A method suitable for monitoring and counting large water surface group of silver carp

Abstract

The present application relates to the technical field of fishery resources monitoring, in particular to a method suitable for big water surface group monitoring and counting of aar fish, according to the type of water area, the grid dynamic distribution of acoustic signal receiving station is carried out, the joint positioning method of more than three points based on TDOA, combined with the weighted least square method to calculate the accurate coordinate of the marked aar fish, and the behavior of the aar fish is judged, so as to master the activity area of the individual fish after the individual fish cluster through the marking of the individual fish, combined with the water depth, water temperature and water quality data, the characteristics of the fish habitat are obtained, and the fish activity range and rule are obtained. Based on the behavior habit of fish, the monitoring range is divided, the population monitoring and counting are carried out through the fish sonar scanning, the problems of big water surface fish population monitoring difficulty and fish population repeated counting are reduced, and the precision of the monitoring and counting result is improved. The scheme of the present application has the characteristics of non-damage, full coverage, high precision and standardization, and is suitable for big water surface aar fish resource investigation, evaluation of aar fish population release effect and ecological fishery management.

Description

Technical Field

[0001] This invention relates to the field of fishery resource monitoring technology, specifically to a method suitable for monitoring and counting bighead carp populations in large bodies of water. Background Technology

[0002] As a core species of freshwater aquaculture and ecological fisheries in my country, bighead carp plays a crucial role in large-scale stock enhancement, water quality control, and fishery resource assessment. Monitoring bighead carp populations mainly includes monitoring environmental indicators of their habitats (ecological capacity, water temperature, pH, dissolved oxygen, conductivity, turbidity, chlorophyll a, and cyanobacteria, etc.), obtaining the density, size distribution, and total biomass of bighead carp populations per unit water body, and assessing the behavior of bighead carp populations.

[0003] The current monitoring of bighead carp in large bodies of water generally faces the following technical bottlenecks: (1) Traditional sampling methods such as netting and electrofishing cause great damage to the fish and have low sampling coverage, making it difficult to reflect the true population size; (2) Existing technologies also use sonar scanning technology to monitor fish populations. In medium to large open water areas of 500 to 5000 mu, due to the limited area and shape of the water and the uncertain factors of fish activity, it is difficult to accurately estimate the precise location coordinates of the fish, and thus cannot obtain the fish's behavioral habits. When monitoring and counting fish populations, problems such as repeated detection and missed scanning of core areas are likely to occur, ultimately leading to a decrease in the accuracy of the monitoring and counting results. Therefore, there is an urgent need for a low-disturbance, full-coverage, high-precision, and continuously operable method for monitoring and counting bighead carp populations in large bodies of water. Summary of the Invention

[0004] The technical solution adopted by this invention to solve its technical problem is: to provide a method suitable for monitoring and counting bighead carp populations in large water bodies, comprising the following steps: Step S1: Dynamically deploy acoustic signal receiving stations according to the type of water area; Step S2: Fix the acoustic beacon to the dorsal fin of the bighead carp and after marking it for a period of time, verify the working status of the acoustic beacon through the signal of the acoustic signal receiving station; Step S3: Based on the three-point joint positioning method of TDOA, the precise coordinates of the marked bighead carp are calculated by combining the weighted least squares method, and the behavior of the bighead carp is judged. Based on the mark-recapture principle, the total population of bighead carp in the monitoring water area is estimated, and the precise coordinates of the bighead carp, the behavior judgment results, and the total population of bighead carp in the monitoring water area are output. Step S4: Monitor water temperature, water quality, food, and water depth in specific waters where bighead carp are active; perform a comprehensive topographic scan of the large water surface and use ArcGIS software to perform a 3D simulation to map the terrain below the water surface; conduct drone aerial photography of the large water surface to map the terrain above the water surface. Step S5: Based on the precise coordinates and behavior judgment results of the bighead carp and the total population of bighead carp in the monitored water area, combined with water depth, water temperature, water quality and topography data, the activity range and pattern of the fish school are obtained and the scanning detection range of the fish finder is determined. The fish school data is detected by scanning the fish school data by the fish finder, and the fish school density and biomass are calculated based on the fish school data detected by the fish finder. Step S6: Verify the results using traditional sampling and multifactor statistical analysis.

[0005] Furthermore, the water types include open waters, rivers flowing into lakes / reservoirs, and special habitats.

[0006] Furthermore, verifying the working status of the acoustic beacon through the acoustic signal receiving station includes: verifying the working status of the beacon through the acoustic signal receiving station within 72 hours after tagging; removing individual data of fish whose swimming speed has continued to decrease by more than 30% or who have exhibited abnormal behavior after tagging; and ensuring that the number of tagged fish in each batch does not exceed 5% of the estimated total population.

[0007] Furthermore, step S3 includes the following steps: Step S31, Data Preprocessing and Verification: Remove data with no signal or abnormal jumps, filter out low signal strength data and mark them; Step S32: The acoustic signal receiving station receives the signal emitted by the acoustic beacon on the bighead carp and records the receiving timestamp, and performs time correction on the receiving timestamp; Step S33: Establish the TDOA distance difference equation and residual function based on the TDOA positioning model; Step S34: Calculate the coordinates of the bighead carp using weighted nonlinear least squares and output the candidate coordinates; Step S35: Filter and classify the candidate coordinates, and output the corresponding processing based on the filtering and classification results; Step S36: Based on the coordinate analysis of the silver carp after screening and classification, determine the behavior type of the silver carp, and estimate the total population of silver carp in the monitored water area based on the mark-recapture principle; Step S37: Close the loop and output data.

