A method for quantitatively evaluating the release range of rare fish based on eDNA macro barcoding

Abstract

The present application relates to a kind of rare fish release range quantitative evaluation method based on eDNA macro bar code, belong to environmental DNA monitoring technical field.By determining initial release site, eDNA sample is collected and processed and macro bar code sequencing obtains clean sequence dataset, then by species identification, information database is constructed, spatial interpolation generates distribution density field, the diffusion range and the aggregation degree of fish release are quantitatively calculated to obtain evaluation result;Then according to the evaluation result, the release target adjustment factor is calculated, whether the adjustment factor reaches the expected threshold value is judged whether new release site needs to be determined again;Finally, according to the specific situation of not reaching threshold, new release site is screened and released again until the adjustment factor is up to standard, the target release site is determined, the quantitative evaluation of rare fish release range and the dynamic optimization of release strategy are completed.The present application can realize the accurate quantitative evaluation of rare fish release range, and can dynamically adjust release strategy, improve the scientificity and effectiveness of rare fish release.

Description

Technical Field

[0001] This invention relates to a method for quantitatively assessing the release range of rare fish species based on eDNA macrobarcoding, belonging to the field of environmental DNA monitoring technology. Background Technology

[0002] Currently, traditional methods for assessing the effectiveness of releasing rare fish species mainly rely on physical fishing, acoustic tagging and tracking, or underwater video observation. These methods have significant limitations: physical fishing is disruptive to rare fish species and inefficient, making it difficult to achieve large-scale monitoring; acoustic tagging and tracking are costly and can only tag a small number of individuals, failing to reflect the overall distribution of the released population; underwater video observation is limited by water visibility and equipment deployment range, resulting in poor spatiotemporal continuity of the acquired data.

[0003] Existing environmental DNA (eDNA) technologies have been applied to biodiversity surveys and species detection, particularly in the discovery and confirmation of rare species. However, conventional eDNA technology applications are typically limited to reporting the presence or absence of a target species at specific sampling points, providing at most relative abundance information based on sequence count. This discrete point data cannot intuitively and accurately depict the continuous distribution of fish throughout the entire aquatic space, and it is even more difficult to quantify the core areas and boundaries of their distribution. Managers cannot directly determine the actual dispersal range, aggregation areas, and spatial variations in distribution density of fish populations from a series of isolated "detection points."

[0004] Current technologies for assessing the effectiveness of fish release programs often qualitatively describe dispersal trends by comparing the locations of species appearance at different times. This method cannot accurately depict the dynamic changes in fish distribution areas. Due to the lack of quantitative analysis of the spatiotemporal dynamics of distribution fields, it is difficult to objectively and accurately assess the settlement success rate, dispersal ability, and habitat preferences of released fish, thus limiting the optimization of release strategies and the accurate evaluation of ecological benefits. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides a method for quantitatively assessing the release range of rare fish species based on eDNA macro barcodes. This invention can achieve accurate quantitative assessment of the release range of rare fish species and can dynamically adjust the release strategy, thereby improving the scientific rigor and effectiveness of rare fish release.

[0006] To achieve the above objectives, the technical solution provided by this invention is: a method for quantitatively assessing the release range of rare fish species based on eDNA macrobarcoding, comprising the following steps:

[0007] (1) Determine the initial release location of rare fish: Collect hydrological characteristics of the release area, ecological habits of rare fish and historical release data of the area, and then determine the initial release location of rare fish based on the hydrological characteristics of the release area, ecological habits of rare fish and historical release data of the area; the hydrological characteristics include water flow velocity and dissolved oxygen content; the ecological habits of rare fish include diurnal vertical movement range and preferred water temperature; the historical release data of the area includes at least the release location and the number of rare fish released; the release area is a lake, reservoir and slow-flowing river section closed or semi-closed water body with an area range of 50–1000 km² and a water depth range of 2–30 m;

[0008] (2) Data acquisition and processing: Environmental DNA samples were collected from the release water area and subjected to macrobarcode sequencing to obtain raw sequence data. Quality control operations were performed on the raw sequence data to remove low-quality reads and contaminated sequences, generating a clean sequence dataset. The clean sequence dataset includes a set of high-quality reads, a sequence quality score vector, and sampling point identifiers.

