated by averaging Chl-a within a 1° x 1° grid within a larger 4° x 4° region. The percentage change in Chl-a was calculated as [(peak value - pre-cyclone value) / pre-cyclone value] × 100. Translational speed (TS) was calculated using the Haversine formula. The influence of El Niño-Southern Oscillation (ENSO) and IOD was assessed using relevant indices, with composites of Chl-a and SSHA created for different ENSO and IOD phases.

Analysis of 51 storm events in the BoB (30 phytoplankton bloom events) and 33 in the AS (18 bloom events) revealed that cyclone-induced phytoplankton blooms are frequent, particularly during the post-monsoon season. Bloom intensity is inversely related to cyclone translational speed (TS); slower storms lead to more intense and longer-lasting blooms (10-14 days in the post-monsoon). The strongest bloom in the BoB reached 3.28 mg/m³ (BOB01 in 2004, pre-monsoon), while the most intense bloom in the AS reached 11 mg/m³ (Gonu in 2007). The presence of cold-core eddies near cyclone tracks significantly enhances bloom intensity. La Niña years showed larger blooms than El Niño or normal years in both basins. The impact of the Indian Ocean Dipole (IOD) varied between basins: positive IOD years showed larger blooms in the BoB, while negative IOD years had stronger blooms in the AS. The majority of blooms (80% in BoB pre-monsoon, 80% in BoB post-monsoon, and 44.4% in AS) were associated with tropical storms and category 1 cyclones, suggesting that slower-moving storms generate more sustained upwelling, leading to greater blooms. The study demonstrates a time lag of 4-12 days between cyclone passage and peak bloom.

The findings demonstrate a clear link between tropical cyclones and phytoplankton blooms in the NIO, with slower-moving cyclones and the presence of cold-core eddies playing crucial roles in enhancing bloom intensity and duration. The contrasting responses to ENSO and IOD events in the BoB and AS highlight the complex interplay of climatic factors affecting primary productivity. The strong correlation between slower translational speeds and increased bloom intensity is particularly relevant in the context of potential global trends towards slower-moving tropical cyclones due to climate change. The results of this study significantly improve our understanding of the ocean's biogeochemical response to tropical cyclones and the influence of large-scale climate variability. The findings offer valuable insights for refining ocean biogeochemical models, which often struggle to accurately simulate observed Chl-a concentrations and primary production.

This study provides a comprehensive assessment of cyclone-induced phytoplankton blooms in the NIO, revealing the significant role of cyclone translational speed and the presence of cold-core eddies in bloom intensity. The contrasting responses to ENSO and IOD in the two basins underscore the complex interplay of climatic factors influencing primary productivity. This research has implications for understanding climate change impacts on the NIO ecosystem and refining ocean biogeochemical models. Future studies could focus on investigating the specific phytoplankton species involved in these blooms, examining the long-term effects of these blooms on the marine ecosystem, and incorporating these findings into predictive models to better forecast bloom events under changing climate conditions.

The study's reliance on satellite data introduces potential limitations due to cloud cover during cyclone events, which can affect data availability. The spatial resolution of the data may also limit the precise identification of smaller-scale bloom events. Additionally, the relatively short period of satellite data (1997-2019) may not capture long-term variability fully. While the study considered various factors influencing blooms, other factors might also play a role, warranting further investigation.