Sensor Technologies for Water Velocity, Flow, and Wave Motion Measurement in Marine Environments: A Comprehensive Review

Abstract

Measuring water motion is essential for oceanography, coastal engineering, and marine environmental monitoring. A wide range of sensing technologies is used to quantify water velocity, wave motion, and flow dynamics, each suited to specific spatial and temporal scales. This paper presents a comprehensive review of modern sensor technologies for marine flow measurement, covering mechanical, electromagnetic, pressure-based, acoustic, optical, MEMS-based, inertial, Lagrangian, and remote-sensing approaches. The operating principles, strengths, and limitations of each technology are examined alongside their suitability for different environments and deployment platforms, including moorings, buoys, vessels, autonomous underwater vehicles, and drifters. Special attention is given to rapidly advancing fields such as MEMS flow sensors, multi-sensor fusion, and hybrid systems that combine inertial, acoustic, and optical data. Applications range from high-resolution turbulence measurements to large-scale current mapping and wave characterization. Remaining challenges include biofouling, performance degradation in energetic shallow waters, uncertainties in indirect velocity estimation, and long-term calibration stability. By synthesizing the state of the art across sensing modalities, this review provides a unified perspective on current technological capabilities and identifies key trends shaping the future of marine flow measurement.

1. Introduction

Measuring fluid flow is a fundamental requirement across science, engineering, and industry. From monitoring discharge in rivers and water distribution in pipelines to optimizing industrial processes and ensuring safety in hydraulic structures, accurate flow characterization underpins decision-making and technological development. In environmental sciences, flow measurement is equally critical: it provides the basis for understanding transport of sediments, nutrients, and pollutants, for assessing habitat dynamics, and for predicting risks associated with floods or extreme weather events.

Within the marine domain, the characterization of currents and water velocity is of particular importance. Ocean currents influence global climate through heat and mass transport, regulate ecosystems by shaping biological productivity, and affect coastal morphology through sediment redistribution [1,2]. At the same time, knowledge of marine currents is vital for a range of practical applications, including navigation safety, offshore operations, marine renewable energy development, and environmental impact assessments [3,4,5].

Measuring water motion in the marine environment poses unique challenges. The harsh and dynamic nature of the ocean—ranging from strong turbulence and variable stratification to extreme pressures and biofouling—demands robust, accurate, and autonomous technologies. Furthermore, the diversity of spatial and temporal scales, from small-scale turbulence near boundaries to basin-wide circulation patterns, requires different measurement approaches and sensor designs. Over the past decades, a wide variety of techniques have been developed, ranging from traditional mechanical and electromagnetic meters to modern acoustic, optical, and remote sensing methods. More recently, autonomous platforms, wireless sensor networks, and low-cost disposable sensors have expanded the range of possibilities for flow monitoring. These measurement approaches have been successfully demonstrated in a wide range of marine environments, including coastal, estuarine, and open-ocean settings, supporting applications such as circulation monitoring, environmental assessment, and hazard analysis [6,7,8,9].

Despite these technological advances, the terminology surrounding flow measurements in the marine context often lacks consistency. Expressions such as flow, current, and velocity are frequently used interchangeably, even though they represent distinct physical quantities. This review, therefore, begins by clarifying these terms and their usage in marine science and engineering, before providing a comprehensive overview of technologies and sensors available for current measurement. The objectives of this review are threefold: (i) to summarize the principles and applications of traditional, modern, and emerging flow measurement technologies in the marine environment; (ii) to evaluate their performance, strengths, and limitations; and (iii) to identify current challenges and future directions for research and development in this field. The review is structured as follows: Section 2 introduces the fundamentals of flow and current measurement, including terminology. Section 3 delves into the different measuring environments and sensing platforms. Section 4 presents different classes of technologies, from traditional mechanical devices to emerging approaches. Section 5 concludes the paper with key findings and recommendations and highlights future trends and perspectives.

2. Fundamentals of Water Flow Measurement

Before reviewing the different technologies available for flow measurement, it is important to establish the fundamental concepts underlying what is being measured. Ocean circulation encompasses a wide range of processes and scales, from turbulent motions in the boundary layer to basin-scale transport, and the choice of measurement technique depends strongly on the physical variable of interest. Moreover, the terminology used to describe these variables is not always consistent across disciplines, which can lead to confusion in interpreting measurements. This section, therefore, introduces the basic definitions, scales of motion, and measurement principles that provide the framework for understanding and comparing the different sensing technologies described in Section 4.

2.1. Terminology and Definitions

The vocabulary used to describe water motion varies across disciplines such as oceanography, hydraulics, and coastal engineering. Terms including flow, current, velocity, transport, and surface motion are often used interchangeably, but they represent distinct physical quantities that must be clearly distinguished in the context of measurement technologies.

• Water Velocity is a vector quantity that describes the speed and direction of water motion at a given point, expressed in units of meters per second (m/s). Instruments that resolve velocity typically provide either a single component (e.g., along a specific axis) or the full three-dimensional velocity vector in the sensor’s own reference frame. This is the primary physical variable measured by the sensors discussed in this review.
• Current is the term most commonly used in oceanography to refer to water motion expressed in the terrestrial (Earth-fixed) reference frame. Depending on the context, it may denote instantaneous velocity at a point, depth-averaged velocity, or even large-scale circulation patterns. Therefore, ocean current measurement requires not only velocity sensing but also orientation information (e.g., compass and tilt) to transform sensor-referenced velocities into Earth-referenced components.
• Flow is more ambiguous. In fluid mechanics, it may refer to the general motion of a fluid, while in hydraulics and hydrology it is often used to mean discharge—the volume of water transported through a cross-section per unit time (m³/s). In marine sciences, the term can refer either to local velocity fields or to the broader concept of water motion.
• Transport refers specifically to the flux of volume, mass, or a tracer (e.g., heat, salt, sediment) through a cross-section. Oceanographers frequently express volume transport in Sverdrups (Sv), where 1 Sv = 106 m³/s, when describing large-scale circulation features.
• Wave or surface motion refers to the oscillatory movement of the sea surface driven primarily by wind. Unlike water velocity, which describes the net movement of water parcels, wave orbital motion is largely periodic and does not correspond to a sustained transport of mass. Wave sensors typically measure parameters such as significant wave height, wave period, and wave direction.

Table 1. Terminology and measured quantities in marine flow and wave studies.

Term	Physical Quantity	Units	Measurement Scale
WATER VELOCITY	Vector speed and direction of water motion at a point	m/s	Point, profile
CURRENT (OCEANOGRAPHIC)	Water velocity expressed in the terrestrial reference frame	m/s	Point, profile, regional
FLOW (GENERAL MOTION)	General fluid motion (non-quantitative)	–	Local to regional
DISCHARGE/VOLUMETRIC FLUX	Volume of water passing through a cross-section per unit time	m3/s	Cross-section
TRANSPORT	Flux of volume, mass, or tracer (e.g., sediment, heat, salt)	m3/s, kg/s	Cross-section to basin
SURFACE WAVE MOTION	Orbital velocity of water particles due to waves	m/s	Near-surface, profile
WAVE ELEVATION	Vertical displacement of the water surface	m	Surface
TURBULENT VELOCITY	Rapid, random fluctuations of water velocity	m/s	mm–m scale

summarizes the terminology adopted in this review, distinguishing between different physical quantities associated with water motion and wave dynamics. These definitions provide a consistent framework for the discussion of measurement technologies presented in the following sections and are essential for interpreting sensor data. For example, a point electromagnetic current meter measures water velocity at a fixed location, whereas an Acoustic Doppler Current Profiler (ADCP) provides velocity profiles that can be integrated to estimate discharge (or even sediment transport with additional echo processing) or ocean currents (using compass and tilt instrumentation). Similarly, remote sensing systems like high-frequency (HF) Radio Detection And Ranging (RADAR) do not measure water velocity directly in terms of water current but infer surface velocities from Doppler shifts. The orbital velocities associated with waves can interfere with water velocity measurements in the upper water column if not properly accounted for. Distinguishing between mean water velocity and wave-induced orbital motion is therefore essential when interpreting data from near-surface sensors.

Many commercially available instruments are referred to in the literature as flow sensors, but the vast majority actually measure water velocity at a point or along a profile rather than true volumetric flux. In this review, when discussing previously published work, the terminology used by the original authors is retained; however, it should be understood that the term “flow sensor” is frequently used referring to a device that measures water velocity, not discharge.

To ensure clarity throughout the remainder of this review, terminology will be used consistently according to the definitions established earlier. Water velocity will designate the physical quantity directly measured by most instruments. Ocean currents will refer to water velocity measurement on the terrestrial frame. Flow will be used only in its general sense to describe water motion without implying a volumetric rate. Discharge will refer specifically to volumetric flux through a defined cross-section. Wave or surface motion will denote the oscillatory movement of the sea surface.

2.2. Scales of Motion in the Marine Environment

Water flow in the marine environment occurs across a wide range of temporal and spatial scales, which strongly influences the choice of measurement techniques and sensor design.

• Temporal scales range from rapid, small-scale turbulence occurring over seconds (e.g., eddies and vortices), to tidal cycles on the order of hours (e.g., semidiurnal or diurnal tides), seasonal variations (e.g., monsoon-driven flows, river discharge, or upwelling cycles), and long-term interannual phenomena such as the El Niño–Southern Oscillation. Each temporal scale imposes different requirements on sensors: fast-sampling instruments are needed to capture turbulence, while longer deployments are necessary to resolve tidal or seasonal variability.
• Spatial scales similarly span several orders of magnitude. Near-boundary layers, velocity gradients may occur over centimeters to meters, whereas estuarine and coastal flows extend over kilometers. Basin-scale currents, such as those in the open ocean, can span thousands of kilometers. The spatial scale of interest dictates sensor placement and coverage: point measurements may suffice for small-scale studies, while profiling instruments or distributed sensor networks are required for larger-scale investigations.

The combination of temporal and spatial scales determines sensor selection, deployment strategy, and sampling frequency. Understanding the scales of motion in the target environment is, therefore, a critical step in designing an effective marine monitoring campaign.

Table 2. Spatial and temporal scales of motion in the marine environment.