[0009] (3) Species identification and database construction: The clean sequence dataset is compared with the rare fish reference macro barcode database to identify matching species sequences, record the detection frequency and sampling point spatial coordinates of each species, and construct a species detection information database; the species detection information database includes species classification code, detection frequency matrix and geographic location label; the rare fish reference macro barcode region is the hypervariable region of mitochondrial COI gene or 12SrRNA gene;

[0010] (4) Generation of distribution density field: Based on the detection frequency and spatial coordinates in the species detection information database, the distribution density of species in the release water area is estimated by spatial interpolation method. Combined with the release time series data, the species distribution density field is generated. The species distribution density field includes spatial grid density values, time slice data and confidence interval index.

[0011] (5) Calculation of release range: Analyze the changing trend of species distribution density field over time, extract distribution boundary and density gradient information, calculate the diffusion range and aggregation degree of released fish, and generate quantitative assessment results of release range; the quantitative assessment results of release range include spatial coverage index, distribution uniformity coefficient and diffusion rate parameter.

[0012] (6) Dynamically adjust the release strategy: compare the diffusion range in the quantitative assessment results of the release range with the preset target range, and calculate the release target adjustment factor;

[0013] If the release target adjustment factor is lower than the expected threshold, the release location is re-determined based on the species detection information database in step (3), the species distribution density field in step (4), and the quantitative assessment results of the release range in step (5), combined with the diffusion pattern of rare fish and the spatial characteristics of the release water area; after the release operation is carried out at the re-determined release location, steps (2)-(5) are repeated to update the species distribution density field and the quantitative assessment results of the release range until the release target adjustment factor reaches the expected threshold; if the release target adjustment factor reaches or exceeds the expected threshold, the location of the current release location is recorded and the current release location is determined as the target release location;

[0014] The specific steps for re-determining the release location are as follows:

[0015] 1) Based on the species detection information database in step (3), extract the detection frequency and geographical location association information of each sampling point of rare fish, and identify the blank distribution area where the fish detection frequency is zero and the core aggregation area with a high detection frequency.

[0016] 2) Retrieve the species distribution density field from step (4), analyze the spatial gradient change characteristics of density values, and clarify the direction of fish diffusion, the short-spot area and the low-density extension area.

[0017] 3) Extract the spatial coverage and diffusion rate parameters from the quantitative assessment results of the release range in step (5) to determine the spatial orientation of fish diffusion restriction and the area with weak diffusion capacity.

[0018] 4) Combining the dispersal patterns of rare fish species, matching suitable dispersal paths and spatial habitat requirements for fish, and considering the spatial characteristics of the release area, identifying hydrologically connected areas and habitat-suitable areas in the water that are suitable for fish survival and conducive to their extension to blank distribution areas and areas with dispersal shortcomings, and combining the results of the coupling analysis of the above multi-dimensional information, selecting areas that can fill the gaps in fish distribution and promote the dispersal of fish to the target range as new release sites, while taking into account the hydrological connectivity with the original core aggregation area to ensure the habitat continuity between the new release site and the existing distribution area.

[0019] The further improvement to the above technical solution is as follows:

[0020] The rare fish mentioned in step (1) must cover at least one of the following categories:

[0021] 1) Fish species under first- or second-class protection;

[0022] 2) Threatened fish species;

[0023] 3) Critically endangered, endangered or vulnerable fish species.

