Role of tropical cyclones and salinity stratification in the expansion of a coastal marine heatwave

doi:10.21203/rs.3.rs-4726556/v1

Download PDF

Article

Role of tropical cyclones and salinity stratification in the expansion of a coastal marine heatwave

https://doi.org/10.21203/rs.3.rs-4726556/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Marine heatwaves (MHWs) during peak hurricane season can impact storm intensification, posing a significant threat to coasts and ecosystems. However, understanding of salinity stratification in the evolution of MHWs in river-dominated, hurricane-prone coastal regions is limited. Here, the event sequence leading to a large-scale coastal MHW in the Gulf of Mexico during the 2019 hurricane season was investigated. Both atmospheric and oceanic events coupled to drive the shelf-wide evolution of this full water-column MHW with surface heat fluxes, vertical mixing, and advection all contributing. In particular, several tropical cyclone events modified the vertical structuring of the MHW, and none intensified over the study region. During the decay phase, low salinity from freshwater discharge resulted in temperature inversions, leaving the bottom layer MHW preserved for an extended period. This study highlights the importance of salinity dynamics on MHWs and nearshore environmental conditions in intensity forecasting of landfalling storms.

Earth and environmental sciences/Ocean sciences/Physical oceanography

Earth and environmental sciences/Natural hazards

Marine heatwaves occur when extreme temperature anomalies persist in the ocean for a prolonged period (Pearce et al., 2011; Hobday et al., 2016), and have severe economic and ecological consequences such as coral bleaching and mass fish mortalities (e.g., Mills et al., 2013; Hughes et al., 2017; Smale et al., 2019). Moreover, there is a growing body of literature connecting interactions between marine heatwaves and tropical storms. Choi et al., (2024) found the first statistical connection between strengthening of the tropical cyclones in the Western North Pacific and Atlantic Ocean and marine heatwave events. Furthermore, marine heatwaves have also been associated with individual storm intensification. In the Bay of Bengal, Rathore et al., (2022) showed the presence of a marine heatwave event under the track of tropical cyclone Amphan (2020) during its rapid intensification. In the East China Sea, Typhoon Bavi (2020) amplified over a marine heatwave in a stratified water column (Pun et al., 2023). Dzwonkowski et al., (2020) showed that a full water column marine heatwave over the shelf formed as result of an atmospheric heatwave and a storm mixing event and created favorable thermal conditions during the intensification of Hurricane Michael (2018) in the northern Gulf Mexico.

Importantly, the intensification of landfalling storms due to the extreme heat content presents enhanced risk, especially when such conditions extend for prolonged durations (i.e., marine heatwaves). Predicted future climate scenarios indicate more intense storms, which are highly likely to undergo rapid intensification in the coastal regions prior to landfall (Emanuel, 2017; Liu et al., 2019, Chu et al., 2020; Knutson et al., 2020). Moreover, the nearshore atmospheric and oceanic conditions in the future will likely be favorable for the increasing storm intensities (Balaguru et al., 2022), which adds vulnerability to coastal populations and natural habitat. Therefore, advancing marine heatwave prediction in coastal regions where storms landfall is of critical importance.

Interactions between marine heatwaves and tropical cyclones should be expected given the importance of sea surface temperatures (SSTs) in turbulent heat fluxes at the air-sea interface. The evolution of SSTs is connected to the subsurface thermal structure which is often quantified through Tropical Cyclone Heat Potential (TCHP). This metric is the vertically integrated ocean temperature above 26℃ (Whitaker,1967; Leipper and Volgenau,1972) and has been correlated to tropical storm intensification in the open ocean (Wada et al, 2012; Jangir et al., 2021). However, on the continental shelf this metric has been suggested to be of limited use because bathymetric constraints which inhibit the accumulation of heat leading to low TCHP despite the presence of favorable ocean thermal conditions (Price, 2009). To resolve this, studies have suggested that depth-averaged temperature or mixing depth-averaged temperature may be a better metric to represent the thermal favorability of ocean waters (Price,2009; Balaguru et al., 2015; Potter et al., 2019, Pun et al., 2019). As such, the depth-averaged temperature provides a reasonable proxy to TCHP in shallow waters and has been shown to have significant impact in storm intensity changes over the shallow continental shelves (Price, 2009; Potter et al., 2019; Dzwonkowski et al., 2020; Dzwonkowski et al., 2022). Thus, extreme thermal conditions over the full water column in coastal zones represent a potentially important component for storm prediction prior to landfall.

Studies on the evolution of full water column marine heatwaves in the coastal ocean are often limited by the availability of long-term observations and inaccuracies in long-term, global-scale ocean models. As a result, the efforts to understand the characteristics of marine heatwaves below the surface have been mostly constrained to a few regions, such as the Australian coasts (Schaeffer and Roughan, 2017), the Mediterranean Sea (Daramaki et al., 2019), and the Northern American coasts (Gawarkiewicz et al., 2019, Chen et al., 2022; Amaya et al., 2023), Moreover, each of these coasts have regional differences and a specific sets of factors involved. For example, the variability associated with large-scale boundary currents tends to modulate the nature of subsurface marine heatwaves in areas lying closer to such systems (e.g., Pearce and Feng, 2012; Elzahaby and Schaeffer, 2019; Li et al., 2023), while other locations are modulated by the wind conditions and localized circulation (Schaeffer and Roughan, 2017; Schaeffer et al., 2023). While Amaya et al. (2023) provide analysis of the Gulf of Mexico and the southeastern United States, most of these studies do not focus on coastal regions regularly impacted by tropical cyclones.

Some tropical cyclones-impacted coastal regions are also strongly influenced by river discharge. The resulting salinity stratification might be expected to have a dynamical role in the evolution of marine heatwaves and their interaction with tropical cyclones. For example, Balaguru et al. (2020) showed that the strong stratification in Eastern Caribbean Sea and the western tropical Atlantic driven by low salinity river plumes from the Amazon–Orinoco River system reduces the vertical mixing and surface cooling, which favors rapid intensification. Similarly, thick barrier layer formation due to freshwater discharge in the northern Bay of Bengal was found to significantly contribute to increasing intensification rates (Fan et al., 2020). Because of the overlap between regions of freshwater influence and tropical cyclones, it is critical to better understand the role of salinity stratification in the evolution of coastal marine heatwaves during hurricane season.

As such, this study focuses on the Gulf of Mexico, which is known for tropical cyclones (i.e., hurricanes) and high river discharge. The continental shelf of Gulf of Mexico is widely known for tropical cyclone activity from June to November. In past years, several notable rapid intensification cases have been observed in the Gulf of Mexico, causing billions of dollars economic loss and substantial number of causalities. Recent examples of rapid intensification in the coastal regions include Hurricane Ida (2021) and Hurricane Idalia (2023), that made landfall as a category 4 storm in southeastern Louisiana and category 3 storm on the west Florida coast, respectively (Beven et al., 2022; Miles et al., 2023; Cangialosi and Alaka, 2024).

Historically, storms entering the Gulf of Mexico commonly make landfall in the northern region of the basin during peak hurricane season that runs between August to October (Kimball and Mulekar, 2004; National Hurricane Center (n.d.)). This is where the Mississippi river, the 9^th largest river in the world, and numerous other regional river systems influence the continental shelf dynamics with the influx of freshwater into the area. The impact of such river discharge on local circulation patterns in this region have been previously documented in many studies (e.g., Walker et al., 2005; Androulidakis and Kourafalou, 2013; Dzwonkowski et al., 2018). Only a few studies have documented marine heatwaves on the continental shelf in the Gulf of Mexico. Based on inner-shelf observations, Dzwonkowski et al., (2020) focused on how a series of compounding events influenced the thermal evolution of a marine heatwave. Other studies in the Gulf of Mexico have focused on long-term statistical assessments of the marine heatwaves (Ameya et al.,2023; Feng et al., 2023). However, the evolution of marine heatwaves and the role of salinity stratification across the continental shelf in the northern Gulf of Mexico remains speculative. With future atmospheric conditions in the nearshore regions predicted to be favorable for intensification (Balaguru et al., 2022), the need for understanding the dynamics of extreme shelf conditions in the northern Gulf of Mexico during peak hurricane season is a fundamental necessity.

To this end, a case study of a large-scale marine heatwave during the peak of hurricane season is investigated in the northern Gulf of Mexico in 2019. This study focuses on the chain of events that drove the evolution of the shelf component in the Mississippi Bight, where a shelf-wide full water column event occurred. The availability of a rare long-term time series of near-surface to bottom observations (Figure 1) in the Mississippi Bight in conjunction with remote sensing and reanalysis datasets allowed for a shelf-wide perspective of the formation, persistence and decay of the marine heatwave event and the role of salinity stratification. Interestingly, neither of the three storms (Imelda, Nestor and Olga, Figure 1) that interacted with this marine heatwave event in this specific region of the coast intensified to hurricanes, which also provided a unique opportunity to better understand how these storms interacted with the evolution and vertical structuring of the marine heatwave. As mentioned earlier, few studies have focused on mechanistic processing impacting marine heatwaves in seasonally stratified, river-dominated systems, particularly during peak hurricane season. This study, therefore, has significant regional and global implications, offering insights into this critical gap in knowledge.