[0008] Furthermore, the candidate coordinate types include valid positioning points, abnormal jump points, and low-confidence points.

[0009] Furthermore, after processing the coordinates of the silver carp according to the screening and classification results, the output includes: The valid location points of the same tagged bighead carp are arranged in chronological order to form a trajectory sequence. Abnormal jump points and low confidence points are removed, downweighted, or marked.

[0010] Furthermore, the behavioral types of the bighead carp include stable aggregation behavior, active migration behavior, long-term residence or habitat loyalty behavior, nocturnal resting or low-activity state at night, and suspected foraging or local patrolling behavior.

[0011] Furthermore, step S4, which involves monitoring water temperature, water quality, food availability, and water depth in the specific waters where bighead carp are active, specifically includes: By deploying an in-situ online water quality monitoring system in the monitored water area, environmental indicators of the monitored water area are collected continuously in real time.

[0012] Furthermore, in step S4, a comprehensive topographic scan of the large water surface is performed, and a 3D simulation is conducted using ArcGIS software to map the subsurface topography, specifically including: Underwater topography was scanned using multibeam and side-scan sonar.

[0013] Furthermore, step S4 involves conducting drone aerial photography of the large water surface to map the terrain above the water, specifically including: Before dynamically deploying acoustic signal receiving stations according to the type of water area, satellite remote sensing technology is used for the main inspection, and drones equipped with high-precision optical cameras and multispectral sensors are used for supplementary inspection to obtain information on water bodies, habitats, pollution and temperature.

[0014] The beneficial effects of this invention are as follows: A fish finder based on sonar scanning technology collects data on underwater bighead carp, reducing damage to the fish. Acoustic signal receiving stations are dynamically deployed in a grid pattern according to the type of water body, ensuring the number and location of the stations adapts to different water areas. Based on the TDOA (True-Difference-Oriented Alignment) three-point joint positioning method, combined with weighted least squares, the precise coordinates of the marked bighead carp are calculated, and the behavior of the bighead carp is determined. By marking individual fish, the activity area of ​​individual fish after grouping can be understood. Combined with water depth, water temperature, and water quality data, the characteristics of the fish's habitat can be derived, thus revealing the range and patterns of fish activity. After dividing the monitoring area based on the fish's behavioral habits, population monitoring and counting are performed through fish sonar scanning, reducing the difficulties of monitoring large fish populations and the problem of repeated counting, thereby improving the accuracy of the monitoring and counting results. The solution of this invention is non-destructive, comprehensive, high-precision, and standardized, and is suitable for bighead carp resource surveys in large water bodies, evaluation of the effects of stock enhancement and release, and ecological fisheries management. Attached Figure Description

[0015] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0016] In the picture: Figure 1 A flowchart of a method for monitoring and counting bighead carp populations in large bodies of water provided by the present invention; Figure 2 This is a flowchart of the in-situ online water quality monitoring steps in an embodiment of the present invention; Figure 3 This is a diagram illustrating the underwater terrain scanning step in an embodiment of the present invention. Detailed Implementation

[0017] To make the technical problem to be solved, the technical solution, and the beneficial effects of this invention clearer, the invention will now be described in detail with reference to the accompanying drawings. This drawing is a simplified schematic diagram, illustrating only the basic aspects of the invention, and therefore only shows the components relevant to the invention. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0018] Please refer to Figure 1 This invention provides a method for monitoring and counting silver carp populations in large bodies of water, comprising the following steps: Step S1: Dynamically deploy acoustic signal receiving stations according to the type of water area.

[0019] Specifically, water types include open waters, rivers flowing into lakes / reservoirs, and special habitats.

[0020] Specifically, when the water area is open water: acoustic signal receiving stations should be deployed using a 1.5km × 1.5km grid unit, with the effective signal coverage overlap of adjacent acoustic signal receiving stations not less than 15%, ensuring that the monitoring blind zone area accounts for less than 5%. Adaptive densification should be implemented based on the seasonal behavioral characteristics of bighead carp: during the spawning season (water temperature 18–22℃), densification should be increased to a spacing of 500m in shallow, hard-bottomed areas; during the warm-meeting migration season, densification should be increased in thermocline distribution areas; and during the overwintering season, densification should be increased in deep-water channels with a depth ≥8m. The location of acoustic signal receiving stations must meet the following requirements: ≥200m from waterways and ≥300m from highways, to avoid noise interference.

[0021] When the water body is a river flowing into a lake / reservoir: a monitoring section is set up every 5 km, and within 500 m upstream and downstream of the confirmed spawning grounds, the density is increased to one section every 1 km. At each section, 3–5 acoustic signal receiving stations are deployed along the water flow direction, with a spacing of 50–150 m, forming a monitoring array to achieve cross-location of migrating individuals, with a positioning accuracy ≤50 m, and supporting time-series trajectory reconstruction and migration rate calculation.