[0024] The specific steps for generating the clean sequence dataset in step (2) are as follows:

[0025] 1) Collect water samples using an environmental DNA sampling device, extract total DNA and amplify fish-specific macrobarcode regions, prepare sequencing libraries, and obtain raw sequence data;

[0026] 2) Use a noise reduction program to process the raw sequence data, correct sequencing errors and remove chimeras, and output the error-corrected sequence set;

[0027] 3) The error-corrected sequence set is length-normalized and abundance-normalized to generate a clean sequence dataset containing a high-quality read length set, a sequence quality score vector, and sampling point identifiers.

[0028] The specific steps for constructing the species detection information database in step (3) are as follows:

[0029] 1) Perform local alignment between the sequences in the clean sequence dataset and the reference macro barcode database, calculate the sequence similarity, identify species sequences that meet the matching threshold, and generate preliminary species identification results; the matching threshold is defined as a sequence similarity ≥ 97% and an alignment coverage ≥ 90%;

[0030] 2) Count the number of times each species appears at different sampling points in the preliminary species identification results, and combine the spatial coordinates of the sampling points to form a species detection frequency distribution map;

[0031] 3) Integrate the species detection frequency distribution map with the timestamp data of release events, mark the detection time intervals and spatial location associations, and construct a species detection information database containing species classification codes, detection frequency matrices and geographic location labels.

[0032] The specific process of generating the species distribution density field in step (4) is as follows:

[0033] 1) Extract species detection frequency and spatial coordinates from the species detection information database to establish a spatial point pattern dataset;

[0034] 2) The kernel density estimation method is used to process the spatial point pattern dataset, calculate the species density value in each spatial grid cell, and generate an initial density distribution map;

[0035] 3) Introduce the time dimension, divide the initial density distribution map into multiple time slices according to the release time series, perform Gaussian smoothing on each time slice, and output the time series density field;

[0036] 4) Combine the behavioral parameters of the released fish to perform spatial correction on the time series density field, and generate a species distribution density field that includes spatial grid density values, time slice data and confidence interval indicators; among which, the behavioral parameters of the released fish include the diurnal vertical movement range or preference for specific water temperatures.

[0037] The specific steps for calculating the diffusion range and aggregation degree of the released fish in step (5) are as follows:

[0038] 1) Based on the density gradient changes in the species distribution density field, detect the location of abrupt changes in density gradient, determine the coordinates of boundary points based on the first and second differences of density values, connect the boundary points to form a distribution contour line, and calculate the area enclosed by the contour line as the diffusion range.

[0039] 2) Calculate the coefficient of variation and spatial autocorrelation index of the density values ​​in the species distribution density field. Then, weight and fuse the reciprocal of the coefficient of variation with the spatial autocorrelation index to obtain the distribution uniformity coefficient, which is the aggregation index.

[0040] The specific steps for generating the quantitative assessment results of the release range in step (5) are as follows:

[0041] 1) Extract the boundary coordinate sequence of the diffusion range, calculate the area of ​​the polygon enclosed by the sequence, and use it as a spatial coverage index;

[0042] 2) Calculate the coefficient of variation and spatial autocorrelation index of the density values ​​in the species distribution density field. Then, weight and fuse the reciprocal of the coefficient of variation with the spatial autocorrelation index to obtain the distribution uniformity coefficient, which is the index of the degree of aggregation.

[0043] 3) Obtain the timestamp of the release event and the current assessment time point, calculate the time interval, and combine it with the spatial coverage index to calculate the diffusion distance per unit time, which is used as the diffusion rate parameter;

[0044] 4) Combine the spatial coverage index, distribution uniformity coefficient, and diffusion rate parameter into a structured data record, add sampling point identifiers and evaluation timestamps, and generate a quantitative evaluation result of the release range containing a quantitative index table and a spatial distribution diagram.

[0045] The preset target range mentioned in step (6) is determined based on the ecological habit parameters of the released fish and the historical carrying capacity data of the water area.

[0046] The formula for calculating the release target adjustment factor mentioned in step (6) is as follows: ,in Represents the target's diffusion range. Represents the actual spread range. It is a dimensionless numerical value, ranging from 0 to 1; when When, it indicates that the release target adjustment factor is lower than the expected threshold; when This indicates that the release target adjustment factor has reached or exceeded the expected threshold; the target diffusion range is determined by statistically analyzing the diffusion range quantiles of historical successful release cases.

