Rapid intensification (RI) of tropical cyclones (TCs), a sudden increase in maximum sustained winds, poses a significant threat to coastal communities due to its difficulty in prediction and the resulting inadequate preparedness. Accurate forecasting is hampered by intensity forecast errors being 2–3 times larger for RI events compared to non-RI events. This unpredictability, combined with the upward trend in the frequency of RI events, increases the risk of catastrophic socioeconomic damage. A key factor contributing to RI is the presence of prolonged high sea surface temperatures (SSTs), known as marine heatwaves (MHWs). While the link between elevated SSTs and TC intensification is established, the extent to which MHWs specifically contribute to RI remains under-explored. This research addresses this knowledge gap by conducting a probabilistic analysis to quantitatively assess the likelihood of RI occurrence during MHWs in the Gulf of Mexico (GoM) and northwestern Caribbean Sea (NWCS). Understanding this relationship is crucial for improving forecasting models and mitigating the associated risks, particularly given the projected increase in both MHWs and TCs due to climate change. Hurricane Ian (2022) serves as a stark example of the devastating consequences of RI, rapidly intensifying from a tropical storm to a Category 4 hurricane in just two days, causing over 150 deaths and $114 billion in economic damage.

Previous research has highlighted the importance of various factors in TC RI, including high SSTs, low wind shear, abundant moisture, and excess ocean heat content. Elevated SSTs are pivotal in driving interannual fluctuations in both TC frequency and intensity, with a critical threshold of SST > 26 °C for tropical cyclogenesis. Even small increases in SST (as low as 1 °C) can significantly boost the maximum total enthalpy (sensible and latent heat fluxes). The concept of MHWs, defined as at least 5 days of warm SSTs exceeding a seasonally varying threshold, has emerged as a critical indicator of prolonged ocean warming that can influence TC development. Studies have shown the importance of MHWs as a catalyst for rapidly intensifying hurricanes, with prolonged pre-storm warming periods serving as a heat source. The co-occurrence of MHWs and RIs has been observed in several TCs (e.g., Amphan in the Bay of Bengal, Hurricane Michael), where even high vertical wind shear (VWS) did not prevent intensification. While some studies have investigated the impact of MHWs on TC intensity change, further research at regional and global scales is needed to clarify the effects of MHWs on RI.

This study employs a probabilistic analysis using historical data from the International Best Track Archive for Climate Stewardship (IBTRACS) dataset, spanning from 1950 to 2022. The analysis focuses on the GoM and NWCS, regions highly susceptible to hurricane activity. The study utilizes a 24-hour minimum separation time (MST) between successive RI events to ensure statistical independence. Rapid intensification events were identified using the National Hurricane Center's definition (a 35 mph increase in maximum sustained winds over 24 hours). Sea surface temperature (SST) data from the ERA5 dataset were used for MHW characterization, employing the Hobday et al. definition (at least 5 consecutive days exceeding a seasonally varying PC80 threshold, with a maximum gap of 2 days). Two baseline periods (1950-1980 and 1981-2022) were considered to analyze long-term MHW trends. The study uses a double-threshold approach to identify MHWs influencing RI, considering both temporal proximity (within 10 days) and spatial distance (within 125 miles) to the onset of RI events. Gridded probabilities of RI occurrence (P(RI)), conditional probabilities of RI given MHW (P(RI|MHW)), and multiplication rates (P(RI|MHW)/P(RI|¬MHW)) were calculated to quantify the impact of MHWs on RI likelihood. Additionally, the study analyzed the spatial distribution of various parameters including Tropical Cyclone Heat Potential (TCHP), vertical wind shear (VWS), and latent heat flux (LHF) to understand the underlying physical mechanisms.

The analysis revealed that approximately 70% of major hurricanes making landfall in the study area experienced at least one RI event. Three hotspots with high RI probabilities were identified: near the Cayman Basin, the Bay of Campeche, and the Yucatán Channel. SSTs in 2023 significantly exceeded historical values, highlighting the increasing trend of anomalous SSTs. The analysis of MHWs showed a significant increase in frequency, intensity, and duration during the warmer period (1981-2022) compared to the cooler period (1950-1980), particularly along the Loop Current path. The conditional probability analysis (P(RI|MHW)) indicated a significant influence of MHWs on RI events, particularly in the NWCS and southern GoM. The multiplication rate showed up to a 5-fold increase in RI likelihood in the presence of MHWs, with an average increase of 1.5-fold across hotspot regions. The spatial patterns of TCHP, VWS, and LHF demonstrated that while high TCHP favors RI, low VWS and high LHF are also crucial, and their combined influence determines the final RI likelihood. Analysis of four costly hurricanes (Harvey, Michael, Ida, Ian) showed a strong association between RI locations and high TCHP values, suggesting the importance of subsurface ocean warming.

The findings demonstrate a clear amplification pattern of RI likelihood during MHWs, particularly in specific hotspots within the GoM and NWCS. The 70% of TCs experiencing MHWs within their impact area highlights the significant influence of these events. The substantial increase in RI likelihood (up to 5-fold) in the presence of MHWs emphasizes the compounding effect of these events and their significant impact on TC intensification. The spatial consistency between RI hotspots and regions with higher MHW frequency, intensity, and duration further supports the relationship. The analysis of TCHP, VWS, and LHF sheds light on the complex interplay of physical factors governing RI, showing that while high TCHP is a favorable condition, it's not sufficient without conducive VWS and LHF conditions. The study's findings have significant implications for improving TC forecasting and risk mitigation, especially in light of the expected increase in both MHWs and TCs due to climate change.

This study provides a quantitative framework for assessing the impact of MHWs on RI events in the GoM and NWCS. The findings strongly suggest that MHWs significantly increase the likelihood of RI, particularly in specific hotspots. These results emphasize the urgent need for improved forecasting models that incorporate MHWs and other relevant physical factors. Future research should focus on refining the methodology by incorporating physically based synthetic TC scenarios and advanced machine learning techniques for improved prediction capabilities. Furthermore, exploring the interplay between El Niño, MHWs, and RI is crucial for a more comprehensive understanding of TC intensification in a changing climate.

The study's reliance on historical data is a limitation, as it cannot fully capture the complexity of interactions and future changes. The double-threshold approach for identifying influential MHWs might not capture all possible influencing factors. The study's focus on a specific geographical region limits the generalizability of the findings to other oceanic basins. While the 24-hour MST is used to ensure independence, there might be some residual correlation between consecutive RI events not fully accounted for. Data scarcity in certain grid cells for some parameters, like TCHP, might affect the results' precision.