2.1. Marine heatwave conditions during peak hurricane season

Extreme thermal conditions were evident across much of the Gulf of Mexico from the peak of hurricane season (August 15) through the tail end of the season (November 15) in 2019. Nearly all of the Gulf of Mexico experienced marine heatwave conditions at surface during this period (Fig. 2) with an exception in the central Gulf, a region that is influenced by the loop current. Most parts experienced average temperatures around 29°C (Fig. 2a) with peak temperatures reaching more than 30°C in some regions (Fig. 2b). It is important to note that most pixels experienced more than one marine heatwave (Supplementary Fig. 1), and the characteristics shown in Fig. 2 are the cumulative properties of all events at each pixel during the period. Furthermore, the maximum deviation of peak SST values from 90th percentile threshold (Fig. 2c) indicated the severity of the events with coastal regions in northern Gulf of Mexico and western Florida experiencing extreme anomalies greater than 1.6°C above the 90th percentile. Moreover, most pixels experienced marine heatwaves for more than a month (Fig. 2d), which is around one- third of the peak hurricane season. The continental shelf regions, defined from the coastline to 300m isobath (green dashed contours in Fig. 1) in Gulf of Mexico experienced 20–40 days of marine heatwave conditions in most parts with intensities exceeding 28.5°C. This basin wide perspective highlights the prevalence of marine heatwaves across the Gulf of Mexico during the hurricane season. However, surface temperature data provided no insight on the subsurface structure of marine heatwaves in the coastal zone. As such, more information on the formation and dynamics of the full water column marine heatwave on the continental shelf in northern Gulf of Mexico (north of 24°N) is needed to better understand these events in river influenced regions.

2.2. Evolution of the marine heatwave event

The space-time progression of marine heatwaves can provide a direct understanding of the stages of their formation and decay and how they are connected in space. The evolution of this marine heatwave event was tracked using the conditions of any pixels that experienced a marine heatwave during different time intervals. Consecutive 2-week periods (referred to as BW1, BW2, etc.) starting 21 August 2019 captured the major developments of the surface (Fig. 3a and 3b) and subsurface conditions (Fig. 3c and 3d). In the first biweekly period (BW1 in Fig. 3), pixels experiencing surface and subsurface marine heatwaves appeared to be scattered in the open ocean (outside shelf margin) except in western Florida shelf and coasts of Texas. In the following week, the shelf wide onset of surface marine heatwaves can be seen in the northwestern and north central parts of the region, extending from the coast to the shelf break, while subsurface marine heatwaves were restricted to the Texas and west Florida shelves. During mid to late September period (BW3 in Fig. 3), the surface marine heatwaves extended across the Gulf covering most of the continental shelves in the northern and western regions. Furthermore, this biweekly period (BW3) was also associated with the onset of subsurface marine heatwaves in nearshore locations eastward of those in BW2, following the surface marine heatwaves.

In early October (BW4 in Fig. 3), both surface and subsurface marine heatwaves continued to develop in the northeastern Gulf, covering the entire northern Gulf during the following period (BW5). Concurrently, a decline of marine heatwaves started over the Texas shelf, as a “detachment” from the nearshore region. In late October and early November (BW5 and BW6 in Fig. 3), most of the continental shelves transitioned out of the marine heatwave state, apart from the west Florida shelf. However, outer shelf regions nearer to the shelf break maintained their subsurface marine heatwave state. During this 12-week period (BW1 to BW6), the weekly intensities of surface marine heatwaves varied from above 30°C in September, decreasing to around 28°C in October, while peak intensities of subsurface marine heatwaves reached above 28℃ within the shelf regions.

The spatial evolution of marine heatwaves showed that a shelf-wide surface marine heatwave event formed in September, which was followed by a subsequent subsurface marine heatwave event in October (Fig. 3). This suggests vertical connectivity between these marine heatwave conditions. However, the two-week period excludes important perspectives of the vertical transition from surface to bottom marine heatwaves as well as the specific timing of the onset and decline of marine heatwave event through the water column. To better understand the vertical evolution of the marine heatwave over the shelf, observations from Fisheries Oceanography in Coastal ALabama (FOCAL) along with time series from Global Ocean Physics Reanalysis (GLORYS) at the mooring location during the period are shown in Fig. 4.

The temperature structure at the mooring location (Fig. 4a and 4b) indicated a surface warming trend in the stratified water column during late August and early September of 2019. This led to the initial marine heatwave event, confined to the surface layer between September 5–19 (Fig. 4d and 4j). During this period, surface temperatures elevated to above 30°C, after which the entire water column shifted to a vertically homogenized layer of ~ 28.5°C (Fig. 4a). In the beginning of October, a reheating of the top layer was followed by a downward transfer of heat to subsurface layers increasing the overall water column temperatures to ~ 30°C. This rewarming resulted in the onset of a full water column marine heatwave that continued until mid-October (Fig. 4c-4j). Around October 16, the surface temperatures sharply declined to ~ 26°C. In the following days, the surface layers started to lose heat at a faster rate compared to the subsurface layers and formed a temperature inversion where the bottom layer was warmer than the surface layer. During this time, the surface marine heatwave faded while the bottom layers remained in an active marine heatwave state (Fig. 4c).

The time series of temperatures at the FOCAL site provided insight on the surface to subsurface migration of the marine heatwave in October. However, this is only a single point on the shelf. A broader perspective of this marine heatwave event can be gained from GLORYS which allows for an expanded analysis over the wider shelf region and how the key events observed at FOCAL affected the broader shelf.

2.3. Evolution of thermal structure across the continental shelf

To expand on how the observations at the FOCAL mooring relate to the broader shelf conditions, GLORYS outputs were used to examine a cross-section of the Mississippi Bight during distinct phases of the marine heatwave event (Fig. 5). The evolution of the thermal structure and the associated marine heatwave event across the shelf involved four transition stages that correspond to those observed at the FOCAL site. During the initial phase between early to mid-September, the upper layers of the shelf experienced warming that resulted in an increase in temperature to around 30°C (Fig. 5a to 5c), with a strong thermocline across majority of the shelf (south of 30°N). In addition to the surface warming, at ~ 50m isobath, the temperatures at mid-depths increased potentially due to the intrusion of warmer water from offshore, where there was a pre-existing marine heatwave in the open ocean region of the Gulf (Fig. 5a to 5c, Supplementary Fig. 2). Around September 9, most of the shelf, except some locations in middle shelf, was experiencing a marine heatwave at the surface, and the outer shelf continued to experience a subsurface marine heatwave that extended further below the thermocline (Fig. 5b and 5c). After the persistent warming period in mid-September, the second stage was characterized by the broader shelf experiencing cooling in top layers and a homogenization of temperatures in upper 15m to ~ 28°C during September 17 to 25 (Fig. 5d to 5f, Supplementary Fig. 2). This temperature drop lead to a decline in the surface marine heatwave state across the shelf, while the more isolated subsurface marine heatwave persisted at depths between 20-40m and below 60m in the outer shelf. Moreover, a tongue of warmer temperatures near ~ 29℃ prevailed between 20-35m across the middle shelf (Fig. 5f).

Through late September and beginning of October, the temperatures across the shelf started to increase with the reintroduction of marine heatwave conditions across the shelf. At the surface, there was significant warming in the inner shelf region (Fig. 5g, Supplementary Figs. 2–3). The subsurface layers experienced marine heatwave conditions, especially north of 29.4°N, where no subsurface marine heatwave conditions occurred in September (Fig. 5g-h). The thermocline deepened to ~ 40m and reached the bottom in shallower locations. By October 15, most of the shelf reached full water column marine heatwave state (Supplementary Fig. 3). This reheating and expansion of the marine heatwave to depth in early to mid-October formed the third stage in the shelf transition.

The fourth and final stage of the marine heatwave event started around 17 October 2019 when the surface cooling begins to accelerate relative to the bottom layers, particularly over the inner shelf (Fig. 4 and Fig. 5j). During this stage between mid to late October, the shelf transitioned into temperatures below ~ 26℃, and the mid-depth intrusion of warmer waters in the outer shelf also ends (~ October 17). The surface layer continued to experience significant temperature declines compared to the subsurface and resulted in a temperature inversion across much of the shelf as the cooling progressed offshore (Fig. 5k-i). Moreover, the cooling phase of the shelf resulted in the decoupling of the marine heatwave into surface and subsurface parts over the mid-shelf (Fig. 5k-i, Supplementary Fig. 3).

Overall temperature variations experienced across the shelf (Fig. 5) were consistent with the marine heatwave stages observed at the FOCAL site (Fig. 4) though there were some differences in marine heatwave duration. Both FOCAL and GLORYS showed a consistent initial onset of the marine heatwave in the upper ocean. This surface event is briefly interrupted by a period of vertical homogenization and heat transfer to depth causing the subsurface expansion of the heatwave. Finally, the decay phase of the marine heatwave ensues which is defined by a thermal inversion allowing the bottom marine heatwave to persist beyond the surface manifestation. This sequence of transitions naturally leads to questions about the mechanisms associated with the extreme thermal conditions during this period. As such, the atmospheric and oceanic processes driving these stages are further examined in subsequent sections.