[0047] In the specific steps of calculating the diffusion range and aggregation degree of released fish, the coefficient of variation is the ratio of the standard deviation to the mean, and the calculation formula is: In the formula, Represents the coefficient of variation. The standard deviation of the density value The mean of the density values; coefficient of variation The larger the value, the greater the relative fluctuation of the density value in space, that is, the more uneven the distribution.

[0048] Spatial autocorrelation index, using the global Moran index. The Moran index assesses whether the density value is spatially clustered, discrete, or random. The Moran index ranges from -1 to 1. A value greater than 0 indicates positive spatial autocorrelation, i.e., spatially clustered, while a value less than 0 indicates negative spatial autocorrelation, i.e., spatially discrete.

[0049] The distribution uniformity coefficient, also known as the clustering index, is determined by the coefficient of variation. reciprocal Spatial autocorrelation index, namely the Moran index The weighting coefficients are obtained through weighted fusion. and satisfy ,in The weights are the reciprocals of the coefficient of variation. The weights of the spatial autocorrelation index; the distribution uniformity coefficient. The higher the distribution evenness coefficient value, the more even the spatial distribution pattern of the species.

[0050] As described above, this invention provides a method for quantitatively assessing the release range of rare fish species based on eDNA macrobarcoding. This method determines the initial release location by combining hydrological characteristics of the release area, the ecological habits of rare fish species, and historical release data. It collects and processes eDNA samples and performs macrobarcoding sequencing to obtain a clean sequence dataset. Then, it constructs an information database through species identification, generates a distribution density field through spatial interpolation, and quantitatively calculates the diffusion range and aggregation degree of the released fish to obtain the assessment result. Next, it compares the diffusion range in the assessment result with the preset target range to calculate the release target adjustment factor. Based on whether the adjustment factor reaches the expected threshold, it determines whether the release location needs to be redefined. Finally, for cases where the threshold is not reached, multi-dimensional coupling analysis is used to screen new release locations and re-release the fish. This process of data collection and assessment is repeated to update the results until the adjustment factor reaches the target, thus determining the target release location and completing the quantitative assessment of the release range of rare fish species and the dynamic optimization of the release strategy. This invention has the following advantages over existing technologies:

[0051] (1) The technical solution adopted in this invention deeply integrates eDNA metabarcode sequencing technology with spatial interpolation analysis, breaking through the limitations of traditional assessment of the release range of rare fish relying on direct observation in the wild and tagging and recapture, and realizing quantitative and refined assessment of the release and diffusion range of rare fish species with low population density and difficult to capture.

[0052] (2) The technical solution adopted in this invention introduces the release target adjustment factor and dynamic iterative optimization mechanism, which realizes the accurate screening of release locations and the dynamic optimization of release strategies, effectively solving the technical pain point that traditional evaluation methods cannot adjust the release plan in real time according to the actual diffusion effect.

[0053] (3) The technical solution adopted in this invention couples key influencing factors such as hydrological characteristics, ecological habits, and historical release data in multiple dimensions, and constructs an assessment model that is more in line with the natural survival law of rare fish. It can more comprehensively and realistically reflect the spatial distribution characteristics of rare fish after release, and greatly improve the scientificity and reliability of the release range assessment results.

[0054] (4) The technical solution adopted in this invention has a simple operation process, efficient data collection and wide applicability. It does not require long-term deployment of large-scale field sampling equipment. Evaluation can be achieved simply by standardized collection and sequencing analysis of eDNA samples, which reduces the technical threshold and implementation cost of rare fish release evaluation. It can be widely used in the evaluation of release effects and release planning of different waters and different rare fish species, and has good industrialization promotion and practical application value. Attached Figure Description

[0055] Figure 1 Flowchart of a method for quantitatively assessing the release range of rare fish species based on eDNA macrobarcoding;

[0056] Figure 2 Flowchart for constructing a species detection information database;

[0057] Figure 3 A flowchart for generating quantitative assessment results of the release range;

[0058] Figure 4 Dynamic changes in the distribution evenness coefficient of rare fish species at different stages after release;

[0059] Figure 5 Dynamic relationship between post-release sampling density adjustment factor and diffusion range. Detailed Implementation

[0060] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments, but the scope of protection of the present invention is not limited to the following embodiments.

