Cold season Arctic strong cyclones enhance Atlantification of the Arctic Ocean

In recent years, as the Arctic Ocean’s warming trend has accelerated, there has been increasing attention on the process of Atlantification in the Arctic Ocean. This study focused on the Arctic Atlantic inflow zone (AAZ) as its research area. Multi-source reanalysis data and in-situ Argo float data were utilized to detect Arctic strong cyclones (ASCs) in the AAZ and analyze the resulting changes in the upper ocean. The findings reveal that during the cold season (October to March), influenced by ASCs’ intensity, frequency, tracks, and the concurrent weakening of ocean stratification, these cyclones can disrupt the cold halocline layer through mechanisms such as mixing and Ekman pumping. This process facilitates the transport of heat from the deep, warm and saline Atlantic Water within the ocean to the subsurface layers. Concurrently, ASCs during the cold season can enhance the process of Atlantification in the Arctic Ocean by intensifying the intrusion of the Barents Sea Branch. Additionally, the attenuation of oceanic stratification during ASCs is primarily driven by changes in salinity, particularly above the 100 m.


Introduction
The Arctic is one of the most crucial ecological and physical systems on Earth, controlling global climate and ocean circulation.The Arctic Ocean lies entirely within the Arctic Circle and is the slowest expanding ocean in the world (Menard and Smith 1966).The Arctic continental shelf area is nearly half the size of the Arctic Basin (Jakobsson 2002).The circulation in the Arctic Ocean is primarily driven by the Norwegian Atlantic Current, Barents Sea Branch, West Spitsbergen Current, East Greenland Current, TransPolar Drift Current, and Beaufort Gyre (figure 1) (Timmermans and Marshall 2020).The Arctic Ocean is covered by sea ice year-round, historically covering approximately half of the Arctic Ocean (Stroeve et al 2007).However, in recent years, accelerated Arctic warming and increased ice melt have resulted in rising sea levels, extreme weather events, and a range of other impacts (Nghiem et al 2006, Previdi et al 2021, Simmonds and Li 2021, Lim et al 2022, Rantanen et al 2022, Sumata et al 2023).
At present, there is a significant focus on climate change in the Arctic Ocean, specifically, the phenomenon of Arctic Atlantification.Arctic Atlantification is primarily characterized by reduced ice cover in the eastern Eurasian Basin, a weakening of the halocline, and an increased winter ventilation within the ocean, which renders this region structurally similar to the western Eurasian Basin (Polyakov et al 2017, Docquier andKoenigk 2021).In the Eurasian Arctic Ocean, seawater temperature increases with depth.The top of the ocean is usually covered by sea ice, followed by a relatively fresh surface mixed layer (SML) and a thick cold halocline layer (CHL), beneath which lies the warm and salty Atlantic Water (Timmermans et al 2003, Polyakov et al 2013, Timmermans and Marshall 2020).The presence of the CHL serves to protect the sea ice from melting due to the influence of Atlantic Water.However, this underlying heat from the Atlantic Ocean's warm water is not permanently trapped.On one hand, CHL has weakened in recent years, leading to increased winter ventilation and reduced winter ice formation (Polyakov et al 2017, Li and Fedorov 2021, Tesi et al 2021).On the other hand, the total area covered by sea ice in the Barents Sea has decreased by nearly half in recent decades, and most of the sea ice currently found in the Barents Sea is blown in from the central Arctic (Onarheim et al 2015, Stroeve and Notz 2018, Smedsrud et al 2022, Heukamp et al 2023, Rieke et al 2023) The melting of sea ice may trigger changes in sea stratification, and when sea ice melts, it replenishes the freshwater layer above the CHL.With reduced sea ice, there is less freshwater, resulting in weakened stratification between ocean layers.This leads to greater ocean mixing, drawing more Atlantic heat to the surface, and this process of 'Atlantification' in turn leads to more ice melting from the bottom (Årthun et al 2012, Smedsrud et al 2013, Carmack et al 2015, Lind et al 2018).Studies based on the CMIP6 model suggest that Arctic Ocean temperatures will significantly increase over the next 100 years, especially in the Barents Sea and Kara Sea, with projected temperature increases of approximately 5.17 • C and 4.4 • C, respectively (Shu et al 2022).
Winter storms can accelerate the loss of sea ice in the Arctic, and strong wind fields have an important impact on the Atlantic Water transport in the Arctic Ocean, further intensifying the process of Atlantification (Polyakov et al 2013, Graham et al 2019, Heukamp et al 2023).However, there is limited research on these phenomena, and the Arctic hosts a highly active weather system known as Arctic cyclones, with lifecycles ranging from a few hours to several days (Keegan 1958, Serreze 1995, Simmonds and Li 2021, Valkonen et al 2021).The centers of Arctic cyclones are often associated with strong upward motion, leading to the formation of stormy weather conditions that significantly impact polar weather and climate (Lang andWaugh 2011, Tanaka et al 2012).In recent years, the Arctic has experienced more intense and long-lasting cyclonic processes, and Arctic strong cyclones (ASCs) are occurring more frequently, further exacerbating the drastic changes in the Arctic Ocean environment (Simmonds and Rudeva 2012, 2014, Rinke et al 2017).Compared to regular Arctic cyclones, ASCs have higher wind speeds and lifecycles that cause upwelling, vertical mixing, and entrainment processes, which more easily break up the stronger CHL and promote heat exchange.
Clearly, the impact of ASCs on Arctic Atlantification holds significant research value, as it plays a crucial role in the reduction of Arctic sea ice and the global climate system.Therefore, this study focuses on the Arctic Atlantic inflow zone (AAZ) as the primary research area (22 • E-125 • E, 70 • N-86 • N) and examines the changes in ocean heat and stratification during ASCs from 1993 to 2020, along with their driving factors.

