Study of Marine Particles Using Submersible Digital Holographic Camera during the Arctic Expedition

: The paper presents the results of in situ studies of marine particles of different nature using a submersible digital holographic camera (DHC) during the Arctic expedition. It also describes the features, performance speciﬁcations, and possibilities of the DHC and the DHC technology. The DHC technology can be used for noninvasive automatic evaluation of spatial and temporal characteristics of plankton, including the distribution of plankton concentrations. The comparison of quantitative analysis of zooplankton net samples and classiﬁcation results using the DHC revealed that the error of the DHC classiﬁcation of mesoplankton at the level of the main systematic orders was about 30%. The results of determining the data on the medium, such as water turbidity, according to the radiation shielding factor (degree) by the particles of the Suspension taxon using the DHC technology are presented; the prospects for studying the size of gas bubbles and their volume content according to the Bubble taxon data are shown. The use of holographic data for in situ point estimates is considered.

Traditionally, marine particles (for example, plankton) are studied through sampling followed by their processing and analysis in the laboratory using a microscope [16,17]. It is obvious that the plankton particles thus studied are removed from the habitat, which makes it impossible to accurately indicate their location in the studied medium. In addition, their state is damaged. Digital holography can be used to solve this problem in situ and in real time [18,19]. In this case, information on suspended particles contained in the fixed seawater volume is recorded as a digital hologram on the CCD/CMOS matrix. The DHC (digital holographic camera) hardware and software complex that we developed for this purpose are described in the works [20,21]. The digital hologram processing algorithms

•
To determine the equipment's performance characteristics; • To estimate the DHC measurement error regarding marine particles; • To assess the validity of a classification.
To address the above tasks, we manufactured the "DHC-probe Arctic", which includes the DHC and a set of additional sensors that measure the medium parameters (temperature, illumination, pressure, electrical conductivity, oxygen saturation, etc.).
This paper describes the results of using the "DHC-probe Arctic" (hereinafter referred to as the "probe") to solve these problems. In addition, for the first time, we tried to use submersible holographic data for in situ point estimates of the size and concentration of gas bubbles in areas of massive methane release offshore the Eastern Arctic Seas, which were identified during 2000-2020 [31][32][33][34].

DHC Engineering
Holography is used as a noninvasive method that registers a hologram and reconstructs an image of an ensemble of particles. Thus, information on each particle (individual plankton, settling non-living particle, oil droplet, gas bubble, etc.) contained in the registered volume of the aqueous medium can be stored, transmitted, and reproduced at a given time. Figure 1 shows an in-line hologram recording scheme used in the DHC (in-line scheme or Gabor scheme) [35][36][37][38].
um. In addition, their state is damaged. Digital holography can be used to solve this problem in situ and in real time [18,19]. In this case, information on suspended particles contained in the fixed seawater volume is recorded as a digital hologram on the CCD/CMOS matrix. The DHC (digital holographic camera) hardware and software complex that we developed for this purpose are described in the works [20,21]. The digital hologram processing algorithms to extract information on marine plankton are presented in the research by the authors of [22][23][24]. However, mesoplankton is one of the suspension components, and the DHC can be used to study other particles [25][26][27][28][29][30].
One of the tasks accomplished during the 82nd Arctic Expedition of the Laboratory of Arctic Research of the V.I. Il'ichov Pacific Oceanological Institute on board the Academician Mstislav Keldysh (AMK-82) from September 28 to November 4, 2020, was to assess the study possibilities and features of such particles.
In addition, in the context of this study, the tasks of the expedition were as follows: • To determine the equipment's performance characteristics; • To estimate the DHC measurement error regarding marine particles; • To assess the validity of a classification.
To address the above tasks, we manufactured the "DHC-probe Arctic", which includes the DHC and a set of additional sensors that measure the medium parameters (temperature, illumination, pressure, electrical conductivity, oxygen saturation, etc.).
This paper describes the results of using the "DHC-probe Arctic" (hereinafter referred to as the "probe") to solve these problems. In addition, for the first time, we tried to use submersible holographic data for in situ point estimates of the size and concentration of gas bubbles in areas of massive methane release offshore the Eastern Arctic Seas, which were identified during 2000-2020 [31][32][33][34].