Spatial Scale	Temporal Scale	Dominant Physical Processes	Representative Environments
MILLIMETERS–CENTIMETERS	Seconds–minutes	Turbulence, shear, boundary-layer processes	Benthic boundary layer
DECIMETERS–METERS	Seconds–minutes	Wave orbital motion, near-bed flows	Surf zone, shallow coastal waters
METERS–TENS OF METERS	Minutes–hours	Mean currents, internal waves, wave–current interaction	Coastal waters, estuaries
HUNDREDS OF METERS–KILOMETERS	Hours–days	Tidal circulation, river plumes, coastal jets	Estuaries, shelves, straits
TENS–HUNDREDS OF KILOMETERS	Days–months	Seasonal circulation, mesoscale variability	Continental shelves, marginal seas
BASIN SCALE (>1000 KM)	Months–years	Large-scale circulation, climate-driven variability	Open ocean

summarizes the characteristic spatial and temporal scales of motion in the marine environment, highlighting the wide range of processes that must be considered when selecting appropriate measurement strategies.

2.3. Measurement Principles and Configurations

Measuring water velocity in the marine environment can be approached using two complementary frameworks: Eulerian and Lagrangian methods. The choice of measurement principle depends on the scales of motion of interest, the desired temporal resolution, and the available platforms.

• Eulerian measurements are taken at fixed points in space. Instruments such as mechanical current meters, electromagnetic sensors, and ADCPs record water velocity as it passes a stationary location. Eulerian measurements are well-suited for capturing temporal variations at a given location, such as tidal cycles or turbulent fluctuations, and are commonly deployed on moorings, seabed frames, or pier-mounted stations.
• Lagrangian measurements follow individual water parcels as they move through the marine environment. Devices such as surface drifters and subsurface floats measure water currents along their trajectories. Lagrangian approaches provide valuable information about transport pathways and large-scale circulation, and they can capture spatial variability over extended areas.
• Indirect or derived methods can be used in addition to direct velocity measurements. For example, remote sensing techniques infer surface velocities from Doppler shifts (RADAR) or surface displacement (satellite altimetry). Pressure gradients, temperature, and density fields can also be combined with hydrodynamic models to estimate water velocity and transport.

Each measurement principle has advantages and limitations. Eulerian sensors provide high temporal resolution at fixed locations but are spatially constrained. Lagrangian devices provide broad spatial coverage by following water parcel trajectories, but they yield lower temporal resolution when referenced to a fixed point in space. Indirect methods can cover very large areas but rely on assumptions and models to infer velocities.

Water velocity measurement can also be organized according to the spatial coverage and resolution of the sensors, which determines the type of information they provide. Three primary configurations are commonly used in marine monitoring:

• Point measurements capture water velocity at a single location, offering high temporal resolution. This configuration is commonly used with Eulerian sensors deployed on moorings, seabed frames, or piers.
• Profiling measurements record water velocity along vertical or horizontal profiles. ADCPs are the most widely used profiling instruments, providing simultaneous velocity data at multiple depths. Profiling allows the characterization of vertical shear, stratification, and boundary layer dynamics, which are critical for understanding sediment transport, mixing processes, and ecological patterns.
• Synoptic or area measurements provide spatially extensive coverage, often near the surface. Remote sensing systems, including RADAR and satellite-based techniques, fall into this category. These measurements capture large-scale circulation patterns over kilometers to hundreds of kilometers, although temporal resolution is typically lower than point or profiling measurements.

Selecting the appropriate measurement configuration depends on the research objectives, the spatial and temporal scales of interest, and the constraints of the deployment environment. Combining multiple approaches is often used to provide a more comprehensive understanding of water motion dynamics.

3. Operational Environments and Deployment Platforms

The successful measurement of water flow depends not only on the choice of sensors but also on the conditions in which they are deployed and the platforms that support them. Environmental factors such as turbulence, stratification, biofouling, pressure, and accessibility impose constraints on sensor operation, maintenance, and data quality. Likewise, the selection of deployment platforms determines the spatial and temporal coverage of measurements. This section provides an overview of typical marine operational environments, presents the challenges they pose, and describes the platforms commonly used to deploy flow sensors in these conditions.

3.1. Operational Environments

Marine environments present a wide range of physical conditions that strongly influence sensor selection, deployment strategies, and data quality. Understanding these environments is essential for designing effective monitoring campaigns and ensuring reliable measurements.

• Coastal zones are characterized by high variability in water velocity magnitude and direction, strong tidal currents, waves and breaking-waves spots, and interactions with river inflows. Sensors deployed in these areas must withstand turbulence, frequent biofouling, and possible sediment resuspension. Typical platforms include moorings, pier-mounted instruments, and shallow-water buoys.
• Estuaries feature strong gradients in salinity and density, combined with complex flow patterns resulting from the interplay of freshwater inflow and tidal forcing. High turbulence and stratification require sensors capable of resolving vertical velocity profiles and rapid temporal changes. Point measurements, profiling instruments, and distributed sensor networks are commonly employed.
• Offshore and deep-sea environments impose challenges such as high hydrostatic pressure, low temperatures, and limited access for maintenance. Long-term autonomous operation is often required, demanding energy-efficient sensors and robust housing. Profiling instruments on moorings, gliders, and AUVs are commonly used, with careful attention to pressure tolerance and reliability under harsh conditions.
• Polar and ice-covered seas introduce additional constraints, including ice-induced damage, extreme cold, and limited light. Deployment platforms must accommodate ice movement and restricted access for maintenance or retrieval. Surface and subsurface drifters, ice-tethered profilers, and autonomous vehicles designed for polar conditions can be employed.

Selecting deployment strategies requires careful consideration of these environments, the spatial and temporal scales of interest, and the robustness and energy requirements of the measurement system. Awareness of environmental constraints ensures reliable data collection and extends sensor lifetime, particularly in remote or harsh marine settings.

In addition to the environments described above, laboratory and controlled settings, such as flumes and wave tanks, deserve a brief mention due to their key role in sensor testing and calibration. These facilities provide precise control over flow velocity, turbulence, and stratification, enabling systematic performance evaluation before deployment in the field.

3.2. Deployment Platforms

The choice of deployment platform is critical for successful measurements, as it determines spatial coverage, sensor stability, accessibility, and maintenance requirements. Different platforms are suited to different marine environments and measurement objectives:

• Fixed moorings are typically small-scale, purpose-built platforms deployed specifically for a monitoring campaign. They are relatively low-cost, deployed temporarily or semi-permanently, and usually host a few sensors at selected depths. They are widely used in coastal and estuarine environments to provide continuous and high-temporal-resolution data. Mooring offers stability and robustness but requires careful design to minimize drag, vibration, and biofouling.
• Coastal and offshore installations include fixed infrastructure such as HF RADAR stations, tidal energy platforms, offshore wind foundations, and marine cables. Like fixed moorings, these platforms provide stable mounting for multiple sensors and can deliver continuous, large-scale monitoring. They are particularly valuable for capturing spatially extensive surface water velocity and supporting long-term environmental observation, though they require significant investment and are constrained to specific locations.
• Surface and subsurface moored buoys provide flexible platforms for point or profiling sensors and are often equipped with telemetry systems for real-time data transmission. Surface buoys are suitable for coastal and open-ocean deployments but can be affected by waves and wind-induced motion. Subsurface buoys reduce surface interference and can host sensors at multiple depths, although they are generally more complex to deploy and maintain. Mooring lines can be used with both types of buoys, allowing multiple single-point or profiler meters to be fixed at different depths, from the seabed to the surface or subsurface.
• Ships and research vessels allow measurements along transects, enabling the collection of spatially extensive data. Ships are versatile and can carry multiple instruments, but they are expensive, require specialized crew, and provide only temporary measurement coverage along their tracks.
• Unmanned Water Vehicles, such as AUVs, gliders, and surface vehicles, offer mobile platforms capable of collecting high-resolution profiles over extended distances and durations. They are particularly valuable in offshore and deep-sea environments where fixed deployments are challenging. Energy efficiency, navigation accuracy, and pressure tolerance are key considerations.
• Drifters and floats are Lagrangian platforms that move with the water, providing information about transport pathways and large-scale circulation. Small surface drifters can be used to study local dispersion in estuaries and wave-breaking coastal areas, while subsurface floats, such as Argo-style profilers, are employed for deep-ocean monitoring. These platforms allow broad spatial coverage but offer limited temporal resolution at a fixed point.
• Aerial vehicles, including drones and crewed aircraft, enable synoptic surveys of coastal and inland waters. They support imaging, thermal mapping, and surface velocity estimation using optical techniques. Flight endurance, sensor payload, and weather constraints influence their operational capability.
• Satellites provide large-scale, repeated observations of ocean and coastal dynamics. They support measurements of sea-surface height, surface currents, waves, winds, and temperature. While limited by spatial resolution and cloud cover for optical techniques, they offer unparalleled global coverage and long-term continuity.

Selecting a platform requires balancing environmental conditions, spatial and temporal coverage, sensor type, and logistical constraints. A combination of platforms can be used to provide comprehensive measurements across multiple scales and environments, maximizing data quality and scientific return.

Table 3. Deployment platforms for water velocity and wave measurement.

