In the following, we describe highly-automated image analysis approaches that help us classify satellite images, and allow us to monitor dynamical changes in image time series. We concentrated on flooding events within the Danube Delta as seen by the European Sentinel-1 and Sentinel-2 satellites, and describe systematic processing approaches to extract pre-defined categories from the image data (being either Synthetic Aperture Radar or multispectral images). One basic tool to monitor dynamical changes is to analyze and compare the compressibility of image patches using their Normalized Compression Distances. These distances can be converted into similarity matrices providing reliable maps of surface changes. The accuracy of these change maps was quantified for several typical test cases. In addition, we analyzed the performance of an alternative active learning approach, where Gabor filters and Weber local descriptors were used to extract features from image patches that were classified and semantically annotated. Then one can perform data analytics and generate maps based on the extracted semantic annotations; again, we used several representative test cases for benchmarking.

Our previous work showed the efficacy neural network (NN) approaches for satellite detection of Karenia brevis (KB) harmful algal blooms (HABs) in the West Florida Shelf (WFS). Applying a multiband NN, trained on a wide range of synthetically simulated inherent optical properties (IOPs), the NN takes Visible Infrared Imaging Radiometer Suite (VIIRS) remote sensing reflectance (Rrs) measurements at 486, 551 and 671 nm bands as inputs and produces related IOPs at 443 nm, including the absorption coefficient of phytoplankton, (aph443) as retrieved output images. We convert the retrieved aph443 values into equivalent chlorophyll-a [Chla] and KB HAB concentration mages, using known empirical relationships. We then compared VIIRS NN retrievals of KB HABs in the WFS with those obtained using other algorithms, and evaluated their accuracy against near co-incident in situ cell count measurements available over the 2012-16 period. Results highlighted the important impact of rapid (15-20 minutes) temporal variations on retrieval accuracy, and showed that the NN technique exhibited the highest accuracy statistics for retrievals of KB HABs in the WFS. This paper presents significant results in three areas. (i) We extend evaluations and comparisons of NN retrievals of aph443 and [Chla] with retrievals from other ocean color (OC) algorithms to waters beyond the WFS, including both complex coastal and open ocean waters, along Florida and Atlantic coasts, with a large dynamic range of chlorophyll-a values. We also now add in situ radiometric measurements to obtain Rrs inputs to retrieval algorithms. In contrast to satellite Rrs retrievals, these are invulnerable to atmospheric transmission correction errors. This permits comparison of in situ radiometric measurement-based retrievals with simultaneous co-located satellite retrievals and validation against in situ sample measurements. This allows us to isolate different factors affecting retrieval accuracy and evaluate the intrinsic merits of different algorithms unencumbered by inadequate/erroneous atmospheric transmission assumptions and/or satellite instrumental calibration limitations. Results obtained extend and demonstrate the efficacy of NN algorithms to widely varying waters beyond the WFS, (ii) Since it is conjectured that in satellite retrievals, it is the deep blue wavelengths that are more detrimentally affected by atmospheric correction inadequacies, we examined impacts on algorithm retrieval accuracy when deep blue wavelengths are used for retrieving Rrs values. Retrievals using NN and OCI/OCx algorithms were compared against in situ sample measurements, using in situ radiometric Rrs measurements and satellite Rrs retrievals as inputs, first at 443 nm (deep blue) and then at 486 nm (non-deep blue). The results unambiguously show that satellite retrieval accuracy, as well as intrinsic retrieval accuracy from in situ radiometric measurements are improved when deep blue wavelengths (443 nm) are avoided in favor of non-deep blue wavelengths (486 nm), raising issues of both atmospheric correction as well as possible underwater spectral interference in CDOM rich and complex waters. Thereby arguing for use of OC algorithms using the longer wavelengths. (iii) Finally, new quantitative analysis of temporal, intra pixel and sample depth variability highlights their important impact on retrieval accuracy.

In the effort to design a more repeatable and consistent platform to collect data for Structure from Motion (SfM) monitoring of coral reefs and other benthic habitats, we explore the use of recent advancements in opensource GPS-guided drone technology to design and test a low-cost and transportable small Unmanned Surface Vehicle (sUSV). The vehicle operates using Ardupilot open source software and can be used by marine managers to map and monitor marine environments in shallow areas (<20m). The imaging system uses two Sony a6300 mirrorless cameras to collect stereo photos that can be later processed using photogrammetry software to create underwater high resolution orthophoto mosaics and digital surface models. The propulsion system consists of two small brushless motors powered by lithium batteries that follow pre-programmed survey transects and are operated by a GPS-guided autopilot control board. Results from our project suggest the sUSV provides a repeatable, viable, and low-cost (<$2000 USD) solution for acquiring images of benthic environments on a frequent basis from near the water surface. These images can be used to create SfM models that provide very detailed images and measurements that can be used for monitoring changes in biodiversity, reef erosion/accretion, and assessing health conditions.

Sun light reflected from sea surface forms a high intensity solar ray. Sun-glint is created by Fresnel reflection which contaminates ocean color products. It covers up the real physical characteristics of water bodies and interferes with the inversion of ocean remote sensing data precision. In this paper, sun glint regional characteristics are simulated and analyzed in consideration of the different satellite attitude angles and time phases based on Cox–Munk model. The numerical simulation shows that through the satellites adjust attitude dynamically, the contamination area decreases by 11.8%.Onorbit test data are used to evaluate the method precision and it will contributes to the estimation of sun glint regional characteristics and strategy of on-orbit use.

This paper attempts to exploit how the geometry of the radiative flux, sensor characteristics, atmospheric effects and particularities inherent to water bodies can affect water quality indicators parameters data retrieved from satellite imagery. This was done by comparing data from bio-optical models applied to Sentinel-2 MSI imagery with field samples from 12 campaigns ranging from 2016 to 2019 at the Jupia Reservoir (Brazil), along with analyzing the spectra of different atmospheric correction algorithms, in the search of possible natural effects that hampered the quality of the satellite-derived data. Compared water quality parameters were: Turbidity, chlorophyll-a, total suspended matter and Secchi disk depth. Results showed that, the data presented very low, if not inexistent, correlation. However, the error values estimated reflected that the data were not too far apart, despite having no correlation. The effects that may have caused the low correlation and the errors were analyzed through the retrieval of the spectra in the field sampling points. The compared spectra showed that, effects of haze caused by aerosols, bottom reflectance from optical depth and the presence of submerged plants were the most critical reflectance altering phenomena, which reflected in the extracted water quality data. Inelastic scattering, fluorescence and ozone layer influence were undetected, while the adjacency effect and Sun glint presented little to no effect on the data. Further focused analysis of these specific effects is a promising field of study in order to improve atmospheric correction and bio-optical algorithms.

A novel polarization camera (Teledyne DALSA) based on the Sony first polarization imaging sensor provides a high resolution broadband image of the object in 400-900 nm spectral range, where each pixel contains four subpixels bearing built-in linear polarizers oriented at the 0, 45, 90 and -45 degrees. With an additional lens it has a field of view (FOV) of about 40° degrees. The camera was equipped with a filter wheel containing five band-pass filters, thus providing multispectral, multi-angular high quality polarimetric data with high spatial resolution. The camera has undergone radiometric calibration and was used in various illumination conditions and water environments in NYC area. Spatial and temporal distribution of water and sky Stokes vector components were characterized at various wind speeds. Polarization images were also used for the estimation of wave slope statistics from the ocean surface. Results are compared with concurrent measurements of total and polarized radiances by a state-of-the-art snapshot hyperspectral imager, which simultaneously acquires spectra with 4 nm spectral resolution in the wavelength range of 450- 750nm, also with a 40° FOV for 20 - 60° viewing angles. A computer-controlled filter wheel was installed in front of the imager, allowing division-of-time Stokes vector images from the ocean surface. The combination of these two instruments allows to observe spectral and polarization properties of the wind-roughened ocean at high spatial and temporal resolution, with the goal of advancing our understanding of the air-water interface and in-water light propagation.