Data
In this study, the following observational/reanalysis products were utilized: (1) The European Center for Medium-range Weather Forecasts ERA5 Reanalysis provided hourly mean sea level pressure and sea surface wind data for the period 1993-2020, which were used for ASCs detection (www.ecmwf.int/en/forecasts/dataset/ecmwf-reanalysis-v5). ( 2) Daily sea ice concentration (SIC), sea ice thickness (SIT), sea surface current, SML, temperature, and salinity data from 1993 to 2020 were obtained from the Copernicus Marine Environment Monitoring Service (https://doi.org/10.48670/moi-00168, https://doi.org/10.48670/moi-00024,https://doi.org/10.48670/moi-00021).( 3) In-situ data were obtained from the Argo program, with Argo float number 6903564 (www.argo.ucsd.edu).This float was located along the ASC tracks near section K and captured temperature and salinity data before and after the ASC passage on 18 November and 21 November 2020.While we obtained in-situ data from the Argo program, the scarcity of Arctic observational data is a prominent challenge when working with limited Arctic observations.Although the data collected from this float offer valuable insights into local conditions along the ASC tracks near section K, it is essential to recognize the inherent limitations of relying on sparse hydrographic data in the Arctic region.Utilizing a single float may not capture the full spatial and temporal variability of oceanic properties in the Arctic.

Arctic cyclone identification and tracking:
The previously described algorithm has been primarily referenced and slightly modified (Zhang et al 2004, Crawford andSerreze 2016): (1) a grid point is designated as a cyclone candidate if the sea level pressure at that point is lower than that of its surrounding eight grid points.(2) For each minimum, a circle with a 1000 km radius is drawn around it, and the average sea level pressure difference between the minimum point and each grid cell intersecting the circle is calculated.If a minimum exhibits an average sea level pressure difference less than 7.5 hPa, it is considered too weak to be classified as a cyclone of interest and is discarded.In cases where multiple cyclone candidates appear within a 1000 km radius simultaneously, they are merged into a single cyclone track.(3) To avoid pseudo-cyclones, the ratio of start-to-end distance to track distance is calculated to determine the cyclone track, which must be greater than or equal to 0.6.(4) A search radius of 450 km is used, corresponding to a maximum allowable propagation speed of 150 km per hour.Beyond this range, a cyclone is regarded as new.If the lifetime of a candidate is less than 12 h, it is removed from the cyclone candidates.
For ASCs determination, in order to include more ASCs within the AAZ, a reference threshold of 972.3 hPa is employed (Liu and He 2023).Arctic cyclones that fall below this threshold during both the cold season (October to March) and the warm season (April to September) are classified as ASCs.
Calculation of relevant physical parameters: The upper ocean heat content (OHC) of the AAZ is calculated by the following formula (Polyakov et al 2013): where T is seawater temperature, T freezing is the freezing temperature, ρ s is seawater density, c p specific heat capacity of seawater under pressure, and z 1 and z 2 are the upper and lower boundary depths.
Ekman pumping velocity (EPV) caused by wind field is an important index for measuring vertical motion of the upper ocean (Kara et al 2007, Gaube et al 2013, Polyakov et al 2017).Ekman layer depth (DE) measures the ability of a cyclone to bring deep water to the surface.EPV can be calculated according to the wind vector using equations ( 2) and (3), while the calculation method for DE is referenced in equation ( 4) (Chen and Tang 2012), Here τ is the wind stress, which is calculated using air density close to the sea surface ρ a , speeddependent drag coefficient for neutrally stable conditions C D and wind speed at 10 m above sea level u.Additionally, SIC is taken into account.The EPV was calculated using equation (3), where ∇ × τ is the curl of the surface stress, f = 2ω sin θ is the Coriolis parameter for latitude θ and Earth rotation rate ω.
The DE can be calculated according to equation (4) as follows: where, V max represents the maximum sea surface wind speed at 10 m.The potential vorticity (PV) and the Brunt-Vaisala frequency squared (N 2 ) can be calculated as equations ( 5) and ( 6).They can reflect the movement of seawater and the stratification properties of the upper ocean (Delman et al 2015, Polyakov et al 2017), where, ζ is the relative vorticity, and g is gravitational acceleration.