[0022] When the water area is a special habitat: a habitat assessment is conducted by taking into account parameters such as substrate type, water depth gradient (shallow area ≤3m, transition area 3–8m, deep area ≥8m), aquatic vegetation coverage, dissolved oxygen (preferably areas with dissolved oxygen content DO ≥5mg / L) and water temperature stratification. An equilateral triangle monitoring array is set up in areas with high fish density, with the triangle side length being 100–300m and the three-point cross-verification positioning error ≤30m.

[0023] Step S2: Fix the acoustic beacon to the dorsal fin of the bighead carp and after marking it for a period of time, verify the working status of the acoustic beacon through the signal of the acoustic signal receiving station; Specifically, the acoustic beacon used in this embodiment has the following specifications: outer diameter ≤ 9mm, length ≤ 32mm, weight in water (after buoyancy approaches neutral) ≤ 2g, and the mass of the acoustic beacon does not exceed 2% of the weight of the marked individual; operating frequency 69kHz (turbid water, long-distance scenario) or 180kHz (clean water, high-precision scenario), signal transmission interval 45–120 seconds (including random jitter ± 15 seconds to prevent signal collision), and rated battery life ≥ 180 days.

[0024] The specific procedure for attaching an acoustic beacon to the dorsal fin of a silver carp is as follows: Before the procedure, anesthetize the fish by placing it in MS-222 anesthetic solution (concentration 50–100 mg / L) until the gill cover movement frequency significantly decreases and the fish rolls to its side. Remove the beacon and insert a sterile puncture needle (14–16G) 1–2 cm behind the base of the first dorsal fin, piercing the dorsal fin base membrane. Secure the beacon to both sides of the dorsal fin with surgical sutures, tie the knots, and apply antibacterial ointment. The entire procedure should be completed within 90 seconds. After the procedure, place the fish in oxygenated water for natural resuscitation until it returns to an upright swimming posture (usually 3–8 minutes) before releasing it.

[0025] The specific steps for verifying the operational status of the acoustic beacon using signals from the acoustic signal receiving station are as follows: Within 72 hours of tagging, the beacon's operational status is verified via an acoustic signal receiving station; data on individuals whose swimming speed decreases by more than 30% or exhibits abnormal behavior after tagging are excluded; the number of tagged fish in each batch does not exceed 5% of the estimated total population to ensure the representativeness and statistical validity of the monitoring data.

[0026] Step S3: Based on the TDOA three-point joint positioning method, combined with the weighted least squares method, calculate the precise coordinates of the marked bighead carp, determine the behavior of the bighead carp, and estimate the total population of bighead carp in the monitored water area based on the mark-recapture principle. Output the precise coordinates of the bighead carp, the behavior determination results, and the total population of bighead carp in the monitored water area. The precise coordinate positioning results of the bighead carp are updated every 60 seconds to form a continuous trajectory.

[0027] Specifically, the following steps are included: Step S31, Data Preprocessing and Verification: Remove data with no signal or abnormal jumps, filter out low signal strength data and mark them; Data preprocessing and verification: Data with no signal or abrupt changes is removed, and low signal strength data is filtered out and marked. The specific process is as follows: Individuals with no valid signals for 7 consecutive days were excluded (considered beacon detachment or fish death); abnormal jumps in data where the displacement speed exceeded the maximum swimming speed of bighead carp (generally 3BL / s) within a single detection interval were excluded; records with signal strength below the acoustic signal receiving station's calibration detection threshold (SNR ≤ 6dB) were marked as low-confidence data and not included in behavioral analysis. The clocks of each acoustic signal receiving station were synchronized via GPS, with a clock error ≤ 1ms, to meet the positioning accuracy requirements of the TDOA (Time Difference of Arrival) algorithm.

[0028] Step S32: The acoustic signal receiving station receives the signal emitted by the acoustic beacon on the bighead carp and records the receiving timestamp, and performs time correction on the receiving timestamp; Let the spatial coordinates of the i-th acoustic signal receiving station be: ;

[0029] The coordinates of the marked fish to be determined are: ;

[0030] The equivalent sound velocity in water is denoted as c. The value of the equivalent sound velocity c in water is generally adopted from the field calibration value, or it is corrected according to the water temperature, salinity, pressure, water depth and sound velocity profile.

[0031] It should be noted that after the acoustic beacon on the bighead carp emits a signal, each acoustic signal receiving station records the reception timestamp, signal strength, signal-to-noise ratio, detection confidence level, and the status of the acoustic signal receiving station. The system uses the tag encoding, transmission sequence, reception time window, signal characteristics, and spatial relationships between the acoustic signal receiving stations to determine whether multiple reception records belong to the same beacon transmission event. Only reception records from the same transmission event are used for location tracking.

[0032] Let the original arrival time of the i-th acoustic signal receiving station be t. i Before calculating the arrival time difference, the timestamps are synchronized, drifted, and corrected for system delays, resulting in: .

[0033] Where, δ i The comprehensive time correction for the i-th acoustic signal receiving station includes clock skew, clock drift, receiver link delay, signal detection delay, and post-processing synchronization error.