Multi-view learning (MVL) is a technique which utilizes multiple views of data simultaneously during training to learn more expressive representations. Multi-view learning has been gaining a large amount of interest in various machine learning applications recently. In this paper, we focus on learning representations prior to classification using multi-view learning via deep canonical correlation analysis (DCCA) in hyperspectral image processing. We propose a classification framework including a proposed view generation approach. The motivation of our proposed view generation approach is to fuse spatial and spectral information. The performance of our proposed view generation approach is compared with the other view generation methods in the literature; namely the uniform band slicing and correlation-partition-based clustering. To evaluate the effectiveness of the proposed approach, we performed experiments on two commonly used hyperspectral image datasets. Experimental results based on two hyperspectral image datasets demonstrate that the proposed classification framework provides satisfactory classification performances.

A water wave glint correction algorithm, light attenuation depths and water depths estimated from satellite imagery in the Indian River Lagoon and Banana River estuarine waters provide a means to evaluate light penetration or water clarity. The shallow water estuarine study area is along Florida’s east coast area known as "space coast” Florida. In-situ data, including water depths obtained from bathymetric charts, in-situ hyperspectral attenuation signatures and satellite- based reflectance factors are obtained from atmospherically corrected visible and near infra-red channels of World View-3 satellite imagery. The high spatial, radiometric and 11bit digital resolution satellite imagery is used to study water visibility that is an important ecosystem quality indicator. In shallow coastal waters as well as deeper water areas and nearby coastal waterways, the effects of water surface gravity wave facets and sun glint influence the estimates of water quality related variables estimated at the pixel level. The results have significant implications related to the estimation of the depth of penetration of light using satellite-based methods in coastal waters. Application of atmospherically corrected satellite imagery followed by glint removal is shown using translating and dilating derivative wavelet algorithms to estimate the attenuation depth of light in water. The research has been conducted in order that results may be transferable to other estuaries that suffer from reduction in light penetration due to factors such as turbidity or harmful algal blooms.

Water surface gravity wave imagery and analyses are presented. The approach is to utilize a protocol to characterize small wind driven surface gravity waves in shallow waters near shorelines or in shallow open coastal waters. Video imagery collected from the littoral zone and a drone are used to better understand and predict wave patch characteristics and the energy (watts m^-2) contained in a patch of small wind driven gravity waves. The analysis of the airborne and ground-based video imagery makes use of space time image analysis techniques. Video imagery collected from ~20 to 120 frames per second, hyperspectral imagery, and wave gauges are presented. The techniques, along with an in-situ line target allow the calibration of wave features such as wave amplitudes and wavelengths. The scientific methods have applications related to management of coastal lagoons, estuaries, near coastal waters. Applications in coastal engineering such as wind farms, protection of structures (canals, seawalls, docks) and vegetated shorelines can benefit from improved understanding of wind driven gravity waves. Data from the Indian River Lagoon and Banana River in Florida are used to demonstrate the protocol and techniques.

Surface emissivity (ε) and Temperature (T_s) are two key parameters to monitor climate and meteorological changes. Because of its high spatial resolution, repeat time and very long period of activity, SEVIRI (Spinning Enhanced Visible and Infrared Imager) on board of MSG (Meteosat Second Generation) geostationary platform, allows us to perform accurate retrieval of these parameters. To take advantage of the data information content, a Kalman filter methodology was developed and implemented for the the simultaneous retrieval of surface emissivity and temperature from SEVIRI infrared radiance measurements (channels at 8.7, 10.8 and 12 μm). That developed is one of the very few physical based approaches for the estimation of surface parameters from infrared instruments on board geostationary platforms and it has shown an accuracy of ±0.005 and ±0.2 K, for surface emissivity and temperature respectively. Based on this Kalman filter methodology a L2 processor has been developed to provide ε and T_s in real time, making it very attractive for application in different fields. The processor was applied and tested comparing its results with other satellites retrievals and analysis and here has been applied to the retrieval of Sea surface temperature retrieval of Mediterranean Basin for the years 2013-2017. First results show for the Mediterranean Sea temperature an increasing trend of 0.23 K/year.

In this work we present the results of internal solitary wave (ISW) observations derived from high-resolution spaceborne synthetic aperture radar (SAR) measurements acquired over the seasonally ice-free Chukchi and Beaufort Seas. The information on internal wave properties and locations was obtained from Envisat Advanced SAR (ASAR) images acquired during May-October 2007 and 2011. Altogether, 233 ISW packets were identified in 727 ASAR images. As observed, there is no significant difference in the normalized number of ISW observations between 2007 and 2011, while most of the waves were detected in July (33%) and August (43%). Geographically, most of the waves were observed over shelf and shelf break regions (94%), while only 6% of them were found over depths exceeding 200 m. In the Chukchi Sea, the waves were frequently observed in the northern part of the sea centered at about 72.6° N, 168° W, in its southwestern part, around Herald Shoal, near the Barrow Canyon, and in the Bering Strait. In the Beaufort Sea, some ISW activity was observed along the shelf break and north of Cape Bathurst, while a major hot spot of ISW generation was found on the Mackenzie Shelf. Maps of selected ISW properties, including their propagation direction, crest length and wavelength of leading waves provide further details about spatial characteristics of the observed internal waves.

In 2016, the first Chinese synthetic aperture radar (SAR), the Gaofen-3 (GF-3) satellite, was launched. Unlike the single-polarized wave mode SARs in Europe, GF-3 is the first satellite acquiring the quad-polarized SAR data in wave mode configuration, which could benefit the ocean wave estimation. Here, the ocean wave spectra estimation from quad-polarized GF-3 wave mode is presented and its performances are evaluated. For the period from January to October in 2017, the ocean wave spectra were inverted from GF-3 wave mode data. The quad-polarized SAR-ocean spectra inversion scheme was utilized, in which the azimuthal and range wave slopes are obtained from the vertically and linearly polarized normalized radar cross sections and then converted to ocean wave slope spectra. The validation was also performed through comparisons against directional wave buoy observations. The spatio-temporal criteria of 100 km and 0.5 h, yield 87 matchups. Two representative cases illustrate the consistency between the GF-3 SAR ocean wave spectra and buoy measurements. Statistical assessment shows the root mean square error (RMSE) of 0.35 m, 19.52 m and 24.89° for the significant wave height, peak wavelength and wave direction, respectively. Evaluation results indicate that the quad-polarized algorithm is suitable for spectral ocean wave estimation from GF-3 wave mode, and encouraging for operational implementation.

River flow is the main source of dissolved and suspended substances entering to the sea, including products of anthropogenic pollution. Coming from the river to the sea, the continental waters form submesoscale structures adjacent to the estuary, distinguished by low salinity and temperature different from the surrounding, and also, increased turbidity, high content of suspended matter and dissolved organics. Such structures are called "Plumes". Traditionally, we can divided two research methods - contact (in-situ/ship measurements) and non-contact (remote sensing data). But, basically the joint use of in-situ measurements and remote sensing data in existing scientific works - is nominally, each source of information is used separately, and more often direct (in-situ) measurements are preferred, and remote sensing data is used as an auxiliary method for qualitative descriptive works. And there is still no reliable information on the extent to which quantitative estimates can be trusted, obtained from remote sensing data, especially in such difficult estuary areas. In this paper, we present the results of verification satellite data in the optical range with the aim of comparing quantitative data in the river-mouth area obtained by contact (field measurements near river mouth in northeastern part of Black sea) and contactless methods.

