The altimeter satellite gridded sea level anomaly (SLA) is estimated by optimal interpolation, merging the measurement from the different altimeter missions available. This product is processed by the DUACS (Data Unification and Altimeter Combination System) multi-mission altimeter data processing system. It processes data from all altimeter missions: Jason-3, Sentinel-3A, HY-2A, Saral/AltiKa, Cryosat-2, Jason-2, Jason-1, T/P, ENVISAT, GFO, and ERS1/2. To produce reprocessed maps in delayed time, the system uses the along-track altimeter missions from products called SEALEVEL_∗PHY_L3_MY_008∗. The SLA computation provides the absolute dynamic topography (ADT) and the surface geostrophic currents. The velocity field spatial resolution is 0.125∘×0.125∘ (available on http://www.aviso.altimetry.fr/duacs/, last access: 20 September 2021).

The daily geostrophic surface velocity fields between 1993 and 2018, from the Herodotus Abyssal Plain and until the easternmost part of the Levantine sea, form the input layer of the SOM. In addition to the zonal and meridional components of the geostrophic velocities, the fluid parameter of Okubo–Weiss (OW) is included in the input layer (see Fig. b). OW measures the relative importance of deformation and rotation at a given point. Positive OW values indicate strain-dominated regions, while negative OW indicates a vortex-dominated region. Accordingly, OW is a physical criterion widely used in the methods of eddy detection.

OW=sn2+ss2-w2, where sn and ss are the normal and the shear components of strain and w is the relative vorticity of the flow defined, respectively, by 1sn=∂U∂x-∂V∂y,ss=∂V∂x+∂U∂y,w=∂V∂x-∂U∂y.

SOM is an unsupervised neural network method used for data visualization. It projects higher-dimensional data into lower-dimensional space by leveraging topological similarity properties. By this method, multidimensional data are clustered into neurons automatically associated into an orderly organization, where similar neurons are adjacent, and the less similar neurons are situated far from each other in the grid. This way allows for obtaining an insight into the topographic relationships of the initial data set .

The SOM is structured in two layers: the input layer (in our case, a 3-D input layer composed of the zonal and meridional components and the Okubo–Weiss parameter) and the resulting neuron grid. Each neuron, representing a cluster with data presenting common characteristics, is associated with a referent vector obtained from a learning data set. Each vector of the input layer will be attributed to the neuron with the closest Euclidean distance to the referent vector. This referent vector is called the best matching unit (BMU), and its associated neuron is the “called” winning neuron. The determination of the referent vectors and the topological order of the SOM maps is done by minimizing the cost function 2JSOMT(X,W)=∑zi∈D∑c∈SOMKT(δ(c,X(zi)))||zi-wc||2, where c∈SOM represents the neuron index in the SOM, X(zi) represents the allocation function that assigns each element zi of the input D to the corresponding referent vector wX(zi), δ(c,X(zi)) represents the discrete distance on the SOM between a neuron c and the neuron allocated to observation zi, and KT is a kernel function parameterized by T (where T stands for temperature in the scientific literature dedicated to SOM) that weights the discrete distance on the map and decreases during the minimization process. During the minimization of the cost function, the topological order is preserved, and thus the more similar neurons are adjacent and the less similar neurons are situated far from each other. To provide an equal-weight distribution of the input parameters, the variables were normalized with their variances. Several tests were conducted to determine the optimal size of the SOM map giving the best representation of the data. These tests were based on the capacity of each map to reproduce the initial data space of our input data with the least error possible. Based on this choice, we opted for a large SOM map of 1400 neurons.

After the training phase, the SOM is well organized where there is a gradient of zonal and meridional velocities. This distribution of U (Fig. a) and V (Fig. b) reflects the OW variation well (Fig. c). The OW in SOM shows clusters of intense positive or negative values. These extreme values represent the characteristics of a vortex (positive) or strain (negative) dominance.

