A method for inverting deep ocean current velocity based on acoustic morphology of methane plume

Abstract

The present application relates to the field of acoustic velocity measurement, in particular to a method for inverting mid-deep ocean current velocity based on acoustic morphology of methane plume, aiming at solving the problems of spatial and temporal limitations, high cost of existing ADCP direct measurement method, and low precision of existing traditional bubble plume inversion method due to the assumption of uniform linear motion of bubbles. Specifically, the method comprises the following steps: extracting the spatial coordinates of the target plume from the screened multi-beam data; obtaining a three-dimensional trajectory curve by weighted fitting coordinates; calculating the curvature and identifying the extreme points to divide the ocean current horizon; determining the horizontal offset of each section; constructing and correcting the bubble rising dynamics model considering temperature and salinity correction and hydrate shell effect to obtain the time required for the bubble to rise to each depth; finally calculating the layered ocean current velocity and direction by the ratio of offset and time. The present application can realize large-scale, low-cost and high-precision indirect inversion of mid-deep ocean current velocity.

Description

Technical Field

[0001] This invention relates to the field of acoustic velocimetry, specifically a method for inverting deep ocean current velocities based on the acoustic morphology of methane plumes. Background Technology

[0002] Ocean structure and current velocity are crucial components of the Earth system's natural evolution, and their long-term monitoring and modeling have consistently been a focus and research hotspot for global marine institutions. Solving scientific problems related to multi-scale ocean dynamic processes relies heavily on acquiring ocean profile structure and current velocity data. From a geological engineering perspective, excessively high-speed mid-deep ocean currents can erode sediments near the seabed, threatening facilities deployed on the seabed. Obtaining mid-deep ocean current velocities is essential for assessing engineering geological risks and conducting engineering maintenance. Acquiring mid-deep ocean current velocities enables underwater "transparency," providing necessary background data support for underwater activities. Therefore, acquiring mid-deep ocean current velocities is of great significance for solving scientific problems and ensuring the safety of engineering facilities.

[0003] Currently, the acquisition of ocean current velocities mainly relies on direct observation equipment such as Acoustic Doppler Current Profilers (ADCPs). The technical solutions are primarily divided into two types: Surfing Current Profiler (SADCP) and Deployed Current Profiler (LADCP), both of which acquire ocean current velocities through direct on-site measurement. SADCPs are installed on the hull of a ship, acquiring ocean current velocity data across the entire depth profile during navigation. This is suitable for large-scale surveys, but it suffers from near-bottom data blind spots and is susceptible to ship attitude. Furthermore, its data quality is often degraded by strong surface noise and complex seabed topography, and it requires dedicated voyages, resulting in high costs. LADCPs are installed on mooring systems, deployed to the seabed, and emit sound waves upwards to achieve continuous observation of the current velocity profile near the bottom. This method is suitable for long-term detailed surveys, but the observation location is fixed, spatial representativeness is poor, the equipment is easily displaced by ocean currents, and in the environment of scarce deep-sea scatterers, the echo signal is weak and the signal-to-noise ratio is low. Its deployment and retrieval operations are complex, with high engineering risks and costs.

[0004] To overcome the spatial and temporal limitations and cost issues of direct observation, scholars have recently begun exploring methods for indirectly inverting ocean current velocities based on natural marine phenomena. For example, some studies have proposed estimating ocean current velocities by measuring the inclination angle of bubble plumes. The basic idea is to treat the vertical rise velocity of the bubbles as constant and vector-synthesize it with the horizontal ocean current velocity. However, this method has significant shortcomings: because the bubbles are dragged and deflected by the ocean current during their ascent, their trajectory can reflect current velocity information. But existing techniques usually assume that the bubble rise velocity is constant and the plume trajectory is a straight line, ignoring the complexity of bubble dynamics and the actual curvature of the plume, resulting in low inversion accuracy.

