Submarine tectonic geomorphology of the Pliny and Hellenic Trenches reflecting the geological evolution of southern Greece

Mapping seafloor geomorphology is a complex task requiring the integration of advanced cartographic technologies and high-resolution spatial data. This paper provides a comparative geomorphological analysis of the Hellenic Trench (HT) and the Pliny Trench (PT) located in the Eastern Mediterranean Sea, southern Greece. These trenches were formed as a result of the tectonic plate subduction in the Eastern Mediterranean Sea: the northward motion of the African and Ara bian plates, complicated by the regional geological settings, such as active faults and earthquakes, which resulted in their different geomorphological forms and bathymetric shapes. Technically, this paper presents an example of the advanced scripting mapping by scripting the cartographic toolset of Generic Mapping Tools (GMT), which performs mapping through shell scripts. The maps are based on the high-quality topographic, geological and geophysical data: GEBCO, EGM96, geoid, and marine free-air gravity fields. The GMT builds upon the modules used for data processing. The region was subsetted by ‘grdcut’, analysed by the Geospatial Data Abstraction Library (GDAL) (gdalinfo utility), and visualized by ‘grdimage’. Two segments of the trenches formed in a condition of varying geological and geophysical settings, have been modelled, visualized and compared, as representative cross-sections. As a result of the automated digitizing, spatial interpolation and sequential aggregating of GMT codes, the segments of the cross-sections were represented. The HT (Ionian segment) has an asymmetric one-sided shape; a steepness of 56.8° on the NE side and 7° on the SW flank. The PT has a more symmetric view; a steepness of 42.14° on its NW flank and 26.66° on its SE flank. The PT has a clear peak of the depths at -2600 to -2800 m and the most representative data range at -5000 to -6000 m. The HT has a bimodal data distribution with two peaks. The most frequent data for HT is -3200 m to -3400 m. Compared to PT, the HT is deeper. The histogram shows the peak of data for HT in the interval between -3200 to -3400 m (135 samples) while the PT shows the peak of data in the interval at -2600 to -2800 m (310 samples). Besides, 105 samples of the HT have depths exceeding 4000 m, while only 20 samples were recorded for PT in the same interval. This paper contributes to the geomorphologi cal studies of the general Eastern Mediterranean Sea region, particularly relating to regional seafloor mapping of the Hellenic and Pliny trenches.


Introduction
The geomorphology of the seafloor, interpreted from bathymetric models, gives information about the geological processes on the Earth's surface. That is why the precise bathymetric mapping contributes to the geological studies and research. Methods of studying the seafloor and its geomorphology are diverse. These include using direct observations by specific software such as Atlas Hydrosweep DS-2, Global Positioning System (GPS) navigation and the Trimble system (Gauger et al., 2007), or indirect modelling through the magnetic and gravimetric measurements with regards to the Earth's crust thickness and geological implications (Ramberg, 1981), geologic measurements of rock samples (Kuhn et al., 2006) or studying natural hazards such as landslides in the geological context (Krkač et al., 2021;Lemenkova et al., 2012).
Various methodological approaches are used in the GIS visualization and data analysis. These include, for example, geostatistical analysis, spatial metrics and statistical modelling in civil engineering (Klaučo et al., 2013a(Klaučo et al., , 2013bLemenkova, 2021a, 2021b;Dobrilović et al. 2021), mining and technological industrial works (Briševac et al., 2021), petrographic and geochemical analyses, studies of sedimentary rocks and seafloor characteristics (Bilić et al., 2021; Nu groho and Putra, 2020), and gravity modelling (Truffert et al., 1993). One of the largest problems for Rudarsko-geološko-naftni zbornik i autori (The Mining-Geology-Petroleum Engineering Bulletin and the authors) ©, 2021, pp. 33-48, DOI: 10.17794/rgn. 2021.4.4 cartographic data visualization is the processing of the multi-source data taken from various sources. This is a cartographically intensive process taking a considerable amount of workflow to process various data formats, to analyze their metadata, accuracy and suitability, often involving re-projecting and reformatting of the data. In many cases, cartographic data processing starts with generating databases that include maps of different scales, topics and statistical analysis (Suetova et al., 2005), involve advanced data analysis and planning (Klaučo et al., 2014(Klaučo et al., , 2017, quantitative mapping (Ra baute and Chamot-Rooke, 2007), along with vectorizing and digitizing (Schenke and Lemenkova, 2008). Some research papers are based on the multidisciplinary analysis for studying the links between tectonic subduction and submarine geomorphology (Vassilakis et al., 2011) or using LiDAR remote sensing data applied for geology (Jagodnik et al., 2020). The advantage of the Generic Mapping Tools (GMT) as a tool for cartographic visualization and mapping is discussed previously (Lemenkova, 2020c;2020d;2019a).