DHC Engineering
Holography is used as a noninvasive method that registers a hologram and reconstructs an image of an ensemble of particles. Thus, information on each particle (individual plankton, settling non-living particle, oil droplet, gas bubble, etc.) contained in the registered volume of the aqueous medium can be stored, transmitted, and reproduced at a given time. Figure 1 shows an in-line hologram recording scheme used in the DHC (in-line scheme or Gabor scheme) [35][36][37][38]. The interference pattern of the reference wave (radiation that passed by the particles) and the object wave (radiation scattered on the particles) is formed as a result of illuminating the studied volume with laser radiation. The camera registers this interference pattern and transmits it to the computer's memory. The distinctive feature of the DHC is the use of a lens for hologram recording to increase the field of view (the diameter of a beam illuminating the volume of the medium with particles). The lens represents a serial movie projection lens, 35KP-1.8/70 (USSR) with the following characteristics: focal length (f = 70 mm), relative aperture (1/1.8), and a diameter of the circle of confusion in the field at a wavelength of 0.55~10 μm. The interference pattern of the reference wave (radiation that passed by the particles) and the object wave (radiation scattered on the particles) is formed as a result of illuminating the studied volume with laser radiation. The camera registers this interference pattern and transmits it to the computer's memory. The distinctive feature of the DHC is the use of a lens for hologram recording to increase the field of view (the diameter of a beam illuminating the volume of the medium with particles). The lens represents a serial movie projection lens, 35KP-1.8/70 (USSR) with the following characteristics: focal length (f = 70 mm), relative aperture (1/1.8), and a diameter of the circle of confusion in the field at a wavelength of 0.55~10 µm. Figure 2 shows the 3D model of the DHC horizontal version. The horizontal design implies the vertical scanning of the water column for free flow of plankton in a vertical direction through the working volume of the camera (3). The design provides for four test objects (4) to be placed on the path of the laser beam in the working volume for further magnification calibration. The laser lighting module (2) and the recording module (1) with the CCD/CMOS camera are the main parts of the DHC, which are connected by a synchronization line. These parts are located in two strong, deep-water cases with portholes and connectors. All other DHC equipment is also placed in these two strong, deep-water sealed cases, which are structurally linked by a frame (6) to the base surfaces for interchangeable assembly. The cases and other devices of the complex are electrically connected by sealed connectors and offshore cables. a synchronization line. These parts are located in two strong, deep-water cases with portholes and connectors. All other DHC equipment is also placed in these two strong, deep-water sealed cases, which are structurally linked by a frame (6) to the base surfaces for interchangeable assembly. The cases and other devices of the complex are electrically connected by sealed connectors and offshore cables.
The lighting module (2) ( Figure 2) has a laser-diode fiber module (9) and an optical system to create an optical radiation beam (7,8). The recording module (1) includes an optical system that receives the optical radiation (7,8), a CCD/CMOS camera (10), and a control system for operating modes (synchronization). The units are optically connected by a mirror-prism system (5,11) that provides the recording of data in 0.75 L of the studied volume per one exposure.
The design of lighting and recording modules ensures the interchangeability of components ( Figure 3). This principle creates the conditions for maintainability and redundancy in unfavorable field conditions as well as upgradability, thus updating or expanding the DHC capabilities or adapting them to various research tasks. The lighting module (2) ( Figure 2) has a laser-diode fiber module (9) and an optical system to create an optical radiation beam (7,8). The recording module (1) includes an optical system that receives the optical radiation (7,8), a CCD/CMOS camera (10), and a control system for operating modes (synchronization). The units are optically connected by a mirror-prism system (5,11) that provides the recording of data in 0.75 L of the studied volume per one exposure.
The design of lighting and recording modules ensures the interchangeability of components ( Figure 3). This principle creates the conditions for maintainability and redundancy in unfavorable field conditions as well as upgradability, thus updating or expanding the DHC capabilities or adapting them to various research tasks.   Specifications of the DHC with a horizontal submersible device: Need to continuously calibrate the increase at depths greater than 500 m due to a significant change in the refractive index of water. In this case, holography is performed with the set calipers (4) (Figure 2).
An information unit of the system represents a set of equipment that performs the functions of communication, calculation, on-board power supply, control, and data switching.
The information unit with a connected DHC makes it possible to install and connect hydrophysical sensors accompanying the plankton measurements. Such an aggregation is traditionally called a probe, and in our case, the hydrobiological "DHC-probe Arctic" (hereinafter referred to as the "probe" (Figure 4)). The design of the DHC probe allows loading it to a depth of 1000 m.
Technical characteristics of the hydrobiological "DHC-probe Arctic": • • Wi-Fi backup channel to run and configure the probe for autonomous work; • 1 Gb Ethernet (to transfer data on plankton and hydrophysics).