Long-term trend changes in the marine environment of the AAZ
With the rapid warming of the Arctic in recent years, the marine environment has also changed dramatically, especially in AAZ.From 1993 to 2020, both SIC and SIT have decreased significantly in the AAZ, although the rate of sea ice loss has slowed since 2007 (Polyakov et al 2023).However, an interesting phenomenon was observed that a downward trend in SIC reaches its maximum during the warm season at −0.063 decade −1 , while a downward trend in SIT reaches its maximum during the cold season at −0.253 m decade −1 (table 1).This may be related to the higher OHC below the subsurface of the AAZ, which reduced ice growth in the cold season.Figure 2 effectively supports this hypothesis, showing a continuous increase in OHC in recent years, with the maximum growth trend observed at the subsurface 100 m, reaching 5.33 × 10 6 J m −2 yr −1 (figure 2(a)).
In recent years, both the SML depth and CHL base depth have significantly deepened.However, since 2013, there has been an upward trend in the CHL base depth.This results in a reduction in CHL thickness, implying a weakening of the vertical gradients of seawater salinity and density (figure 2).Weakened ocean stratification favors vertical mixing and the upward transport of heat carried by Atlantic Water.Furthermore, figures 2(b) and (d) show obvious seasonal variations in OHC, SML and CHL.During the cold season, the SML and CHL are deeper, whereas SML is shallower during the warm season, resulting a thicker ocean stratification that hinders upward heat transport in the warm season.In contrast, during the cold season, N 2 tends to weaken, further enhancing upward heat transport.Overall, there is higher OHC during the cold season in the AAZ, which accelerates the reduction of SIT.
The increase in OHC during the cold season is also likely influenced by atmospheric environment.In the cold season, AAZ experiences more active ASCs (figure 3), which are prone to disrupting the already weakened ocean stratification, facilitating the upward transport of heat from the deep ocean layers.Consequently, we conducted the following research.

Cold season ASCs enhance the OHC in the AAZ through mixing
To quantify the impact of ASCs on the upper ocean, we calculated the average physical parameters of the strongest position of each ASC in the AAZ using a 2 • × 2 • grid, a method enabling the assessment of the extent of ASCs' impact on the upper ocean.Due to the influence of atmospheric circulation such as the Arctic Oscillation and the Iceland Low, the cold season in the AAZ is generally susceptible to generate more and stronger cyclones (Serreze 1995, Zhang et al 2004).Our findings indicate that between 1993 and 2020, a total of 188 ASCs were detected.During the cold season, there were 153 ASCs with an average intensity of 965.06 hPa, average sea surface wind speed of 9.53 m s −1 , and average DE of 69 m.In the warm season, there were 35 ASCs with an average intensity of 970.82 hPa, average sea surface wind speed of 9.03 m s −1 , and average DE of 65 m (figure 3).The upper ocean mixing caused by ASCs is significantly stronger during the cold season compared to the warm season.The DE in the cold season far exceeds the CHL base depth and can reach a maximum average EPV of 8.9 × 10 −6 m s −1 , while DE in the warm season slightly exceeds or is near the CHL base depth, with a maximum average EPV of 8 × 10 −6 m s −1 (figures 2(d) and 3). Figure 4 shows the environmental changes in the upper ocean of AAZ before and after the passage of ASCs, with the background climate trend removed.During the main impact period of ASCs in the warm season, the OHC, temperature, and salinity at 100 m significantly decreased by −1.05 × 10 −6 J m −2 , −0.0027 • C, and −0.0036 psu, respectively.In contrast, during the cold season, they exhibited an increasing trend, with the average OHC increasing by as much as 2.1 × 10 5 J m −2 .Both SIC and SIT significantly decreased during the passage of ASCs in both seasons.Additionally, although the SML deepened significantly during the warm season ASCs (figure 4(d)), the overall ocean stratification remained thick during ASCs due to the shallow SML base depth during the warm season (figure 2(d)), which hindered ocean mixing.Therefore, influenced by ASCs' frequency, intensity, and basic ocean stratification environment, ASCs during the cold season resulted in stronger ocean mixing and higher OHC in the AAZ.