The SOM allowed for classifying the velocity field into neurons that represent the different circulation patterns of the targeted grid based on U, V, and OW. To simplify the representation of the physical processes obtained from the different situations captured by each neuron, we applied the HAC to group these neurons into a reduced number of clusters. HAC is a cluster analysis that seeks to build a bottom-up hierarchy of clusters. From the initial partition containing the neuron groups of the SOM map, two neurons of the same neighborhood were clustered at each iteration. The used criterion was Ward's minimum variance method, which provides a partition that minimizes the within-cluster inertia while respecting the topological organization. As a result, the neurons were separated into five different clusters (see Fig. d). Consequently, clusters 4 and 5 (denoted as C4 and C5) were characterized by a negative OW, compared to positive OW for C1 and C2, while C3 had an average OW of 0. MKE boxplots of the neurons in each cluster show that C3 had the weakest flow intensity, while C4 and C1 represented the highest MKE intensities, followed by C2 and C5 (Fig. e and f). In summary, C1 and C2 are clusters of strong flow with high vorticity (high MKE and positive OW), C4 and C5 are clusters of strain-dominated strong flow (negative OW and high MKE), and C3 is the cluster of the weakest velocities.

To analyze the activity of the most dominant features in the Levantine basin, we targeted the area from the Herodotus Abyssal Plain until the easternmost part of the Mediterranean Sea. The Mersa Matruh Eddy (MME) (see Fig. ) is an eddy generated by the coastal instabilities near the eastern Libyan and Egyptian coasts. These eddies drift seawards while being guided by the 3000 m isobath. Indeed, the 3000 m Herodotus Plain traps and prevents the coastal vertically extended eddies from propagating further to the east, thus causing an accumulation of these eddies in this area . MME is one of the most dominant mesoscale features in the eastern Mediterranean, where their mean kinetic energy exceeds 300 cm2 s-2 . In the easternmost part of the basin, the Shikmona Eddy (ShE) system (see Fig. ) represents another complex system composed of several cyclonic and anticyclonic eddies with varying sizes, positions, and intensities . Similar to MME, ShE is an area where previously formed eddies tend to accumulate and/or merge . Another important mesoscale feature existing in the eastern part is the Cyprus Eddy (CE) (see Fig. ). It is an intense dynamic feature occurring in the open sea. Unlike the MME and ShE, eddies are formed in this area and do not accumulate . We should note that other mesoscale structures exist in the eastern Levantine basin but are less frequently observed. Among these, we mention the “Lattakia Eddy” (LE) taking place between Cyprus and Syria. LE is a cyclonic eddy generated by the interaction of the northward current along the Lebanese and Syrian coasts with a Mid-Mediterranean Jet or between ShE and the coastline or generated by the topography .

The frequency variation of the five clusters in each of the selected boxes, i.e., Bei, Shik, MME, AMC, Nile, and CE, is seen in Fig. . This frequency variation reflects the percentage of pixels assigned to each of the five clusters in a designated box. As a result, except for C4 in the AMC, all the clusters permanently occurred, with different proportions highly variable with time in each box. Moreover, cluster frequency significantly varies from one box to another.

Although clusters of strain-dominated flow (C1 and C2) were not frequent everywhere, C1 and C2 were frequently observed in MME and AMC, respectively, during the entire period. Such a high frequency occurred at the expense of other clusters, especially the cluster of weak flow C3, which was less observed in these two boxes. Regarding the vortex-dominated clusters (C4 and C5), C5 was the most frequent. C4 was quasi-absent in AMC and scarcely existed in the other boxes. When comparing between boxes, C5 was most frequent in the MME. Overall, all the cluster occurrences fluctuated greatly over time. To better analyze such a frequency variations, we presented the daily dominant cluster in each box in Fig. . On the one hand, C3 was the main cluster in CE, Nile, Bei, and Shik boxes. Moreover, between 1993 and 2000, C3 dominated CE almost exclusively. On the other hand, C3 dominance was rare or quasi-nonexistent in the MME and AMC, where instead C1 and C2, respectively, were the most frequent clusters. While C2 dominance was not observed in MME, an increasing periodic C1 domination was observed in AMC, starting from 2000. C5 was frequently dominating all the boxes, albeit intermittently. This dominance was rare in the CE, Nile, and AMC, compared to the MME, Bei, and Shik. Indeed, in these last three boxes, C5 dominance was frequent and showed a long lifetime that could last for several months, such as observed in Bei in 2003 and 2014 and Shik in 2005.