Platform Type	Spatial Coverage	Deployment Duration	Power and Communication	Typical Applications	Key Advantages	Main Limitations
FIXED MOORINGS	Local (point to profile)	Weeks to years	Battery; acoustic or cabled	Coastal and estuarine monitoring, turbulence studies	Stable reference frame, high temporal resolution	Limited spatial coverage, biofouling, maintenance
COASTAL AND OFFSHORE INSTALLATIONS	Local to regional	Long-term (years)	Grid-powered; cabled or RF	Long-term monitoring, energy sites, coastal dynamics	High power availability, continuous data streams	Fixed location, high installation cost
SURFACE BUOYS	Local to regional	Months to years	Battery/solar; RF or satellite	Waves, surface flows, meteorology	Easy deployment, surface access	Wave-induced motion, vandalism, weather exposure
SUBSURFACE BUOYS	Local to profile	Months to years	Battery; acoustic telemetry	Subsurface currents, stratified flows	Reduced surface interference	Complex recovery, limited real-time access
SHIPS AND RESEARCH VESSELS	Regional to basin scale	Hours to weeks	Ship power; real-time links	Surveys, transects, calibration campaigns	Flexibility, large payloads	High operational cost, limited temporal coverage
UNMANNED UNDERWATER VEHICLES	Local to regional	Days to months	Battery; acoustic/RF links	Offshore surveys, adaptive sampling	Mobility, access to remote areas	Energy constraints, navigation uncertainty
DRIFTERS AND PROFILING FLOATS	Regional to basin scale	Weeks to years	Battery; satellite	Surface and subsurface transport, dispersion studies	Lagrangian sampling, low cost	Limited control, sensor drift, recovery challenges
AERIAL VEHICLES (UAVS)	Local to regional (surface)	Minutes to hours	Battery; RF	Surface flow mapping, coastal surveys	Rapid deployment, non-intrusive	Weather sensitivity, limited endurance
SATELLITES	Regional to global	Years to decades	Solar-powered; downlink	Large-scale circulation, wave fields	Global coverage, long-term consistency	Limited resolution, indirect measurements

summarizes the main deployment platforms used in marine environments, highlighting their spatial coverage, operational duration, power and communication characteristics, and typical applications.

4. Sensor Technologies for Water Flow Measurement

A wide variety of sensor technologies have been developed to measure water flow. These instruments differ in their measurement principles, configurations, robustness, and suitability for specific spatial and temporal scales. While some technologies, such as mechanical propeller meters, have a long history of use, others, such as Microelectromechanical Systems (MEMS) based sensors, represent recent advances in miniaturization and low-power designs. The choice of sensors depends not only on the measurement objective—whether resolving turbulence at centimeter scales or profiling basin-scale flows—but also on the deployment platform and operational environment. Factors such as cost, maintenance requirements, calibration stability, susceptibility to biofouling, and energy consumption strongly influence their applicability.

Table 4. Comparison of sensor technologies for water velocity, flow, and wave measurements in marine environments.

Sensor Technology	Measurement Principle	Primary Measured Quantity	Spatial Coverage	Temporal Resolution	Typical Environments	Key Strengths	Main Limitations
MECHANICAL	Drag or rotation induced by flow	Water velocity	Point	High	Rivers, coastal waters	Simple, low power, historically well established	Moving parts, biofouling, limited turbulence and low-flow response
ELECTROMAGNETIC	Voltage induced by conductive fluid motion in a magnetic field	Water velocity, currents (with compass)	Point	High	Coastal, estuarine, freshwater	No moving parts, robust in sediment-laden flows	Sensitive to conductivity variations, alignment required
PRESSURE-BASED	Pressure differences related to dynamic or wave-induced motion	Velocity (inferred), wave elevation	Point	High	Coastal zones, channels, wave buoys	Simple, low power, suitable for shallow water	Indirect measurement, calibration sensitive
ACOUSTIC DOPPLER	Doppler shift of backscattered acoustic signals	Velocity profile, currents (with compass)	Point, Profile (up to a dozen meters)	Moderate to high	Coastal to deep ocean	Profiling capability, mature technology	Power demand, acoustic interference, sidelobe effects
ACOUSTIC TIME-OF-FLIGHT	Differential acoustic travel time	Water velocity (path-averaged)	Point/short path	Very high	Channels, pipelines, controlled flows	High precision, low processing complexity	Limited spatial coverage, alignment constraints
OPTICAL AND IMAGING	Tracking of particles or patterns in illuminated flow	Velocity field, surface velocity	Point to 2D/3D fields	Very high	Laboratory, near-surface, clear waters	High spatial resolution	Limited range, turbidity sensitivity, lighting requirements
MEMS-BASED	Thermal, resistive, piezoelectric, or capacitive response to flow	Water velocity	Point	High	Distributed networks, low-cost platforms	Compact, low power, scalable	Drift, calibration sensitivity, limited robustness
INERTIAL	Integration of acceleration and rotation	Motion-corrected velocity, wave motion	Platform dependent	High	Wave buoys, AUVs, drifters	Essential for wave measurements and motion correction	Indirect velocity inference, drift
LAGRANGIAN	Tracking of drifting platforms	Trajectory derived velocity	Regional to basin scale	Low to moderate	Coastal to open ocean	Captures transport and dispersion	Limited control, sparse sampling
REMOTE SENSING	Electromagnetic backscatter or surface roughness	Surface velocity, wave parameters	Regional to global	Low to moderate	Coastal and open ocean	Synoptic coverage, long-term monitoring	Indirect measurement, limited subsurface insight
HYBRID	Combination of sensing principles or data fusion	Velocity, waves, transport	Multi-scale	Variable	Research and operational settings	Overcomes single sensor limitations	System complexity, limited standardization

provides a comparative overview of the main sensor technologies used for water velocity, flow-related quantities, and wave motion measurements, summarizing their measurement principles, spatial coverage, strengths, and limitations. In the following subsections, the main classes of water flow sensors are reviewed according to their measurement principle. For each class, the working concept, representative technologies, advantages and limitations, and typical applications in marine research and monitoring are presented. Each subsection expands upon the corresponding row.

4.1. Traditional Mechanical Sensors

Traditional mechanical sensors represent the earliest generation of instruments for measuring water velocity in marine environments. Their fundamental operating principle is based on the direct interaction between moving water and a mechanical component, most commonly a propeller, rotor, or vane. As the flow passes over the sensor, it induces rotation or deflection, and the rate of this mechanical movement is proportional to the local water velocity. By recording the number of rotations or the angular displacement over a given time interval, an estimate of water velocity can be derived.

Among the most well-known instruments are propeller-type current meters (Figure 1), such as the Ekman current meter introduced in the early twentieth century [10]. These and similar devices typically consist of a propeller mounted on an axis aligned with the flow. The propeller spins as water moves through it, and the rotations are registered mechanically (in early designs) or electronically (in more modern adaptations). Another widespread design is the Savonius rotor [11], which employs a cup or vane-shaped rotor that is driven by flow from any horizontal direction, enabling omnidirectional velocity measurement without alignment mechanisms. Variations in these designs have been deployed on fixed moorings, towed instruments, and profiling platforms.

Mechanical sensors offer several advantages that contribute to their long history of use. Their design is conceptually simple and easy to understand, making them accessible to a broad range of users, from early oceanographers to engineers testing flow conditions in laboratory flumes. They are also relatively inexpensive to manufacture and deploy compared to modern electronic instruments, and calibration procedures are well established and widely documented. For these reasons, mechanical flow meters remain a practical choice for quick initial assessments, especially where operating conditions can be carefully controlled, and long-term reliability or automation is less critical.

Despite these strengths, mechanical sensors have notable limitations that restrict their utility in contemporary oceanographic practice. The reliance on moving parts introduces vulnerability to wear, drag, and mechanical failure, particularly in long-term deployments exposed to harsh conditions. They also require a minimum flow to start moving, limiting their ability to measure low velocities. Biofouling is a significant problem, as the growth of organisms on propellers or rotors can alter their hydrodynamic properties and reduce measurement accuracy. Similarly, debris or suspended particles can obstruct moving components, leading to measurement errors or sensor failure. These issues are particularly pronounced in estuarine and coastal environments, where turbulence, sediment resuspension, and biological activity are high.

Another limitation is their restricted temporal resolution. Mechanical sensors tend to integrate velocity over a given period, which smooths out short-term fluctuations but makes them unsuitable for resolving turbulence or high-frequency variability. Furthermore, they are inherently point sensors, meaning they provide measurements only at a specific location. To capture spatial variability, multiple instruments must be deployed, which increases logistical complexity. As a result of these limitations, mechanical current meters have largely been supplanted by technologies that avoid moving parts and offer improved robustness, higher temporal resolution, and greater suitability for long-term autonomous deployments.

4.2. Electromagnetic Sensors

Electromagnetic (EM) sensors constitute another traditional class of instruments for measuring water velocity in marine and freshwater environments (Figure 1). The method of measuring seawater velocity using the electromagnetic induction principle was first systematically described by R.W. Guelke in 1947 [12]. The operation is based on Faraday’s law of electromagnetic induction, which states that when a conductive fluid such as seawater moves through a magnetic field, an electric potential is induced that is proportional to the velocity of the fluid. By measuring this induced voltage across pairs of electrodes positioned in the sensor, it is possible to obtain a direct estimate of the local water velocity. A typical electromagnetic flowmeter consists of a magnetic field generator, usually implemented with permanent magnets or electromagnetic coils, and electrodes mounted orthogonally to both the flow direction and the magnetic field [13]. As seawater passes through the sensing volume, the interaction generates an electric field whose strength depends on the water’s velocity. This signal is captured by the electrodes and converted into an electronic output, which can then be calibrated against known flow conditions.

Electromagnetic sensors offer several advantages that make them attractive for marine applications. Unlike mechanical devices, they contain no moving parts, which eliminates wear and reduces the risk of mechanical failure. This makes them less sensitive to turbulence-induced noise, since the signal arises from the induced potential rather than mechanical inertia. Their response time is generally fast, allowing them to resolve rapid changes in velocity. Furthermore, electromagnetic sensors provide direct measurements of water velocity at a point, without requiring assumptions about particle scattering (as in acoustic sensors) or tracer concentration (as in optical methods). They can operate equally well in turbid or sediment-laden environments where acoustic or optical methods may fail, and are less affected by the presence of bubbles or suspended matter.

Nevertheless, electromagnetic sensors also have important limitations. A key challenge lies in their sensitivity to electrical and magnetic interference. Nearby sources of electromagnetic noise, such as power cables, electronic equipment, or even geomagnetic fluctuations, can cause signal interference. Additionally, in instruments that use a compass to resolve ocean currents, the sensor’s own generated magnetic field can interfere with the compass, affecting the measured orientation. Careful shielding and signal processing are required to ensure reliable measurements. The sensor output is influenced not only by velocity but also by the electrical conductivity of the water. Since seawater is highly conductive, this issue is less significant in open-ocean applications. However, in estuarine, tidal environments and freshwater contexts, where conductivity can change considerably over short time scales, measurements may lead to erroneous velocity estimates if the data are not carefully calibrated and interpreted.