[0034] Step S33: Establish the TDOA distance difference equation and residual function based on the TDOA positioning model; Using the first acoustic signal receiving station as the reference station, calculate the time difference of arrival: ;

[0035] The corresponding propagation distance difference is: ;

[0036] The distance function from the marked fish position P to the i-th acoustic signal receiving station is: .

[0037] Therefore, the TDOA distance difference equation is: ;

[0038] The residual function is: ;

[0039] To solve for the complete 3D coordinates For a single signal to be received, at least four spatially distributed acoustic signal receiving stations are required, i.e., n≥4. If the depth z is given by a depth label, pressure sensor, water depth model, or other conditions, the horizontal coordinate can be solved. At this point, at least three effective acoustic signal receiving stations are required, i.e., n≥3. It is preferable to use more than the minimum number of acoustic signal receiving stations to form redundant observations.

[0040] Step S34: Calculate the coordinates of the bighead carp using weighted nonlinear least squares and output candidate coordinates.

[0041] Due to the distance function Regarding the nonlinearity of coordinate P, the weighted nonlinear least squares method is used for solution: ;

[0042] Among them, Ω represents the spatial constraint area where the tagged fish may appear, such as the boundary of the monitored water area, the water depth range, the boundary of the river or reservoir area; Let be the weight of the i-th TDOA observation.

[0043] Weight It can be taken as the reciprocal of the variance of the observation error, and the specific formula is as follows: ;

[0044] in, This is the estimated variance of the TDOA observation error. To prevent positive numbers with a denominator of zero.

[0045] The weighting can be determined by a combination of factors including received signal strength, signal-to-noise ratio, detection confidence level, time synchronization error, sound velocity error, multipath interference, acoustic signal receiving station coordinate error, and acoustic signal receiving station geometry. In practice, observation targets with low signal quality, large time errors, or strong multipath interference are generally assigned lower weights.

[0046] During the solution process, the geometric center of the acoustic signal receiving station array, the effective positioning point of the previous moment, or the coarse positioning point of the grid search can be used as initial values. The coordinates are then updated using the Gauss-Newton method, the Levenberg-Marquardt method, or other iterative optimization methods. When the coordinate change between two adjacent iterations is less than a preset threshold, the change in the objective function is less than a preset threshold, or the number of iterations reaches the upper limit, candidate coordinates p* are output.

[0047] Step S35: Filter and classify the candidate coordinates, and output the corresponding processing based on the filtering and classification results; The candidate coordinate types include valid positioning points, abnormal jump points, and low confidence points.

[0048] Whether a candidate coordinate is a valid location point is determined based on a quality threshold. Candidate coordinates that meet the quality threshold are considered valid location points; candidate coordinates that do not meet the quality threshold are eliminated, downweighted, or marked as low-confidence points.

[0049] The quality threshold is determined comprehensively based on the weighted sum of squared residuals, root mean square residuals, number of effective acoustic signal receiving stations, geometric configuration of acoustic signal receiving stations, location reliability, received signal quality, time synchronization status, and whether the location point is located within the preset water boundary and water depth range.

[0050] A trajectory sequence is formed by arranging the effective location points of the same tagged bighead carp in chronological order: Among them, P k Let t be the Kth positioning coordinate. k For the corresponding time, q k To determine quality indicators or confidence levels. The determination results can be output according to a preset cycle. Alternatively, the time window can be updated; in this embodiment, Seconds. If multiple valid positioning points exist within the same window, the point with the highest quality, the weighted average point, or the filtered and smoothed point can be selected as the representative coordinates.

[0051] To avoid abnormal positioning affecting behavior determination, instantaneous velocity is calculated for adjacent positioning points: ;

[0052] Let the length of the marked fish be BL, and the maximum reasonable swimming speed threshold be: ;

[0053] Where 'a' is an empirical coefficient related to fish species, body length, water temperature, current velocity, and environmental conditions. For large freshwater fish, 'a' can be taken as 3, i.e. When V k >V maxIf the location point crosses an inaccessible area, or if the location residual is too large, the confidence level is too low, or the trajectory acceleration is abnormal, the corresponding location point will be identified as an abnormal jump point or a low confidence point, and will be removed, downweighted, or marked.

[0054] Step S36: Based on the coordinate analysis of the silver carp after screening and classification, determine the behavior type of the silver carp, and estimate the total population of silver carp in the monitored waters based on the mark-recapture principle.

[0055] Specifically, in this embodiment, the behavior types of bighead carp include stable aggregation behavior, active migration behavior, long-term residence or habitat loyalty behavior, nocturnal resting or low-activity state at night, and suspected foraging or local patrolling behavior.

[0056] The specific process for determining the behavior is as follows: When the behavior of bighead carp is stable aggregation, calculate the spatial distance within the same time window for any two marked fish a and b. The specific formula is as follows: ;

[0057] Let the cluster distance threshold be... Number of close co-occurrences The specific formula is as follows: ; in For indicator functions, when the condition is met One instance of close co-occurrence is counted within a preset time window. Inside (Threshold for the number of times close-range co-occurrence occurs), or no less than a preset proportion. When labeled individuals consistently appear in the same area, and their actual co-occurrence intensity is significantly higher than that of the random background model, it is determined to be a stable cluster, stable co-occurrence, or aggregation behavior. In this embodiment, , sky, , The significance level is If ecological evidence such as spawning season, water temperature, current velocity, water depth, historical spawning site location, or fish egg sampling is simultaneously met, it can be further determined as a suspected spawning aggregation.