These results showed that MME and AMC are two zones of a special regime of flow. This latter zone is represented by clusters of intense current, the so-called C1 and C2. The other boxes are zones of relatively weaker currents. In all the boxes, there were sporadic events of intense eddy activity, exhibited by the intermittent periods of C5 dominance.

The daily mean kinetic energy of the mean flow per unit of mass (MKE) computed in each box (see Fig.  in the Appendix) shows that the lowest MKE values were observed in the boxes where C3 dominated the most (CE, Nile, Shik, and Bei). In these boxes, MKE values were less than 150 cm2 s-2. On the other hand, the highest values were in AMC and MME, with values consistently exceeding 300 cm2 s-2. Hence, the boxes dominated by C1 and C2 had MKE values that were 2 times higher than those observed in boxes dominated by C3.

The tendency of the dominant cluster could change with time, where previous results show, for example, that C5 was rare in Bei before being more frequently observed as a dominant cluster between 1993 and 1997. Figure  shows the seasonal variation of the C1, C3, and C5 average frequencies in all the boxes between 1993 and 2018. The general trend of C5 frequency increases over time. The most intense C5 positive tendency was in MME, where C5 increased by 10 % in 26 years. In all the seasons, the C5 frequency average in MME increased from 25 % in 1993 to 35 % in 2018. There was a similar increase in all the seasons in the Bei, MME, and Nile boxes. The weakest increasing tendency was in AMC and Shik. In terms of seasonality, there was no significant difference in C5 frequency between seasons, except in CE, where in summer and fall the values were higher, with values closer to the MME. MME registered the highest seasonal average frequency of C1, meaning that this box is the most dominant in terms of intensity and vorticity (C5 and C1). In terms of trends, C1 showed similar results to C5. The MME and Nile boxes showed the sharpest increases, followed by AMC, Nile, CE, and Bei, while C3 was continuously decreasing in all the boxes.

These results show that the activity of the dominant mesoscale is increasing with time. Previous altimetric data observations from 1993–2003 revealed increasing variability of the Mediterranean Sea activity that is maximal in the Levantine Sea, especially in the Mersa Matruh area, where increasing energetic structures were observed . Studies have discovered evidence that eddies are becoming more energetic. This increase was related to several reasons, such as the changes in winds or large-scale horizontal temperature gradients or the changes in the shear of the ocean currents . This current evolution is well highlighted where C3 is being progressively replaced by C1. Nevertheless, we should not exclude the impact of the altimetry increasing accuracy in describing the surface circulation, with an increasing number of satellite tracks in both time and space. Satellite along-track sampling accurately estimates the sea surface height (SSH) from which the surface circulation is derived . The interpolation of the gaps could lead to the underestimation of the mesoscale features. However, with the increasing number of satellites over time, the merging of altimeter missions provides a better description of the mesoscale activity . In terms of seasonality, there was no significant difference in C5 frequency between seasons, except in CE, where in summer and fall the values were higher, with values closer to the MME. Although previous studies have mentioned that eddies in MME are mainly formed in summer and spring , there was no clear seasonality of C5 frequency. This is explained by the frequent eddy apparitions and their relatively long lifetime. Consequently, eddies permanently occur in the MME box .

It should be mentioned that C2 and C4 did not show a clear tendency. However, C2 was only significantly present in AMC with values around 40 %, while in all other boxes the frequency was less than 20 % (see Fig.  in the Appendix).