During five years, in the frame of the NAOMI (New Advanced Observation Method Integration) research project, Total and ONERA have worked on radar and optical imagery to detect, characterize and quantify slicks at sea. Laboratory and pool measurements, physical modelling and offshore experiments have been combined to fully understand the signal collected over slick-covered area. As the measured signal is analytically expressed according to the geophysical parameters of the imaged slick, it enables to fully monitor the ocean surface: is a slick present? What kind of slick is it (extremely thin or not)? Is it a known product (existing in the data base)? Can the thickness be probed by the used of optical or radar device? What is the slick volume? In the Health Security and Environment (HSE) context, an exhaustive measurement campaign can be done in order to create a data base with hydrocarbon or hydrocarbon emulsion signatures, extinction coefficients, skin depths, minimum thicknesses perceptible thanks to extinction and thickness values. Thus, it offers more processing options in the optic branch of the tool to monitor the slick. Depending on the available data, optical and/or radar imagery, the capability of slick detection, characterization and quantification will be presented. After a recall of the HSE specificity, the paper will give an overview of the main features of the input data that is to say SAR and optical images. Then, based on modelling results, the optimal observation conditions for radar and optical imagery will be introduced. Afterwards, capability of detection will be described and illustrated for both the radar and the optical case. In the optical domain, the process will distinguish at least two classes: thin and thick. In the HSE context, a database can be used to identify some detected products. The last step is quantification. A sophisticated method, relying on L band radar imagery, will be used to identify pixels covered by a film, meaning presence of oil at the surface, and the ones for which the oil may be as droplets in the volume. The traditional use of SAR data is also extended to the estimation of the oil concentration within an oil and seawater mixture. For optical data, the most direct quantification process relies on automatic Bonn code classification. The code links a class with a range of thickness and computes a minimum and a maximum volume of product in each class. If the product is in the data base a more suited classification and volume assessment can be done. If the thickness is too thin (spectral signature due to absorption is too weak) or too thick (only the upper part of the product layer contributes to the signal), a thickness estimated thanks to pool experiment is associated to each class enabling to compute a volume per class and a global volume. In the other cases, in a near future, modelling would enable to assess the thickness. Concerning hydrocarbon emulsions, modelling in the optical domain is in progress in order to predict skin depth and to derive water content.

Evolution of film slicks on the sea surface is a very important problem, particularly in application to monitoring of pollution transport in the ocean and in the coastal areas. The geometry of film slicks is determined by physical characteristics of surface films and by environmental parameters. At present our understanding of the role of geophysical processes controlling the slick dynamics is still insufficient. This is partly because of the luck of systematic studies of film slicks in controlled experiments. A perspective approach to the problem solution is organization of special experiments with artificial slicks formed by films with known characteristics. Previously we reported on the results of some experiments and proposed a simple model of spreading of surfactant spills accounting for the surface stresses induced by wind waves. In this paper new results of experiments on oil spill evolution are presented and physical mechanisms of this evolution are discussed. The experiments were conducted on the Gorky Water Reservoir. Surfactants (oleic acid) were spilled out from a motor boat. The slick shape was studied using aerial photography and a methodology of contouring slicks using a motor boat with a GPS receiver onboard. It is obtained that the cross-wind slick size grows quite slowly with time being almost independent on wind, while the growth rate of the along-wind axis increases rapidly with wind velocity. To explain the observed effects a mechanism of “oil overflow” is hypothesized. It describes the movement of surfactants from an upwind slick side to its downwind side due to the action of along-wind wave induced stresses. This process should contribute also to some additional drift velocity of slicks. The conclusion about the drift is consistent with a hypothesis, discussed in the literature that oil slicks move faster than the surrounding water surface microlayer.

Damping of gravity-capillary waves on the water surface covered with surface films has been well studied for the case of monomolecular films. In this case, the damping is determined by the elasticity of the film and the surface tension coefficient. The parameter of elasticity itself was introduced for a monomolecular film. The elasticity is determined by dependence of the surface tension on surfactant concentration. Therefore, the usage of this parameter when describing an oil film of nonzero thickness is generally not indisputable. In this paper, we justified the capability of the usage of a phenomenologically introduced parameter - elasticity for quantitative description of the wave damping due to oil films in the wide range of oil film thickness. It is shown that the value of the elasticity dramatically affects dependences of wave characteristics on film thickness. The analysis is based on comparison between results of our laboratory measurements and numerical calculations carried out in the framework of a model of two viscous layer fluids with elastic interface.

The salinity of the Arctic Ocean is influenced by the fresh water input from sea ice melting, river discharge and precipitation, and its variation could provide a useful reference for the study of changes in Arctic environment. Since the SMOS satellite launched in 2009, there have been three microwave salinity satellites (Aquarius, SMOS and SMAP satellites) that could provide us sea surface salinity (SSS) data from space. In this study, we compared the SSS products derived from these three salinity satellites: the Aquarius satellite data have a limited spatial resolution and a short time series of less than 5 years, the SMOS SSS product for the Arctic from Barcelona Expert Center provide a 7-year record of the Arctic SSS in 2011-2017, the SMAPSSS data is more sensitive in low salinity regions. With the data from the SMAP and SMOS satellites, we presented a general view on summertime SSS variation during 2011-2017 in the Arctic area. The SSS maps showed that the SSS mainly varied in the plume area and were stable in the Arctic open sea. Satellite-derived salinity could detect the extremely low SSS area caused by sea ice melting and continental runoff. We extracted the ice-melt water area by combing the microwave salinity satellite data and the ocean color satellite data, which could exclude the continental fresh water input. The result showed consistence with sea ice concentration and former literature. The distribution of the ice-melt water could be of great significance for the study of Arctic sea ice anomaly and relative biogeochemical changes.

An increasing number of underwater construction projects demand a solution for precise and highly resolved infrastructure monitoring which cannot be addressed by conventional sonar systems. We present an optical ranging system based on the pulsed time-of-flight method incorporating a signal-processing toolchain especially equipped for mapping technical infrastructure. This allows us to handle multiple returns from objects which are only separated by a 15 cm gap. A high lateral resolution of 4 mm in combination with a high measurement rate of 50 000 measurements per second allows the reconstruction of fine details. The measurement range can reach several 10 m, depending on the water quality. Further, we present the system design of our scanner and discuss first results obtained in an underwater measurement range.

LED mini lidar has been developed and installed into the sea wave dynamics visualization. This study aims mini-lidar installation to small fishing ship and pleasure boat. In the viewpoints of its safety and robustness of the installed system, LED pulse beam was selected as the lidar light source. Its shallow angle observation visualizes the sea wave motion from just a front of the ship to 300m ahead. At the coast experiment, two types of LED mini lidars were used for polarization measurement and Raman scattering measurement. They were almost the same optical characteristics except for their LED wavelength, that is, 365nm for polarization lidar and 365nm for Raman lidar. To visualize the sea wave dynamics, the summation time was shortened to 0.2s. The shallow angle (depression angle) of the lidar observation was set in 2 – 10 degrees against the horizontal direction. The fundamental information of sea wave, which were wave speed, height, period and wavelength, was deduced from the lidar echo. The polarization echo measurement revealed its dependency due to the LED beam incident angle against the sea surface. The Raman echo measurement visualized the correlation between the surface atmosphere activity and the sea wave dynamics. The surface atmosphere activity on water vapor Raman echo related to the sea wave motion from the fluorescent backscattered light of sea water, while it had its original change. These results are valid for safe cruise of small boat. This additional sea wave activity information will be effective for low fuel consumption operation and the automatic control in safety cruise. They are also of interest in physical oceanography.