Biofouling also poses a concern, as the electrodes must remain in direct contact with the water. The growth of organisms or deposition of material on electrode surfaces can alter the effective sensing area, leading to drift or signal attenuation. Regular cleaning or the application of antifouling measures is often required for long-term deployments. In addition, electromagnetic sensors typically measure velocity at a single point and in a single direction, meaning that multiple sensors or multi-axis electrode arrangements are needed to resolve the full three-dimensional velocity vector.

In terms of deployment, electromagnetic flowmeters are often mounted on fixed platforms such as moorings, buoys, or coastal installations, where they can provide continuous records of velocity at one or several depths. They are particularly useful in shallow waters, estuaries, and regions with high turbidity where acoustic instruments are challenged. Portable, hand-held electromagnetic current meters are also widely used in hydrological studies and engineering surveys for spot measurements. In oceanography, their role is sometimes complementary to acoustic Doppler devices, providing robust point measurements in environments where acoustics are unreliable.

4.3. Pressure-Based Sensors

Pressure-based sensors are widely used in marine monitoring to infer water velocity, wave motion, and vertical water-level fluctuations from measurements of hydrostatic and hydrodynamic pressure. Unlike sensors that directly measure velocity, these instruments estimate flow or surface motion by relating pressure variations to fluid dynamics using physical laws, empirical relationships, or numerical models.

Classical flow measurement devices, such as Pitot tubes, determine velocity from the pressure difference between the stagnation (total) pressure at the tube opening and the static pressure of the surrounding fluid [14]. According to Bernoulli’s principle, this pressure difference scales with the square of the flow velocity, enabling simple and robust point measurements. Differential-pressure sensors extend this concept by measuring pressure drops across constrictions or shaped flow elements (e.g., Venturi tubes, orifice plates [15]), with the induced pressure gradient proportional to the local velocity. These systems are mechanically simple, low-cost, and tolerant to harsh conditions, and remain widely used in pipelines, tidal channels, and shallow coastal flows where profiling is unnecessary.

Pressure-based instruments are typically simple, robust, and low power, allowing battery-operated deployments over long periods. They are also cost-effective, making them suitable for small-scale monitoring campaigns. However, they also have limitations. Careful calibration and alignment with the flow are required, and biofouling or debris can alter pressure readings. They provide indirect velocity measurements, so turbulence, flow separation, or complex flow patterns can introduce errors. Additionally, they cannot provide full velocity profiles without multiple sensors. Recent research has explored miniaturized differential pressure sensors, including MEMS-based designs, for distributed marine monitoring.

In addition to velocity sensing, pressure technology plays a central role in wave and water-level measurements (Figure 1). High-resolution bottom pressure recorders (BPRs) are a standard tool for quantifying surface gravity waves, tsunami signals, and storm surges. By measuring small variations in the overlying water column height, they can estimate wave height, wave period, and spectral energy distributions with high accuracy. Accurate wave measurements with BPRs require deployment at relatively shallow depths or coastal locations and the use of high-resolution sensors. BPRs form the backbone of several national and international tsunami early-warning systems. Examples include the Deep-ocean Assessment and Reporting of Tsunamis (DART) network operated by NOAA [16], the Seafloor Observation Network for Earthquakes and Tsunamis (S-NET) and Dense Ocean floor Network for Earthquakes and Tsunamis (DONET) systems deployed around Japan [16], and the North-East Pacific Time-series Undersea Networked Experiments (NEPTUNE) Canada cabled observatory [16]. Additional pressure-based wave and tsunami monitoring systems exist in the Pacific, Indian, and Atlantic Oceans, including long-term observatories in Hawaii and integrated networks that support numerical nowcasting and rapid-response warnings.

Pressure-based sensors are frequently mounted on bottom frames, fixed moorings, pier pilings, and benthic landers to obtain high-frequency measurements of velocity-derived pressure fluctuations, wave-induced orbital motions, and tide-driven water-level changes. Because they are not affected by turbidity, suspended sediments, or bubble clouds, pressure sensors remain reliable in highly energetic or shallow zones where acoustic systems experience signal degradation [17].

4.4. Acoustic Sensors

Acoustic techniques are currently the most popular approach for measuring water velocity in marine environments. These instruments operate based on either the acoustic time-of-flight method or the Doppler shift principle (Figure 2), with Doppler-based technology being the most adopted.

4.4.1. Doppler Acoustic Sensors

Doppler shift operation describes the change in frequency of an acoustic signal when it is scattered by particles suspended in a moving fluid. By transmitting sound waves into the water and detecting the frequency shift of the backscattered signal, it is possible to estimate the velocity of the water along the acoustic beam [18]. Since natural seawater contains sufficient scatterers such as plankton, sediment, and microbubbles, no artificial tracers are required.

There are two main categories of Doppler acoustic instruments in oceanographic practice. The first are Acoustic Doppler Velocimeters (ADVs), which are designed to measure water velocity at a single point with very high temporal resolution [19]. An ADV consists of a single transmitter and three or more receivers arranged in a known geometry. The device measures the Doppler shift in a small sampling volume located a short distance from the instrument, typically a few centimeters. ADVs are widely used in laboratory flumes, estuarine studies, and turbulence research, as they can resolve fluctuations at frequencies up to several tens of hertz [20].

The second and most important category in large-scale monitoring is Acoustic Doppler Current Profilers (ADCPs). These instruments employ multiple acoustic beams to measure velocity along a vertical or horizontal profile or to perform single-point measurements [21,22]. By dividing the acoustic return into depth bins, ADCPs can reconstruct velocity profiles across the water column. The use of multiple beams allows for the three-dimensional velocity vector to be resolved, assuming homogeneity within the sampling volume. ADCPs have become a cornerstone of oceanographic research and monitoring due to their ability to provide spatially extensive and temporally continuous velocity data. Modern ADCPs and ADVs are equipped with compasses, which are essential for transforming the measured velocities along the instrument’s beam axes into the Earth reference frame. Tilt sensors are also included to compensate for any inclination of the instrument, ensuring accurate current measurements [23].

In addition to their central role in measuring ocean currents, ADCPs have proven invaluable for studying suspended sediments and particle dynamics. The intensity of the acoustic backscatter, often treated as a secondary data product, can be calibrated and interpreted to estimate suspended sediment concentration and particle size distribution [24,25,26]. This capability enables researchers to investigate processes such as sediment transport, resuspension events, turbidity dynamics, and the impact of storms or river plumes on coastal and shelf systems.

In theory, any suspended particle in the water can be characterized using ADCPs. For example, recent studies have applied this approach to biological targets, such as zooplankton [27,28,29], air bubbles [30,31], ice particles [32,33,34], and plastics [35,36]. Although an advantage, this wide variety of targets also presents a challenge, potentially leading to identification errors, reduced accuracy, and complex calibration needs. The performance of Doppler sensors depends on the presence of sufficient acoustic scatterers. In very clear, particle-free waters, signal strength can be too low for reliable measurements. Conversely, environments with high sediment concentrations or strong bubble entrainment (e.g., surf zones, near-breaking waves) can produce spurious signals. Beam geometry and assumptions about homogeneity introduce additional uncertainties, particularly in highly stratified or turbulent flows. In addition, Doppler equipment is often considered relatively expensive, and its power requirements can be high for long-term autonomous deployments.

Still, by simultaneously providing velocity and suspended matter measurements, ADCPs offer a powerful tool for coupling hydrodynamic, geological, and biological observations, which is critical for understanding sediment dynamics, seabed evolution, and the health of benthic ecosystems. ADCPs can also be deployed in a variety of configurations, including bottom-mounted upward-looking frames, downward-looking installations on surface buoys, hull-mounted systems on research vessels, and even integration into autonomous platforms such as gliders and AUVs [37]. The versatility of acoustic Doppler–based technology is reflected in the wide range of commercial products available. Prominent manufacturers such as Teledyne RDI, Nortek, SonTek, and Rowe Technologies offer Doppler instruments designed for applications ranging from estuarine studies to offshore profiling. This technology is supported by decades of development and research, with robust calibration procedures and widespread user communities.

Table 5. Representative specifications of commercially available Acoustic Doppler Current Profilers (ADCPs) commonly used in marine and coastal monitoring. Values are typical and may vary with deployment configuration and environmental conditions.

Instrument	Operating Frequency	Measuring RANGE	Accuracy	Resolution	Max. Ping Rate	Maximum Profile Range	Maximum Depth
AANDREAA SEAGRADII DCP	600 kHz	±4 or 5 m/s	0.3 cm/s or 1.5%	0.1 cm/s	10 Hz	80 m	300 m
SONARDYNE ORIGIN 600 ADCP	625 kHz	±2 or 3.75 m/s	0.5%	1 cm/s	4 Hz	60 m	150 m
TELEDYNE SENTINEL V V50	500 kHz	±5 m/s	1 cm/s	0.1 cm/s	4 Hz	80 m	200 m
NORTEK SIGNATURE 500	500 kHz	±2.5 or 5 m/s	0.3%	0.1 cm/s	8 Hz	70 m	300 m
ROWE SEAWATCH	600 kHz	±5 m/s	0.25% or 2 mm/s	0.01 cm/s	10 Hz	78 m	300 m

summarizes representative specifications of similar widely used commercial ADCPs, highlighting their operating frequency, velocity range, accuracy, resolution, ping rate, profile range, and depth rating.

4.4.2. Time-of-Flight Acoustic Sensors

Time-of-flight (ToF) acoustic sensors are used for point measurements of water velocity. Their principle of operation relies on transmitting an acoustic pulse between two transducers placed at a fixed distance apart and measuring the travel time of the signal through the moving water. Because the flow either accelerates or delays the pulse depending on its direction relative to the acoustic path, the difference in travel time allows the velocity component along the path to be calculated [38,39].

Like acoustic Doppler instruments, ToF sensors provide reliable measurements without moving parts, but they differ in several important ways. ToF instruments are usually more compact, capable of higher temporal resolution, and involve fewer complex data processing compared to Doppler devices, making them attractive for both laboratory experiments and field deployments on fixed moorings. They are particularly advantageous in highly turbulent flows, where short sampling volumes are required, or in environments where bubbles, suspended sediments, or weak scatterers may interfere with Doppler backscattering. Another important feature is that, unlike Doppler sensors, the relationship between acoustic travel time and water velocity can be derived directly from geometry and sound speed, often reducing the need for empirical calibration. While variations in fluid density (through changes in salinity, temperature, or pressure) can affect acoustic propagation speed, this limitation can be mitigated using bi-directional transmission (i.e., transducer A to B and B to A), which mathematically cancels propagation effects and isolates the flow contribution.