Cold season ASCs enhance the Arctic Ocean's Atlantification by promoting the intrusion of Atlantic water
Barents Sea Branch is one of the main branches through which Atlantic Water flows into the AAZ (figure 1) (Smedsrud et al 2022), and it predominantly carries Atlantic Water on its western side.In recent years, Atlantic Water has been steadily advancing eastward, altering the seawater physical properties on the eastern side of Barents Sea Branch.ASCs as a strong storm have strong mechanical energy, are likely to accelerate Atlantic Water intrusion through their movement.To better reveal the phenomenon and mechanisms of Atlantic Water intrusion under the influence of ASCs, section K ( 22• E-48 • E, 73.5 • N) located on the path of Barents Sea Branch was selected for main research.Section K effectively characterizes the contrasting physical properties and intrusion processes of seawater on both the eastern and western sides of Barents Sea Branch.From 1993 to 2020, a total of 52 ASCs occurred near section K, with 45 in the cold season and 7 in the warm season.ASCs primarily moved in a northeast direction, intensifying as they progressed eastward (figures 5(a) and (b)).Due to the infrequent occurrence of ASCs near section K during the warm season, making them non-representative, and previous analyses highlighting the dominant influence of cold season ASCs on AAZ's OHC, our research centered on the impact of cold season ASCs on section K.
During the cold season, ASCs strengthened the eastward intrusion of Barents Sea Branch.While the passage of ASCs led to some heat loss at the sea surface, it notably increased the OHC in the oceanic sub-surface layer at section K, especially east of 35 • E where significant eastward heat transport was observed (figures 5(c)-(e)).ASCs enhanced the PV at Barents Sea Branch during ASCs periods.The location of Barents Sea Branch exhibited a notable increase in PV and affected the upper ocean from the oceanic sub-surface upwards.Compared to the PV during the cold season average, the eastern side of Barents Sea Branch experienced a substantial PV increase in the upper ocean, peaking at 4 × 10 −9 s −1 m −1 (figures 5(f)-(h)).Atlantic Water clearly intruded further eastward.The mixing and intrusion driven by ASCs disrupted the stable oceanic stratification, significantly reducing density and N 2 below the sub-surface layer at section K (figures 5(i)-(n)).

Validation based on measured in-situ Argo float
In the 2020 cold season, there was an ASC that entered the research area on November 18, passing through section K on 19 November.Argo float platform number 6903564 captured the oceanic stratification changes before and after the ASC affected section K on 18 November and 21 November, respectively.When the ASC passed through section K, it induced strong upwelling, with Ekman upwelling reaching a maximum of 5.09 × 10 −5 m s −1 on 19 November.This upwelling likely facilitated subsurface warm water rise and intensified sea surface heat loss.After the ASC passage on 21 November, Ekman upwelling significantly weakened and was replaced by Ekman downwelling (figures 6(a)-(c)).This was corroborated by the In-situ Argo buoy near the ASC.After the ASC passage on 21 November, the sea surface temperature was significantly reduced by 0.22 • C, while the subsurface ocean temperature increased by approximately 0.8 • C. Salinity in the 0-200 m depth range primarily exhibited an increasing trend.On November 21, the Brunt-Vaisala frequency increased due to the influence of Ekman downwelling at the float location (figures 6(d)-(f)).
The passage of the ASC intensified the eastward transport of subsurface OHC in section K (figures 7(a)-(c)).Since the ASC primarily affected the area west of 33 • E in section K, on November 18, the PV displayed a significant positive-negative vorticity gradient west of 33 • E, causing substantial hindrance to the eastward Atlantic Water flow.However, after the ASC passage on 21 November, the vorticity gradient west of 33 • E notably decreased, evident in the increased negative PV and decreased positive PV.This reduction reduced the resistance to Atlantic Water flow eastward (figures 7(d)-(f)).Following the ASC passage, seawater density below 100 m significantly decreased, while surface water density notably increased (figures 7(g)-(i)).This change was associated with the decrease in sea surface temperature and the increase in salinity (figure 6).Furthermore, the oceanic stratification west of 33 • E in section K significantly weakened under the influence of the ASC (figures 7(j)-(l)).