The spatial variation of clusters' frequencies is shown in Fig. . The intensity of the along-slope coastal flow showed a spatial variability. Indeed, the high kinetic energy clusters, C1 and C2, frequently occurred off the Libyo-Egyptian coast and between Turkey and Cyprus, respectively, while the weak current cluster, C3, dominated the easternmost part of the basin. The C5 predominance is mostly situated in the western part, revealing that the mesoscale activity is more intense in that area. No clear jet was observed in the middle of the basin, including the MMJ pathway, where C1 dominance was disconnected by high C3 occupancy between 30 and 32∘ E (see C1 in Fig. ).

To further investigate the time evolution of the potentially existing MMJ, we present a Hovmöller diagram in Fig. a (along 31.5625∘ E) that shows the temporal variation of the clusters across its potential path. The longitude selection was based on the previous results showing an area dominated by weak flow cluster C3 in the MMJ potential pathway between the Eratosthenes Seamount and the coast of Cyprus (see C1 frequency in Fig. ). Although there was frequent C3 domination, there were years where clusters revealing higher kinetic energy (KE) were frequently occurring, such as C1 in 2015 or C5 between 2013 and 2014. An example of this variation is presented in Fig. b, which shows the cluster frequency distributions during the EGYPT/EGITTO (September 2005–July 2007) and CINEL campaigns (September 2016–August 2017) that provided in situ drifter observations in the MMJ potential pathway. During the EGYPT/EGITTO campaign, C3 frequency was more than 50 %, and the other clusters did not exceed 15 %, while during the CINEL campaign, clusters of stronger KE increased at the expanse of C3, decreasing to 27 %. All of these results reveal that the MMJ is present in the middle of the basin, but it is only present sporadically and is not one of the most remarkable features, and strong flow clusters were frequently observed here but were masked by C3, the dominant cluster in this area. In , the drifter data set used from the EGYPT/EGITTO campaign showed no distinct jet between 2005 and 2007, a period of C3 domination. On the other hand, in the drifter tracks showed a clear MMJ crossing the basin from west to east between late 2016 and 2017, a period when there was a sharp shift in C3 frequency, decreasing by more than half compared to EGYPT/EGITTO. The Hovmöller diagram also showed that MMJ variability was without any seasonal or periodic nature, in agreement with .

Both C4 and C5 are vortex-dominated clusters. However, the previous results showed that C4 is a peripheral cluster that is scarcely observed, dominating only very few pixels close to the coast. C5 was the main cluster that mainly reflected the eddies' presence. Here we present a more detailed analysis of the C5 evolution that reveals eddy activity in the Levantine Sea.

Figure a shows the spatial distribution of C5, whose occurrence exceeded 40 % (C5P). The highest persistent C5P number was at the borders of the Herodotus Plain. In addition, another group of C5P occurred in areas of the extended continental shelf, more precisely the shelf located offshore from Egypt and between Cyprus and Turkey in the northern part of the basin. A small number of C5P followed the bathymetric isobath of 1000 m in the western part of the basin. On the other hand, C5P was absent in the eastern part of the Levantine Sea. In Fig. b we present the variation of C5P numbers regarding their distances to the closest main bathymetric structures (iso-lines of 1000, 2000, and 3000 m). More than 80 % of C5P were located at a distance less than 60 km from these main features. In the zones of extended continental shelf, such as offshore from the Libyo-Egyptian coast, there is a strong flow dominated by C2 and C1. Previous studies have shown that a current becomes unstable when it is wider than the bathymetry. Thus, it presents favorable conditions for eddy formation . The C5P absence off the Lebanese coasts is explained by the tight continental shelf being almost absent. Indeed, the weak current dominated by C3 near the coastline is not strong enough to permanently create instabilities. The group of C5P observed in MME is due to the impact of the Herodotus Abyssal Plain. The vertically extended eddies pinch off from the coast and propagate to the east before being trapped, thus accumulating eddies . These results show that the main bathymetric features could potentially influence eddy creation and persistence in the Levantine basin.