The goal of this paper is to explore how high-resolution data can be sensibly represented as a series of the thematic maps by means of GMT, and to demonstrate the technical functionalities of GMT as useful and feasi-ble analysis and visualisation methods in geological and geomorphological studies. The paper discusses the difference in the topographic patterns of the two trenches of the Eastern Mediterranean Sea, the Hellenic Trench, and the Pliny Trench. Study of the deep-sea trenches is limited to direct observations due to their inaccessible location, and replaced by remote sensing approaches, data modelling, visualization and statistical analysis by machine learning methods (Lemenkova, 2019d(Lemenkova, , 2019a. The Hellenic Trench is placed in the eastern part of the Mediterranean Sea, the Aegean Sea (see Figure 1).
The Hellenic Trench is formed along the Mediterranean seafloor, subducted toward the northeast, below the Aegean Sea. The area surrounding the Hellenic Trench includes Southern Greece, Western Anatolia in Turkey and the basin of the Aegean Sea (McKenzie, 1970;Maravelis et al., 2016;Le Pichon and Angelier, 1979). The seafloor of the Aegean Sea has a very complex block structure. It is composed by a system of elevated blocks with steep submarine slopes. Numerous islands are located on top of these block between the depressions and elongated sub-latitudinal troughs, formed along the geologic fault system. Minor local troughs and a Hellenic Volcanic Arc are formed in the northern and southern parts of the Aegean Sea, with depths over 1200 and less than 1500 m, respectively. Tectonically, the Hellenic Trench is located in the zone of the lithospheric plate subduction (see Figure 2) where the African Plate is subducting under the Eurasian Plate (Angelier and Le Pichon, 1980;Le Pichon et al. 1982;Angelier et al., 1982). The Hellenic Island Arc and a trench are formed as a result of the complex interrelation of the tectonic processes and seafloor spreading (Philippon et al., 2014;McPhee and van Hinsbergen, 2019;Elsasser, 1971;Confal et al., 2016). The island arc stretches from the Ionian islands, including the Peloponnese Peninsula, the islands of Crete, Karpathos Island, Rhodes and others along the southern part of the Aegean Sea (see Figure 2). The geomorphology of the external (southern) slope of this arc is characterized by a steep slope and a complex stepwise block structure.
A system of narrow depressions is stretching along the foot of the Hellenic Island Arc forming the Hellenic Trench with dominating depths at 3500-4500 m, and the maximum depth at 5121 m. Le Pichon et al. (1982) reports distribution of the vertical cliffs with hard rock exposure over a depth range of 1000-1500 m of the Hellenic Trench. Following geographical specifics, the structure of the Hellenic Trench allows the differentiation into three minor trenches, sometimes referred as Ionian, Pliny and Strabo (Le Pichon et al., 1981;Peters and Huson, 1985;Özbakır et al., 2013).