The airborne lidar with two segmented field-of-view (FOV) receivers was used to detect the subsurface scattering layers. Significant differences were observed in the waveforms from one channel with small FOV of 6 mrad and the other channel with larger FOV of 40-6 mrad. The larger FOV of 40-6 mrad was to provide a larger dynamic range for the deep-water signal detection. A small-angle approximation based Lidar waveform simulation model was developed, and found that these differences are owing to the narrow beam divergence of laser pulse of only 0.3 mrad. Next, an algorithm, which incorporates a waveform-decomposition technique and a lowpass digital differentiator, was then used to detect the scattering layers from both small- and large- FOV channels. The observation of scattering layer along the coastal region of Sanya Bay of China shows that, more than three thin scattering layers can be found in the same water column close to the coasts, and the maximum depth of the scattering layer detected by the large FOV Channel can be up to 35m, and internal waves can be detected from spatial distributions of scattering layer. It can be found that the airborne bathymetry lidar with segmented field-of-view receivers can also be a great tool for the subsurface scattering layer detection.

This paper is devoted to the study of the possibilities of restoring the concentrations of chlorophyll-a (Chl-a) and total suspended matter (TSM) in fresh waterbodies using the images of Sentinel-2 / MSI of the Gorky reservoir as an example, validated via the high-resolution ground-truth data obtained by fluorescence lidar UFL. This water area in the period of intense chromaticity of cyanobacteria is distinguished by their completely heterogeneous spatial-temporal distribution with scales of patchiness from several tens of meters and high temporal variability. For this reason, direct methods of “spot” measurements in place on water samples in a small part of the reservoir for 3 hours relative to the satellite’s span do not provide us with reliable and statistically proved data. In this regard, the work proposed an original method of high-performance sub-satellite synchronous measurements of the brightness of the water column, concentrations of Chl-a and TSM with a fluorescent lidar onboard a high-speed vessel. This technique is provided with a spatial (8 m) and a temporary (1 s) resolution. Based on this, 4 independent regional models for the recovery of Chl-a concentrations, as well as TSM were developed based on in-situ measurements of the brightness of the water surface, or from Sentinel-2 / MSI images. To ensure the possibility of using satellite images, the analysis of atmospheric correction algorithms was performed and the best one was determined. The proposed models can be used for regular environmental monitoring of the Gorky Reservoir according to the in-situ ship measurements of the brightness of the water surface or from Sentinel-2 / MSI images. When using the proposed methodology for lidar high-performance collection of sub-satellite data, the developed models can be adapted for limnological monitoring of other freshwater bodies with high spatial and temporal variability of the optical properties of water.

The results of multi-year satellite monitoring of ice cover in the Caspian Sea are presented. The basis of the study was remote sensing data in the optical range, such as Sentinel-2 MSI, Landsat-8 OLI, Landsat-7 ETM+, Terra/Aqua MODIS, as well as archival Landsat-5 TM data. Also, Sentinel-1 SAR-C and archival ERS-1/2 SAR and Envisat ASAR radar data were used. The main objectives of the study were: determination of areas occupied by ice, identification of trends in ice formation and clearing periods, their inter-annual variability. To solve the problem of determining the parameters of ice cover, we used the multisensory approach whenever possible, i.e. comprehensive integrated analysis of heterogeneous data. The analysis of a vast array of satellite data, differing in nature and spatial resolution, and ice mapping was carried out using the toolkit of the See the Sea (STS) information system developed at the Space Research Institute of the Russian Academy of Sciences (IKI RAS). STS enables working with multi-year archival data, conduct joint analysis of different satellite data together with hydrometeorological information. It was found that during the study period, from the 1998-1999 winter to the present time, the largest area of ice cover of over 90,000 sq. km was observed in the 2011/2012 winter period. Ice was observed even in the bays of the Southern Caspian. Another direction of research, the results of which are presented, is the identification of capabilities and limitations in the use of satellite data obtained in different ranges (radar and optical) and at different polarizations.

In preparation for a research project focused on the risk of the sea level rise (SLR), and on the basis of previous studies, we started to investigate the ice loss by some marine-terminating ice tongues in Antarctica in the last 10 years. The first test site selected for this study was the Erebus Ice Tongue (EIT) which stretches from Ross Island into the Ross Sea. The analysis was carried out using Cosmo-SkyMed (CSK) images from the catalogue of the Italian Space Agency (ASI) and from a recent acquisition. The scene under analysis includes various elements such as sea-ice, open-sea and sections of the Ross Island coast. Over the years, the area of the EIT underwent various changes. Thus, to avoid misinterpretations caused by the temporal variations of the scene elements, it was decided to restrict the analysis to the tongue itself. The first step of the analysis defined a methodology for automatically detecting the EIT in a delimited window. This was accomplished using a Template Matching technique which samples the input image with a manual training pattern of the EIT structure. The result shows the matches of the EIT patch. In the second step, a relaxation method for pixel labelling was developed in order to compare the area of the ice tongue as it was 10 years ago and today. In the case of the Erebus Ice Tongue, the analysis shows a modest decline of the area in the last 10 years, hence a modest ice loss.

Here we analyse high-resolution spaceborne synthetic aperture radar (SAR) measurements to retrieve statistical and dynamical properties of ocean eddies in the ice-free regions and the marginal ice zone (MIZ) of Fram Strait. Analysis of historical Envisat ASAR images allowes to obtain detailed spatial statistics of eddy generation sites, their diameters and vorticity sign. As shown, the total number of cyclones strongly dominates over anticyclones. Eddies are frequently observed over continental slope and shelf regions around Svalbard, however, about 80% of them are identified over the deep Fram Strait regions. In general, eddies range in size between 1.5 km and 64 km with a mean value of 8 km. Larger eddies of 10-30 km in diameter are usually found over deep water, while smaller eddies of 3-7 km in diameter prevail around Svalbard. As observed, the overall pattern of eddy field in the strait attributes well to the position of West Spitsbergen Current jet and its southern recirculation branch known from literature. Analysis of sequential Sentinel-1 images allows to retrieve surface current velocity field having a clear signature of an anticyclonic eddy in the Fram Strait marginal ice zone. Mean orbital velocity of the observed eddy is estimated to be around 0.4-0.5 m/s, in good agreement with earlier field observations. Presented results demonstrate the potential of spaceborne SAR data to retrieve critical properties of eddies of various scales, and further investigate their role in the Arctic Ocean.

On-site high-resolution sea ice images from the imaging instruments on airplane, helicopter, ship and unmanned aerial vehicle (UAV) platforms have been used as the reference datasets for validation of the sea ice properties. Among the low altitude remote sensing platforms, helicopters usually carried onboard icebreakers for scientific research activities and logistics have been recognized as a reliable remote sensing platform from enhanced endurance and verified stability, and have flexibility for the attachment of multiple sensors. Although areal reference datasets, e.g., helicopter-borne high-resolution images, can be used to validate lower resolution sea ice information from the characteristics of enough coverage and fine spatial resolution, continuous drift of sea ice causes distorted locations of image acquisition along the drift; thus, becomes an obstacle for precise image mosaicking and matching with the sea ice information extracted from lower-resolution remote sensing imagery. This study presents methods for the cost-effective acquisition of helicopterborne high-resolution images over drifting Arctic sea ice using a simple configuration of imaging sensor and GPS logger, and for the compensation of the effect from sea ice drift from each imaging location using the drift trajectory of sea ice. After mosaicking the sea ice drift compensated high-resolution helicopter-borne images with structure-from-motion technique, the applicability of the mosaicked image was assessed by comparing with lower resolution satellite synthetic aperture radar image. The results suggest that the proposed methods can be applicable to the high-resolution images from other low altitude remote sensing platforms, e.g., UAVs, and can be used for precise comparison with various sea ice products from satellite remote sensing.