2.4. Mechanisms driving the onset of the marine heatwave event

To understand the processes responsible for the onset of the marine heatwave, various atmospheric factors are considered. As observed previously in Fig. 4 and Fig. 5, the shelf experienced two periods of surface warming, one in September and the other in October. Both of these surface marine heatwave periods were preceded by extreme environmental conditions. Extended atmospheric heatwaves were observed at the FOCAL mooring (Fig. 6), representative of basin-wide conditions that were sustained for more than a month in this region (Fig. 6a, 7a, and 7b.). These extreme air temperatures resulted in small air-sea temperature differences that were at times positive, i.e., air warmer than the ocean. Accompanying these high air temperatures were high relative humidity with conditions above 75 percent (Supplementary Fig. 4b and 7c) and relatively weak wind conditions with magnitudes below 5 ms^− 1 (Fig. 6b and 7d). These conditions favored reduced heat loss due to low latent and sensible heat fluxes (Fig. 6c) and resulted in net positive heat flux into the ocean (Fig. 6c and 7f). These atmospheric conditions and the resulting surface heat flux were consistent with the upper ocean heating observed in both the FOCAL data (Fig. 4) as well as the GLORYS reanalysis transect.

2.4.1. Offshore intrusion

In addition to the surface layer warming through atmospheric heating, the warm mid-depths over the shelf break were observed to be connected to the warmer temperatures over the outer shelf (Fig. 5a to f, Supplementary Fig. 2). The outer edge of the mid-shelf region, i.e., just before the shelf slope rapidly steepens (~ 50m isobath), was a location where key variations in the warm water intrusion from the outer shelf can be observed from the thermal structure (Fig. 5). Hence, the along-shelf and cross-shelf heat advective transports were examined at this location (Fig. 8, Supplementary Fig. 5). In September, the net advective transport was dominated by the cross-shelf transport over the outer shelf. Moreover, the strongest cross-shelf temperature gradient was observed in the subsurface layers below 10m (Fig. 8b) during that period, with a warmer outer shelf. During early September, the northward onshore flow from the outer shelf resulted in an increase in advective heating in layers below 10m (Fig. 8c). However, the temperatures in the subsurface layers between 5-30m at this location only increased significantly after September 7 (Fig. 8d), when the eastward along-shelf current transported warmer water from west (Supplementary Fig. 5). The subsurface warming below 15m continued with the contributions from the cross-shelf transport during September 16–20. During this period and just after, there were some intermittencies in the subsurface heat transport associated with strong surface forcing from the transit of Tropical Cyclone (TC) Imelda and a subsequent cold front (Fig. 8c and 8d). However, the advective heat transport across the outer shelf resumed shortly thereafter and continued until around October 2, at which point the offshore contribution to the shelf diminished substantially for the rest of the month. Therefore, the inbound transport of heat from offshore played a key role in elevating subsurface conditions over the outer shelf in September.

2.5. Inner shelf expansion of the marine heatwave to depth

At the opposite end of the shelf, other processes carried the extreme conditions from the surface to lower layers of the water column. The transition began in mid-September when the initial warming of surface layers was interrupted by the passage of TC Imelda and a subsequent cold front. At this time, there was a homogenization of the upper water column during September 15–23 that was associated with the wind conditions from these systems. While TC Imelda did not form until reaching the Texas shelf (Fig. 1), well west of the study site, the disorganized precursor low pressure system originated in the eastern United States and moved southwestward during September 10–14 passing through the eastern Gulf of Mexico. The relatively strong winds associated with the transit resulted in high winds around ~ 8 ms^− 1 over the shelf (Fig. 6b). This wind event led to significant cooling of the surface layer (Fig. 5d-e) which was primarily driven by mixing. This can be seen from the surface heat flux, which remained positive, i.e., heat gain, while at the same time the lower layer temperatures experienced warming.

After the pre-Imelda low pressure system passed through eastern Gulf of Mexico, there was a brief reduction in the wind forcing (Fig. 6b, September 18). However, the remnants of TC Imelda over Texas (i.e., southwestern United States), in conjunction with the passage of a cold front over the Mississippi Bight led to a reinvigoration of the wind forcing (September 19). For the next 36 hours the shelf experienced winds with magnitudes of ~ 10 ms^− 1 (Fig. 6b) which led to significant changes in the thermal conditions over the shelf. During this wind event there was some minor heat loss through surface fluxes with values peaking at ~-80 W m^− 2. Despite the minor surface heat flux, the surface layer still exhibited notable cooling as well as warming of the lower layers, indicative of mixing during these high wind events. Additional evidence of mixing during this period can be seen at the FOCAL mooring where the salinity time series showed an increase in the surface salinity (~ 1–2 psu) during each wind event (Fig. 9b).Thus, these wind events transferred upper ocean heat to depth, warming the subsurface layers setting up conditions that contributed to the full water column marine heatwave (October 2) that began during the second atmospheric heatwave.

The expansion of the marine heatwave to depth was further supported by advective transport in the bottom boundary layer. Immediately after the period of tropical storm activity, the shelf temperature structure showed gradual warming across the bottom (Fig. 5f-i, Fig. 9d). The change in temperature structure was coincident with the offshore transport of inner shelf water along the bottom (Fig. 9c). Observational evidence of this transport can be seen in the FOCAL velocity data, where onshore transport begins on September 28 (Supplementary Fig. 6) and continues till the end of the time series on October 3. While there was some discrepancy in the across-shelf velocity between FOCAL and GLORYS during this post tropical activity period, there is offshore shore flow over the mid-shelf south of 30°N in the GLORYS product. After a brief, weak reverse of bottom flow, the offshore transport over the mid-shelf resumes on October 4 and continues through October 21.

This bottom boundary transport of heat was initially driven by wind forcing. On September 28, wind conditions shifted to an easterly direction (Fig. 6b). While the wind forcing is relatively weak, the FOCAL mooring does show a characteristic downwelling response consistent with Ekman coastal circulation where the depth-average flow was westward and was accompanied by onshore surface velocities and offshore bottom flow (Supplementary Fig. 6). By October 4, stronger wind conditions impact the area (Fig. 6b), driving a similar response across the shelf from October 5–13 (Fig. 9). After this point, the wind forcing becomes weak and directionally variable. However, across the shelf the sea surface slope over the mid-shelf maintained westward flow over this region (Fig. 9a and 6d). This along-shelf flow continued to support the offshore flow through Ekman transport in the bottom boundary layer. As such, the offshore advective transport of heat along the bottom, as seen by the ‘tongue’ of warm water, maps onto the mid-shelf expansion of the subsurface marine heatwave (Figure. 9d).

2.6. Mechanisms driving the decay of marine heatwave.

After the peak expansion of the full water column marine heatwave between October 15–17 (Fig. 5j, Supplementary Fig. 4), the inner shelf surface layers started cooling and the associated heat loss resulted in the decline of the marine heatwave at the surface (Fig. 4). However, the warmer subsurface layers persisted, maintaining the marine heatwave state (Fig. 4). While GLORYS does not capture the marine heatwave statistics precisely as observed in Fig. 4, the temperature inversion observed in the inner shelf is captured to some degree. In the following days, the surface cooling and temperature inversion extended offshore and resulted in the ‘decoupling’ of the surface and bottom part of marine heatwaves in the mid-shelf (Fig. 5k-i).

This cooling period in late October was dominated by heat loss at the air-sea interface. This heat loss was associated with a series of weather events including the transit of two successive tropical storms TC Nestor and TC Olga over the northern Gulf of Mexico on October 19 and 26 as well as several cold fronts. These events were characterized by an extended period of persistent negative heat fluxes with several peaks larger than ~ -400 W m^− 2. The negative net heat fluxes were driven by negative latent and sensible heat fluxes (Fig. 6, Supplementary Fig. 7). At the same time, advective heat transport was limited (Fig. 8, Supplementary Fig. 5).

As surface temperatures decreased rapidly in response to the negative heat flux, stratification insulated the subsurface layers allowing heat loss to occur at a much slower rate. This stratification and the associated temperature inversion was supported by vertical gradients in salinity that resulted from extensive river discharge delivered to this region. The low salinity layer spread across the Mississippi Bight in October (Fig. 10a, Supplementary Fig. 8) and resulted in a distinct surface layer with increased buoyancy relative to the saltier, deeper water column. Moreover, there was a notable salinity-driven restratification around October 16 that separated the surface and bottom layers in the inner shelf location (i.e., FOCAL mooring, Fig. 10b). Qualitatively, the salinity structure across the shelf from GLORYS indicated signs of stratification and associated temperature inversion (Fig. 10c and 10d). Hence, the freshwater discharge from the Mississippi river and regional river systems contributed to the evolution of the marine heatwave at depths through stratification.

3.1. Marine heatwave continuum

The evolution of marine heatwaves in the coastal ocean have been studied in different ways. Most of the previous studies focused on the spatial evolution of marine heatwave events at the surface due to availability of satellite data and numerical models that replicate surface conditions at relatively high accuracy (Oliver et al., 2021). On the other hand, most studies of subsurface extensions of marine heatwaves rely on point in situ observations to describe the changes at depths (Schaeffer and Roughan, 2017; Elzahaby and Schaeffer 2019, Zhang et al., 2023). This study provides a broad understanding of a large-scale marine heatwave event making use of multiple data sources including satellite (OISST), ocean reanalysis (GLORYS), and mooring observations (FOCAL), which allows for the visualization of how marine heatwaves evolve in space and time across the Eastern Gulf of Mexico continental shelf. These spatial and temporal mappings highlighted an event ‘continuum,’ where one event consisted of the three types of marine heatwaves defined by Schaeffer et al., (2023) (i.e., surface, extended, and subsurface), depending on the developmental stage.