The southward retreat of the Hellenic Trench resulted in the Aegean Sea extension in a two-stage period. From Middle Eocene to Middle Miocene it was accommodated by deformation leading to the exhumation of metamorphic rocks, high-temperature rocks in core complexes and the deposition of Paleogene sedimentary basins. Since Middle Miocene, the Aegean Sea extension results in a development of the Neogene sedimentary basins (see Figure 3). The process of the Aegean Sea formation is driven by slab rollback (Brun et al., 2016;Brun and Sokoutis, 2010).
The geophysical settings of the Eastern Mediterranean Sea are notable for a complex gravitational field (Tirel et al., 2004) as reflected in geoid undulations (see  of the geologic setting of a study area, which contains following elements: front lines (yellow front lines), borders of the tectonic plates (thick red lines), volcanoes (red triangles), seismic active places (focal mechanisms), ophiolites (purple small circles) showing the portions of Earth's oceanic crust uplifted and exposed above sea level and emplaced on the Anatolian and Eurasian Plates onto continental crustal rocks.
Bouguer anomalies in the eastern part of the Mediterranean Sea are mainly from 80 to 200 mGal, decreasing to zero in some northern areas (see Figure 5). The Aegean Sea is characterized by a sharply dissected gravitational field, where Faye anomalies range from 40 to 100 mGal, and Bouguer anomalies range from 40 to 160 mGal. The negative Faye anomalies (-40 mGal to -80 mGal) are observed in the eastern part of the Mediterranean Sea, except for the region of the Nile River where anomalies are positive (Malovitsky et al., 1982).

Methods
In order to examine and build a link between the tectonic and geological settings and the geomorphology of the Hellenic Trench in its western and eastern parts (the Pliny Trench) in the southern Aegean Sea, the geospatial data were visualized and a comparative analysis was divided into several phases performed by GMT (Wessel and Smith, 1991;. The GMT utilizes native scripting language with its own syntax similar to the programming languages to process data. Compared to the traditional Graphical User Interface (GUI)-based GIS, the process of generating maps in GMT is a bit more complex, since it requires the understanding of the logic of its syntax which, due to the insufficient examples with open publications, may present certain difficulties in utilization. To give some examples of the GMT based coding, the following illustration of the technical approach is as follows. Selecting the study area and subset of the necessary region was done using a 'grdcut' module of GMT: 'grdcut GEBCO_2019.nc -R19/37/ 30.5/41.5 -Ght_relief.nc'. Here the -R19/37/30.5/41.5 shows the geographic coordinates of the area in West-East-South-North (WESN) format. Visualizing the raster image of the General Bathymetric Chart of the Oceans (GEBCO) was done by the 'grdimage' module which reads the 2D grid file in NetCDF and produces a coloured map using selected colour palette tables.
The core idea of raster image visualization by a 'grdimage' consists in plotting a series of rectangles centered on each grid node and assigning them with a colour based z-value, which corresponds to the topographic elevation value in meters. The colour palette table (cpt) in this case (see Figure 1) was defined by a module 'makecpt' through a command 'gmt makecpt -Cgeo.cpt -V -T-  Divins (2003). The 5-arc-minute total sediment thickness grid is based on high-resolution new data and several regional observations of the oceanic sediment thickness. The isolines are placed every 500 m and show the increase of data (orange colours) near of the Nile River mouth. 5119/3505 > myocean.cpt'. The colour palette was placed vertically on the left side of the image. It shows a range of data from -5119 to 3505 m. The image was then visualized by a code 'gmt grdimage ht_relief.nc -Cmyocean.cpt -R19/37/30.5/41.5 -JM6i -P -I+a15+ne0.75 -Xc -K > $ps'. Here the '-JM6i' flag stands for the Mercator projection, '-P' flag is a portrait orientation of the output, the -I option was used for the illumination of the image. The same principle was applied to all other cartographic elements plotted on the presented maps by GMT.