[0005] Therefore, existing technologies still have shortcomings in terms of observation coverage, inversion accuracy, implementation cost, and physical model realism. There is an urgent need for a new method that integrates high-resolution acoustic observations with a bubble dynamics model that is more in line with physical reality, in order to achieve large-scale, high-precision, and low-cost indirect inversion of deep-sea current velocities and make up for the deficiencies of the existing technology system. Summary of the Invention

[0006] To address the shortcomings of existing deep-sea current velocity measurements, such as small range, low accuracy, and high cost, this invention proposes a method for retrieving mid-deep ocean current velocities based on the acoustic morphology of methane plumes. Specifically, it includes the following steps:

[0007] (1) Target methane bubble plume was selected from multibeam acoustic data and its spatial coordinates were extracted;

[0008] (2) Perform position-weighted fitting on the spatial coordinates to obtain the three-dimensional spatial trajectory curve;

[0009] (3) Calculate the curvature of the trajectory curve, identify the extreme points of curvature, and use them to represent different ocean current layers in segments;

[0010] (4) Determine the horizontal offset of the plume in each segment based on the curvature extrema;

[0011] (5) Establish a numerical model for the rise of a single bubble;

[0012] (6) Correct the parameters of the numerical model;

[0013] (7) Based on the modified numerical model, obtain the time required for the bubble to rise to the depth of each curvature extreme point;

[0014] (8) Calculate the velocity and direction of each ocean current based on the horizontal offset and the corresponding rise time of each segment.

[0015] This invention achieves large-scale, high-precision, and low-cost quantitative inversion of the velocity and direction of deep-sea currents by extracting the spatial coordinates of the plume from the multibeam profile, constructing a three-dimensional trajectory with weighted polynomial fitting, identifying curvature extrema to divide the ocean current layer, and coupling a temperature-salinity corrected bubble dissolution kinetics model to calculate the rise time.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0017] (1) In view of the problem that direct measurement methods such as ADCP have spatiotemporal limitations near the bottom layer, this invention proposes an indirect inversion method based on the acoustic morphology of methane bubble plumes: the three-dimensional trajectory of the plume is reconstructed using multibeam acoustic data, and the velocity is inverted by combining the physical dynamics model. This effectively solves the problem of "small range" in the observation of mid-deep ocean currents and significantly expands the spatiotemporal coverage of ocean current information.

[0018] (2) In view of the problem that existing ocean current observation requires the organization of separate high-cost scientific expeditions and the cost of a single expedition is high, the present invention proposes a "piggyback" flow field screening strategy: embedding plume identification and analysis into conventional marine geological or geophysical survey expeditions, and using existing equipment such as shipborne multibeam or side-scan sonar to simultaneously obtain flow field information, effectively solving the problem of "high cost" and significantly reducing the threshold for monitoring mid-deep ocean currents.

[0019] (3) To address the problems of model distortion and insufficient accuracy caused by the assumption of uniform bubble rise, straight trajectory, and neglect of dynamic evolution in existing indirect inversion methods, this invention proposes a high-fidelity inversion strategy that integrates simulation of real physical paths: on the one hand, weighted polynomial fitting is used in plume trajectory modeling to accurately characterize its curvature; on the other hand, in bubble rise dynamic modeling, the simplified assumption of average velocity throughout the entire process is abandoned, and instantaneous rise velocity is calculated based on depth integral units, coupled with the dynamic change of bubble diameter with depth. As a result, ocean current velocities can be inverted separately for water masses in different depth ranges, significantly refining the effective water mass thickness from hundreds of meters in traditional methods to tens of meters, greatly improving the vertical resolution and calculation accuracy of ocean current velocity inversion.

[0020] (4) In view of the problems of poor regional adaptability and reliance on local empirical parameter calibration of existing indirect methods, this invention proposes a standardized inversion strategy based on global universal ocean physical relations: the core parameters of the model are driven by the general ocean state equation, without the need for regional specialization adjustment, which effectively solves the problem of difficulty in promoting the method and has good global sea area applicability and engineering portability. Attached Figure Description

[0021] Figure 1 This invention provides a recording of bubble plumes bent by horizontal ocean currents on a continuous multibeam water body scanning acoustic time-series fan-shaped profile.