[0058] When the behavior of bighead carp is active migration, calculate the displacement distance between adjacent positioning points for the continuous trajectory of the same marked fish. and average speed The specific formula is as follows: ;

[0059] In the time window Within, calculate the net displacement distance and the cumulative path distance. The specific formula is as follows: , .

[0060] Directional persistence indicators for: ; When the net displacement distance D net Greater than the migration distance threshold, average speed If the speed exceeds a threshold, the directional persistence index exceeds a preset threshold, and the trajectory is not caused by abnormal jump points, it is determined to be active migration or directional migration. In this embodiment, the migration distance threshold is 2km, and the speed threshold is 0.5BL / s. If the net displacement distance D within 24 hours... net If the distance is greater than 2km, the average speed Vk is greater than 0.5BL / s, and the trajectory direction is continuous, then it is determined to be active migration.

[0061] When the behavior of bighead carp is resident or habitat-loyal, the target area is set as The cumulative effective occurrence time of the marked fish in the area was statistically analyzed. : Total effective monitoring time for: Regional residency ratio for: ;when If the tagged fish's residence rate exceeds a preset threshold, and it repeatedly returns to or remains in the same area across multiple independent time periods, it is considered a long-term resident behavior or habitat loyalty behavior. In this embodiment, the residence rate threshold is 60%.

[0062] When the bighead carp exhibits nocturnal resting or low-activity behavior, the sunset and sunrise times are determined based on the geographical location and date of the monitored waters. The period from sunset to sunrise the following day is defined as the night window. Within this night window, if the tagged fish is continuously captured by the acoustic signal receiving station or receives valid positioning results continuously, the change in positioning coordinates does not exceed the nighttime activity radius threshold, the average nighttime speed is lower than the low-activity speed threshold, and the water depth is not less than the water depth threshold, then it is determined to be in a nocturnal resting or low-activity state. In this embodiment, the nighttime activity radius threshold is 100m, and the water depth threshold is 5m.

[0063] When the behavior of bighead carp is suspected to be foraging or localized patrolling, the analysis time window is set to [value missing]. Its length can be set from 30 min to 6 h; the candidate region can be selected by radius. Delineation of spatial grids, receiving station coverage areas, or ecological function patches in m. (In the window) Within, calculate the cumulative distance of the path. Net displacement distance Path tortuosity and average speed : ; ; ; ;

[0064] in, To prevent positive numbers with a denominator of zero, The larger the value, the more tortuous the trajectory and the more obvious the local search features.

[0065] When the same tagged fish meets the following conditions simultaneously within the same candidate area, it can be determined as suspected foraging or localized patrolling behavior: (1) Cumulative distance along the path It is 100–3000 m, or 1.5–3.0 times greater than the median path length of the same type of time window for that individual; (2) Net displacement percentage Not greater than 0.10 to 0.40, or net displacement distance Not greater than 50–500 m; (3) Path tortuosity Not less than 2.5 to 10; (4) Average speed The range is 0.05–0.50 BL / s; (5) The number of times a person enters, leaves and returns to the same area within 24 hours is not less than 2 to 5 times, or the cumulative stay time in the area is not less than 30 to 180 minutes.

[0066] The aforementioned thresholds can be selected within a given range based on the target fish species, body length, water temperature, current velocity, receiving station deployment density, positioning error, and water area scale. If food resource distribution, water quality parameters, underwater images, feeding observations, or other ecological evidence are simultaneously obtained, the confidence level of the behavior assessment can be improved; if only the positioning trajectory is used, the output will be "suspected foraging" or "local patrolling" behavior.

[0067] Step S37: Close the loop and output data.

[0068] The aforementioned distance threshold, time window, velocity threshold, significance level, body length normalization coefficient, water depth threshold, positioning quality threshold, and output cycle are all configurable parameters that can be set or updated based on the target fish species, body length range, water type, acoustic signal receiving station deployment density, season, water temperature, flow velocity, on-site acoustic conditions, and calibration experiment results.

[0069] Population counting estimation steps: In this embodiment, the total population size N is estimated based on the label-recapture principle (Lincoln-Petersen method). The minimum sample size requirements are: the number of tagged individuals released M ≥ 5% of the estimated population, the number of recaptured individuals C ≥ 50, and the target error ±20% (95% confidence interval). The formula is: N = (M×C) / R, where R is the number of tagged individuals in the recapture. This method is suitable for waters with relatively closed populations during the monitoring period; for open, large water bodies, a correction estimate needs to be made using the Jolly-Seber open population model.

[0070] Step S4: Monitor water temperature, water quality, food, and water depth in specific waters where bighead carp are active; perform a comprehensive topographic scan of the large water surface and use ArcGIS software to perform a 3D simulation to draw the terrain below the water surface; and take aerial photos of the large water surface using drones to draw the terrain above the water surface.

[0071] The specific steps for monitoring water temperature, water quality, food, and water depth in specific waters where bighead carp are active are as follows: By deploying an in-situ online water quality monitoring system in the monitored water area, the water temperature, pH value, dissolved oxygen, conductivity, turbidity, chlorophyll a, and blue-green algae in the monitored water area are collected continuously in real time.