The main limitation of ToF systems is that they provide only point or path-averaged measurements, resulting in reduced spatial coverage compared to profiling ADCPs. Characterizing complex or multi-directional flow fields often requires deploying multiple ToF sensors in an array, which increases deployment complexity. Proper alignment of transducers with the flow direction is also critical. This is relatively straightforward in fluvial or estuarine studies, where flow direction is strongly constrained along a channel, but becomes impractical in open-ocean settings, where currents vary in both magnitude and direction. For these reasons, acoustic ToF sensors have found greater application in industrial and engineering domains, such as pipeline and process flow monitoring, than in large-scale oceanographic deployments. Nevertheless, their compactness, high temporal resolution, and robustness make them valuable for specialized marine applications, particularly in semi-controlled fluvial or estuarine monitoring where flow geometry is well defined.

4.5. Optical and Imaging Sensors

Optical and imaging techniques provide complementary approaches to mechanical, electromagnetic, and acoustic methods for measuring water velocity, with particular strengths in shallow-water and coastal marine environments where visual access to the flow is possible. These techniques rely on the interaction of light with suspended particles or tracers in the water (Figure 2), or on the direct observation of surface patterns. While less widely applied offshore than acoustic systems, they offer unique advantages for high-resolution, non-intrusive measurements.

4.5.1. Particle Image Velocimetry

One important family of techniques is Particle Image Velocimetry (PIV) [40]. PIV tracks the motion of tracer particles suspended in the flow. The technique relies on illuminating a plane of the fluid with a thin sheet of light, typically generated by a pulsed laser. Tracer particles, which are assumed to faithfully follow the flow, scatter the laser light and are recorded by a camera. Two successive images of the illuminated plane are captured at a known time interval. By comparing the positions of particle patterns between the two images, the displacement of the particles is determined. This comparison is usually carried out through statistical cross-correlation within small interrogation windows of the image. Dividing the measured displacement by the time interval gives the velocity vector of the flow at each interrogation window. By repeating this process across the entire field of view, a two-dimensional velocity map of the illuminated plane is obtained. Extensions of the method include stereoscopic PIV [41], which uses two cameras to reconstruct three-dimensional velocity components in a plane, and tomographic PIV [42,43], which uses multiple cameras to resolve full three-dimensional flow structures.

Although originally developed for laboratory applications, adaptations of PIV have been successfully applied in coastal and estuarine waters. For example, underwater PIV has been used to study turbulence and near-bed flows around coral reefs [44], helping to quantify how these ecosystems modify local hydrodynamics and nutrient exchange [45]. PIV has also been applied near man-made structures, such as bridge piers and turbine blades, to characterize wake structures and flow–structure interactions [46,47]. Despite these advances, its use in the field remains constrained by challenges such as light attenuation, water turbidity, and the need for tracer particles. The accuracy of PIV depends on several factors: the ability of tracer particles to follow the flow without lag, the quality and thickness of the laser sheet, the resolution and frame rate of the imaging system, and the robustness of the cross-correlation algorithms.

4.5.2. Laser Doppler Velocimetry

Another well-established approach is Laser Doppler Velocimetry (LDV). The method is based on the Doppler shift of laser light scattered by small tracer particles moving with the flow [48] (similar to acoustic Doppler technology). In a typical LDV setup, a coherent laser beam is split into two beams of equal intensity. These beams are directed to intersect at a measurement volume within the flow, creating an interference fringe pattern. When tracer particles pass through this illuminated volume, they scatter light that is modulated by the Doppler effect. The scattered light is collected by a photodetector, which records a signal oscillating at a frequency proportional to the particle velocity component perpendicular to the fringes. The measured Doppler frequency shift is directly related to the particle velocity according to the known laser wavelength, the geometry of the intersecting beams, and the scattering angle. By using different optical arrangements, LDV systems can measure one, two, or all three components of velocity at the measurement point.

LDV provides precise velocity measurements with temporal resolutions up to several kilohertz, making it especially valuable for turbulence studies and validation of numerical models. Unlike PIV, which measures velocity fields, LDV gives point-based measurements, but with a higher frequency response and accuracy. This method requires the presence of tracer particles to scatter the laser light and sufficient water clarity to allow optical access to the measurement volume. This makes LDV particularly well-suited for controlled laboratory experiments, flume studies, and field deployments in clear estuarine or coastal waters, but more challenging in turbid or highly absorbing environments.

4.5.3. Large-Scale Particle Image Velocimetry

In recent years, large-scale optical surface velocimetry techniques have gained significant attention in marine and estuarine settings. Methods such as Large-Scale Particle Image Velocimetry (LSPIV) and related video-based approaches estimate surface water velocity from the movement of tracers like foam, bubbles, or floating debris observed in video recordings [49,50]. These measurements can be expressed in an Earth-fixed reference frame and interpreted as surface currents if combined with camera calibration and georeferencing. When deployed from drones, fixed coastal cameras, or vessels, these methods can cover large spatial areas at relatively low cost. Case studies include the use of drone-based LSPIV to map surface circulation in tidal inlets, river plumes, and surf zones, providing valuable information for sediment transport modelling and water quality assessment [51,52,53]. These techniques are particularly promising in intertidal and nearshore zones where conventional in-water sensors are difficult to deploy. Their performance, however, depends on stable lighting and the presence of visible tracers.

Commercial availability of optical and imaging systems for marine applications is more limited than for acoustic sensors. Specialized companies (e.g., Dantec Dynamics, TSI Incorporated) provide PIV and LDV systems that have been adapted for underwater and field use, while off-the-shelf software for LSPIV and drone-based imaging is increasingly applied in estuarine and coastal monitoring. Recent research emphasizes advances in high-resolution cameras, underwater imaging systems, and automated image processing. Integration with machine learning and computer vision holds the potential to improve tracer detection and velocity estimation, making optical measurements more robust in turbid or dynamic marine conditions. Overall, optical and imaging sensors are not yet as widely used in the open ocean as acoustic instruments, but they are gaining space in estuaries, tidal channels, and nearshore coastal environments. Their ability to provide spatially extensive, high-resolution velocity data makes them especially valuable for studying surface circulation, sediment transport, and turbulence in dynamic marine systems.

Table 6. Representative configurations of PIV and LDV systems (typical experimental setups reported in the literature).

Technique	Typical Hardware	Measurement Volume/Field	Camera Rate	Velocity Range	Spatial Resolution (Order)	Notes
PIV (LAB SCALE)	LaVision FlowMaster + dual-head laser	~100 mm × 100 mm	1–5 kHz	0–10 m/s	10 µm	Controlled lab setups for turbulence/wake studies
PIV (LARGE SCALE)	High-speed cameras + laser	~0.5–2 m field	500–1000 Hz	0–5 m/s	100 µm	Coastal flume or field optical setups
LDV (LAB)	TSI LDV with 532 nm laser	Small focused volume (~1 mm3)	Continuous	0–20 m/s	10 µm	Point measurement with high temporal resolution
STEREO PIV	Dual cameras + laser sheet	2D vector field	500–2000 Hz	0–10 m/s	10–50 µm	Captures two velocity components
TOMOGRAPHIC PIV	Multi-camera PIV	3D volume (~10 × 10 × 10 cm)	500–1000 Hz	0–5 m/s	50–100 µm	3D flow field reconstruction

summarizes representative PIV and LDV system configurations commonly used in marine flow characterization studies. Because these systems are highly configurable and often assembled for specific experimental conditions, the values listed are indicative rather than absolute.

4.6. MEMS Sensors

Microelectromechanical Systems (MEMS) represent one of the most recent and dynamic approaches to water velocity measurement. By miniaturizing mechanical, electrical, and in some cases optical components onto a single chip, MEMS devices offer unique advantages for flow sensing. Their compact size, low power consumption, and potential for low-cost mass production make them particularly attractive for emerging applications such as distributed sensing networks, autonomous platforms, and Internet-of-Things (IoT)-based ocean monitoring [54]. The strengths of MEMS flow sensors include their suitability for integration into underwater and surface vehicles and infrastructures, as well as their potential for large-scale deployments where traditional instruments would be impractical or prohibitively expensive. In addition, their ability to deliver high-frequency measurements makes them promising tools for turbulence studies and fine-scale hydrodynamic observations.

Despite these advantages, MEMS-based flow sensors face important challenges when transitioning from laboratory to field use. They are generally more fragile than conventional instruments and remain vulnerable to biofouling, corrosion, and pressure effects in harsh marine environments. Furthermore, calibration can be complex due to cross-sensitivity to temperature, salinity, and pressure variations. For these reasons, most MEMS water velocity sensors remain at the prototype or experimental stage, though rapid progress is being made toward robust, field-ready devices. Also, many MEMS flow technologies reported in the literature are tested with gases. While the operating principle remains the same for water, ensuring the watertightness of the electronic components introduces additional challenges.

Recent advances include the development of protective coatings and antifouling strategies to extend operational lifetimes, the integration of sensors into multi-parameter lab-on-chip platforms, and their combination with wireless communication technologies for real-time IoT-enabled monitoring. These innovations position MEMS-based sensors as a hot topic in marine instrumentation, with the potential to complement and eventually rival established acoustic flow measurement techniques.

Depending on the underlying physical principle, MEMS flow sensors can be broadly categorized into thermal, piezoresistive, piezoelectric, and capacitive types, each offering distinct advantages and limitations for flow measurement.

4.6.1. Thermal

Thermal MEMS flow sensors are the most extensively developed microfabricated devices for fluid velocity measurement. They rely on the interaction between a heated element and the surrounding fluid, where the flow modifies heat transfer and, consequently, the sensor’s electrical output. Three main working principles are in use: hot-wire/hot-film anemometry, calorimetric sensing, and thermal time-of-flight approach (Figure 3).