DHC Software
The principle of building a complex (probe) as an information measuring system assumes the intelligent functionality of the sensor. In the DHC, these functions are fulfilled in the information unit using an on-board single-board computer with the basic DHC software supplemented by software focused on a measurement problem. This separation makes it possible to easily adapt the DHC to a particular task within the framework of one technical solution. The problem-oriented software can have three options:

•
To solve operational tasks according to the assessment of the integral characteristics of plankton with a high degree of averaging by volume; • To solve monitoring tasks related to the classification of plankton individuals and other particles according to preliminary defined criteria, a set of features, and databases of plankton in the studied water area; • To solve monitoring tasks related to the analysis of plankton behavior by identifying preliminary signs of behavioral responses.
The result of problem-oriented software is a stream of digital data on marine particles. In this regard, the DHC serves as the same data "provider" as other sensors of the information measuring system. The information unit has an SSD drive to store data on board the probe.  Four channels to connect with hydrophysical sensors (RS485 and RS232)temperature, pressure, microwave conductivity, and CTD (Valeport Mini); • Communication channels: • Wi-Fi backup channel to run and configure the probe for autonomous work; • 1 Gb Ethernet (to transfer data on plankton and hydrophysics).

DHC Software
The principle of building a complex (probe) as an information measuring system assumes the intelligent functionality of the sensor. In the DHC, these functions are ful- Problem-oriented software (DHC technology) makes it possible to automatically register a digital hologram (or a sequence of digital holograms) according to specified parameters and then, by mathematical processing, restore the image of each particle according to special computational algorithms. The mathematical processing of a digital hologram makes it possible to obtain images containing information on the distribution of both amplitude and phase of light waves [36,37]. The phase-contrast image of a particle gives information on the volume of the particle, but in order to obtain it, we need to use one of the unwrapping methods [36,37,40,41]. The three-dimensional coordinates of each particle (spatial distribution of particles in the studied volume), dimensions, space orientation, speed, and direction of movement of each particle can be determined within the in situ tasks according to the amplitude images of particles. Thus, the DHC technology allows for the creation of a virtual 3D image of the volume with the studied particles.
A lens makes the magnification ratio during the reconstruction of holograms dependent on the position of a particle and the refractive index of water. This dependence is described using four coefficients determined during calibration. The magnification ratio taken for different particle positions is one of the functions of the DHC software [42].
The DHC software includes a particle recognition and classification unit that is used to obtain information on individual particles of the medium and their belonging to various taxa. The values of features (decision tree) for automatic classification of particles are presented in Table 1. A rectangle is circumscribed around the reconstructed holographic image of a particle to ensure its automatic classification. Here it is critical to consider the length of this rectangle H and the morphological parameter M, which is the width-to-length ratio of the rectangle, as well as the presence of outgrowths in the zooplankton individual. The values of the relevant criteria for the taxonomy of certain particles are taken on the basis of an experiment, which is one of the tasks of the expedition measurements. Table 1. Features of taxa for automatic classification of particles: H-length of the rectangle circumscribed around the particle image; M-morphological parameter (width-to-length ratio of the rectangle).