[0072] Specifically, the in-situ online water quality monitoring system is either a buoy station or a fixed multi-parameter monitoring station, as is currently known in technology. The workflow of the in-situ online water quality monitoring system is as follows: Figure 2 As shown, real-time monitoring of water-related indicators is conducted using photovoltaic power to collect water quality data; simultaneously, meteorological element monitoring is carried out to collect meteorological environmental data. All monitoring data collected by the water quality monitoring and meteorological monitoring modules are then uniformly aggregated to the information transmission module, and subsequently transmitted to the data storage module to complete data preservation and archiving, enabling the retention and subsequent retrieval of monitoring information.

[0073] By deploying buoy stations or fixed multi-parameter monitoring stations at key points in the waterway, and utilizing sensor technology, real-time and continuous data collection of indicators such as water temperature, pH, dissolved oxygen, conductivity, turbidity, chlorophyll a, and cyanobacteria can be achieved. Compared to traditional manual sampling, this method overcomes the shortcomings of large-scale water surface monitoring, such as numerous blind spots and poor timeliness, and accurately captures the fluctuation patterns of water quality parameters under diurnal and seasonal variations, providing precise data support for subsequent aquatic environment monitoring and fish surveys.

[0074] The specific steps for conducting a comprehensive topographic scan of a large water surface are as follows: Underwater topography is scanned using multibeam and side-scan sonar, as shown in the diagram below. Figure 3 As shown.

[0075] Addressing the challenges of complex underwater topography in large bodies of water and the low efficiency and limited comprehensive coverage of traditional single-beam measurements, multibeam bathymetry systems are employed for underwater topographic mapping. This multibeam bathymetry system is a key technology for achieving high-precision, full-coverage topographic awareness. By projecting fan-shaped acoustic beams into the seabed, it can acquire strip-shaped water depth data perpendicular to the course of navigation in a single pass, forming a seamless, high-density point cloud. This allows for the precise identification of underwater micro-topographic features such as steep banks, channels, reefs, and submerged ancient river channels, enabling the construction of high-resolution digital water depth models and underwater 3D topographic maps. Furthermore, it accurately delineates shallow, hard-bottomed slopes required for fish spawning habitats, suitable gentle-slope transition zones for feeding grounds, and deep-water channels required for overwintering sites. This provides crucial underwater dimensional supplementation to the water area plan and offers a solid and reliable data foundation for optimizing the deployment of in-situ water quality monitoring systems and identifying fish habitats.

[0076] The specific steps for conducting drone aerial photography of large water surfaces and mapping the topography above the water surface are as follows: satellite remote sensing technology is used for the main inspection, and drones equipped with high-precision optical cameras and multispectral sensors are used for supplementary inspection to obtain information on water bodies, habitats, pollution, and temperature.

[0077] Satellite remote sensing technology, as the primary inspection method, utilizes high-resolution optical imagery and radar data, along with high-resolution multispectral imagery, to rapidly generate large-scale basic planar maps of water bodies through water body extraction and geometric correction, providing accurate base maps for subsequent work. Simultaneously, based on the spectral characteristics of the remote sensing images, indicators such as chlorophyll a, suspended solids concentration, and transparency of the water body are retrieved to achieve spatial distribution mapping of water quality parameters, forming a macroscopic distribution of aquatic vegetation and shoreline wetland plants, as well as a planar map of the entire water body. In this embodiment, satellite remote sensing technology mainly employs NDWI and NDVI dual-track analysis of water body area with an accuracy of ±3%; the MNF algorithm identifies sewage outlets with a positioning accuracy of 50m; and random forest is used to assess habitat encroachment, with a Kappa coefficient >0.85.

[0078] Drone patrols serve as a supplementary method for targeted verification and intensive observation. High-precision optical cameras and multispectral sensors mounted on drones acquire centimeter-level thermal infrared and hyperspectral data. For key shorelines, water areas, and vegetation transition zones, drones equipped with high-precision optical cameras and multispectral sensors obtain centimeter-resolution images. Aerial data allows for detailed identification of vegetation community structure and water quality, correcting boundary details in remote sensing interpretation, and verifying high-precision water depth and water quality parameters in local water areas. The two methods work together to form a complete technical closed loop of "macro-level control - micro-level verification," providing detailed and reliable data support for water area mapping and water environment surveys. Both satellite remote sensing and drone-based supplementary patrols are existing technologies.

[0079] Step S5: Based on the precise coordinates and behavior judgment results of the bighead carp and the total population of bighead carp in the monitored water area, combined with water depth, water temperature, water quality and topography data, the activity range and pattern of the fish school are obtained and the scanning detection range of the fish finder is determined. The fish school data is detected by scanning the fish school data by the fish finder, and the fish school density and biomass are calculated based on the fish school data detected by the fish finder. Specifically, based on water depth, water quality, water temperature, and topographic data, and according to the precise coordinates and behavioral analysis of bighead carp, as well as the total population of bighead carp in the monitored waters, the diurnal activity patterns and feeding patterns of the fish can be determined, and the main habitat areas of the fish, such as spawning grounds, feeding grounds, and water sources, can be identified.