Topographic visualization
The first research phase has been designed as a geographical visualization of the topographic General Bathymetric Chart of the Oceans (GEBCO) and geological data. The actual processing of the raster GRD data consists of editing functions of the script where each line of the code generates a certain element on a map using a specific GMT module (e.g. 'grdimage', 'pscoast' and many more) and provides a layer-style mapping, which finally results in the design of a map. The data include high-resolution shoreline vector database (Wessel and Smith, 1996), tectonic plate boundaries and seismic events (focal mechanisms). The visualization aimed to show the consequences of the geological processes on geomorphological form of the deep-sea trench. The analysis of the data grids (their extent and spatial queries) was performed by Geospatial Data Abstraction Library (GDAL) library (GDAL/OGR contributors, 2020). The importance of the quality of the bathymetric and topographic grids has been discussed previously (Smith, 1993). To achieve a high standard of mapping, this research is based on the topographic grid of famous GEBCO map (GEBCO Compilation Group, 2020; Schenke, 2016) that provides the most authoritative, publicly available bathymetry data sets for the world's oceans. The GEBCO map is the most precise exiting global bathymetric map developed through the Nippon Foundation and is widely used in geophysical, geological and topographic mapping (Vrdoljak et

Geophysical mapping
In the second phase, the variations of the geophysical conditions, such as geoid (see Figure 4) and marine free-air gravity (see Figure 5), have been analyzed using the National Oceanic and Atmospheric Administration (NOAA) grid and Earth Gravitational Models of 1996 (EGM96) data (Lemoine et al., 1998). The GMT modules used for geophysical mapping implement a set of commands and defined settings ('flags') where a command is entered and defined. For example, the command 'gmt grdimage gravHT.grd -Ccolors.cpt -R19/37/30.5/ 41.5 -JM6i -P -I+a15+ne0.75 -Xc -K > $ps' was used to visualize the map in Figure 5 Marine free-air gravity map of the Aegean Sea Basin. Here the 'grdimage' is the main GMT module visualizing the raster grid, and the rest are auxiliary commands used to refine mapping and add elements on the map.
Analysis of the high-resolution data from the Cryo-Sat-2, Jason-1 (Sandwell et al., 2014; Wessel and Watts, 1988) revealed regional differences in the rock properties that lead to diverse patterns of the gravity anomaly fields alongside existing geomorphic disparities of the submarine relief (Sandwell & Smith, 1997). Published reports on geophysical and geological settings of the Aegean Sea in general and the Hellenic Trench system as its particular landform have been reviewed and considered for critical analysis of the above-mentioned issues (Papazachos and Comninakis, 1971

Geomorphological modelling
In the third phase, the geomorphological modelling has been performed using scripting technologies (Le menkova, 2020a, 2020b) for visualizing cross-section profiles of the Hellenic Trench (see Figure 6) and the Pliny Trench (see Figure 7), their comparative plotting and statistical analysis of histograms (see Figure 8). Automated digitizing of the cross-section profiles by GMT (Wessel et al. 2013) presents effective and advanced spatial data processing and information retrieval. The approach is based on the idea to speed up the vectorizing raster grids by machine learning: 1) first, the GMT is commanded to create a series of profiles running with set up parameters along a selected segment with defined coordinates; 2) the generated profiles create a table with three input columns (XY-coordinates of each cell and the measured elevation depth); 3) the GMT is commanded to use this table to plot a graph visualizing the profile as an output from a dataset.
Hence, the cross-section modelling allows for the plotting of a 2D transect for visualizing a geomorphological cross-section of the mountain relief in a given segment of the transect line of the relief to be mapped as a series as parallel-going lines crossing the target area in a perpendicular direction. The resulting geomorphological profiles can then be used for analysis of the topographic variations of the landforms. The representative  Figure 6A shows the cross-sectional segments of the trench digitized automatically using GMT. Figure 6B shows the location and topography of the Hellenic Trench extending in the NW direction from SW Crete towards the Peloponnese Peninsula. The topographic pattern of Kythira Island and the adjacent region shows the increase in elevations on the right side of the flank of the segment. The greyed area (see Figure 6A) denotes the data deviations and thin vertical lines show error bars, while the thick red line shows the median value of the trench's talweg computed based on data from the 12 of profiles (thin yellow line in Figure 6B). The Aegean Sea Plate (minor tectonic plate) is located eastward of the trench, while the Africa Plate is on its western flank (see Figure 6B). The width of the profiles is 200 km with the talweg of the trench in the middle (see Figure 6A).