In hot-wire and hot-film anemometry, a microheater is electrically heated, and its resistance varies with temperature [54]. The surrounding fluid removes heat through convection, and the rate of heat loss is proportional to the flow velocity. The system can operate under constant-temperature or constant-power modes. These sensors are highly sensitive and respond quickly, especially at very low flow velocities, which makes them attractive for turbulence studies. Their main drawbacks are cross-sensitivity to temperature, density, and viscosity, as well as the fragility of the thin elements, which limits long-term deployment in harsh marine environments.

Calorimetric sensors operate with a central heater placed between two temperature sensors, one upstream and one downstream [55,56]. In a still fluid, the temperature distribution is symmetric. When flow is present, the downstream sensor records a higher temperature, and the difference between the two is proportional to both the magnitude and direction of the flow. Calorimetric devices are more robust than hot-wire sensors and provide directional information, but their sensitivity at lower fluid velocities is lower. In addition, they depend strongly on the thermal properties of the medium, which complicate calibration when salinity and temperature vary, as is the case in estuarine environments.

Thermal time-of-flight is a more recent approach that mimics acoustic ToF but uses temperature convection. In this technology, a heater generates periodic thermal pulses that propagate downstream and are detected by a temperature sensor [57]. The time delay between pulse generation and detection is directly related to flow velocity. This method provides absolute measurements that are less affected by heater drift and offers the potential for greater accuracy under changing conditions. However, it requires more complex and precise electronics, is less mature than the other two methods, and may suffer from signal instability in turbulent or highly conductive fluids such as seawater.

In summary, thermal MEMS sensors offer high sensitivity and the possibility of multi-axis measurement, but their reliance on fluid thermal properties presents calibration challenges in real marine environments. Recent research has focused on closed-loop temperature control, protective coatings to mitigate corrosion and biofouling, and packaging strategies to improve pressure resistance. Thermal MEMS remain the most mature class of MEMS flow sensors, yet their widespread use in field deployments will depend on improving robustness, simplifying calibration, watertight meets, and applied field results.

4.6.2. Piezoresistive

Piezoresistive flow sensors exploit the change in electrical resistance of a material under mechanical strain [58]. When a flow interacts with a mechanical structure, the induced deformation produces strain in the integrated piezoresistive elements. This strain alters their resistance, which can then be measured and correlated with flow velocity or direction. The principle is simple and robust, and because piezoresistive readout circuits are straightforward to implement, these sensors are also among the most widely studied MEMS-based flow devices. Different structural designs have been explored (Figure 3). Cantilever- and beam-based sensors are the most common, where fluid drag bends a suspended element and strain gauges record the deflection [59]. Diaphragm structures use thin membranes that deform under flow-induced pressure differences, offering high sensitivity but often limited bandwidth [60]. Hair-like structures, frequently inspired by the fish lateral line system, use artificial cupula-like components coupled with piezoresistive bases to measure both flow intensity and direction [61].

The main advantages of piezoresistive MEMS flow sensors are their simplicity of fabrication, direct electrical readout, and good scalability. They are also less dependent on external power compared to thermal sensors, making them attractive for low-power and long-term monitoring systems. However, piezoresistive sensors are generally less sensitive than piezoelectric-based counterparts and susceptible to drift caused by temperature and pressure variations. Another drawback is that they typically require mechanical elements exposed to the flow, which can be vulnerable to biofouling, corrosion, or mechanical damage in harsh marine environments. Recent studies have sought to improve these limitations by integrating full Wheatstone bridge circuits to minimize temperature effects, optimizing the geometry of cantilevers or diaphragms for specific flow ranges, and developing biomimetic cupula structures to enhance sensitivity at low Reynolds numbers [62]. MEMS-based resistive technologies, such as strain-gauge sensors, have also been used in larger (non-microscale) cantilever-type structures for in situ measurements of water velocity and discharge [63]. While most piezoresistive MEMS flow sensors remain in the laboratory stage, their potential for dual-axis flow measurement, low energy consumption, and compatibility with complementary metal–oxide–semiconductor (CMOS) circuitry make them promising candidates for future environmental and marine applications.

4.6.3. Piezoelectric

Piezoelectric MEMS flow sensors rely on the ability of piezoelectric materials to generate an electrical charge when subjected to mechanical stress [64]. When a flow induces vibrations, bending, or pressure on a piezoelectric element, the resulting strain creates a measurable voltage signal without the need for an external excitation source. This self-generating capability is one of the most important advantages of piezoelectric sensors, as it enables very low-power or even passive operation. In fact, the piezoelectric mechanism has also been investigated for energy harvesting from induced flow [65].

Like piezoresistive devices, several structural configurations have been explored. Cantilever-based sensors are common, where flow drag produces oscillations that are directly converted into electrical signals by the embedded piezoelectric film [66,67]. Diaphragm structures have also been used, taking advantage of uniform pressure distributions to induce strain across the piezoelectric layer [68]. Many hair-like biomimetic designs have been used [69,70,71], in which a flexible cupula transmits flow-induced forces to an underlying piezoelectric element, enabling detection of both flow velocity and environmental vibrations such as pressure waves. These biomimetic devices are particularly interesting because they extend sensor utility beyond flow measurement into object detection and navigation, which is particularly relevant for gliders and AUVs’ operations.

The main advantages of piezoelectric MEMS flow sensors are their high sensitivity, low power consumption, and wide dynamic range. They can respond quickly to rapid changes in flow and are especially well-suited to detecting oscillatory or turbulent signals. However, piezoelectric sensors can only measure dynamic rather than static signals, meaning they underperform with steady-state flow, and can be sensitive to environmental factors such as temperature. The electronic instrumentation is more challenging than for their resistive counterparts, and their fabrication is often more complex, requiring deposition and patterning of piezoelectric thin films and specialized electric wiring. Recent developments have focused on biomimetic piezoelectric arrays, where multiple hair-like sensors are arranged to provide spatial resolution of flow fields, and on improving material stability for underwater applications. These advances suggest that piezoelectric MEMS flow sensors could complement or surpass piezoresistive sensors in applications where high-frequency flow variations are critical.

4.6.4. Capacitive

Capacitive MEMS flow sensors exploit the principle that fluid-induced deformation of a mechanical structure (such as a diaphragm, cantilever, or beam) alters the capacitance between conductive electrodes [72]. This variation in capacitance can be measured with high precision and directly related to flow velocity, pressure gradients, or turbulence intensity. A typical design consists of a flexible cantilever structure suspended above a fixed electrode [73]. When flow acts on the structure, its displacement changes the gap distance or overlapping area between electrodes, producing a measurable capacitance variation. Capacitive sensing offers high sensitivity, low noise, and good linearity, making it attractive for flow measurement in both air and water.

One of the main advantages of capacitive sensors is their low power consumption, since they only require small AC excitation signals rather than continuous electric current. They also exhibit good thermal stability compared to piezoresistive or piezoelectric counterparts, which makes them less prone to environmental drift. In addition, they are compatible with CMOS integration, enabling compact, low-cost, and multifunctional sensor arrays. However, capacitive MEMS flow sensors are susceptible to parasitic capacitances from packaging and electric wiring, which can degrade signal-to-noise ratios, particularly in marine environments where humidity and salinity may interfere. Their performance is also strongly dependent on the dielectric properties of the surrounding medium, meaning that calibration can be complex. Furthermore, while capacitive sensors are excellent at detecting small displacements, their measurement range is often narrower than that of other MEMS devices. Improvements in packaging and antifouling strategies hold the potential to make capacitive MEMS flow sensors significant future tools for real-world aquatic monitoring.

4.6.5. Other Approaches

Beyond thermal, piezoresistive, piezoelectric, and capacitive designs, other alternative MEMS-based approaches have been explored for fluid velocity measurement. These methods, although less widespread, demonstrate the versatility of MEMS technology and its potential to leverage different transduction principles.

One promising category is optical MEMS flow sensors, which use light transmission or interference to detect structural deformation caused by fluid flow. Micro-mirrors, waveguides, fiber-optical or interferometric structures can be integrated into MEMS devices, allowing extremely sensitive displacement detection [74,75]. The optical principle of operation is inherently immune to electromagnetic interference and can achieve high resolution, but optical sensors typically require more complex fabrication and packaging, as well as precise alignment, which limits their robustness.

Another approach involves resonant MEMS sensors, in which flow-induced forces alter the natural frequency of a vibrating microstructure [76,77]. By monitoring shifts in resonance, flow velocity, or pressure changes can be inferred with high sensitivity. Resonant sensors have the advantage of excellent long-term stability and compatibility with digital signal processing, but they can be highly susceptible to damping effects in liquid environments, which reduces their performance in water compared to air.

Many other alternative MEMS flow sensors use the cantilever bending principle but rely on transduction mechanisms other than piezoresistive, piezoelectric, or capacitive methods. Examples include measuring cantilever deformation through cameras [78], electromagnetic coupling [79], Coriolis effect [80], or inertial systems [81]. Although these alternative MEMS approaches remain largely experimental, they highlight the rapid diversification of flow-sensing technologies. As fabrication techniques and protective strategies continue to advance, such alternative MEMS sensors may gain increasing relevance in specialized applications, complementing more established technologies.

4.7. Inertial Systems

Inertial systems determine motion using accelerometers, gyroscopes, and sometimes magnetometers, which together form an Inertial Measurement Unit (IMU). Although they do not measure water velocity directly, they are essential for estimating flow, correcting platform motion, and resolving wave dynamics and turbulence. Their importance has grown as marine platforms increasingly operate in environments where direct velocity measurements are difficult or where instrument motion significantly affects observations. IMUs measure linear acceleration, rotational velocity, and the Earth’s magnetic field, which can be integrated to reconstruct platform trajectory, orientation, and heave. Strapdown inertial navigation algorithms perform this reconstruction, which becomes more accurate when aided by external references. By comparing the platform’s absolute trajectory with its water-relative motion (measured or inferred), inertial systems enable the estimation of surrounding flow.

In wave measurements, inertial sensors are also central to modern buoys and floaters (Figure 4), which typically rely on tri-axial accelerometers and gyroscopes to estimate heave, pitch, and roll [82,83]. These quantities are transformed into wave spectra using reconstruction techniques such as Maximum Likelihood Estimation (MLE) and Fast Fourier Transform (FFT)-based analyses. Inertial measurements support the estimation of significant wave height, periods, directional spreading, crest elevation, and near-surface orbital velocities.