Taxa Presence of Outgrowths
Taxa 1-7 in Table 1, characterizing mesoplankton, are chosen based on their highest prevalence in the aquatic medium. The Suspension taxon is used to determine the turbidity of the medium. It includes particles whose size is less than 0.2 mm, and the section shape is close to a circle, i.e., the height-to-width aspect ratio of the circumscribed rectangle is M ≥ 0.9. Their size and concentration are used to calculate the turbidity of water. Taxon 9 (Marine snow) includes particles that satisfy the definition of marine snow. It includes particles of aggregated phytoplankton, feces, and other detritus components [43][44][45]. To determine the size and concentration of floating bubbles in areas where methane is released from bottom sediments, the DHC software uses data on the Bubble taxon in the studied volume. Here we include data on particles that represent gas bubbles larger than 0.2 mm and whose section shape is close to a circle, i.e., M ≥ 0.9 (size H for the Bubble taxon means the diameter of a gas bubble). Offshore the East Siberian Arctic Seas, bubble methane is released because of the loss of continuity of submarine permafrost, which subsequently creates taliks and destabilizes hydrates [31][32][33][34]. Note that the diffusional transfer occurs during the upward movement between the primary methane bubble (in a thin bottom layer) and nitrogen and oxygen dissolved in water, thus increasing the nitrogen and oxygen fractions as the bubble approaches the water-air interface [33,46,47].
The processing results of each hologram are registered as a 2D image with sharp images of particles and a table with data for each particle detected by the algorithm in this hologram. Examples of such data are shown in Figure 5 and Table 2  The processing results of each hologram are registered as a 2D image with sharp images of particles and a table with data for each particle detected by the algorithm in this hologram. Examples of such data are shown in Figure 5 and Table 2. Figure 5. Example of the DHC data after processing in problem-oriented software at the AMK-82 station No. 6939. (a) Digital holograms and particle images reconstructed therefrom. The particle number is the ID number in Table 2. (b) A 2D view of the studied volume. (c) Patterning of the 2D view for particle recognition and classification (the color of the circumscribed rectangle indicates belonging to a particular taxon). Table 2 shows the degree of completeness and level of detail of the data recorded in a digital holographic experiment in situ.  Table 2. (b) A 2D view of the studied volume. (c) Patterning of the 2D view for particle recognition and classification (the color of the circumscribed rectangle indicates belonging to a particular taxon). Table 2 shows the degree of completeness and level of detail of the data recorded in a digital holographic experiment in situ.