The independent effects of salinity and temperature on ocean stratification during ASCs
Although the OHC is mainly affected by temperature, breaking through the thick CHL is essential for transferring heat from the deep ocean to above the subsurface layer.Ocean stratification stability is mainly determined by the Brunt-Vaisala frequency and seawater density, which are primarily influenced by temperature and salinity fields.So, during ASCs, did the weakening of ocean stratification still result mainly from temperature?To quantify the individual contributions of temperature and salinity trends to stratification in the study area, the relative contributions of temperature and salinity were separated for  In summary, although temperature plays a primary role in OHC, salinity appears to have a more significant impact on ocean stratification, especially above 100 m.

Conclusions
In recent years, with the accelerated warming and melting of sea ice in the Arctic, a new phenomenon has emerged known as the Atlantification of the Arctic Ocean.Specifically, the AAZ is particularly susceptible to the influence of atmospheric circulation patterns such as the Arctic Oscillation, North Atlantic Oscillation and the Iceland Low, leading to an increased prevalence of stronger cyclones (Alexeev et al 2017, Luo et al 2017).Since 2007, the atmospheric Arctic Dipole has predominantly been in a positive phase, further contributing to the progressive weakening of ocean stratification in the AAZ, making it more susceptible to the influence of ASCs (Polyakov et al 2023).Unlike long-term climate changes, this study focused on short-lived but intense weather systems known as ASCs and investigated their impact on the Atlantification of the Arctic Ocean.The research primarily focused on the AAZ as the key study area and detected ASCs from 1993 to 2020.Multiple reanalysis datasets and in-situ Argo float data were used to analyze the changes in the upper ocean under the influence of ASCs.
The cold season with more and stronger ASCs and weaker ocean stratification results in more pronounced ocean mixing.In the cold season, the average DE caused by ASCs was 69 m, and the average EPV was 8.9 × 10 −6 m s −1 (figure 3).This allowed ASCs to affect the depth below the CHL base depth, facilitating the upward transfer of heat due to Ekman pumping and mixing effects.This enhanced the OHC of the subsurface layer in the AAZ, with an average increase of up to 2.1 × 10 5 J m −2 (figure 4).Moreover, during the cold season, a significant portion of ASCs moved from west to northeast (figures 5(a) and (b)).This eastward movement promoted the intrusion of the Barents Sea Branch, further contributing to the increase in subsurface OHC in the AAZ.During ASCs in the cold season, the subsurface OHC east of section K at 35 • E significantly increased, accompanied by a noticeable increase in PV and a significant decrease in density and N 2 (figures 5(i)-(n)).In-situ Argo float data also confirmed an increase in subsurface temperature and salinity after the passage of ASC (figure 6).Additionally, although OHC is primarily influenced by temperature, the weakening of ocean stratification caused by the disruption of the thick CHL by ASCs in the cold season is mainly driven by salinity changes, especially above 100 m (figure 8).
In conclusion, ASCs during the cold season can significantly enhance the OHC in the subsurface layer of the AAZ through mixing and intrusion effects, thus accelerating the Atlantification process of the Arctic Ocean.

Figure 1 .
Figure 1.Schematic diagram of Arctic Ocean topography and major ocean surface circulation.Major ocean surface circulations are indicated by red arrows, NAC: Norwegian Atlantic Current, BSB: Barents Sea Branch, WSC: West Spitsbergen Current, EGC: East Greenland Current, TPD: TransPolar Drift Current, BG: Beaufort Gyre.The black line represents section K (position: 22-48 • E, 73.5 • N), which is located on Barents Sea Branch intruding into the Arctic Ocean.The location is also known for its high frequency of ASCs.
All trends have passed a significance test at the 1% level.