This continuum notion may be important for predicting/anticipating the evolution of marine heatwave events. For example, recent work by Zhang et al., (2023) speculated that “…a shallow marine heatwave could be the start or end of a deep or subsurface marine heatwave event…” in the open ocean. Here, direct evidence of this coupling was found in a coastal system. Heating during a surface marine heatwave in September was mixed to depth, after which a second surface heating event in conjunction with downwelling circulation drove a full water column event over the inner to mid-shelf region of the shelf. While statistical analysis of ‘types’ of shelf events have yielded important understanding of marine heatwaves (e.g., Hu 2021, Amaya et al., 2023, Schaeffer et al., 2023), these approaches have a more limited ability to capture the connections between such extreme events. As in this case study, the connections between surface and subsurface events were important in understanding the extreme thermal conditions over the shelf.

Furthermore, this study showcases an event that is characterized by the co-occurrence of surface and bottom marine heatwaves during the formation to peak stages. The detailed understanding of the evolution of this event suggests that even though the ‘types’ of marine heatwaves are asynchronous at certain points in the evolution, they are linked though mixing and advection. Therefore, it is important to understand the physical processes that drive each evolutionary stage in transforming the shelf to extreme temperatures.

3.2. Driving mechanism for the marine heatwave

Marine heatwave events can be driven by a combination of atmospheric and oceanic processes during each evolutionary stage. Here, multiple drivers lead to a shelf-wide, full water column event including surface heat flux, vertical mixing, and advection over both the inner shelf to mid-shelf as well as the shelf break to outer shelf. During the initial stage of the event, the persistent warming at the surface was driven by the surface heat fluxes, which has been identified as important in other coastal marine heatwave studies (e.g., Holbrook et al, 2019; Sen Gupta et al., 2020; Oliver et al., 2020; Großelindemann et al., 2022). As subsequent wind events paused the development of this marine heatwave, the transfer of heat to depth through mixing was critical over the inner to mid-shelf region. The role of wind forcing in subsurface heat transfer was previously recognized in marine heatwave studies in continental shelves across the globe (e.g., Schaeffer and Roughan, 2017; Benthuysen et al., 2018; Dzwonkowski et al., 2020). In addition, the importance of advection via downwelling was also identified an essential component of this event sequence, which has been noted in other coastal marine heatwave studies (e.g., Schaeffer et al., 2017, Gröger et al., 2024). In addition, a pre-existing subsurface marine heatwave at the outer edge of the study region proved to be significant in transporting subsurface heat across the shelf break onto the outer shelf. The shelf break region of the Gulf of Mexico is regularly impacted by large-scale circulation patterns associated with the Loop Current such as mesoscale eddies and/or frontal eddies (e.g., Walker et al., 2005, Le Hénaff et al., 2012, Androulidakis and Kourafalou 2013). While beyond the scope of this work, it is expected that the setup of such features, as demonstrated in other marine heatwave studies in shelf break regions near western boundary currents (e.g., Benthuysen et al., 2014, Gawarkiewicz et al., 2019, Großelindemann et al., (2022), Schaeffer et al., 2023), likely contributed to the pre-existing offshore subsurface marine heatwave. While previous studies have typically identified individual mechanisms that correlate with coastal marine heatwaves, fewer studies have focused on event sequences that lead to extreme conditions at depth.

3.3. Influence of river discharge in marine heatwave evolution

Stratification in the water column limits the heat transferred between layers and therefore can play a dominant role in the expansion of marine heatwaves. Previous studies on marine heatwaves focused on different aspects of stratification associated with the temperature and/or salinity structure in the water column. Studies in Australian shelves suggest that weak thermal stratification in the coastal region due to downwelling conditions can potentially allow the marine heatwaves to penetrate to subsurface layers (Schaeffer and Roughan 2017; Schaeffer et al.,2023). The presence of strong salinity stratification can lead to extended durations in surface marine heatwaves (Liu et al., 2023; Rathore et al., 2022), especially in marginal seas like the Baltic Sea with a permanent halocline (Gröger et al., 2024). Furthermore, Großelindemann et al., (2022) found that the occurrence of subsurface marine heatwaves on continental shelf in the northeast United States coincided with salinity extremes and salt influx from the outer shelf. However, limited studies have focused on the role of salinity in the evolution of marine heatwaves at depths, specifically in river-dominated shelves where large variability in stratification can be expected. In this study, salinity stratification was found to be crucial in the extended duration of the bottom marine heatwave phase of the event.

The low salinity top layer created across the shelf by the discharge from the Mississippi River and other regional river systems resulted in strong vertical gradient in salinity, which impacted the marine heatwave at surface and subsurface. The salinity stratification in early October led to shallow mixed layer depth offshore (Fig. 10c), trapping the heat received in the upper layers while limiting the transfer heat to lower layers (Fig. 10d). The trapping of heat in a shallower upper layer is consistent with previous open ocean studies (e.g., Kumar et al., 2023) and allowed the marine heatwave to regain the surface signal. However, this same shallow salinity-driven stratification contributes to the rapid loss of upper ocean heat in late October. This is clearly demonstrated at FOCAL where the rapid cooling generated a significant temperature inverse that resulted in the decay of the marine heatwave at surface (Fig. 4, Fig. 10b). However, this same stratification isolated the bottom layers from losing heat, sustaining the subsurface marine heatwave state in both the inner shelf and offshore locations.

3.4. Interaction with tropical storms

Extreme heat content in the ocean favors intensification of tropical cyclones and so the prolonged duration of intense temperature anomalies or marine heatwaves therefore provide higher probabilities for severe storms. In the open ocean, surface marine heatwaves and strengthening of tropical storms have been found to be directly correlated (Choi et al., 2024). Moreover, the precursor marine heatwaves along cyclone tracks were found to be a decisive factor in rapid intensification of storms across the globe (Rathore et al., 2022; Pun et al., 2023). In the coastal ocean, several recent studies report a growing number of rapid intensity changes in storms that are expected to continue to increase in the future (e.g., Balaguru et al.,2022; Li et al.,2023; Li et al.,2024). These studies rely, in part, on the connections of SST as well as atmospheric conditions, to explain the increase in coastal intensification. Coincidentally, several studies using SST data have shown that coastal marine heatwaves are becoming more frequent and longer (Marin et al., 2021, Hu, 2021). However, little is known about trends in subsurface marine heatwave conditions in coastal regions, an important aspect for the prediction of tropical cyclone-ocean interactions.

The potential for interaction between such extreme events requires an enhanced understanding of the drivers of prolonged events of extreme depth-average temperatures in the coastal ocean that are affected by landfalling storms. This study provided one sequence of mechanisms that drove an extreme depth-averaged temperature event with temperatures above 26.5℃, the critical ocean temperature for sustaining tropical cyclones (Palmen,1948, Gray, 1968), and that lasted more than a month (Fig. 4). We suspect that other sequences of shelf and shelf break processes will allow shelf-wide, full water column marine heatwaves during hurricane season. Expanding our knowledge of precursor mechanisms may be useful to estimate subsurface coastal ocean conditions prior to landfalling storm events, given the lack of real-time observational data on bottom shelf conditions.

The three tropical storms during our study period modified the thermal structure across the shelf and impacted the evolution of marine heatwaves differently. Tropical storm Imelda and a subsequent cold front during mid-September homogenized the upper layers of the water column through mixing while pausing the atmospheric heatwaves for a week (Fig. 4a, Fig. 5d-f, Fig. 6a). This is consistent with the mechanisms in the case study by Dzwonkowski et al., (2020), where tropical storm Gordon (2018) favored the subsurface transfer heat before an atmospheric heatwave initiated a full water column marine heatwave. On the other hand, the successive storms Nestor and Olga on October 19 and 26 in conjunction with cold fronts resulted in a period of enhanced heat loss in the surface layer and a decline in temperatures. These wind events also put an end to the favorable atmospheric conditions for marine heatwave that existed prior to their transit (Fig. 5j-l, Fig. 6c). Though favorable oceanic conditions were present for all three storms as they passed through the Mississippi Bight (the precursor low pressure system for Imelda), none of the storms intensified in this region. This is likely because of the unfavorable atmospheric conditions. Interestingly, Imelda intensified from a low-pressure system into a named tropical storm overnight when it moved over the Texas Shelf which was in marine heatwave state during the period. However, details of these atmospheric responses are beyond the scope of this study. This does reinforce the need for understanding interactions between marine heatwaves and tropical cyclones within the coastal zone during hurricane season.

Marine heatwaves are characterized by extreme thermal conditions in the ocean that last for an extended period. These events are significant in coastal regions where tropical cyclones make landfall, due to their favorable thermal conditions for intensification. Limited studies examined marine heat wave events in river-dominated regions where salinity dynamics can be important to their evolution. This study explores a largescale, full water column marine heatwave event during peak hurricane season in 2019 that affected the entire continental shelf in the northern Gulf of Mexico. To understand the driving mechanisms across the shelf, the study focused on the Mississippi Bight which is regularly affected by tropical cyclones and Mississippi River discharge. A rare set of long-term (16 years) full water column data, in conjunction with remote sensing and reanalysis datasets were used to expand our understanding of the complex interaction between surface heating, mixing, and advection during the initiation, persistence, and decay of a shelf-wide marine heatwave.