Data Analysis
The integral characteristics of clusters of marine particles calculated based on the information contained in the classification tables are used to analyze and compare the DHC data with the measured values accepted in oceanography. The integral data for plankton usually include the concentration ( ssp /m 3 ) and biomass (1 mg/m 3 ) for each taxon. Some literature sources [48,49] provide various methods to determine plankton biomass, including those based on images of plankton individuals [50,51]. The DHC software can be adapted to the requirements of the chosen biomass counting technique. This study determines the maximum size H of the rectangle circumscribed around the reconstructed image of each plankton particle suspended in water, i.e., in its wet state. In this case, the density of a plankton individual of any taxon can be taken as equal to the density of water: g ≈ 1 mg/mm 3 [52]. To calculate the biomass by wet weight, each plankton particle is represented as an ellipsoid, whose volume is as follows: where a, b, c-semi-axis of the ellipsoid circumscribed around the particle, Mmorphological parameter, and H-maximum size of the rectangle circumscribed around the reconstructed image of a plankton particle. The total biomass of k taxon particles by wet weight will be as follows: where WW k i = gV k i = 1 6 π · H k i 3 · M k i 2 -wet weight (mg) of the i particle of the k taxon in the DHC measured volume-V 0 . The total biomass of all particles is the sum of the biomass of planktonic individuals of certain taxa.
Turbidimetric and sedimentological measurements are typically performed to clarify the hydrological state of the water area (for example, turbidity measurements). According to the DHC data, turbidity can also be estimated by making a histogram of the Suspension particle size distribution ( Figure 6). Data on the size and number of particles in the Bubble taxon are used to determine the volume content of methane. A laboratory bubble test was performed to verify the correct bubble size. The test used square test particles with a size of 500 × 500 μm for calibration (test objects in Figure 2). The Hailea ACO-5501 tank compressor was used as the bubble generator during the test. Capacity: 80 L per hour. Outlet pressure: 0.10 bar. The Hailea ACO-5501 is designed to ensure the effective aeration of water tanks.
Digital holograms of bubbles were registered against the background of test particles. A 2D display of holographic images of the water volume with bubbles was ob- Turbidity is taken equal to the total fraction of the volume section area covered by the sections of the Suspension particles: where S DHC -area of the DHC entrance pupil, S i -area of the image of the i-particle section, and n-number of particles of the Suspension taxon. Data on the size and number of particles in the Bubble taxon are used to determine the volume content of methane. A laboratory bubble test was performed to verify the correct bubble size. The test used square test particles with a size of 500 × 500 µm for calibration (test objects in Figure 2). The Hailea ACO-5501 tank compressor was used as the bubble generator during the test. Capacity: 80 L per hour. Outlet pressure: 0.10 bar. The Hailea ACO-5501 is designed to ensure the effective aeration of water tanks.
Digital holograms of bubbles were registered against the background of test particles. A 2D display of holographic images of the water volume with bubbles was obtained from the focused images of particles and bubbles reconstructed from digital holograms using the DHC technology ( Figure 7). Data on the size and number of particles in the Bubble taxon are used to determine the volume content of methane. A laboratory bubble test was performed to verify the correct bubble size. The test used square test particles with a size of 500 × 500 μm for calibration (test objects in Figure 2). The Hailea ACO-5501 tank compressor was used as the bubble generator during the test. Capacity: 80 L per hour. Outlet pressure: 0.10 bar. The Hailea ACO-5501 is designed to ensure the effective aeration of water tanks.
Digital holograms of bubbles were registered against the background of test particles. A 2D display of holographic images of the water volume with bubbles was obtained from the focused images of particles and bubbles reconstructed from digital holograms using the DHC technology ( Figure 7).  The size of the bubbles studied in the laboratory experiment was established by comparing the size of the bubble images with the known size of the test particles (model particles), and a histogram of the distribution along the bubble diameter was compiled ( Figure 8).
The formula for calculating the volume density of the Bubble taxon is used to calculate the volume content of bubbles: where V DHC -studied volume by the DHC, H i -size of the i particle (bubble) of the Bubble taxon, and n-number of particles of the Bubble taxon.
The gas flow delivered by the bubble, estimated at a given depth of the probe position, can be calculated using the following formula: where v i = g·H 2 i 18v S -ascending speed of I bubble [53], S =1 m 2 -sea water surface area accepted for analysis, v S = 1.787 · 10 −6 m 2 /s-sea water kinematic viscosity for 0 • C temperature [54], and g = 9.8 m/s 2 -gravity acceleration. It should be noted that the conversion of the bubble gas flow (a mixture of methane, nitrogen, and oxygen) into methane requires independent studies of the composition of bubbles coming from various types of bottom sediments in the Eastern Arctic Seas, as well as the study of the methane fraction dynamics in floating bubbles of different sizes under various hydrological conditions [47]. Appl The size of the bubbles studied in the laboratory experiment was established by comparing the size of the bubble images with the known size of the test particles (model particles), and a histogram of the distribution along the bubble diameter was compiled (Figure 8). The formula for calculating the volume density of the Bubble taxon is used to calculate the volume content of bubbles: where -studied volume by the DHC, -size of the i particle (bubble) of the Bubble taxon, and n-number of particles of the Bubble taxon.
The gas flow delivered by the bubble, estimated at a given depth of the probe position, can be calculated using the following formula: where =

Methods and Equipment for Verification and Comparison
Data for verification and comparison are obtained by standard methods, namely: • For plankton-by vertical net catching followed by sample fixation, taxonomic determination, counting of individuals, and laboratory measurement under a microscope; • For suspension-by processing turbidimetric turbidity measurements; • For floating bubbles-by using hydroacoustic data of seeps-places of massive release of bubble methane-previously discovered in the Eastern Arctic Seas [31,33,47,55].

Net Sampling of Plankton
Zooplankton was collected by vertical net sampling in the water column using the Juday net with a mesh size of 180 µm and a net inlet diameter of 0.37 m. Zooplankton samples were concentrated and fixed with a neutral 4% formaldehyde solution. Subsequently, Rose Bengal dye-stained samples were analyzed in the Bogorov chamber under a stereomicroscope. The organisms were identified according to types, classes, and orders using determinants [56,57].