The research is devoted to the problem of estimations of CO2 fluxes between the hydrosphere and atmosphere. Hurricane-force winds lead to intensive wave breaking, with formation of spray in the air, and bubbles in the water. It strongly intensifies gas flux characterizing by power dependence of the transfer rate on the 10-m height wind speed used for approximation of the empirical results. But available data demonstrate wide variation which leads large confidence limits for coefficients in empirical approximations. On the other hand there is an obvious problem of obtaining reliable data on the wind speed. Widely used reanalysis data typically underestimate wind speed magnitude, due to the low spatial and temporal resolution. One of the most promising ways to measure near water wind speed is the use of the data of remote sensing. The present study used technique to achieve wind speed based on the processing sea surface images obtained in cross-polarized mode with C-band (5.4 GHz) radar with synthesized aperture (RSA) of RADARSAT satellite. To this propose geophysical model function (GMF) which binds values of wind speed and normalized radar cross section in cross-polarized mode was used. This GMF was developed in a special laboratory experiment on the wind-wave flumes for a wide range of wind speeds, including hurricanes. In turn, for parameterization of gas transfer rate results of recent laboratory experiment on high speed wind-wave flume was used.

Determination of satellite-based sea surface temperature (SST) dates back to the 1970s, and it was derived using measured brightness temperature (BT) of 11 and 12 μm channels only. Although triple-window algorithm (TWA) including shortwave infrared (SWIR) channel/s is a proven better option for SST retrieval, generation of linear coefficients including shortwave channels during daylight hours is extremely challenging due to the highly nonlinear contribution of solar reflection and scattering. On the other hand, SWIR channel/s can be easily incorporated in physical deterministic SST (PDSST) retrieval method. A successful implementation of SWIR channels for daytime SST retrieval in operational environment from MODIS-AQUA using PDSST method is discussed here. The performances of newly developed PDSST are validated by two different ways using: a) collocated in-situ measurements (buoys/Argos) quantitively and b) microwave SST from AMSR2 qualitatively. This study mainly focuses on the Indian Ocean region that is known to be a most oceanographic dynamic region among all Oceans. Also, the performances of newly developed PDSST are compared with the quality of the currently NASA-distributed MODIS-AQUA SST, obtained from Physical Oceanography Distributed Active Archive Center (PO.DAAC). An enormous improvement in the quality and coverage for daytime SST data by PDSST using SWIR channels as compared to currently operational PO.DAAC SST product that is regression based without SWIR channels is reported in this paper.

Waters from a closed Vistula lagoon actively influences the southeastern part of the Baltic Sea (offshore Kaliningrad region). When entering the Baltic Sea lagoon waters become a perfect tracer to track different hydrodynamic and biological processes with the help of Ocean Color Data. The research is primarily based on remote sensing data being a first step to determine properties and propagation boundaries of the outflow into the open Baltic Sea. During research we could analysis different types of remote sensing data including color composite images from MODIS Terra/Aqua, OLI/TIRS Landsat 8; MSI Sentinel 2; OLCI Sentinel 3. The oceanographic experiments include CTD transects and determination of turbidity and CHL-a concentration in outflow waters to describe water properties of the outflow and provide verification of satellite data. Results of a combined experiment for determination of the Vistula Lagoon outflow distinctive features with the use of satellite images from optical sensors and oceanographic in-situ data are shown.

Coastal Zone Imager(CZI) was designed for coastal water body and islands. To analyze the potential application of Chinese ocean color satellite in coastal zone area and evaluate the image quality, we used subjective and objective evaluation method and took a comparison between Chinese ocean color satellite and Gaofen series satellite. Subjective evaluation showed that Chinese ocean color satellite images had a better performance than Gaofen series satellite images in water bodies. Based on statistical information ,objective evaluation showed that each band gray distribution of Chinese ocean color satellite images was more dispersed and had a higher separability feature. Applying evaluation showed that Chinese ocean color satellite also performed better in terms of texture property and classification accuracy. Overall Chinese ocean color satellite had a high image quality and potential applications in coastal areas.

Characteristics of corals spectral from different species are expected have optically different characters. This study classified the dominant substrate of shallow water base on spatial resolution of imagery and in situ measurement and analyzed the accuration of Landsat 8 OLI_TIRS and Sentinel-2A satellite imagery to determine health of coral reefs. The image processing are atmospheric correction, cropping, masking, Depth Invariant Index, Unsupervised classification, ground truthing, reclassify, accuracy assessment, and spectral reflectance analysis. Unsupervised classification used IsoData method with Lyzenga application to detection of coral reefs condition. Spectral measurement by spectroradiometer underwater, photo underwater, and geotagging are conduct as in situ measurement. Spectral reflectance of medium spatial resolution image and in-situ measurement are integrated to discriminate of live coral, dead coral cover with algae, rubble and algae. The results of this study show a baseline for develop the Di gi t a l Coral Health Chart as an approach to determine living coral condition using remote sensing techniques. It can be used as an effective way for detecting and monitoring of dynamic changes of coral reefs on small islands in Spermonde Archipelago. Image analysis integration and in situ survey results show that rubble and dead coral with algae were indicating as coral death due to either damaging human activity and natural death.

In recent years, the golden tide, which is caused by the explosive proliferation of Sargassum, has occurred frequently in China Seas. It has made a great negative impact on the marine ecosystem, aquaculture, and coastal tourism. Fortunately, satellite observation can monitor and track the growth of large algae such as Sargassum in a timely and effective manner, providing scientific basis for disaster prevention and mitigation in fisheries and environmental protection departments. Most of traditional extraction methods of macroalgae are pixel-oriented. Although these methods can be performed easily, they loss the rich texture information of the natural objects. The Sargassum seen from remote sensing imageries tends to aggregate in groups, like strips, covering several to dozens of pixels. Therefore, this paper considered distinguishing Sargassum from a certain area based on scene by utilizing contextual relationships among pixels and the diversity of spatial and structural features. In this paper, the image acquired by GF-1 during the golden tide disaster in the sea area near Jiangsu Province of China on December 31, 2016 were used. We adopter an unsupervised feature learning method to distinguish Sargassum. The Voting method was used to divide the original image into small image blocks guided by the corresponding saliency image. After 0-meanization and ZCA whitening, the initial weights were obtained by training the sparse autoencoder, then these weights were convolved as the convolution kernel to obtain the local features of the image, the features convoluted were passed. We pooled them to reduce the eigenvectors of the convolutional layer output so that the global statistical features of the image could be extracted. Finally, the Softmax classifier was used to distinguish the regions of Sargassum in the original image. The experimental accuracy was 77.79% and superior to the threshold extraction methods compared with the result of manual labeling.

Katabatic winds generally flow from mountains or hills down to their lee side in the paths of depression. If the mountain or hill is near a coast, the katabatic winds may cause imprints on the sea surface. The katabatic wind pattern shown on a synthetic aperture radar (SAR) image is a bright-dark region that mirrors the coastal mountain topography. In this study, bright regions on SAR images caused by katabatic winds are found in the west of Hengchun Peninsula where is located in the southern Taiwan. The katabatic winds cause the sea state variations and then the sea surface temperature changes. Relationships between normalized radar cross section (NRCS) and sea surface temperature (SST) as well as the temperature difference between air and sea in the west of Hengchun Peninsula are investigated to find out the air-sea heat transfer. The results show that 1) the SST decreases when the NRCS increases, that is, the higher wind speed would cause the SST lower; and 2) the gradient of linear relationship between NRCS and SST is related to the temperature difference between air and sea, that is, the higher temperature difference could increase the release of heat from the ocean to the atmosphere.