Event sequences proved to be critical to the evolution of the marine heatwave. Here, the interruption of consecutive atmospheric heatwaves by a mixing event (i.e., coupling of TC Imelda and a cold front passage) lead to the transfer of heat to depth. This was coupled with the import of heat at depth from a pre-existing subsurface marine heatwave offshore of the shelf break as well as the across-shelf advection of excessive inner shelf heat from a wind-driven downwelling event. These coupled of events drove the development of a full water column marine heatwave across the shelf with depth-averaged temperatures reaching well above 26.5℃ during the peak of hurricane season. In a rapidly changing climate, with future near-shore conditions predicted to be more favorable to rapid intensification, the mechanistic understanding of marine heatwave evolutions will be expected to be increasingly important for improved storm forecasting accuracy prior to landfall.

Moreover, this case study highlights the significant role that salinity stratification can have in the evolution of coastal marine heatwaves. After the peak expansion, the marine heatwave started to decline due to surface cooling and heat loss from a series of weather events in mid to late October. However, low salinity from freshwater discharge allowed for a temperature inversion to develop, which preserved the bottom layer anomalies for an extended period. As such, additional consideration is needed when examining river-dominated continental shelves in the context of surface to subsurface linkages and the available heat content.

While previous studies showed marine heatwaves as a driver for intensification of tropical storms, this study points out that the correlation between them does not imply causation as other environmental conditions also need to be favorable. Here, three tropical storms interacted with the marine heatwave event in the Mississippi Bight, but none intensified in this region. Only TC Imelda intensified during its Louisiana/Texas shelf passage over a surface marine heatwave occurring over the broader northern Gulf of Mexico. This highlights the need for better understanding of the relationships between the relative importance of oceanic and atmospheric conditions in intensification as storms approach landfall.

The primary focus of this study is on the continental shelf across the northern Gulf of Mexico (Fig. 1) for reasons mentioned the above. As such, a cross-shelf transect in the Mississippi Bight is used as a representative subregion that is highly affected by both landfalling storms and seasonal river discharge (Fig. 1). In addition, the availability of long-term surface and subsurface measurements (ORB and FOCAL in Fig. 1) allow for direct observations of the evolution of marine heatwaves and provide ground truthing of GLORYS ocean reanalysis for shelf wide analysis. Using these datasets, we address the broader goal of understanding marine heatwave evolution with a case study of 2019, one of the warmest years on record (NCEI Climate Report) with extreme SSTs across the Gulf of Mexico. Moreover, the 2019 peak hurricane season featured a basin wide long-lasting marine heatwave event during peak months (Fig. 2).

5.1. In situ and satellite datasets

Optimum Interpolation Sea Surface Temperature (OISST): To identify the surface marine heatwaves, long-term sea surface temperatures (SST) from National Oceanic and Atmospheric Administration (NOAA) OISST version 2.1 (Banzon et al., 2016; Huang et al., 2021) was used. This is a 0.25° resolution product that integrates observations from different platforms (satellites, ships, buoys, and Argo floats), which extends from September 1, 1981, to the present and is therefore widely used in marine heatwave studies (e.g., Hobday et al., 2016; Sen Gupta et al., 2020; Dzwonkowski et al., 2020; Pun et al., 2023). The OISST is accessible via (https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/).

Soil Moisture Active Passive (SMAP) remote sensing data: To track the surface salinity patterns during the evolution of the marine heatwave event, SMAP data was used. SMAP uses a passive sensor to collect signals in the microwave band to determine the sea surface salinity (SSS). Here, the 8-day running averages of SSS were obtained from the SMAP mission L3 product (https://www.remss.com/missions/smap/salinity/). This product has a spatial resolution of 1/4° (~ 25km) and extends from 2015 to the present.

Fisheries Oceanography in Coastal ALabama (FOCAL) mooring: To understand the aspects of the hydrographic characteristics of the shelf, observations from the FOCAL mooring located 25 kilometers southwest of the mouth of the Mobile Bay, Alabama (Fig. 1, 30.0902° N, 88.2116° W) were used. This mooring provides a rare long-term time series of water column temperature (4.8-20.3m), near-surface (~ 4.8m) and bottom (~ 20.3m) salinity and water column current velocity data collected from 2004 to 2020. A detailed description of mooring configuration and processing can be found in various studies (e.g., Dzwonkowski et al., 2018, Dzwonkowski et al., 2022).

Orange Beach Surface Buoy (ORB): For additional insight on the near-surface temperatures as well as surface heat fluxes during the study period, data from the National Oceanic and Atmospheric Administration (NOAA) National Data Buoy Center (NDBC) surface buoy (42012) located offshore of Orange beach, AL (Fig. 1, 30.060°N, 87.548°W) were used (https://www.ndbc.noaa.gov/station_page.php?station=42012). Standard meteorological information including air temperature, dewpoint temperature, wind speed and wind direction as well as water temperature at 2m below surface were obtained at hourly intervals beginning in 2009.

5.2. Reanalysis datasets

North American Regional Reanalysis (NARR): To understand the atmospheric conditions, 3-hourly values of key variables including air temperature at 2m, relative humidity at 2m, windspeed, and heat flux terms were obtained from the National Centers for Environmental Prediction (NCEP) NARR (https://www.esrl.noaa.gov/psd/data/gridded/data.narr.html). This atmospheric reanalysis extends from 1979 to present covering the North American regions and outputs are recorded in a Northern Lambert Conformal Conic grid, which were interpolated and converted to rectilinear lat-lon grid for visualization and analyses. This product is previously used in marine heatwave studies in the region to analyze the role of atmospheric processes (Chen et al., 2015; Dzwonkowski et al., 2020).

Global Ocean Physics Reanalysis (GLORYS): In order to track the evolution of the marine heatwave as well as to assess the mechanisms driving the changes in the marine heatwave, GLORYS12v1 was used. GLORYS is a global eddy-resolving physical ocean and sea ice reanalysis from Copernicus Marine Environment Monitoring Service (CMEMS)(https://data.marine.copernicus.eu/product/GLOBAL_MULTIYEAR_PHY_001_030/description). This dataset provides daily mean files for temperature, salinity, and currents from the top to the bottom on a standard regular grid at 1/12° (~ 8km) and on 50 standard levels from 1993 to present. This product was highlighted for marine heatwave research by Oliver et al., (2021) because of its extended duration. While the resolution is still coarse for shelf processes, a recent study of marine heatwaves (surface and bottom) on North American shelves has demonstrated its capacity in these coastal regions (Amaya et al., 2023). Additional validation of this product during this specific study period is also provided in Section 5.4.

5.3. Tropical Cyclone Data

The best track data to evaluate the transit of three tropical storms Imelda, Nester and Olga were acquired from those provided by NOAA National Hurricane Center (https://www.nhc.noaa.gov/gis/archive_besttrack.php?year=2019).

5.4. Methods

Identification of heatwaves: In this study, widely used definitions of heatwaves were used. Atmospheric heatwaves were defined as a period when the air temperature persists above 90th percentile threshold for three or more days (e.g., Collins et al., 2000; Pezaa et al., 2012; Perkins and Alexander, 2013). Marine heatwave requires the temperatures to exceed the climatological 90th percentile threshold for more than 5 days with a maximum gap of less than 3 days as defined in Hobday et al, (2016). We acknowledge the ongoing debate on different baselines in defining temperature extremes put forward by Amaya et al., (2023b) and Sengupta (2023). However, we follow a fixed baseline approach according to climatological 90th percentile threshold by Hobday et al., (2016) to be consistent with recent studies on marine heatwaves (Schaeffer and Roughan, 2023) and tropical storm interactions (Dzwonkowski et al., 2020; Rathore et al., 2022; Choi et al., 2024). At each pixel, the marine heatwaves are detected using daily SST values from OISST, GLORYS, or the long-term in situ datasets using the python module from https://github.com/ecjoliver/marineHeatWaves. Although longer time-scale definitions for marine heatwaves exist (e.g., Jacox et al., 2022), the daily time-scale definition aligns more closely with the time frame relevant to the potential event interaction, such as tropical cyclones, which is a focus of this study.

Depth-averaged temperature: Upper ocean heat content (or TCHP) is an important metric for assessing the risk of tropical cyclone intensification (e.g., Leipper and Volgenau, 1972; Shay et al., 2000; Lin et al., 2013). However, the relevance of TCHP calculation may be limited in the shelf regions where bathymetry constraints the water column depth. In such regions, depth-averaged temperature calculated across entire water column is an alternative measure for assessing the thermal favorability of the coastal ocean for storm intensification (Price, 2009; Potter et al., 2019; Dzwonkowski et al., 2021). In this study, the depth-averaged temperature is calculated by integrating the top to bottom temperature within the shelf and the upper 300m beyond the shelf break.

Evaluation and use of reanalysis products: To gain confidence in the large-scale reanalysis products in this shelf region, comparison to in-situ based estimates were derived. For NARR, the sensible and latent heat fluxes were compared to those at ORB using TOGA-COARE version 3.5 algorithm (Supplementary Fig. 4). Relative humidity, which is a required input to this algorithm, was determined from the dew point temperature using standard equations (https://bmcnoldy.rsmas.miami.edu/Humidity.html) (Alduchov and Eskridge, 1996). This methodology for calculating surface heat fluxes has been previously used in studies in the region (e.g., Coogan et al., 2018; Dzwonkowski et al., 2022).