Results
An analysis has been carried out based on a series of thematic maps (topographic, geologic, geophysical) and geomorphological models supported by the review of the existing extensive literature on the geology of the Eastern Mediterranean and the Hellenic Trench  Figures 2 and 3, while the geophysical maps (see Fig  ure 4 and 5) contain categories of marine free-air anomaly fields and geoid undulation.
The geologic data included several categories of objects related to geologic settings of the areas, that is, trenches and ridges, volcanoes, slabs, ophiolites, tecton-  Figure 7B) represent the 17 cross-section profiles plotted as 200 km long parallel segments, sampled every 2km, spaced 10 km and stacked using the median (the thick red line in Figure 7A). Raw bathymetric data originated from GEBCO (Schenke, 2016). The Pliny Trench extends from southern Crete in the NE direction (see Figure 7B). The African Plate is located southwards of the trench, which is formed in the subduction zone. The map in Figure 7B shows the distribution of 17 transects of the cross-sections.
ic plate boundaries, and focal mechanisms (see Figure  2). The sediment thickness map visualizes the variations in sediment by categories divided by each 1000 m, which corresponds to the previously obtained results (Erba et al., 1987;Feldhausen and Stanley, 1980). Thus, the highest sediment thickness can be seen (see Figure 3) near the Nile River discharge (over 9,000 m, beige colours on Figure 3) while the northern Aegean Sea seafloor and the region northward of Crete are notable for the lowest values of <3,000 m (dark blue coloured areas). The Hellenic Trench system has varying values ranging between 7,000 and 8,000 m (light green areas) followed by 8,000 to 9,000 m (light orange-coloured areas). More detailed studies on the sediment structure of the Hellenic Trench can be found in Blan pied and Stanley (1981) who describe the depositional origin of fine-grained marine sediments caused by the well-exposed mudstone formations in the mountain chains in Greece and the Eastern Mediterranean Sea. The map of the geoid regional model (see Figure 4) shows higher categories of the geoid undulation (35-45 m) in the northern region of the Aegean Sea Basin con- tinuing over the terrestrial area of Bulgaria. The area to the southeast of Crete and the central part of the basin of the Eastern Mediterranean Sea demonstrate a notable regional decrease in geoid values (50-10 m, blue-coloured area). The region of the Hellenic Trench has dominating values of 11-12 m (light blue colour, Figure 4). A zone of the intense negative anomalies of Faye (-120 to -200 mGal) extends along the Hellenic Trench, which continues northward to the Adriatic Sea.
The lowest values of the free-air gravity are notable over the regions of the deep-sea trenches clearly depicting the physiographic isolines of the topographic map and showing correlation between the distribution of the depths and the anomaly fields (see Figure 5, dark blue colours). In contrast, the highest values (>180 mGal) can be seen over the Cyprus Island area, as well as over the south-eastern territory of Turkey (light-orange to red colours in Fig  ure 5). The majority of the terrestrial area of Turkey demonstrates a clear correlation between the gravity fields and topographic isolines of the mountain areas (light green to yellowish coloured areas as a hilly relief on Figure 5).
A comparison of the Hellenic Trench (see Figure 6) against the Pliny Trench (see Figure 7) shows that the Pliny Trench has a more symmetrical shape while the Hellenic Trench has an asymmetric one-sided shape with a steepness of 56.8° on the northeastern side of the studied segment with coordinates 22.2°E 35.9°N to 23.2°E 35.3°N (see Figure 6B). The opposite southwestern flank is almost flat with a steepness of 7° (see Figure 6A). The cross-track profiles are plotted as 200 km long, sampled every 2 km, spaced 10 km. The Pliny Trench has a segment stretching from 25.3°E 34.2°N to 26.9°E 34.8°N (see Figure 7B). It has a rather symmetric view in a cross-section with the following steepness: 42.14° on its northwestern flank (crossing the eastern part of the Crete Island) and 26.66° on its southeastern flank (see Figure  7A). Despite the certain difference in slope steepness of both the flanks of the Pliny Trench, it has a more symmetric geometric form comparing to the cross-section of the Hellenic Trench (the Ionian segment).