Harmful algal blooms (HAB) is one of the most serious marine disasters, which not only reduce fishery production, deteriorates the marine environment, affects coastal tourist industry, but also cause human poison. The satellite remote sensing technology has the characteristics of large-scale, synchronized, low cost and rapid monitoring, it is become an important method for HAB observation. In this paper, we describe the HAB monitoring system based on remote sensing data which developed by the Second Institute of Oceanography, China. The system can achieve the whole procedure automatically from the satellite remote data acquirement and satellite image process, products generation, and it can automatically identify the HAB region using the spectral reflectance and inherent optical properties derived from remote sensing data. Currently, the National monitoring department has adopted this system for the operational monitoring of HAB in the East China Sea.

Sea surface salinity of ocean waters has been obtained from L-band microwave radiometer on satellite, such SMOS, Aquarius and SMAP. It provides a new technology and way to observe the salinity over large region from space. However, it is still difficult to get the sea surface salinity of the coastal waters due to that the effect on satellite-observed microwave radiance of radio frequency interference over near-shore regions. Fortunately, the colored dissolved organic matter (CDOM) has conservative or semi-conservative property in the freshwater areas of coastal rivers and it has a good linear correlation between the absorption coefficient of CDOM and salinity. Based on those, a conservative mixing relationship between the fresh water from the Pearl River and the sea water from the South China Sea are established and the distribution of sea surface salinity from MODIS-retrieved absorption coefficient of CDOM (a_CDOM) at 355nm wavelength in this study. The linear relationship between salinity and aCDOM at 355 nm under conservative mixing conditions can be established from the in-situ observed salinity and MODIS-retrieved a_CDOM at 355 nm at the central water mass in the Pearl River Estuary and the northern basin region of South China Sea. By comparing daily observed MODIS-retrieved sea surface salinity with the in-site observed salinity, it shows that the salinity retrieved from satellite-based a_CDOM has good agreement with observations, and it can better capture the location of low salinity fresh water in the Pearl River. For the summer averaged satellite-based salinity from 2002 to 2018, sea surface salinity are gradually increases from the vertical coastline to the sea in the Pearl River Estuary region. The salinity in coastal region is low by the effect of the Pearl River diluted water, and the salinity in offshore region is high. On average, the distribution of salinity shows that the diffusion of fresh water from the Pearl River generally spread along the coastline to the east and west in summer, while the runoff on the westward is larger and on the eastward it is affected by some flushing water of other rivers. In the future, MODIS-retrieved salinity can be as a supplement to salinity inversion of Microwave radiometer in the coastal region.

Since 2008, macroalgal blooms of Ulva Prolifera (also called green tide) have occurred every summer in the Yellow Sea (YS), which has caused environmental and economic problems. In recent years, a variety of detection algorithms for green tide have been proposed. However, the extraction thresholds of each algorithm are uncertain because of atmospheric conditions, the distribution of green tides, etc. In this paper, Geostationary Ocean Color Imager (GOCI) data and Landsat- 8 data were used to explore the threshold stability of some common detection algorithms for green tide, including the AFAI, DVI, EVI, IGAG, and NDVI. Four scenes of GOCI satellite data from 2016 to 2018 were selected for the experiments. The first step was to extract the green tide areas in one region to determine the threshold for each algorithm. In this step, the extraction results of the Landsat-8 data, which has a resolution of 30 m, was seen as the true value of the green tide coverage. Then, we determined the threshold value for each algorithm by visual inspection. The thresholds determined in the first step were used to extract the green tide area in the other three regions, and the extraction results were compared by visual contrast. A comparison of the extraction precision for each algorithm in the other three regions indicated that the threshold stability of the AFAI algorithm was the best among these data in the YS region.

This study uses the MODIS-Aqua satellite data provided by the National Aeronautics and Space Administration (NASA) and matches with the time and location of the chlorophyll-a concentration data measured by SeaBASS to select the satellite data time and observation area. The number of matched data is 924. Firstly, remote sensing reflectance (Rrs) is used to classify satellite remote sensing data into different water bodies, and then the best chlorophyll-a concentration algorithm is established. The results show that the mean percentage difference (MPD) in Case 2 water is 131.2% through comparing the percentage of chlorophyll-a provided by MODIS with the in-situ observations. In addition, the chlorophyll-a concentration of the new algorithm compared with the in-situ chlorophyll-a concentration are also calculated. The mean percentage difference in Case 2 water is 26.6%, and the average chlorophyll-a is 6.16 mg/m³ , which is much closer to the in-situ value,7.22 mg/m³ than the average chlorophyll-a of MODIS, 13.7 mg/m³. The chlorophyll-a concentration deduced by the new algorithm of this study is consistent with the in-situ values in Case 2 water, and it is much more convergent than the data of MODIS. Obviously, the new algorithm established in this study can be used to improve the chlorophyll-a concentration estimation results in Case 2 water. When the new algorithm is applied to calculate the chlorophyll-a concentration of the marginal Northwestern Pacific, the value is still higher than the offshore waters. Additionally, the chlorophyll-a concentration calculated by this new algorithm is lower than the value provided by MODIS, but the difference between them in the offshore waters is small. However, the algorithm of this study can improve the overestimation of the original MODIS value.

In this paper, we discuss applications of satellite remote sensing for detection and analysis of spatiotemporal characteristics of oil showings on the sea surface due to natural hydrocarbon seafloor seeps in deep waters of the southeastern Black Sea. The study is based on the data obtained in the course of a satellite survey of the study region. The study region includes two areas of the heaviest oil pollution of the Black Sea surface. These are the areas of natural seepages off the Georgian coast near the town of Poti as well as the shelf area off the eastern coast of Turkey near the town of Rize. Our main result is the discovery of significant seasonal variability in the spatial distribution of natural oil films affected by local winds and surface currents prevailing in given periods. Further, we show that meso- and sub-mesoscale processes significantly affect the drift of natural oil spills. In the case of moderate winds and weak currents, the influence of these processes on the oil spills drift becomes paramount. We found that in 10-15 percent of cases, the oil film gets involved in vortical motions after its emersion on the sea surface, which often radically changes the trajectory of the slick propagation. Risks of oil surface pollution due to natural hydrocarbon showing are assessed for the test areas.

Red tide is an ecological anomaly that phytoplankton in seawater suddenly proliferation or aggregation under certain environmental conditions and within a period of time, resulting in seawater discoloration. Red tide not only endangers marine fisheries and aquaculture, deteriorates the marine environment, affects coastal tourist industry, but also causes human health problems. East China Sea (ECS) is a high incidence region of red tide disasters. Remote sensing is an effective means of monitoring red tides. In this paper, the high-incidence area of the red tide in the East China Sea is selected as the study area, MERIS L2 data is used as the data source to analyze and compare the normalized water radiation (nlw) spectral difference between the red tide water body and the non-red tide water body in the red tide event. Based on the spectral difference, this paper develops nlw560/nlw490>1.25 and nlw681-nlw665>0 algorithm to extract the red tide information of ECS. Applying the algorithm to ECS, the results show that the developed model can effectively determine the location of the red tide and correspond well with the results of the official bulletin. This indicates that the algorithm can effectively extract red tide information.