In order to gain confidence that GLORYS was capturing the main processes occurring over the shelf during the study period, the behavior of the temperature, salinity and velocity were compared to various in-situ and remote sensing data. For temperature, the surface values of GLORYS were compared against observations from OISST (Fig. 3, Supplementary Fig. 9a) and ORB (Supplementary Fig. 9b) and the subsurface structure was validated against FOCAL (Fig. 4). Both spatial and temporal characteristics of the surface temperatures were captured remarkably well in GLORYS, which was expected as it assimilates the remote sensing and surface buoy observations. For the comparison to the FOCAL mooring observations, GLORYS captured major changes in the temperature structure during the period (Fig. 4a,b). Both FOCAL (Fig. 4a) and GLORYS (Fig. 4b) indicated surface warming of a stratified water column in late August, which continued until mid-September. Furthermore, both datasets showed enhanced cooling of surface layers during late October, while bottom layer temperatures remained warmer resulting in temperature inversion in the water column. However, some of the exact details of the higher frequency patterns in the FOCAL data were underrepresented.

In relation to capturing the marine heatwave events, the time series of near-surface temperature from GLORYS (Fig. 4d) indicated an initial marine heatwave in September with a two-week duration, followed by a second event in October that lasted around three weeks. This was consistent with those observed in OISST and ORB at the same location (Supplementary Fig. 9). On the other hand, time series of FOCAL temperature at ~ 4m (Fig. 4c) indicated heat spikes in September and a marine heatwave event in October that had similar onset and decline to that observed in GLORYS. A full water column marine heatwave event was observed in the time series of depth-averaged temperatures in October, which was characterized by a month-long event in FOCAL (Fig. 4e) and seventeen-day event in GLORYS (Fig. 4f). Significant differences existed in bottom temperatures, in which the FOCAL timeseries (Fig. 4g) displayed a month-long heatwave event in October while GLORYS (Fig. 4h) showed a twelve-day event. While GLORYS captured the surface events consistent with the observations, it showed shorter duration of subsurface events compared to the observations. However, the general patterns and key features including onset of the events were represented in GLORYS.

The vertical structure of the overall deviations of temperatures from 90th percentile from FOCAL (Fig. 4i) and GLORYS (Fig. 4j) revealed the notable features which GLORYS captured. These included the extreme surface conditions in September and the vertically extended marine heatwave event in October. It is important to note that the differences observed in GLORYS could be attributed to the shallow nature of the site compared to the resolution of GLORYS, that is FOCAL is ~ 2 grid cells away from the coastal boundary in GLORYS. In addition, there is a considerable difference in duration between FOCAL and GLORYS with FOCAL having more than 10 fewer years of observation. This difference in durations between FOCAL and GLORYS would be expected to lead to some statistical differences in thresholds. Despite these limitations, GLORYS showed the ability to capture the key features associated with extreme conditions and allows for a broader examination of the temperature variations across the continental shelf where observations are lacking.

Since Mississippi Bight is a river-influenced region where salinity dynamics are expected to be a critical element of the evolution of the marine heatwave, the salinity performance of GLORYS was also important to assess. In order to validate the salinity for the study period, GLORYS values were compared against SMAP (Supplementary Fig. 10) and FOCAL (Fig. 9b, Supplementary Fig. 8). The sea surface salinity comparison with SMAP suggests that GLORYS captures the qualitative extent of low salinity plumes from the Mississippi river during the period, despite quantitative disagreements in values (Supplementary Fig. 10). Similarly, surface and bottom salinities from GLORYS showed quantitative differences when compared to FOCAL (Fig. 9b, Supplementary Fig. 8a). However, qualitatively the salinity variations captured the key features during the period including the stratification, mixing, and downwelling events. Furthermore, the difference between surface and bottom salinities (salinity stratification) in GLORYS followed a similar pattern to that observed at FOCAL suggesting the salinity dynamics at the site were captured well in GLORYS despite the differences in the actual salinity (Supplementary Fig. 8b). This qualitative agreement in the vertical salinity difference between GLORYS and FOCAL was notable during the subsurface warming and temperature decline in October (Supplementary Fig. 8c). Furthermore, the overall qualitative similarities of GLORYS to SMAP and FOCAL suggest that the product can be used to study the changes in salinity stratification across the shelf during the study period. The quantitative difference in salinity in GLORYS could be due to the use of climatological river discharge forcing, which would be expected to lead to an underestimation in the amount of fresh water delivered to the coastal zone during the summer 2019, as the Mississippi River discharge in Spring 2019 significantly exceeds the climatology (Supplementary Fig. 11). However, in August and September 2019, the discharge values approach the climatology, likely improving the regional salinity representation in GLORYS during the study period.

Finally, the inner shelf currents in GLORYS were found to be mostly in agreement with that of available observations at FOCAL (Supplementary Fig. 6). The overall agreement with the FOCAL mooring, data not assimilated into GLORYS, provides confidence that the main physical process and forcing response over the shelf are reasonably represented in GLORYS during the study period. Hence, the selected reanalysis products allow for the quantification of the spatial forcing conditions (via NARR) and system wide response (via GLORYS) so that spatial and temporal changes in the thermal structure during a shelf-wide marine heatwave can be fully explored.

Competing Interests

The authors declare no competing interests.

Author Contributions

D.R. contributed to conceptualization, investigation, formal analysis, and writing (original draft). B.D. provided guidance to D.R. and contributed to conceptualization, data curation, investigation and writing. S.F. provided guidance to D.R. and contributed to investigation. G.L. contributed to data curation. All authors contributed to the writing of the final version (review and editing).

Acknowledgements

This work would not have been possible without the many folks that have passed through the Tech Support Group at the Dauphin Island Sea Lab to help maintain site FOCAL over the years. Authors thank University of South Alabama and Dauphin Island Sea Lab for graduate student (D.R.) support and resources. The authors are grateful to Zhilong Liu, Harikrishnan Sreeshylam, and Aravind Puzhankara for their support. A portion of this work was conducted at the Jet Propulsion Laboratory, California Institute of Technology, under contract with NASA. This research was made possible by the NASA Physical Oceanography program (80NSSC21K0553) and NOAA IOOS program via the Gulf of Mexico Coastal Ocean Observing System (NA21NOS120092).

Data Availability

Data are publicly available from various sources stated in the “Methods” section. All data that were not downloaded from public sources have been made publicly available through the Dauphin Island Data Management Center (https://data.disl.edu/ ) and (or have been submitted to) the NOAA National Center for Environmental Information (NCEI). In particular, FOCAL data is available through the following links: https://accession.nodc.noaa.gov/0203749; https://accession.nodc.noaa.gov/0211052; https://accession.nodc.noaa.gov/0243995; https://accession.nodc.noaa.gov/0243655; https://accession.nodc.noaa.gov/0243653; https://accession.nodc.noaa.gov/00243652.

Code Availability

The codes used in this study are available upon request to the corresponding author. The references in the “Methods” section provide speciﬁc details of the calculations in the codes.