The comparative analysis of the statistical histograms (see Figure 8) shows the frequency of depths as data distributed along the segments of the Hellenic and Pliny trenches (see Figure 8). The analysis of the cross-section profile based on the summarized data shows the following results. The Pliny Trench (see Figure 8, upper part) has a clear peak of the most repetitive depths in an interval with a range -2,600 to -2,800 m (310 samples) followed with a large gap by the intervals within depths -2800 to -3000 (173 samples) and an interval of -2400 to -2600 (138 samples). All other depths have values <100 samples with rather insignificant variations comparing to the depth at an interval of -2400 to -3000 m, which points at the most representative data range for the Pliny Trench between -5000 to -6000 m.
The Hellenic Trench demonstrates a different pattern of data variations. Thus, in contrast with the Pliny Trench showing a clear peak (see Figure 8), the Hellenic Trench has a bimodal data distribution. The data distribution has a bimodal character with two peaks corresponding to the two intervals: 1) a distinct peak at -2800 m to -4000 m; and 2) a less pronounced peak at 0 to -1400 m. Other data values are located on the shelf and slopes. The most frequent bathymetric data for the Hellenic Trench correspond to the interval of -3200 m to -3400 m (135 samples). Comparing to the Pliny Trench, the Hellenic Trench is deeper with 105 values below the -4,000 m while there are only 20 values at such depths for the Pliny Trench. Similarly, the range of data between the -4000 to -3000 m has 241 values for the Pliny Trench while there are 498 data values for the Hellenic Trench.
This accordingly illustrates the difference in the depth distribution for the Hellenic Trench against the Pliny Trench. The comparison of shelf areas (depths of 0 to -1,000 m) shows relatively similar patterns: the Pliny Trench covers 281 observation points while the Hellenic Trench has 240 samples for the same data range. Thus, in the context of regional topographic analysis, the comparison of the histograms points at the difference in data distribution for the two trenches: a relatively wider, bimodal character of data distribution for a Hellenic Trench and a single-peaked data distribution for the Pliny Trench. The Hellenic Trench-Arc system has been investigated in only a few studies Royden and Papanikolaou, 2011;Huchon et al., 1982) comparing to the wide variety of other oceanic trenches of the Pacific Ocean, e.g., the famous Mariana Trench, the Peru-Chile Trench, the Kuril-Kamchatka Trench, the trenches of Oceania and the region of Australia (Lemen kova, 2019b, 2019c). This is mainly explained by the complex tectonic and geophysical settings of the Pacific Ocean comparing to the Mediterranean Sea (Chapple and Tullis, 1977;Lemenkova, 2021a;McClusky et al., 2000).
The existing papers on the Hellenic Trench are methodologically based on various methods. Among others, these include topographic mapping, visualization of slabs with a vertical tear or a lateral edge, GNSS-derived horizontal velocity field and seismicity, modelling geometry of the Hellenic subduction zone, visualizing Svelocity (Özbakır et al., 2020), studying structural trends in the Hellenic Trench system in context of the distribution of faults, folds and décollement types displacement (Kenyon et al., 1982). Other interesting publications include research focused on earthquakes and seismic reconstructions in the Hellenic Trench and the Aegean Sea area both in the past and in recent time (Pa pazachos and Papazachou, 2002; Evagelatou-Notara, 1993; Kiratzi et al., 2018;Di Vita, 1995).