For underwater vehicles, inertial systems provide continuous estimates of position and velocity when global positioning system (GPS) is unavailable. Fusing IMU data with Doppler Velocity Log (DVL) measurements allows separation of vehicle motion from ambient water movement, enabling accurate water-relative velocity estimation during transects in spatially varying currents [84,85]. The performance of these systems depends on bias stability, drift compensation, and the frequency of external updates. High-grade fiber-optic and ring-laser gyros offer enhanced precision but are costly and energy-intensive, while modern MEMS IMUs have improved substantially through better calibration, thermal compensation, and sensor fusion techniques. Drifters and profiling floats increasingly incorporate IMUs to resolve wave-induced drift, Stokes drift, and small-scale velocity variations. Combining GPS-derived trajectories with inertial measurements allows reconstruction of motion between intermittent satellite fixes and captures high-frequency components that GPS alone would smooth out. This capability is especially important in coastal and estuarine regions, where wave–current interactions strongly influence the motion of drifting platforms.

Despite their versatility, inertial systems face inherent limitations. Stand-alone IMUs experience drift, bias instability, and long-term integration error, which makes unaided inertial navigation unsuitable for extended velocity estimation. Underwater platforms lack continuous GPS access and therefore require frequent aid from DVLs or acoustic beacons. Because inertial systems infer water motion indirectly by removing platform movement from absolute motion, uncertainties arise when hydrodynamic forces affect the platform itself. For wave measurements, low-frequency drift must be filtered out, while high-frequency noise must be controlled to avoid distorting the spectral estimates. Ongoing developments include improved multi-sensor fusion, machine-learning-based drift correction, and the use of high-rate inertial measurements to resolve fine-scale flow features that are difficult to capture with conventional sensors.

4.8. Lagrangian Approaches

Lagrangian approaches measure water velocity by tracking the movement of objects or particles as they are carried by the flow, rather than measuring velocity at a fixed point [86]. This perspective contrasts with Eulerian methods, where sensors provide data at a fixed location. Lagrangian measurements are particularly useful in environments with complex, time-varying flows, such as tidal inlets, estuaries, and coastal regions, where flow structures can vary over multiple spatial and temporal scales.

The most common implementation is through drifters and floats. These devices are deployed at the surface or at depth and are allowed to drift with the current while logging position via GPS, inertial, or acoustic tracking [87,88]. By analyzing the trajectories, researchers can derive surface or subsurface velocities, transport pathways, and Lagrangian statistics such as dispersion and residence times. Examples include GPS-tracked surface drifters in coastal regions, as well as subsurface floats used to study estuarine exchange or internal tidal currents. The ARGO program remains the most important pioneering case of success in Lagrangian ocean current measurements conducted in open waters, using floaters that can descend to depths of up to 2000 m and remain submerged for several days before resurfacing to transmit the collected data [89,90]. AUVs and gliders can also be set to operate in Lagrangian mode (Figure 4), adjusting their propulsion to minimize deviation from the flow and effectively acting as “smart drifters” [91]. This allows the measurement of water parcels over longer periods and distances, providing insights into three-dimensional flow structures, mixing, and transport processes.

Lagrangian approaches capture the integrated effect of flow over space and time, providing insights into transport, dispersion, and connectivity that point measurements cannot. They can be deployed over large spatial scales, making them suitable where deploying dense arrays of fixed sensors is impractical. They are particularly useful for validating numerical models, which often simulate Lagrangian trajectories or particle transport. However, Lagrangian methods also have limitations. They provide limited spatial resolution of instantaneous velocity fields, as measurements are only available along the trajectory of the device. Deployment and recovery can be logistically challenging and costly, especially in open-ocean or high-energy environments, and surface drifters are particularly affected by windage and waves, which must be accounted for when interpreting results [92,93].

Recent advances include the integration of high-frequency GPS, high-resolution IMUs, and multi-parametric sensors to improve trajectory accuracy and collect additional environmental data (temperature, salinity, pressure). Additionally, networked drifter fleets and real-time communication allow researchers to track and adapt deployments dynamically, enabling studies of tidal dynamics, estuarine flushing, and coastal circulation at unprecedented temporal and spatial scales [94,95,96].

4.9. Remote Sensing

Remote sensing provides an alternative to in situ sensors by measuring water velocity fields from above the surface, either using land-based, airborne, or satellite platforms (Figure 4) [97]. These approaches are particularly valuable for large-scale observations, covering spatial domains that are inaccessible or impractical for direct sensor deployment. While typically limited to surface velocities, they play an essential role in monitoring coastal dynamics, circulation patterns, and riverine inflows to the ocean.

The most widely adopted technique is the High Frequency RADAR, which has been extensively deployed in coastal regions worldwide [98,99]. Systems such as CODAR (Coastal Ocean Dynamics Applications Radar) and WERA (Wellen Radar) operate by transmitting radio waves that undergo Bragg scattering from ocean surface waves [100,101]. By analyzing the Doppler shift of the Bragg peaks, radial surface velocities along the antenna beams can be obtained; combining data from two or more stations allows the reconstruction of two-dimensional surface velocity fields over tens to hundreds of kilometers offshore.

Table 7. Specifications of CODAR SeaSonde and WERA Helzel systems for surface current monitoring.

System	Operating Frequency	Typical Range (Surface Current)	Spatial Resolution	Temporal Resolution	Typical Velocity Accuracy
CODAR SEASONDE	4.3–11 MHz (long range) 11–25 MHz (standard) 20–50 MHz (high-resolution)	160–220 km (long range) 50–100 km (standard) 10–20 km (High resolution)	3–12 km (long range) 500 m–3 km (standard) 200–500 m (high-resolution)	~hourly	~<7 cm/s total current; ~2 cm/s tidal component
WERA HELZEL	4.4–9.3 MHz (long range) 13.5–27 MHz (medium range) 39–50 MHz (VHF system)	170–400 km (long range) 40–100 km (medium range) 18–25 km (VHF system)	1.5–5 km (long range) 1–3 km (medium range) 230–1000 m (VHF system)	5–9 min (long and medium ranges) 3–5 min (VHF system)	~2–5 cm/s (vector current)

summarizes the principal specifications of each system for surface current velocity measurement.

HF RADAR technology provides continuous, real-time monitoring and Earth-referenced data, making it indispensable for coastal circulation studies, search-and-rescue operations, and oil spill response. It is typically used to access ocean surface motion but can also be used for near-surface and ice shell inspection [102]. However, its accuracy decreases in strong wave conditions or in regions with complex coastal geometry, and its spatial resolution is typically on the order of kilometers. RADAR technology requires a stable, well-grounded site with sufficient space for the antennas, so shore-based fixed installations are typically used. Yet, experiments employing motion-correction algorithms have been conducted to enable their use on vessels and offshore structures [103,104].

Satellite-based RADAR altimetry also contributes to surface motion measurement, though indirectly. By measuring sea surface height with centimeter accuracy, altimetry allows the derivation of geostrophic currents at basin scales [105,106]. While powerful for large-scale circulation, this approach is limited in spatial resolution and does not capture small-scale coastal or estuarine dynamics. Recent satellite missions, such as Sentinel-6 and Surface Water and Ocean Topography (SWOT), have significantly improved spatial and temporal resolution, creating new opportunities for coastal and nearshore applications. For example, modern satellite altimetry has expanded its capabilities beyond traditional open-ocean sea-level measurements to include observations in more challenging environments, such as ice-covered regions [107].

Emerging airborne approaches are also expanding the capabilities of remote sensing for marine flow measurement [108,109,110]. Unmanned Aerial Vehicles (UAVs or drones) equipped with cameras, optical or RADAR sensors can rapidly acquire high-resolution surface velocity data in estuaries, tidal channels, and nearshore areas. These platforms allow targeted measurements in regions that are difficult to access with boats or moorings, and their mobility enables repeated surveys over short timescales to capture tidal and storm-driven variability. Advanced drones can integrate Light Detection and Ranging (LiDAR) or high-frame-rate imaging with LSPIV techniques, providing centimeter-scale resolution over hundreds of meters. While these systems are still largely experimental, they offer significant potential for complementing HF RADAR and satellite observations, particularly for estuarine and coastal management applications.

Remote sensing techniques complement in situ measurements by providing broad spatial coverage and continuous monitoring of surface flows. Their limitations include restriction to surface velocities, dependency on atmospheric or sea-state conditions, and the need for calibration and validation within in situ data. Despite these constraints, remote sensing approaches have become indispensable in integrated observing systems, especially when combined with moorings, drifters, and numerical models to create a comprehensive picture of marine circulation.

4.10. Emerging and Hybrid Approaches

In addition to established technologies, other emerging and hybrid approaches are being explored. Although most of these approaches remain in early stages of development or deployment, they highlight the ongoing innovation in marine flow sensing. These methods aim to overcome limitations of traditional sensors by combining multiple physical principles, leveraging recent advances in materials science, optics, and data processing, or integrating low-cost sensing into broader ocean observing systems.

4.10.1. Hybrid Approaches

One promising direction is the development of hybrid acoustic-optical systems, which combine the range and robustness of acoustic sensors with the high-resolution capabilities of optical methods. For example, prototype systems have coupled acoustic with optical principles to provide both turbulence measurements at millimeter scales and flow profiles over several meters [111,112]. These hybrid approaches are being tested in coastal laboratories, and tidal channels were resolving both fine-scale turbulence and larger-scale circulation is essential.

Another example of a hybrid approach is the RADAR–optical fusion [113,114]. It combines the broad, all-weather coverage of RADAR systems with the high spatial detail of optical imaging (drone- or shore-based LSPIV, dense optical-flow methods, or satellite optical imagery) to produce more accurate and spatially complete surface velocity fields over limited areas and under favorable illumination and visibility. Data fusion is implemented using geostatistical or variational methods (e.g., optimal interpolation, Kalman filters, or variational data assimilation) or multi-sensor blending schemes that respect each dataset’s error characteristics and sampling geometry. These approaches improve gap-filling, reduce noise near HF RADAR baselines, and enhance representation of small-scale features such as tidal jets, eddies in inlets, and plume fronts by injecting high-resolution optical information into the RADAR-derived fields.