Turbidimetric Measurements
This is a standard integral method used to determine the turbidity on a study scale of about 1 cubic meter by measuring the scattering of light passed through the water column. The Sea-Bird Scientific USA CTD probe SBE 911plus was used for these purposes and was additionally equipped with a turbidity sensor by WETLabs ECO NTU. The comparison of turbidity versus suspended matter concentrations measured by the weight method shows that the sensor data most likely reflects the real long-term variability of the suspended matter content in the Eastern Arctic Seas.

Acoustic Survey
This is an integral method that uses standard or specialized echosounders and sonars. This method, which is based on the generation of an acoustic pulse and the reception of a scattered acoustic signal, is the most effective for observing floating bubbles and obtaining estimates on the amount of methane delivered by them to the water and surface layers of the atmosphere (Figure 9). Modern hydroacoustic tools are able to detect single floating bubbles at distances of up to 2 km or more, estimate their size, and track their ascending speed [12,47,55,[58][59][60][61][62][63][64]. The experiment in places marked with an arrow was carried out by simultaneously holographing gas bubbles using the DHC. The comparison of the DHC and acoustic data for methane bubbles is not presented in this paper, since the issue of synchronizing acoustic and holographic data on this expedition could not be addressed. Some preliminary results from the expedition are discussed below.

Results and Discussion
The classification features of the algorithm were sufficient to explicitly identify the terrigenous suspension, mesoplankton, zooplankton, marine snow, and floating methane bubbles released from the bottom sediments of the Eastern Arctic Seas. We compared the corresponding DHC results with independent data obtained under the same conditions by other research teams during the AMK-82 expedition to assess the correctness of the processing algorithm (and therefore the choice of the pre-set values of the H and M criteria).
The objects of interest shown in Table 3 were described depending on the geo- The experiment in places marked with an arrow was carried out by simultaneously holographing gas bubbles using the DHC. The comparison of the DHC and acoustic data for methane bubbles is not presented in this paper, since the issue of synchronizing acoustic and holographic data on this expedition could not be addressed. Some preliminary results from the expedition are discussed below.

Results and Discussion
The classification features of the algorithm were sufficient to explicitly identify the terrigenous suspension, mesoplankton, zooplankton, marine snow, and floating methane bubbles released from the bottom sediments of the Eastern Arctic Seas. We compared the corresponding DHC results with independent data obtained under the same conditions by other research teams during the AMK-82 expedition to assess the correctness of the processing algorithm (and therefore the choice of the pre-set values of the H and M criteria).
The objects of interest shown in Table 3 were described depending on the geographical location of the stations. Furthermore, the data are given in accordance with these descriptions. A map of the stations considered in this work is shown in Figure 10.  The operating characteristics of the DHC probe at the stations are given in Table 4.  The operating characteristics of the DHC probe at the stations are given in Table 4.
If the performance characteristics of the DHC are evaluated by the station No. 6961 with the largest number of holograms, then the performance characteristics per 1 m of profile are as follows: • Time for recording and reading the holographic sample data~3.5 min/m; • Required memory, taking into account the post-processing data~400 MB/m; • Data processing time~1.2 h/m. Figure 11 shows the results of the DHC's performance on the registration of changes in the depth of concentrations of Mesoplankton, Marine snow, and Suspension taxa at two stations. The comparison of the graphs shows that:

•
The measurements in the Kara Sea were taken near the Ob River estuary, which showed a very high content of the river-borne terrigenous suspension [65]. These factors determine the high turbidity of the water area and the reduced content of plankton; • The water was more transparent, and there was more plankton in the Laptev Sea outside the area of the river flow.   Tables 5 and 6 compare the results of plankton number measurements by sampling and analysis with the DHC measurements of plankton.
The analysis of these tables shows that, along with some differences, the classification performed by different methods matches in most cases. Below are the most probable sources of mismatch related to the peculiarities of selection:

1.
Different submergence depths of the net and the DHC, different averaging volumes, and different sea states affect the camera shooting quality, as illustrated in Figure 12; 2.
The blur of the shape parameter of the Others taxon. The mesh size of the net here is 180 µm. For the DHC, the minimum size is H = 200 µm adopted in the classification algorithm (Table 1). Unlike other taxa, this group includes organisms of various shapes. In addition, particles of non-living matter may also belong here.      The analysis of these tables shows that, along with some differences, the classification performed by different methods matches in most cases. Below are the most probable sources of mismatch related to the peculiarities of selection: 1. Different submergence depths of the net and the DHC, different averaging volumes, and different sea states affect the camera shooting quality, as illustrated in Figure 12; The analysis of these tables shows that, along with some differences, the classification performed by different methods matches in most cases. Below are the most probable sources of mismatch related to the peculiarities of selection: 1. Different submergence depths of the net and the DHC, different averaging volumes, and different sea states affect the camera shooting quality, as illustrated in Figure 12; The analysis of these tables shows that, along with some differences, the classification performed by different methods matches in most cases. Below are the most probable sources of mismatch related to the peculiarities of selection: 1. Different submergence depths of the net and the DHC, different averaging volumes, and different sea states affect the camera shooting quality, as illustrated in Figure 12; Due to these reasons, the DHC performance error is more accurately established through the comparison of data from the automatic and expert classification of images of plankton individuals reconstructed from holograms. For station No. 6961 (Table 6), this single variation does not exceed 30%, depending on the taxon. Figure 13 shows the comparison of turbidity measurements using the DHC technology and standard turbidimeter data. A good correlation in the data at a significance level of 0.05 indicates the high potential of digital holography for turbidimetry purposes. The correlation coefficient between turbidity measurements using the turbidimeter and the DHC was 75.5%.
Data on methane bubbles away from gas flames in the East Siberian Sea at station No. 6962 and in the Laptev Sea at station No. 6947 are not presented in this paper, since the issue of synchronizing acoustic and holographic data on this expedition could not be addressed. The estimated submergence trajectory of the digital holographic camera shown in Figure 9 by the green line is not certain due to the drift in relation to the research vessel. Highly scattered holographic survey data should be averaged not over five samples at one point taken adrift of the vessel, but over a large area~1 m 2 of the water surface along the vessel journey. The number of samples in this case increases to 1000. Then the holographic flow measurements will be more representative.
The reverse acoustic dispersion profiles shown in Figure 9 confirm that the holographic results are applicable only for in situ point measurements performed at different distances from gas jets (probe position is marked by a black arrow) and, so far, cannot be used for a representative estimation of the integral bubble flow. Figure 14 shows the results of the study conducted in a shallow seep in the Laptev Sea at station No. 6975 in the surface layer. plankton individuals reconstructed from holograms. For station No. 6961 (Table 6), this single variation does not exceed 30%, depending on the taxon. Figure 13 shows the comparison of turbidity measurements using the DHC technology and standard turbidimeter data. A good correlation in the data at a significance level of 0.05 indicates the high potential of digital holography for turbidimetry purposes. The correlation coefficient between turbidity measurements using the turbidimeter and the DHC was 75.5%. Figure 13. Turbidity results using the DHC technology (red dots) and a standard turbidimeter (blue dots).
Data on methane bubbles away from gas flames in the East Siberian Sea at station No. 6962 and in the Laptev Sea at station No. 6947 are not presented in this paper, since the issue of synchronizing acoustic and holographic data on this expedition could not be addressed. The estimated submergence trajectory of the digital holographic camera shown in Figure 9 by the green line is not certain due to the drift in relation to the research vessel. Highly scattered holographic survey data should be averaged not over five samples at one point taken adrift of the vessel, but over a large area~1 m 2 of the water surface along the vessel journey. The number of samples in this case increases to 1000. Then the holographic flow measurements will be more representative. The reverse acoustic dispersion profiles shown in Figure 9 confirm that the holographic results are applicable only for in situ point measurements performed at different distances from gas jets (probe position is marked by a black arrow) and, so far, cannot be used for a representative estimation of the integral bubble flow. Figure 14 shows the results of the study conducted in a shallow seep in the Laptev Sea at station No. 6975 in the surface layer. At this point, the Academician Mstislav Keldysh research vessel was kept above the seep, which made it possible to obtain new interesting facts on the dynamics of the shape of floating bubbles, including the aggregation effects, which are one of the least studied areas in the field of bubble gas transfer in the bottom sediment-water column-