[0080] The fish finder used in this embodiment is based on sonar scanning technology. It conducts mobile detection according to a preset cross section. By emitting modulated ultrasonic pulses and receiving the echo signals reflected by the fish, it analyzes the time difference to determine the distance of the fish and continuously records the echo images of fish in the entire water layer. Compared with traditional nets, it does not require catching fish and has a large coverage area, while not causing damage to the fish.

[0081] In this embodiment, the fish finder is a BioSonics DT-X multi-functional echo sounder, used in conjunction with a Garmin S17xHVS receiving antenna and sensor kit to synchronously collect and store GPS data. Underwater acoustic data is acquired using BioSonics Acquisition 6.0 software installed on a portable computer. During acquisition, the transducer pulse frequency is 8pps, the pulse width is 0.5ms, and the data collection threshold is -130dB. The fish finder operates on the principle of sonar scanning technology, conducting mobile detection along a preset cross-section. It emits modulated ultrasonic pulses and receives the echo signals reflected by fish schools, analyzing the time difference to determine the distance to the fish school. It continuously records the echo images of fish throughout the water column. By combining the echo integration method and the target intensity-body length conversion model, the density, size distribution, and total biomass of fish per unit water volume can be accurately estimated.

[0082] During the surveying process, factors such as weather and water depth were considered. The fish finder primarily used a zigzag route around the center of the reservoir. Due to the large area of ​​the reservoir, a single-measurement method was employed. The measurement route was adjusted based on theoretical coverage values ​​and actual conditions. The theoretical coverage value was calculated using the Aglen coverage formula, as follows: , In the formula: L is the distance traveled during the underwater acoustic survey (m); A is the surface area of ​​the reservoir (m²). 2 ); D C For underwater acoustic survey coverage, the theoretical coverage value is usually required to be above 6.

[0083] Fish body length is calculated using an empirical formula relating TS value and body length: ;

[0084] In the formula: TL is the body length of the target fish (cm); TS is the target intensity of the fish (dB). The obtained target signal intensity TS, water depth, depth from the bottom, and other parameters were used for mathematical statistics, data statistical analysis, and graphical plotting using SPSS 19.0 software.

[0085] The density formula is Where V is the actual volume of water being scanned.

[0086] Biomass formula is In the formula This represents the statistical average of TS; Let W be the statistical average of W, where W is the weight of a single target fish. The formula for calculating W is W=aTL b Where 'a' is the fish body size coefficient (conditional factor constant), which is generally determined by the fish species and growth environment. 'b' is the power exponent of body length and weight (condition index), with b≈2.5~3.5 for most fish; if the fish grows at a uniform rate and proportionally, then b=3.

[0087] Compared to traditional methods, fish finders based on sonar scanning technology offer advantages such as all-weather operation, no harm to fish, and wide coverage, making them particularly suitable for rapid surveys in deep, turbid waters. The spatial distribution and biomass estimation results of fish resources obtained by fish finders can be correlated with water depth parameters obtained from multibeam and side-scan sonar topographic scanning and environmental factors obtained from in-situ online water quality monitoring systems. This provides crucial data support for accurately identifying fish population habitats, assessing aquatic ecological carrying capacity, and developing scientific fisheries management strategies.

[0088] Step S6: Verify the results using traditional sampling and multifactor statistical analysis.

[0089] Integrated monitoring stations were deployed in key areas such as spawning grounds, feeding grounds, and water sources to collect data continuously and synchronously. Pearson / Spearman correlation analysis and canonical correspondence analysis / redundancy analysis were used to identify the dominant environmental factors affecting the distribution and behavior of bighead carp and to verify the reliability of the monitoring results.

[0090] Specifically, the large water surface area mentioned in this embodiment refers to medium to large open water areas ranging from 500 mu to 5000 mu. It should be noted that the population monitoring and counting method described in this invention is applicable to large water surface areas of less than 5000 mu. If monitoring is to be carried out on water areas of more than 5000 mu, it is necessary to conduct separate detection.

[0091] This invention utilizes a fish finder based on sonar scanning technology to collect data on underwater bighead carp, reducing harm to the fish. It employs a combination of satellite remote sensing and drone patrols to assess habitat and pollution on the water surface; multibeam and side-scan sonar to scan underwater topography; and an in-situ online water quality monitoring system to monitor the aquatic environment. These three elements work together to monitor the bighead carp's habitat and, compared to manual patrols and conventional acoustic monitoring, improve the monitoring coverage of the entire water area. This reduces the impact of large water surfaces with numerous blind spots, complex terrain, weather interference, and issues affecting data accuracy and timeliness. Based on the water type, acoustic signal receiving stations are dynamically deployed in a grid pattern, adapting the number of stations to different water areas. Using a three-point or higher joint positioning method based on TDOA, combined with weighted least squares, the precise coordinates of marked bighead carp are calculated, and their behavior is assessed. By marking individual fish, the activity area of ​​fish after grouping is determined. Combined with water depth, temperature, and quality data, the characteristics of the fish's habitat are derived, thus revealing the range and patterns of fish activity. After defining the monitoring area based on fish behavior, fish sonar scanning is used for population monitoring and counting, reducing the difficulties of monitoring large fish populations and avoiding duplicate counting, thereby improving the accuracy of monitoring and counting results. This invention features non-destructive, full-coverage, high-precision, and standardized characteristics, making it suitable for bighead carp resource surveys in large water areas, evaluation of stock enhancement and release effects, and ecological fisheries management.