The present paper reveals the practical possibilities of the solar path observation from underwater as an instrument for remote sensing of wavy surface. Paper includes: a theoretical model of underwater solar path image and its statistical moments; algorithm for solving inverse problem; description of full-scale experiment and results of slope frequency spectrum retrieval. Presented spectra were obtained for random selected data from continues field measurements in the coastal zone of the Black Sea. On their example it was shown that spectra obtained by image processing and wave gauge data are close over the entire frequency interval. At frequencies corresponding to capillary waves, where wave gauge not applicable, underwater vision system still continues to measure wave spectrum. Obtained results complement previously published results of retrieval wind wave characteristics and water optical properties using underwater solar path images, thereby confirming advantages of applying underwater optical systems for remote sensing of different waters.

The article discusses the results of theoretical and experimental studies of Doppler velocity on the sea surface. Doppler velocity is measured by radar sensing at moderate and low grazing angles. The experiments were performed using coherent X-band and Ka-band panoramic digital radars operating with the horizontal polarization for transmission and reception. These radars have high spatial resolution. An algorithm for reconstructing Doppler velocities is proposed and estimates of the fluctuation sensitivity of the method are carried out. The method is applied to the conditions of the fetch-limited wind wave growth, which is typical for enclosed waters and the sea nearshore, where the dominant wavelength is of the order of ten meters. Based on the two-scale model, the dependences of the Doppler velocity on the parameters of the ocean-atmosphere interface are considered. The effect of shadowing by the crests of wind waves on the magnitude of the Doppler velocity at low grazing angles is discussed. The manifestation of the orbital velocity of wind waves is demonstrated.

Surfactant films on the sea surface can appear due to pollutions, river and collector drains, as well as biological processes. Film slicks can indicate different processes in the upper ocean and in atmosphere. In particular, slick signatures in SAR-imagery of the sea surface at low and moderate wind speeds are often associated with marine currents. Other factors such as wind and physical characteristics of films can significantly influence the dynamics of slick structures. A perspective approach aimed at measuring surface currents is developed. Based on the approach an impact of wind on the kinematics of artificial slick bands is determined. Simulation of slick band propagation from the localized source of surfactant in the field of wind and eddy-shape current is performed. As a result of simulation the shape of surface slick structure, which is close to the observed on SAR image of water surface, is obtained. It is shown that the possibility of spiral bands formation due to presence of marine submesoscale eddies is determined by near-surface wind. Moreover, it is declared that a traditional estimation of scales of marine eddies based on the scales of spiral slick structures is not basically correct.

The possibilities of the definition of physical characteristics of organic films on the water surface, including oil thickness estimation, were studied in laboratory conditions based on the novel opto-acoustical approach. It was shown that the continuous infrared irradiation of film on the water surface leads to the generation of ultrasonic wave with frequencies of the order of several kHz which is the result of local heating of subsurface layer of water. The preliminary results regarding a relation between physical characteristics of surfactant films and the characteristics of acoustic wave are presented. A promising approach of remote sensing definition of surfactant films characteristics in real sea conditions was proposed on the base of these results. The development of this approach will contribute to the development of modern systems of strategic remote sensing of the ocean and inland basin.

The work is devoted to the experimental study of Doppler velocity in artificial surfactant films on the sea surface. Such films simulate the oil spills. The paper develops a method of remote detection of oil spill pollution on the water surface. The method is based on a joint analysis of amplitude and velocity radar images of the water surface. It is shown that the surfactant films lead to a significant change in the Doppler velocity, which can be used to increase the detection probability of pollution on the sea surface. A statistical analysis of the two-dimensional distribution of RCS and Doppler velocity in artificial surfactants films and pure water in a wide range of weather conditions is performed. It is demonstrated that the difference between the measured Doppler velocity in the upwind direction and the phase velocity of the Bragg waves corresponds with the slick drift velocity.

Understanding of physical mechanisms of gravity capillary waves (GCW) damping due to an impact of turbulence is important for developing methods of ship wake remote sensing, especially for tracking and characterization of ships. Analysis of literature reveals the necessity of setting a reliable experiment to study the effect of damping of GCW due to turbulence. Available laboratory studies are based on significantly different experimental methods with some disadvantages which have caused a large scatter of experimental data. The previously proposed method, which is based on the simultaneous independent generation of surface waves and turbulence in a wave tank, is free from inherent disadvantages of previously used methods. The method is used to conduct a series of experiments in order to measure the dependence of the damping coefficient of GCW on their frequency at different intensities of turbulence in a wide frequency range. Due to the proposed technique the range of surface wavelengths is extended in order to investigate the case when the surface wavelength is comparable to the scale of turbulence. It is shown that the frequency dependence of the eddy viscosity coefficient is characterized by the presence of a maximum for both turbulent regimes. The maximum value of eddy viscosity coefficient is proportional to the velocity of turbulent pulsations. The results are discussed in application to radar imaging of turbulent ship wake.

The presented paper is aimed at the possibilities of detecting surfactants fluorescence on a water surface using a portable UV diode light source and spectrometer. Under controlled conditions of film thickness, a series of laboratory fluorescence measurements for some surfactant most commonly found in inland waters like crude oil and diesel were performed. Obtained fluorescence spectra were registered for thin films with a thickness from 0.6 to 33 μm. Estimates of the fluorescence intensity in relation to film thickness are obtained.

In paper presents a comparative study of the short-scale wind waves spectra, received by means of optical and radiometric methods. The optical method for recording the wind waves spectrum is based on spectral analysis of the optical images of the sea surface with non-nadir observations. An incoherent optical spectrum analyzer developed at the Institute of Applied Physics of the Russian Academy of Sciences is described that allows one to record two-dimensional spectra of the sea surface in real time in the wave number range from 0.3 to 5.0 rad/cm. Wind-wave spectrum retrieval using microwave radiometers was carried out by the method of Non-linear Radiothermal Resonance Spectroscopy (NRRS), which was developed and are improving at present at the Space Research Institute of the Russian Academy of Sciences. Using a set of radiometers-polarimeters of the 3, 5 and 8-mm ranges, provides restoration of the spectrum in the range of wave numbers from 0.39 to 15.0 rad/cm. Synchronous microwave and optical measurements of wind wave spectra were carried out on a marine hydrophysical platform in the Black Sea. The results of the sea wave's spectrum restoration obtained by means microwave and optical technique gave almost identical estimates of the spectral curves level, which also coincide with the model spectra. This fact confirms the consistency of described wind wave's spectra recording methods and the possibility of their use for monitoring the sea surface state in natural conditions.

The problem of characterization of marine currents using ocean remote sensing data is very challenging and has not been completely resolved by now. Optical and IR satellite images of the ocean have been traditionally used to estimate the current velocities when comparing color or temperature inhomogeneities in co-located subsequent scenes. SAR as all-weather and all-day instrument with high spatial resolution is very perspective for ocean remote sensing, and the procedure similar to the optical/IR observations seems to be usable for SAR for the current velocity estimation, too. Marine biogenic film slicks often observed on the sea surface as systems of “filamentary” structures at low/moderate wind conditions can be considered as appropriate features for marine current tracking. However, very few attempts have been made to study the current velocity field when studying slick features in SAR images acquired from different satellites at a comparably short time interval. In this paper two sequential satellite SAR images acquired with Envisat ASAR and ERS-2 SAR have been analyzed in order to estimate the surface marine currents. The acquisition time difference between the images was nearly 30 min. The images were characterized by a number of slick features which were nearly identical within the 30 min time shift, so that it was rather easy to track any chosen slick structure and to retrieve the velocity field. A Maximum Cross-Correlation (MCC) method has been used for the current retrieval, when analyzing correlation between the sequential images. It has been obtained that for some slick filamentary structures or for their parts the retrieved current velocities were directed nearly along the filaments, so that the slicks can be considered as the current streamlines. On the contrary, for some other slicks the retrieved current velocity vectors were directed at quite large angles to the filament tangent lines. We believe that the latter effect appears for varying currents due to the “memory” of slicks which cannot change their orientation or appear/disappear instantaneously according to fast changes of environmental conditions, in particular according to wind speed velocity/direction changes.