[0022] Figure 2 The spatial location coordinates of the plume extracted by this invention are projected onto a map.

[0023] Figure 3 A schematic diagram of the plume trajectory fitting function in this invention;

[0024] Figure 4 A schematic diagram illustrating the numerical simulation principle of a single methane bubble rising in water in this invention;

[0025] Figure 5 Numerical simulation results of the methane bubble rising process in this invention Figure 1 ;

[0026] Figure 6 Numerical simulation results of the methane bubble rising process in this invention Figure 2 ;

[0027] Figure 7 Comparison of simulated bubble rising trajectory results. Detailed Implementation

[0028] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0029] A method for inverting deep ocean current velocity based on the acoustic morphology of methane plumes includes the following specific implementation steps:

[0030] S1: Filter the target methane bubble plume from multibeam acoustic data and extract its spatial coordinates.

[0031] Select target methane bubble plumes that meet the following conditions (1), (2), (3), and (4) from multibeam acoustic data:

[0032] (1) The plume is located in the central beam coverage area. The difference in backscattering intensity between its top and bottom ends and the surrounding background seawater is obvious, and the boundary is clear, which makes it easy to accurately determine the starting and ending depths of the plume.

[0033] (2) The plume can extend for more than 200 meters in the water body and is recorded by multiple time-series multibeam profiles at the same location, so as to improve the reliability of the identification results through cross-validation of data;

[0034] (3) The trend of the plume's lateral sway with depth is significantly greater than 20 meters, in order to avoid the error caused by the lateral resolution of multibeam detection (usually no more than 10 meters);

[0035] (4) Select the location where multiple time points of the same location appear, that is, a single plume is covered and scanned multiple times by multiple beams, which can eliminate the interference of other accidental events (such as the flow field disturbance of local deep-sea fish schools) on the bubble movement.

[0036] After screening, multibeam acoustic processing software (such as QPS FMMidwater) is used to identify the target plumes, and each plume is analyzed on an acoustic profile (e.g., Figure 1 Derive the three-dimensional coordinates representing the spatial location of the plume from the data shown. Figure 2 As shown in the figure, the sampling points in each profile are spaced at the same interval to ensure the uniformity of subsequent fitting.

[0037] S2: Perform position-weighted fitting on the spatial coordinates to obtain the three-dimensional spatial trajectory curve.

[0038] The extracted spatial coordinates are arranged in a monotonically decreasing order of depth to form a discrete point sequence. Since the sampling of the plume by the multibeam system in different profiles is not equidistant, and the edge region is easily affected by noise or bubble diffusion, if a conventional fitting method is used, the obtained trajectory is prone to deviating locally from the high-density core region inside the plume, resulting in geometric distortion and affecting the accuracy of subsequent ocean current inversion.

[0039] To overcome the above problems, this invention employs a third- or fourth-order polynomial to perform a weighted fitting of the discrete point sequence. The weighting coefficient of each discrete point is positively correlated with its corresponding backscattering intensity. Regions with higher backscattering intensity in a single methane bubble plume typically have higher bubble density and are closer to the core rising channel of the plume. Therefore, high-intensity points are more likely to represent the high-probability locations where bubbles actually exist and should be assigned higher weights to dominate the trajectory fitting direction.

[0040] In addition, to further enhance physical plausibility, a weighted strategy is implemented:

[0041] (1) Assign weights to the coordinate points at the top and bottom of the plume, with weight values ​​between 10 and 20, to ensure that the curve fitting function mainly changes along the z-axis direction, thus better reflecting the bubble rising process.

[0042] (2) Assign corresponding weights to other coordinate points based on their corresponding backscattering amplitude intensity. :

[0043] ,

[0044] in, It extracts the backscatter intensity corresponding to point n in the plume data. It extracts the minimum backscatter intensity from the plume data. It extracts the maximum backscatter intensity from the plume data.