Figure 2 .
Figure 2. Interannual variation and climatology of OHC and ocean stratification in the AAZ from 1993 to 2020.Figures (a) and (b) represent the interannual variations and climatology of OHC during this period, while (c) and (d) represent the interannual variations and climatology of N 2 .The black dashed line and solid black line indicate SML and CHL respectively.The CHL base depth is defined as the depth at which the ratio of the density gradient caused by temperature to the density gradient caused by salinity equals 0.05 (R = (α∂θ/∂z) / (β∂S/∂z) = 0.05, where α is thermal expansion coefficient, β is the haline contraction coefficient).

Figure 3 .
Figure 3.The distribution of ASCs in the AAZ from 1993 to 2020.Figures (a) and (b) represent the frequency distribution of ASCs during the cold and warm seasons, respectively.Figures (c)-(e) display the average sea surface wind, DE and EPV when ASCs reach their maximum intensity in the cold and warm seasons, respectively.The '0' on the x-axis represents the time when ASCs pass through that location.

Figure 4 .
Figure 4.The average marine environment changes during the passage of ASCs in the AAZ from 1993 to 2020.Figures (a)-(f) represent the results after removing the background trend.Figures (a)-(c) reflect the changes at a depth of 100 m.Blue and yellow ribbons are the standard errors in cold and warm seasons.The '0' on the x-axis represents the time when ASCs pass through that location.Between the two gray bars represents the main impact period of ASCs.

Figure 5 .
Figure 5. Enhanced Atlantic Water Intrusion during cold season ASCs.(a) ASC tracks near section K during the cold season from 1993 to 2020.(b) ASC tracks near section K during the warm season from 1993 to 2020.Red dots indicate the location of maximum ASC intensity, while blue dots indicate the endpoints of ASCs.(c)-(e) Variations in OHC at section K, (f)-(h) variations in PV at section K, (i)-(k) Variations in density at section K, (l)-(n) variations in N 2 at section K.Among these, (b), (f), (i), and (l) are during ASCs periods; (d), (g), (j), and (m) are during multi-year cold season averages; (e), (h), (k), and (n) are the differences between ASCs periods and multi-year cold season averages.

Figure 6 .
Figure 6.Strong mixing during ASC.(a) EPV before ASCs affected section K and the nearby Argo float, (b) EPV during ASC passing section K and the nearby Argo float, (c) EPV after ASCs passed section K and the nearby Argo float.Positive (negative) values indicated upwelling (downwelling).(d)-(e) Changes in observed temperature, salinity, and Brunt-Vaisala frequency at the Argo float before and after ASC passage.Blue lines represent ASC tracks, and the red diamond indicates the position of the Argo float.

Figure 7 .
Figure 7. Enhanced Atlantic water intrusion during ASC.Figures (a)-(c) represent the changes in OHC at section K before ASC passage, after ASC passage, and the difference between them.Figures (d)-(f) represent the changes in PV at section K before ASC passage, after ASC passage, and the difference between them.Figures (g)-(i) represent the changes in density at section K before ASC passage, after ASC passage, and the difference between them.Figures (j)-(l) represent the changes in N 2 at section K before ASC passage, after ASC passage, and the difference between them.

Figure 8 .
Figure 8. Diagnosed effects of temperature and salinity changes during ASCs on seawater density and N 2 .Figure (a) represents the impact of diagnosed temperature changes on density when salinity changes during ASCs were ignored.Figure (b) represents the effect of diagnosed salinity changes on density when temperature changes during ASCs were ignored.Figure (c) depicts the impact of diagnosed temperature changes on N 2 when salinity changes during ASCs were ignored.Figure (d) depicts the impact of diagnosed salinity changes on N 2 when temperature changes during ASCs were ignored.

Table 1 .
Trends of SIC and SIT in the AAZ from 1993 to 2020.

Table 2 .
Diagnostic methods for temperature and salinity contributions.
clim ) is used for temperature analysis, maintaining the temperature field during ASCs but using climatological salinity; F(T clim , SASC) is used for salinity analysis, maintaining the salinity field during ASCs but using climatological temperature; F(T clim , S clim ) represents the fundamental climatological analysis, utilizing both climatological temperature and salinity fields.