Claims

1. A method suitable for monitoring and counting silver carp populations in large bodies of water, characterized in that, include: Step S1: Dynamically deploy acoustic signal receiving stations according to the type of water area; Step S2: Fix the acoustic beacon to the dorsal fin of the bighead carp and after marking it for a period of time, verify the working status of the acoustic beacon through the signal of the acoustic signal receiving station; Step S3: Based on the three-point joint positioning method of TDOA, the precise coordinates of the marked bighead carp are calculated by combining the weighted least squares method, and the behavior of the bighead carp is judged. Based on the mark-recapture principle, the total population of bighead carp in the monitoring water area is estimated, and the precise coordinates of the bighead carp, the behavior judgment results, and the total population of bighead carp in the monitoring water area are output. Step S4: Monitor water temperature, water quality, food, and water depth in specific waters where bighead carp are active; perform a comprehensive topographic scan of the large water surface and use ArcGIS software to perform a 3D simulation to map the terrain below the water surface; conduct drone aerial photography of the large water surface to map the terrain above the water surface. Step S5: Based on the precise coordinates and behavior judgment results of the bighead carp and the total population of bighead carp in the monitored water area, combined with water depth, water temperature, water quality and topography data, the activity range and pattern of the fish school are obtained and the scanning detection range of the fish finder is determined. The fish school data is detected by scanning the fish school data by the fish finder, and the fish school density and biomass are calculated based on the fish school data detected by the fish finder. Step S6: Verify the results using traditional sampling and multifactor statistical analysis.

2. The method for monitoring and counting silver carp populations in large bodies of water according to claim 1, characterized in that: The water types include open waters, rivers flowing into lakes / reservoirs, and special habitats.

3. The method for monitoring and counting silver carp populations in large bodies of water according to claim 1, characterized in that: Verifying the operational status of an acoustic beacon using signals from an acoustic signal receiving station includes: verifying the beacon's operational status via signals from an acoustic signal receiving station within 72 hours of tagged fish; removing individual data that show a continuous decrease in swimming speed exceeding 30% or exhibit abnormal behavior after tagged fish; and ensuring that the number of tagged fish in each batch does not exceed 5% of the estimated total population.

4. The method for monitoring and counting silver carp populations in large bodies of water according to claim 1, characterized in that: Step S3 includes the following steps: Step S31, Data Preprocessing and Verification: Remove data with no signal or abnormal jumps, filter out low signal strength data and mark them; Step S32: The acoustic signal receiving station receives the signal emitted by the acoustic beacon on the bighead carp and records the receiving timestamp, and performs time correction on the receiving timestamp; Step S33: Establish the TDOA distance difference equation and residual function based on the TDOA positioning model; Step S34: Calculate the coordinates of the bighead carp using weighted nonlinear least squares and output the candidate coordinates; Step S35: Filter and classify the candidate coordinates, and output the corresponding processing based on the filtering and classification results; Step S36: Based on the coordinate analysis of the silver carp after screening and classification, determine the behavior type of the silver carp, and estimate the total population of silver carp in the monitored water area based on the mark-recapture principle; Step S37: Close the loop and output data.

5. The method for monitoring and counting silver carp populations in large bodies of water according to claim 4, characterized in that: The candidate coordinate types include valid positioning points, abnormal jump points, and low confidence points.

6. The method for monitoring and counting silver carp populations in large bodies of water according to claim 5, characterized in that: After processing the coordinates of the bighead carp according to the screening and classification results, the output includes: The valid location points of the same tagged bighead carp are arranged in chronological order to form a trajectory sequence. Abnormal jump points and low confidence points are removed, downweighted, or marked.

7. The method for monitoring and counting silver carp populations in large bodies of water according to claim 1, characterized in that: The behavioral types of the bighead carp include stable aggregation behavior, active migration behavior, long-term residence or habitat loyalty behavior, nocturnal resting or low-activity state at night, and suspected foraging or local patrolling behavior.

8. The method for monitoring and counting silver carp populations in large bodies of water according to claim 1, characterized in that: Step S4 involves monitoring water temperature, water quality, food availability, and water depth in specific waters where bighead carp are active. This includes: By deploying an in-situ online water quality monitoring system in the monitored water area, environmental indicators of the monitored water area are collected continuously in real time.

9. The method for monitoring and counting silver carp populations in large bodies of water according to claim 1, characterized in that: Step S4 involves performing a comprehensive topographic scan of the large water surface and then using ArcGIS software to create a 3D simulation, specifically mapping the subsurface topography, including: Underwater topography was scanned using multibeam and side-scan sonar.

10. The method for monitoring and counting silver carp populations in large bodies of water according to claim 1, characterized in that: Step S4 involves conducting drone aerial photography of the large water surface to map the terrain above the water, specifically including: Before dynamically deploying acoustic signal receiving stations according to the type of water area, satellite remote sensing technology is used for the main inspection, and drones equipped with high-precision optical cameras and multispectral sensors are used for supplementary inspection to obtain information on water bodies, habitats, pollution and temperature.

Images