[0045] The parameterized three-dimensional space curve obtained after fitting is as follows:

[0046] ,

[0047] The fitting order is typically third or fourth, and should not exceed fourth, to avoid introducing non-physical inflection points in higher-order polynomials, which would violate the smoothness characteristics of the bubble ascent process. This embodiment, by introducing a weighted strategy, significantly improves the modeling accuracy and physical plausibility of the trajectory curve. It enhances the identification capability of the core ascent channel while avoiding non-physical inflection points, providing a reliable spatial path basis for subsequent accurate extraction of curvature extrema and stratified inversion of ocean current velocity.

[0048] S3: Calculate the curvature of the trajectory curve, identify the extreme points of curvature, and use them to represent different ocean current layers in segments.

[0049] The curvature of the above three-dimensional curve can be calculated using the following formula:

[0050] ,

[0051] in, It is the first derivative. It is the second derivative. It is the cross product of vectors. It represents the length of the product of vectors.

[0052] The spatial coordinates of curvature extrema are identified using the curvature formula. The curve segment between two adjacent extrema represents the segment influenced by ocean currents at the same stratum, exhibiting the same current velocity characteristics on the horizontal xy plane. Based on the statistical regularity of global ocean plume morphology, local curvature extrema typically do not exceed three. Therefore, the continuous plume curve is decomposed into several segments with different curvature characteristics. Each segment corresponds to an area dominated by a specific stratum current, thus physically linking the three-dimensional morphology of the plume to the vertical stratification of ocean currents. This provides crucial geometric segmentation basis for subsequent hierarchical and refined inversion of ocean current velocities at each stratum.

[0053] S4: Determine the horizontal offset of the plume in each segment based on the curvature extrema.

[0054] The curvature extrema are projected onto the trajectory curve function and filtered to select local curvature extrema within the target depth range as inflection points. For example... Figure 3 As shown, the vector formed from the bottom of the plume to the target inflection point. The vector length D projected onto the xy plane is the horizontal offset of the plume under the influence of the bottom ocean current; if there are multiple obvious target inflection points, the horizontal offset of each trajectory segment is calculated separately to support multi-layer ocean current inversion.

[0055] This embodiment achieves precise quantification and hierarchical extraction of plume horizontal offset by screening and projecting curvature extrema points: the method transforms the geometric features of the three-dimensional spatial trajectory into quantifiable and calculable horizontal displacement parameters, thereby directly inverting the intensity of the influence of ocean currents on bubble motion from a morphological perspective; by calculating the horizontal offset D of different trajectory segments in segments, it is possible to separate and identify the contributions of multiple ocean currents to the plume morphology, providing key displacement inputs for subsequent calculation of layered ocean current velocities, effectively improving the hierarchical resolution and physical rationality of the inversion results.

[0056] S5: Establish a numerical model for the rise of a single bubble

[0057] To obtain the correspondence between bubble depth and time in the vertical direction (z-axis), a numerical model of single bubble rise considering bubble dissolution dynamics is constructed.

[0058] Existing methods typically assume that bubbles rise at a uniform average velocity throughout the entire process. However, current research shows that the bubble's rising velocity has a non-linear relationship with its diameter and changes dynamically as the radius decreases during dissolution. Figure 4As shown, the bubble shrinks continuously during its ascent due to methane diffusion, and its trajectory is deflected by ocean current shear, eventually dissolving completely. This dynamic characteristic indicates that using the assumption of uniform velocity will cause the calculated depth of the bubble at any given time to deviate significantly from the actual value, thereby reducing the accuracy of ocean current inversion.