However, the existing papers do not present geomorphological cross-sectioning of the trench using GMT, which fully illustrates the profile of the trenches as a comparative view, based on the new generation of cartographic scripting technologies. Previous studies in the Greek region (Crete Island and Hellenic Trench-Arc system) were restricted by the tectonic application and studies of the geological setting without a detailed focus on its submarine relief (Gautier et al., 1999;Jolivet et al., 2013;Konca et al., 2019;Papazachos, 1996). In this study, the geomorphology of the two trenches in the context of the regional geological settings was investigated using a dataset of high-resolution grids processed by the GMT approach. A comparison of the geomorphology of the Pliny and Hellenic trenches revealed significant variations in their cross-sections in the selected segments (plotted using 'grdtrack' modules of the GMT). This indicates the impact of the geological regional settings and rock stratification as well as the differences in speed of slab subduction in the two segments of the trenches. These factors sculpt the current geometric shape of the trenches.

Conclusions
The contextualization of this research in the era of big Earth data analysis and machine learning application presents a multi-disciplinary approach of geologic data processing by the GMT-based machine algorithms of geospatial data processing that allow to better understand the interaction between the seafloor geomorphology and tectonic-geologic setting sculpting the shape of a trench. The demonstrated GMT-based approach for mapping deep-sea trenches combines the scripting functionality that enables the performance of a high-quality mapping and cartographic visualization with the flexibility of programming. The existing scripts can be reused for the next steps of mapping and data analysis to save time and computer's memory and to decrease the repeatability of the monotonous process. The other advantage of the scripting based cartography is found in the significant automatization of the cartographic routine which enhances both the precision of the plotting and the quality of the output through avoiding humaninduced errors possible through the hand-made digitizing and mapping. Machine-made visualization of the geophysical grids, topographic and geological vector maps enable a comparative analysis of the multi-source data through their overlay with changed transparency settings and the location of layers within the cartographic script.
The actuality of the presented research on deep sea trenches is caused by the specific approach of this study. High-resolution geospatial datasets are one of the major drivers used for big Earth data analysis that consists of the processing of millions of cells and records in tables. The combination of high-resolution data (GEBCO, EGM96, GlobSed) and machine learning methods for processing of this data expands the functionality of the research projects of seafloor geomorphology. Since these unique submarine landforms are formed in the region of the tectonic plate subduction, active seismicity and submarine volcanism, their bathymetry is generally extremely deep and unavailable for direct observations. Therefore, studying deep-sea trenches was proven impractical until recent years. However, the rapid development of the machine learning methods and the onset of the big Earth data era enabled complex data analysis in marine geosciences. Some detailed studies of the submarine topography have been made over the mid-ocean ridges since the 1970s, but fewer studies were focused on the oceanic trenches.
The GMT scripting toolset is an effective cartographic visualization tool, alternative to traditional GIS. The power of automatization achieved through scripting facilitates the generation of maps as batch processing and enables the visualization and processing of large datasets in addition to mapping, statistical analysis, and data conversion. In contrast to GIS, GMT presents a completely console-based (no visual interface) scripting system for spatial analysis and cartographic mapping using a smart combination of modules and scripting approach using its native language. The advanced functionality of UNIX-compatible GMT enables the exportation of data from multiple formats and reprocesses them with GDAL processing tools. Furthermore, shell scripting by GMT automates the cartographic workflow, which becomes accurate and rapid, in contrast to the menu-based Graphical User Interface (GUI) GIS.
As a result, this increases the effectiveness of the cartographic methodology and presents fine print-quality cartographic output. Finally, another important feature of GMT consists in its open source availability, which presents perfect solutions to the distance-based (onlinebased) research, which is currently very useful for students and researchers alike. Unlike the commercial GIS, free GMT is available to 'work-at-home' projects, however, it has all the powerful cartographic functionalities: data import and formatting, data operation and organizing, data modelling, visualization and analysis. Besides, GMT is compatible with key libraries of the advanced statistical analysis such as Julia, Python (PyGMT for processing GMT's plots), R and Octave /MATLAB.