River confluence is a ubiquitous phenomenon which plays an important role in river dynamics, mixing processes, pollution transport etc. It can be often visually observed that two converging rivers continue to flow as two parallel weakly mixing streams separated by a relatively thin transition region – a mixing zone (MZ), which can reach out for some distance downstream the junction apex. A typical example of a river confluence is the merging of the Volga and Oka rivers (Russia). This paper presents some new results on the Volga/Oka rivers confluence based on both ‘in situ’ measurements of hydrological characteristics in the MZ and on satellite observations. During the ‘in situ’ measurements a large set of data regarding velocity fields, chlorophyll-a concentration, water temperature, turbidity, parameters of organic surface films etc. was obtained. It is found that significant differences between the hydrological characteristics of the Volga and Oka flows are observed at sufficiently large distances downstream the junction apex (about 10 river widths or more); and the mixing zone remains quite narrow. Film sampling and further analysis of the surfactant films in the MZ were carried out. It is shown that the surface tension decreases and the film elasticity grows in the foam/slick bands separating the Volga and Oka flows thus indicating the increased concentration of surfactants in the bands. Satellite images of the Volga-Oka MZ are collected indicating that the confluence area is clearly seen in satellite optical imagery (MSI Sentinel-2 satellite) due to strong difference in color between the flows. The radar imagery shows the mixing zone in the form of a slick/foam band (SAR Sentinel-1) which manifests on the water surface due to enhanced damping of short wind waves.

Aquarius is a satellite designed for measuring sea surface salinity (SSS). The Aquarius measurements may be influenced by marine environmental factors. The result is the inconsistency of data quality under different conditions. Although the data qualities have been considered in some previous studies, they have only been used for data screening. Base on this, a quality weighting method is proposed in this paper. The key differences between our method and traditional method, is way of weighting the data. In the present paper, both distance and data quality are considered in the weighting process. After the weight is determined, the weighted average fitting (WAF) method is used to calculate the grid SSS value. Then weekly 0.25°×0.25°gridded SSS fields between 40°S and 40°N are generated, covering the period from September 2011 through May 2015. The error statistics are calculated and the result shows that the root-mean squared difference (RMSD) is about 0.18 psu, which can improve the accuracy by about 31%. Therefore the method proposed in this paper could improve the precise of SSS filed with higher temporal and spatial resolution significantly.

One of the urgent problems of gas pipeline transport today is the detection of gas leaks in underwater sections of gas pipelines and reducing their negative impact on the environment. This paper is devoted to the development of remote detection methods of gas leaks from underwater gas pipelines by images of slicks above them. A series of laboratory experiments was carried out, in which the structure of average flows in the water column created by bubble flows with different gas flow rates was obtained. The simulation of surfactant removal to the water surface by bubbles was performed by adding sodium dodecyl sulfate (SDS) to the laboratory cuvette at the concentration of 0.5 mg/l. For the surfactant film on the water surface, experimental dependences of its characteristics on the operation time of the "bubble pump", such as the surface tension coefficient, elasticity and the relative damping coefficient of small- scale waves under the film, were obtained. It was shown that there are flow convergence zones at some distance from the gas outlet area, which vary depending on the intensity of the gas flow. Surfactant film samples taken in these areas confirmed that over time, the accumulation of surfactants occurs on the water surface. Field experiment conducted in one of the backwaters of the Oka river in the city of Nizhny Novgorod allowed to spot and evaluate the size and shape of the film slick formed around the gas outlet area. The conducted experiments confirmed the formation of a surfactant film spot near the removal area, which indicates the possibility of its remote detection in full-scale conditions, provided the wind near water surface is not more than 2-10 m/s. The obtained data made it possible to get a general idea of the morphology and characteristics of the surfactant spot around the gas outlet from the water.

In developing the on-board equipment of aircraft, used radar maps for navigation, there is a need for mathematical model's signals, reflected from the Earth's surface, sea surface and coastal edge. Traditionally, using a theoretical construct, stochastic signals were used as such models and its fluctuations were described by Rayleigh and Rayleigh- Rice. These models are used both for simulation signals, reflected from the Earth's surface, and for signals, reflected from the surface of the sea. At low resolution capability of on-board radars, the similar models describe quite well the statistical characteristics of fluctuating signals. Modern on-board locator has high resolution capability and the Rayleigh and Rice’s models can no longer be used in the synthesis and simulation of modern on-board navigation systems. In this paper, we propose an approach to the construction of models of radar signals using both theoretical constructs and experimental data that allows you to take into account the features of reflection of radar signals from small plots of the Underlying Surface of earth and sea. Along with it the correlations between the individual sections and anisotropic reflections are taken into account when observing sites with different angles. The reflections from the sea surface approximated by a log-normal law, reflections from the earth's surface at the sight of the manifold types of surface by the Beckmann and Weibull laws, special cases of which are the laws of Rice, Rayleigh and Hoyt. As the reflection model of the edge uses the distribution law of the vector sum of the signals, reflected from the elementary areas of earth and sea, getting in the resolution cell. In this case, the law of distribution of the total vector is subject to a law similar to Huber's law, in which the above-mentioned laws of distribution of reflections from the earth and the sea are used as the basic distributions. Previously used reflection models turned out as special cases of the proposed models. These models and the modeling algorithms developed for them can be used in the development and research of high- precision methods of radar monitoring for the purposes of environmental reconnaissance, forecasting and prompt prevention of natural and man-made emergency situations. In addition, for testing the operating modes of the equipment of unmanned aerial vehicles, including for multi-position radar systems. Algorithms for modeling location signals based on mathematical models using experimental data of reflections from various types of underlying land and sea surfaces, as well as coastal edges (coastal waters), allow us to bring the results of computer experiments to the results of actual tests of radio-electronic equipment. This reduces the time and reduces the cost of design by reducing the semi-natural and full-scale tests.

Producing very extensive blooms in the world’s oceans in both hemispheres, a coccolithophore E. huxleyi affects both marine ecology and carbon fluxes at the atmosphere-ocean interface. In turn, it is subject to impacts of multiple co-acting environmental forcings responsible for spatio-temporal dynamics in E. huxleyi blooms. To reveal the individual importance of each forcing factor (FF) that is known to significantly control the extent and intensity of E. huxleyi blooms, the 1998-2016 spaceborne time series of sea surface temperature and salinity, incident photosynthetically active radiation, and the Ekman depth relevant to the North, Norwegian, Greenland, Labrador, Barents and Bering seas were employed. The descriptive statistical approach showed that E. huxleyi phytoplankton blooms were capable of arising and developing within wide but expressly sea-specific FFs ranges. Sea-specific FFs ranges, within which the blooms are particularly extensive were identified. The Random Forest Classifier (RFC) allowed to reliably rank the FFs in terms of their role in E. huxleyi bloom spatiotemporal dynamics in each target sea. High prediction ability of RFC modelling (>70%) confirms the adequacy of the developed FFs prioritization models. Although the parameters of the carbon chemistry system per se were beyond consideration, however, over the twenty years of observations, the prioritized FFs have not failed to explain the registered patterns of the spatial extent of and particulate inorganic carbon content in E. huxleyi blooms. Also, several verifications (pastcasts) showed a high degree of their consistency with the observations. Collectively, these results tell in favor of sufficiency of the FFs employed.