[0059] To address this issue, the numerical model abandons the simplified treatment of average velocity throughout the entire process, instead calculating the instantaneous ascent velocity based on depth integral units and coupling it with the dynamic evolution of bubble diameter with depth to accurately establish the depth-time mapping relationship. The core physical assumption of the model is that during the bubble's ascent, methane diffuses into the seawater through the gas-liquid interface, leading to a decrease in the number of bubble molecules and a reduction in bubble radius, eventually resulting in complete dissolution. Simultaneously, under deep-water conditions, a methane hydrate shell may form on the bubble surface, inhibiting the diffusion process. The governing equations of the numerical model are:

[0060] ,

[0061] in, It is the molar amount of a single methane bubble. It is the Henry's constant. It is hydrostatic pressure. This is the background methane concentration in seawater. It is the radius of a single methane bubble. It is the rising speed of methane bubbles in seawater. It is the mass transfer coefficient at the gas-liquid interface, and its value depends on the bubble diameter. It is determined by the following piecewise function:

[0062] , 0 < < 0.5cm,

[0063] , 0.5cm ≤ ≤ 1.3cm,

[0064] , >1.3cm,

[0065] in, The diffusion coefficient of methane in seawater is given by the exponent n, which depends on the presence of a hydrate shell.

[0066] like If the bubble size is ≤3.5cm and the bubble is located within the hydrate stability zone, then =0.667;

[0067] In other cases =0.52;

[0068] The upper boundary depth of the hydrate stability zone can be calculated using CSMHyK software or determined by the intersection of the methane hydrate phase equilibrium equation and the measured temperature and pressure profile.

[0069] This embodiment significantly improves the computational accuracy of the depth-time mapping relationship by constructing a numerical model that considers bubble dissolution dynamics and the variable diameter effect. This method abandons the simplistic assumption of uniform bubble ascent in traditional methods. By coupling the dynamic decrease in bubble diameter with depth with the diffusion process of methane in water, it achieves accurate calculation of the instantaneous bubble ascent velocity within the integral depth unit. Furthermore, the model incorporates the inhibitory effect of hydrate shells on mass transfer processes in the deep-sea environment, making the simulation of bubble ascent trajectories more consistent with the complex physicochemical environment of deep-sea cold seep areas. This provides a reliable time-scale benchmark for the subsequent accurate conversion of plume spatial geometry into ocean current velocity.

[0070] S6: Correct the parameters of the numerical model

[0071] With respect to Henry's constant Perform temperature and salinity correction:

[0072] Temperature correction uses the Van 't Hoff equation:

[0073] ,

[0074] in, It is temperature The corresponding Henry's constant, These are the parameters for the experimental fitting.

[0075] Salinity correction uses the Setschenow equation:

[0076] ,

[0077] in, and Henry's constant in pure water and salt solution, respectively, and the parameter. This is the salting-out constant. This represents the concentration of salt.

[0078] S7: Based on the corrected numerical model, obtain the time required for the bubble to rise to the depth of each curvature extreme point.

[0079] Solving the corrected master control equations yields numerical simulation results of the methane bubble rise process, including the bubble depth-radius relationship (e.g., ...). Figure 5 (as shown) and the bubble radius-time relationship (as shown) Figure 6(As shown). The simulated maximum bubble rise height was compared with the actual observed plume tip depth, and the numerical model was further corrected by adjusting the initial bubble radius. After the model correction was completed, the depth of the curvature extremum point was calculated. corresponding time This is the time it takes for the bubble to rise to this depth.

[0080] S8: Calculate the velocity and direction of the ocean currents in each layer based on the horizontal offset and corresponding rise time of each segment.

[0081] For each ocean current layer, the ocean current velocity v is calculated using the following formula:

[0082] ,

[0083] Where D is the horizontal offset corresponding to this layer. The rise time. The direction of the ocean current is determined by the displacement vector. The projection direction on the xy plane is determined. If the trajectory contains multiple significant curvature extrema, the above calculation is performed segment by segment to obtain the velocity and direction of multiple layers of horizontal ocean currents.

[0084] S9: Effect Verification

[0085] To verify the accuracy and improvement effect of the present invention, under the same set ocean current conditions (bottom layer 0.2 m / s, middle layer -0.01 m / s), the horizontal displacement simulated by the method of the present invention and the existing method at a water depth of 800 meters differed by 50 meters, with a relative deviation of 37%. Figure 7 This is a comparison of simulation results considering the variation in bubble rising velocity and the traditional uniform velocity assumption (see figure). This comparison visually demonstrates the significant improvement in simulation accuracy of this invention.