4.10.2. Distributed Fiber Optical Sensing and Marine Cables

Distributed fiber optic sensing is also gaining attention. Techniques such as Distributed Acoustic Sensing (DAS) or Brillouin-based Distributed Temperature Strain systems use optical-fiber cables as continuous, dense arrays of acoustic and strain sensors. A laser is periodically injected into the fiber, and backscattered light (via Rayleigh scattering) is analyzed to detect changes in phase or amplitude that correspond to mechanical strain along the fiber. Because the fiber spans large distances, DAS delivers spatially distributed strain measurements at meter-scale or better resolution, making it a powerful tool for monitoring mechanical vibrations induced by oceanographic processes. In marine and coastal environments, DAS deployed on subsea telecommunication cables enables the detection of surface gravity waves, vortex-induced vibrations, and ocean currents [115]. For example, surface waves impart pressure variations on the seabed that deform the fiber, and DAS measures the resulting strain. By applying ambient-noise interferometry and frequency-domain signal-processing techniques (such as waveform-stretching or beamforming), researchers have derived current velocity and depth profiles along long cable sections [116,117]. At present, such approaches are primarily suited for qualitative or large-scale flow characterization and structural–hydrodynamic interaction studies, as uncertainties in the retrieved current velocities remain significant and are not yet compatible with operational ocean current monitoring.

DAS technology faces significant challenges: fiber-seabed coupling can vary, which affects signal fidelity, and the strain response must be carefully calibrated and interpreted to infer hydrodynamic processes (e.g., translating measured vibration into current velocity), which requires sophisticated signal processing and physical modelling. Despite these hurdles, DAS is rapidly maturing, with growing applications in tsunami early-warning, ambient-noise oceanography, and structural monitoring of marine installations [118,119].

The integration of sensing capabilities directly into submarine cables is an emerging frontier in ocean observation. Traditionally, marine telecommunication cables were a passive infrastructure whose role was limited to data transmission. Recent technological advances have enabled these cables to act simultaneously as communication backbones and distributed sensor arrays capable of monitoring seismic activity, temperature, strain, oceanographic processes, and structural health. A new generation of “smart” marine cables is under development, which will allow water velocity and surface motion estimations at strategic locations along the cable route [120,121].

4.10.3. Data-Driven Approaches

Finally, data-driven approaches are increasingly being used to complement—or in some cases partially replace—traditional physical sensing. Recent work has shown that even sparse in situ measurements, when combined with machine-learning or variational data-assimilation frameworks, can yield flow reconstructions with spatial and temporal resolutions far beyond the footprint of the sensors themselves [122,123,124]. In estuaries and coastal systems, where strong nonlinear dynamics and energetic boundary processes make dense observations difficult, these hybrid observational–modelling systems have proven especially valuable.

Machine-learning methods such as deep neural networks, Gaussian Process regression, and physics-informed neural networks have been trained to infer full velocity fields from partial ADCP transects, fixed-point moorings, or limited surface drifters. These models effectively learn the underlying dynamical structure of the flow and interpolate across spatial gaps that would otherwise require costly deployments. In parallel, 4D-Var and Ensemble Kalman Filter schemes are increasingly being used to ingest ADCP observations into high-resolution hydrodynamic models, constraining model uncertainty while allowing predictions to be propagated forward in time [125,126].

Because these approaches embed sensor data within a physically guided or statistically learned framework, they offer a powerful extension to traditional monitoring: sparse observations maintain physical realism, while the model or learning architecture fills in structure where direct measurements are unavailable. As a result, hybrid data-model systems are becoming a practical solution for reconstructing estuarine circulation, quantifying transport pathways, forecasting tidal and subtidal flows, and supporting operational management tasks such as contaminant tracking and navigation safety [127,128].

5. Conclusions

The accurate measurement of water flow in marine environments relies not only on sensor technology but also on strategic deployment, rigorous calibration, and integration within observational networks. Over recent decades, significant improvements have been achieved in all these aspects, enabling measurements that are more accurate, spatially extensive, energy-efficient, and operationally robust than earlier generations of instruments. Sensor performance is governed by multiple factors, including accuracy, precision, spatial and temporal resolution, environmental robustness, and energy requirements. Selection of sensors should therefore be guided by the spatial and temporal scales of interest, balancing measurement fidelity with coverage, feasibility, and monitoring objectives. For example, high-resolution turbulence and fine-scale variability are effectively captured by modern acoustic Doppler instruments and optical velocimetry, which have benefited from advances in signal processing, beam geometry, and onboard computation. In contrast, pressure-based and Lagrangian approaches generally provide lower-frequency or integrated observations but have seen substantial gains in long-term stability and deployment simplicity. MEMS-based sensors, while still limited in absolute precision, now offer unprecedented opportunities for high spatial density measurements due to their reduced size, cost, and power consumption.

Environmental factors continue to significantly affect sensor reliability and longevity, but notable progress has been made in mitigating these challenges. Biofouling, corrosion, suspended sediments, bubbles, and variations in temperature or salinity can alter readings and reduce instrument lifespan. Acoustic and optical sensors remain sensitive to turbidity, whereas mechanical, pressure-based, and MEMS devices may be compromised by debris accumulation or fouling on exposed surfaces. Electromagnetic-based sensors require careful calibration in estuarine and tidal regions with strong salinity gradients. Nevertheless, improvements in protective housings, low-toxicity antifouling coatings, sensor encapsulation, and materials engineering have extended deployment durations and reduced maintenance requirements [129,130,131]. While MEMS sensors are still far from being consolidated instruments capable of withstanding the harsh conditions of the ocean environment, their rapid development trajectory suggests increasing robustness and reliability in the near future. Protective housings, antifouling measures, and optimized deployment strategies are thus essential to maintain data quality over long-term monitoring campaigns.

Energy autonomy has emerged as a central advancement in marine flow monitoring. Battery-powered sensors now benefit from ultra-low-power electronics, adaptive sampling strategies, and intelligent duty-cycling, allowing long-duration deployments that were previously impractical. Energy harvesting approaches—drawing from solar, wave, tidal, or microbial sources—have transitioned from experimental concepts to operational components in surface buoys and fixed coastal installations [132,133]. These developments have directly enabled persistent monitoring and real-time data transmission in remote or hard-to-access environments.

Calibration and validation remain fundamental to ensure data reliability and comparability across platforms. Laboratory calibration establishes baseline sensor performance, while field calibration increasingly accounts for environmental variability through in situ reference measurements and cross-comparison among co-located instruments. Integration with numerical models and data assimilation frameworks allows sparse observations to be extrapolated over larger spatial domains, enhancing the representativeness of collected data. These practices mark a new shift from isolated measurements toward coordinated, system-level observation strategies.

Cost reduction and accessibility represent one of the most transformative improvements in recent years. While high-end acoustic or optical systems provide exceptional accuracy, their high cost can limit dense spatial coverage. In contrast, low-cost MEMS, pressure-based, and optical sensors have enabled distributed networks with broader spatial and temporal coverage at reduced expense. This democratization of sensing technology has expanded monitoring capabilities to smaller research groups, developing regions, and citizen-science initiatives, fundamentally changing how ocean monitoring is conducted [134].

Future developments in marine flow measurement are driven by emerging technologies and hybrid approaches. Acoustic-optical systems can offer high-resolution turbulence measurements with broad profiling capabilities, while imaging-based methods, such as LSPIV deployed via drones or shore-based cameras, provide non-intrusive surface velocity mapping over larger areas. Distributed fibre-optic sensing leverages existing submarine cables for long-range monitoring, and networked low-cost MEMS sensors mounted on drifters, buoys, or autonomous vehicles enable dense, real-time coverage. These advances reflect a shift from single-instrument solutions toward coordinated, multi-sensor systems.

Data-driven and model-integrated approaches represent a major conceptual advancement. Machine learning, data assimilation, and hybrid methodologies allow sparse in situ measurements to be combined with remote sensing and numerical models, generating high-resolution velocity fields and predictive insights. Applications include coastal management, estuarine flushing studies, and hazard forecasting. The emergence of digital twins—virtual, continuously updated representations of marine environments—illustrates this evolution [135,136,137]. Their implementation involves (i) acquisition of multi-scale observations from in situ and remote platforms, (ii) data fusion and assimilation into physics-based numerical models, and (iii) continuous updating and validation using real-time observations. Despite their transformative potential, several bottlenecks remain, including limited spatial and temporal coverage of in situ measurements, sensor reliability and calibration drift, high computational cost, and challenges in real-time data interoperability. Potential solutions include the deployment of distributed and low-power sensor networks, standardized data architectures, advances in edge computing, and adaptive sampling strategies. By integrating real-time sensor data with physics-based models, digital twins can provide persistent, high-fidelity reconstructions of flow and wave dynamics. High-frequency measurements from ADCPs, HF RADARs, inertial units, and distributed sensor networks underpin these systems. Adaptive digital twins may autonomously direct observational effort, activating dormant sensors or redeploying platforms in regions of high uncertainty to optimize data collection.

Sustainability and ecological impact are increasingly shaping the design of next-generation monitoring systems. Research is advancing biodegradable or bio-inspired materials, low-toxicity antifouling solutions, modular and recyclable electronics, and compact sensor geometries that minimize hydrodynamic disturbance. These improvements support the expansion of observation networks while mitigating the environmental footprint. Despite progress, challenges persist, including long-term reliability, biofouling mitigation, power autonomy, and calibration under highly variable conditions. Continued advances in materials science, self-calibration strategies, edge computing, and data standardization are expected to further enhance durability, interoperability, and global integration. These developments support the expansion of observation networks while reducing environmental footprint and enabling long-term physical-to-digital implementations.

In summary, recent improvements in sensor miniaturization, energy efficiency, robustness, data integration, and cost-effectiveness have significantly advanced the state of the art in marine water flow monitoring. By combining established technologies with emerging approaches and integrated observational frameworks, it is now possible to achieve higher-resolution, larger-scale, and more sustainable monitoring than ever before. These advancements not only improve scientific understanding of coastal, estuarine, and open-ocean dynamics but also support operational decision-making for marine management, infrastructure protection, and hazard mitigation.