Pearce, A., Lenanton, R., Jackson, G., Moore, J., Feng, M. & Gaughan, D. The “marine heat wave” off Western Australia during the summer of 2010/11. Fisheries Research Report No. 222. Department of Fisheries, Western Australia. 40 (2011). https://www.fish.wa.gov.au/documents/research_reports/frr222.pdf
Hobday, A. J., Alexander, L. V., Perkins, S. E., Smale, D. A., Straub, S. C., Oliver, E. C., Benthuysen, J. A., Burrows, M. T., Donat, M. G., Feng, M. & Holbrook, N. J. A hierarchical approach to defining marine heatwaves. Prog. Oceanogr. 141, 227–238 (2016). https://doi.org/10.1016/j.pocean.2015.12.014.
Mills, K. E., Pershing, A. J., Brown, C. J., Chen, Y., Chiang, F.-S., Holland, D. S., Lehuta, S., Nye, J. A., Sun, J. C., Thomas, A. C. & Wahle, R. A. Fisheries management in a changing climate: Lessons from the 2012 ocean heat wave in the Northwest Atlantic. Oceanography 26, 191–195 (2013). https://doi.org/10.5670/oceanog.2013.27.
Hughes, T., Kerry, J., Álvarez-Noriega, M. et al. Global warming and recurrent mass bleaching of corals. Nature 543, 373–377 (2017). https://doi.org/10.1038/nature21707.
Smale, D. A., Wernberg, T., Oliver, E. C. J. et al. Marine heatwaves threaten global biodiversity and the provision of ecosystem services. Nat. Clim. Chang. 9, 306–312 (2019). https://doi.org/10.1038/s41558-019-0412-1.
Choi, H. Y., Park, M. S., Kim, H. S. et al. Marine heatwave events strengthen the intensity of tropical cyclones. Commun Earth Environ 5, 69 (2024). https://doi.org/10.1038/s43247-024-01239-4.
Rathore, S., Goyal, R., Jangir, B., Ummenhofer, C. C., Feng, M. & Mishra, M. Interactions between a marine heatwave and tropical cyclone Amphan in the Bay of Bengal in 2020. Front. Clim. 4, 861477 (2022). https://doi.org/10.3389/fclim.2022.861477.
Pun, I. F., Hsu, H. H., Moon, I. J. et al. Marine heatwave as a supercharger for the strongest typhoon in the East China Sea. npj Clim Atmos Sci 6, 128 (2023). https://doi.org/10.1038/s41612-023-00449-5.
Dzwonkowski, B., Coogan, J., Fournier, S., Lockridge, G., Park, K. & Lee, T. Compounding impact of severe weather events fuels marine heatwave in the coastal ocean. Nat. Commun. 11, 4623 (2020). https://doi.org/10.1038/s41467-020-18339-2.
Emanuel, K. Will Global Warming Make Hurricane Forecasting More Difficult? Bull. Am. Meteorol. Soc. 98, 495–501 (2017). https://doi.org/10.1175/BAMS-D-16-0134.1.
Liu, M., Vecchi, G. A., Smith, J. A. et al. Causes of large projected increases in hurricane precipitation rates with global warming. npj Clim Atmos Sci 2, 38 (2019). https://doi.org/10.1038/s41612-019-0095-3.
Chu, J.-E., Lee, S.-S., Timmermann, A., Wengel, C., Stuecker, M. F. & Yamaguchi, R. Reduced tropical cyclone densities and ocean effects due to anthropogenic greenhouse warming. Sci. Adv. 6, eabd5109 (2020). https://doi.org/10.1126/sciadv.abd5109.
Knutson, T., Camargo, S. J., Chan, J. C. L., Emanuel, K., Ho, C.-H., Kossin, J., Mohapatra, M., Satoh, M., Sugi, M., Walsh, K. & Wu, L. Tropical Cyclones and Climate Change Assessment: Part II: Projected Response to Anthropogenic Warming. Bull. Am. Meteorol. Soc. 101, E303–E322 (2020). https://doi.org/10.1175/BAMS-D-18-0194.1.
Balaguru, K., Foltz, G. R., Leung, L. R., Xu, W., Kim, D., Lopez, H. & West, R. Increasing hurricane intensification rate near the US Atlantic coast. Geophys. Res. Lett. 49, e2022GL099793 (2022). https://doi.org/10.1029/2022GL099793.
Whitaker, W. D. Quantitative determination of heat transfer from sea to air during passage of hurricane Betsy. Master's thesis, Texas A&M University (1967). Available electronically from https://hdl.handle.net/1969.1/ETD-TAMU-1967-THESIS-W578.
Leipper, D. F. & Volgenau, D. Hurricane Heat Potential of the Gulf of Mexico. J. Phys. Oceanogr. 2, 218–224 (1972). https://doi.org/10.1175/1520-0485(1972)002<0218>2.0.CO;2
Wada, A., Usui, N., & Sato, K. Relationship of maximum tropical cyclone intensity to sea surface temperature and tropical cyclone heat potential in the North Pacific Ocean. J. Geophys. Res. 117, D11118 (2012). https://doi.org/10.1029/2012JD017583.
Jangir, B., Swain, D., & Ghose, S. K. Influence of eddies and tropical cyclone heat potential on intensity changes of tropical cyclones in the North Indian Ocean. Adv. Space Res. 68, 773–786 (2021). https://doi.org/10.1016/j.asr.2020.01.011.
Price, J. F. Metrics of hurricane-ocean interaction: Vertically-integrated or vertically-averaged ocean temperature? Ocean Sci. 5(3), 351–368 (2009). https://doi.org/10.5194/os-5-351-2009.
Balaguru, K., Foltz, G. R., Leung, L. R., Asaro, E. D., Emanuel, K. A., Liu, H., & Zedler, S. E. Dynamic Potential Intensity: An improved representation of the ocean's impact on tropical cyclones. Geophys. Res. Lett. 42, 6739–6746 (2015). https://doi.org/10.1002/2015GL064822.
Potter, H., DiMarco, S. F. & Knap, A. H. Tropical cyclone heat potential and the rapid intensification of Hurricane Harvey in the Texas Bight. J. Geophys. Res. Oceans 124, 2440–2451 (2019). https://doi.org/10.1029/2018JC014776.
Pun, I.-F., Chan, J. C. L., Lin, I.-I., Chan, K. T. F., Price, J. F., Ko, D. S., Lien, C.-C., Wu, Y.-L. & Huang, H.-C. Rapid Intensification of Typhoon Hato (2017) over Shallow Water. Sustainability 11, 3709 (2019). https://doi.org/10.3390/su11133709.
Dzwonkowski, B., Fournier, S., Lockridge, G., Coogan, J., Liu, Z. & Park, K. Hurricane Sally (2020) shifts the ocean thermal structure across the inner core during rapid intensification over the shelf. J. Phys. Oceanogr. 52, 2841–2852 (2022). https://doi.org/10.1175/JPO-D-22-0025.1.
Schaeffer, A. & Roughan, M. Subsurface intensification of marine heatwaves off southeastern Australia: The role of stratification and local winds. Geophys. Res. Lett. 44, 5025–5033 (2017). https://doi.org/10.1002/2017GL073714.
Darmaraki, S., Somot, S., Sevault, F. et al. Future evolution of Marine Heatwaves in the Mediterranean Sea. Clim. Dyn. 53, 1371–1392 (2019). https://doi.org/10.1007/s00382-019-04661-z.
Gawarkiewicz, G., Chen, K., Forsyth, J., Bahr, F., Mercer, A. M., Ellertson, A., Fratantoni, P., Seim, H., Haines, S., & Han, L. Characteristics of an advective marine heatwave in the Middle Atlantic Bight in early 2017. Front. Mar. Sci. 6, 712 (2019). https://doi.org/10.3389/fmars.2019.00712.
Chen, K., Gawarkiewicz, G. & Yang, J. Mesoscale and submesoscale shelf-ocean exchanges initialize an advective Marine Heatwave. J. Geophys. Res. Oceans 127, e2021JC017927 (2022). https://doi.org/10.1029/2021JC017927.
Amaya, D. J., Jacox, M. G., Alexander, M. A. et al. Bottom marine heatwaves along the continental shelves of North America. Nat. Commun. 14, 1038 (2023). https://doi.org/10.1038/s41467-023-36567-0.
Pearce, A. F. & Feng, M. The rise and fall of the “marine heat wave” off Western Australia during the summer of 2010/2011. J. Mar. Syst. 111, 139–156 (2013). https://doi.org/10.1016/j.jmarsys.2012.10.009.
Elzahaby, Y. & Schaeffer, A. Observational insight into the subsurface anomalies of marine heatwaves. Front. Mar. Sci. 6, 745 (2019). https://doi.org/10.3389/fmars.2019.00745.
Li, J., Roughan, M. & Hemming, M. Interactions between cold cyclonic eddies and a western boundary current modulate marine heatwaves. Commun. Earth Environ. 4, 380 (2023). https://doi.org/10.1038/s43247-023-01041-8.
Schaeffer, A., Sen Gupta, A. & Roughan, M. Seasonal stratification and complex local dynamics control the sub-surface structure of marine heatwaves in Eastern Australian coastal waters. Commun. Earth Environ. 4, 304 (2023). https://doi.org/10.1038/s43247-023-00966-4.
Balaguru, K., Foltz, G. R., Leung, L. R., Kaplan, J., Xu, W., Reul, N. & Chapron, B. Pronounced impact of salinity on rapidly intensifying tropical cyclones. Bull. Am. Meteorol. Soc. 101, E1497–E1511 (2020). https://doi.org/10.1175/BAMS-D-19-0303.1.
Fan, K., Wang, X., & He, Z. Control of salinity stratification on recent increase in tropical cyclone intensification rates over the postmonsoon Bay of Bengal. Environ. Res. Lett. 15(9), 094028 (2020). https://doi.org/10.1088/1748-9326/ab9690.
Beven J. II, Hagan A., Berg R. Tropical cyclone report: Hurricane Ida (National Hurricane Center). (2022). Available at: https://www.nhc.noaa.gov/data/tcr/AL092021_Ida.pdf (Accessed January 6, 2024).
Miles, T. N., Coakley, S. J., Engdahl, J. M., Rudzin, J. E., Tsei, S., & Glenn, S. M. Ocean mixing during Hurricane Ida (2021): the impact of a freshwater barrier layer. Front. Mar. Sci. 10, 1224609 (2023). https://doi.org/10.3389/fmars.2023.1224609.
Cangialosi, J. P. & Alaka, L. Tropical cyclone report: Hurricane Idalia (National Hurricane Center), (2024). Available at: https://www.nhc.noaa.gov/data/tcr/AL102023_Idalia.pdf (Accessed March 10, 2024).