[0086] Furthermore, this method has been field-verified at a depth of 950 meters in the eastern waters of my country. The ocean current velocity and direction obtained by its inversion are closer to the ocean current profile data measured by shipborne ADCP during the same period than existing methods based on simplified assumptions. This result further confirms the reliability and technical superiority of the present invention in a real marine environment.

[0087] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A method for retrieving deep ocean current velocities based on the acoustic morphology inversion of methane plumes, characterized in that, Includes the following steps: S1: Filter the target methane bubble plume from the multibeam acoustic data and extract its spatial coordinates; the extraction of spatial coordinates specifically includes: (1) using multibeam acoustic processing software to identify the target plume; (2) exporting the coordinates representing the spatial position of the plume from each profile to ensure that the sampling points in each profile are close in interval; S2: Perform position-weighted fitting on the spatial coordinates to obtain a three-dimensional spatial trajectory curve; use a third- or fourth-order polynomial to perform position-weighted fitting on the above spatial coordinates, and assign higher weights to the following coordinate points: the coordinate points at the top and bottom of the plume and the coordinate points with high backscattering intensity. S3: Calculate the curvature of the trajectory curve, identify the extreme points of curvature, and use them to represent different ocean current layers in segments; the formula for calculating the curvature of the trajectory curve is as follows: ， in, Represents a parameterized three-dimensional space curve. It is the first derivative. It is the second derivative. It is the cross product of vectors. Represents the length of the product of vectors; S4: Determine the horizontal offset of the plume in each segment based on the curvature extremum points; the process of determining the curvature extremum points includes: (1) Projecting the curvature extremum points onto the trajectory curve function and filtering them, selecting the local curvature extremum points within the target depth range as inflection points; (2) The length of the vector projected onto the plane from the bottom of the plume to the target inflection point is the horizontal offset of the plume. S5: Establish a numerical model for the rise of a single bubble; The master equations of the model are as follows: ， in, It is the molar amount of a single methane bubble. It is the Henry's constant. It is hydrostatic pressure. This is the background methane concentration in seawater. It is the radius of a single methane bubble. It is the rising speed of methane bubbles in seawater. It is the mass transfer coefficient at the gas-liquid interface; The gas-liquid interface mass transfer coefficient K is based on the bubble diameter. Segmentation determination: (1) When When the value is greater than 0 and less than 0.5 cm: , (2) When When the value is greater than 0.5cm but less than 1.3cm: ， (3) When When it is greater than 1.3cm: , in, It is the diffusion coefficient, and n is the exponent; S6: Correct the parameters of the numerical model; S7: Based on the modified numerical model, obtain the time required for the bubble to rise to the depth of each curvature extreme point; S8: Calculate the velocity and direction of the ocean currents in each layer based on the horizontal offset of each segment and the corresponding rise time.

2. The method for retrieving deep ocean current velocity based on the acoustic morphology inversion of methane plumes according to claim 1, characterized in that: In step S6, when correcting the numerical model parameters, the effects of temperature and salinity are considered, and the Henry's constant is corrected using the van der Hoff equation and the Sechenov equation, respectively. .

3. The method for retrieving deep ocean current velocity based on the acoustic morphology inversion of methane plumes according to claim 2, characterized in that: The van der Hoff equations are as follows: ， in, It is temperature The corresponding Henry's constant, parameter Provided by experiments; The Sechenov equation is as follows: ， in, and These are the Henry's constants in pure water and salt solutions, respectively. This is the salting-out constant. This represents the concentration of salt.

4. The method for retrieving deep ocean current velocity based on the acoustic morphology inversion of methane plumes according to claim 1, characterized in that: In step S8, The ocean current velocity is calculated by dividing the horizontal offset by the bubble rise time. The direction of the ocean current is the direction of the vector projected onto the plane.

Images