Kimball, S. K., & Mulekar, M. S. A 15-Year Climatology of North Atlantic Tropical Cyclones. Part I: Size Parameters. J. Climate 17, 3555–3575 (2004). https://doi.org/10.1175/1520-0442(2004)017<3555>2.0.CO;2.
Tropical Cyclone Climatology. Available at: . Accessed 2 July 2024.
Walker, N. D., Wiseman, W. J., Rouse Jr., L. J., & Babin, A. Effects of river discharge, wind stress, and slope eddies on circulation and the satellite-observed structure of the Mississippi River Plume. J. Coast. Res. 21(6), 1228–1244 (2005). http://www.jstor.org/stable/4299530.
Androulidakis, Y. S., & Kourafalou, V. H. On the processes that influence the transport and fate of Mississippi waters under flooding outflow conditions. Ocean Dyn. 63, 143–164 (2013). https://doi.org/10.1007/s10236-012-0587-8.
Dzwonkowski, B., Fournier, S., Park, K., Dykstra, S. L., & Reager, J. T. Water column stability and the role of velocity shear on a seasonally stratified shelf, Mississippi Bight, northern Gulf of Mexico. J. Geophys. Res. Oceans 123, 5777–5796 (2018). https://doi.org/10.1029/2017JC013624.
Feng, Y.-T., Bethel, B. J., Tian, Y., et al. Marine heatwaves in the Gulf of Mexico 1983–2021: Statistics, recent intensifications, and threats on coral reefs. Adv. Clim. Change Res. 14, 560–572 (2023). https://doi.org/10.1016/j.accre.2023.08.006.
Oliver, E. C. J., Benthuysen, J. A., Darmaraki, S., Donat, M. G., Hobday, A. J., Holbrook, N. J., Schlegel, R. W., & Sen Gupta, A. Marine heatwaves. Annu. Rev. Mar. Sci. 13, 313–342 (2021). https://doi.org/10.1146/annurev-marine-032720-095144.
Zhang, Y., Du, Y., Feng, M., & Hobday, A. J. Vertical structures of marine heatwaves. Nat. Commun. 14, 6483 (2023). https://doi.org/10.1038/s41467-023-42219-0.
Hu, L. A global assessment of coastal marine heatwaves and their relation with coastal urban thermal changes. Geophys. Res. Lett. 48, e2021GL093260 (2021). https://doi.org/10.1029/2021GL093260.
Holbrook, N. J., Scannell, H. A., Sen Gupta, A., Benthuysen, J. A., Feng, M., Oliver, E. C. J., Alexander, L. V., et al. A global assessment of marine heatwaves and their drivers. Nat. Commun. 10, 2624 (2019). https://doi.org/10.1038/s41467-019-10206-z.
Sen Gupta, A., Thomsen, M., Benthuysen, J. A., et al. Drivers and impacts of the most extreme marine heatwave events. Sci. Rep. 10, 19359 (2020). https://doi.org/10.1038/s41598-020-75445-3.
Großelindemann, H., Ryan, S., Ummenhofer, C. C., Martin, T., & Biastoch, A. Marine heatwaves and their depth structures on the northeast US continental shelf. Front. Clim. 4, 857937 (2022). https://doi.org/10.3389/fclim.2022.857937.
Benthuysen, J. A., Oliver, E. C. J., Feng, M., & Marshall, A. G. Extreme marine warming across tropical Australia during austral summer 2015–2016. J. Geophys. Res. Oceans 123, 1301–1326 (2018). https://doi.org/10.1002/2017JC013326.
Gröger, M., Dutheil, C., Börgel, F., & Meier, M. H. E. Drivers of marine heatwaves in a stratified marginal sea. Clim. Dyn. (2024): 1–13. https://doi.org/10.1007/s00382-023-07062-5.
Benthuysen, J., Feng, M., & Zhong, L. Spatial patterns of warming off Western Australia during the 2011 Ningaloo Niño: Quantifying impacts of remote and local forcing. Cont. Shelf Res. 91, 232–246 (2014). https://doi.org/10.1016/j.csr.2014.09.014.
Liu, H., Nie, X., Cui, C., & Wei, Z. Compound marine heatwaves and low sea surface salinity extremes over the tropical Pacific Ocean. Environ. Res. Lett. 18, 064001 (2023). https://doi.org/10.1088/1748-9326/acd0c4.
Kumar, S., Chakraborty, A., Chandrakar, R., Kumar, A., Sadhukhan, B., & Roy Chowdhury, R. Analysis of marine heatwaves over the Bay of Bengal during 1982–2021. Sci. Rep., 13(1), 14235. (2023). https://doi.org/10.1038/s41598-023-13358-z.
Li, Y., Tang, Y., Wang, S., Toumi, R., Song, X., & Wang, Q. Recent increases in tropical cyclone rapid intensification events in global offshore regions. Nat. Commun., 14(1), 5167. (2023). https://doi.org/10.1038/s41467-023-40605-2
Li, Y., Tang, Y., Wang, S., Toumi, R., Song, X., & Wang, Q. Recent increases in tropical cyclone rapid intensification events in global offshore regions. Nat. Commun., 14(1), 5167. (2023). https://doi.org/10.1038/s41467-023-40605-2.
Palmen, E. H. On the formation and structure of tropical cyclones. Geophysics, Univ. of Helsinki, Vol. 3, 1948, pp. 26–38. (1948). https://www.geophysica.fi/pdf/geophysica_1948_3_1_026_palmen.pdf
Gray, W. M. Global view of the origin of tropical disturbances and storms. Mon. Wea. Rev., 96, 669–700. (1968). https://doi.org/10.1175/1520-0493(1968)096<0669:GVOTOO>2.0.CO;2
NOAA National Centers for Environmental Information, Monthly Global Climate Report for Annual 2019, published online January 2020, retrieved on May 13, 2024 from https://www.ncei.noaa.gov/access/monitoring/monthly-report/global/201913.
Banzon, V., Smith, T. M., Chin, T. M., Liu, C., & Hankins, W. A long-term record of blended satellite and in situ sea-surface temperature for climate monitoring, modeling and environmental studies. Earth System Science Data 8, 165–176 (2016). https://doi.org/10.5194/essd-8-165-2016.
Huang, B., Liu, C., Banzon, V., Freeman, E., Graham, G., Hankins, B., Smith, T., & Zhang, H. Improvements of the Daily Optimum Interpolation Sea Surface Temperature (DOISST) Version 2.1. J. Climate 34, 2923–2939 (2021). https://doi.org/10.1175/JCLI-D-20-0166.1.
Chen, K., Gawarkiewicz, G., Kwon, Y.-O., & Zhang, W. G. The role of atmospheric forcing versus ocean advection during the extreme warming of the Northeast U.S. continental shelf in 2012. J. Geophys. Res. Oceans 120, 4324–4339 (2015). https://doi.org/10.1002/2014JC010547.
Collins, D. A., Della-Marta, P. M., Plummer, N., & Trewin, B. C. Trends in annual frequencies of extreme temperature events in Australia. Australian Meteorological Magazine 49, 277–292 (2000).
Pezza, A. B., van Rensch, P., & Cai, W. Severe heat waves in Southern Australia: synoptic climatology and large scale connections. Clim Dyn 38, 209–224 (2012). https://doi.org/10.1007/s00382-011-1016-2.
Perkins, S. E., & Alexander, L. V. On the Measurement of Heat Waves. J. Climate 26(13), 4500–4517 (2013). https://doi.org/10.1175/JCLI-D-12-00383.1.
Amaya, D. J., Jacox, M. G., Fewings, M. R., Saba, V. S., Stuecker, M. F., Rykaczewski, R. R., Ross, A. C., Stock, C. A., Capotondi, A., Petrik, C. M., Bograd, S. J., Alexander, M. A., Cheng, W., Hermann, A. J., Kearney, K. A., & Powell, B. S. Marine heatwaves need clear definitions so coastal communities can adapt. Nature 616(7955), 29–32 (2023). https://doi.org/10.1038/d41586-023-00924-2.
Sen Gupta, A. Marine heatwaves: Definition duel heats up. Nature 617(7961), 465 (2023). https://doi.org/10.1038/d41586-023-01619-4.
Jacox, M. G., Alexander, M. A., Amaya, D., et al. Global seasonal forecasts of marine heatwaves. Nature 604, 486–490 (2022). https://doi.org/10.1038/s41586-022-04573-9.
Shay, L. K., Goni, G. J., & Black, P. G. Effects of a warm oceanic feature on Hurricane Opal. Mon. Wea. Rev. 128(5), 1366–1383 (2000). https://doi.org/10.1175/1520-0493(2000)128<1366>2.0.CO;2.
Lin, I. I., Goni, G. J., Knaff, J. A., et al. Ocean heat content for tropical cyclone intensity forecasting and its impact on storm surge. Nat Hazards 66, 1481–1500 (2013). https://doi.org/10.1007/s11069-012-0214-5.
Alduchov, O. A., & Eskridge, R. E. Improved Magnus form approximation of saturation vapor pressure. J. Appl. Meteor. Climatol. 35(4), 601–609 (1996). https://doi.org/10.1175/1520-0450(1996)035<0601>2.0.CO;2.
Coogan, J., & Dzwonkowski, B. Observations of wind forcing effects on estuary length and salinity flux in a river-dominated, microtidal estuary, Mobile Bay, Alabama. J. Phys. Oceanogr. 48(8), 1787–1802 (2018). https://doi.org/10.1175/JPO-D-17-0249.1.

There is NO Competing Interest.

SupplementaryInformationRaoetal2024.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Role of tropical cyclones and salinity stratification in the expansion of a coastal marine heatwave

Status:

Version 1

Abstract

Figures

1. Introduction

2. Results

2.1. Marine heatwave conditions during peak hurricane season

2.2. Evolution of the marine heatwave event

2.3. Evolution of thermal structure across the continental shelf

2.4. Mechanisms driving the onset of the marine heatwave event

2.4.1. Offshore intrusion

2.5. Inner shelf expansion of the marine heatwave to depth

2.6. Mechanisms driving the decay of marine heatwave.

3. Discussion

3.1. Marine heatwave continuum

3.2. Driving mechanism for the marine heatwave

3.3. Influence of river discharge in marine heatwave evolution

3.4. Interaction with tropical storms

4. Conclusion

5. Data and Methods

5.1. In situ and satellite datasets

5.2. Reanalysis datasets

5.3. Tropical Cyclone Data

5.4. Methods

Declarations

Competing Interests

Author Contributions

Acknowledgements

References

Additional Declarations

Supplementary Files

Status:

Version 1