[0061] The flowchart of a method for quantitatively assessing the release range of rare fish species based on eDNA macrobarcoding provided in this invention is as follows: Figure 1 As shown, it includes the following steps:

[0062] (1) Determine the initial release location of rare fish: Collect hydrological characteristics of the release area, ecological habits of rare fish and historical release data of the area, and then determine the initial release location of rare fish based on the hydrological characteristics of the release area, ecological habits of rare fish and historical release data of the area; the hydrological characteristics include water flow velocity and dissolved oxygen content; the ecological habits of rare fish include diurnal vertical movement range and preferred water temperature; the historical release data of the area includes at least the release location and the number of rare fish released; the release area is a lake, reservoir and slow-flowing river section closed or semi-closed water body with an area range of 50–1000 km² and a water depth range of 2–30 m;

[0063] The rare fish species must cover at least one of the following categories:

[0064] 1) Fish species under first- or second-class protection;

[0065] 2) Threatened fish species;

[0066] 3) Critically endangered, endangered or vulnerable fish species.

[0067] (2) Data acquisition and processing: Environmental DNA samples were collected from the release water area and subjected to macrobarcode sequencing to obtain raw sequence data. Quality control operations were performed on the raw sequence data to remove low-quality reads and contaminated sequences, generating a clean sequence dataset. The clean sequence dataset includes a set of high-quality reads, a sequence quality score vector, and sampling point identifiers.

[0068] The specific steps for generating the clean sequence dataset are as follows:

[0069] 1) Collect water samples using an environmental DNA sampling device, extract total DNA and amplify fish-specific macrobarcode regions, prepare sequencing libraries, and obtain raw sequence data;

[0070] 2) Use a noise reduction program to process the raw sequence data, correct sequencing errors and remove chimeras, and output the error-corrected sequence set;

[0071] 3) The error-corrected sequence set is length-normalized and abundance-normalized to generate a clean sequence dataset containing a high-quality read length set, a sequence quality score vector, and sampling point identifiers.

[0072] (3) Species identification and database construction: The clean sequence dataset is compared with the rare fish reference macro barcode database to identify matching species sequences, record the detection frequency and sampling point spatial coordinates of each species, and construct a species detection information database; the species detection information database includes species classification code, detection frequency matrix and geographic location label; the rare fish reference macro barcode region is the hypervariable region of mitochondrial COI gene or 12SrRNA gene;

[0073] The flowchart for constructing the species detection information database is as follows: Figure 2 As shown, it includes the following steps:

[0074] 1) Perform local alignment between the sequences in the clean sequence dataset and the reference macro barcode database, calculate the sequence similarity, identify species sequences that meet the matching threshold, and generate preliminary species identification results; the matching threshold is defined as a sequence similarity ≥ 97% and an alignment coverage ≥ 90%;

[0075] 2) Count the number of times each species appears at different sampling points in the preliminary species identification results, and combine the spatial coordinates of the sampling points to form a species detection frequency distribution map;

[0076] 3) Integrate the species detection frequency distribution map with the timestamp data of release events, mark the detection time intervals and spatial location associations, and construct a species detection information database containing species classification codes, detection frequency matrices and geographic location labels.

[0077] (4) Generation of distribution density field: Based on the detection frequency and spatial coordinates in the species detection information database, the distribution density of species in the release water area is estimated by spatial interpolation method. Combined with the release time series data, the species distribution density field is generated. The species distribution density field includes spatial grid density values, time slice data and confidence interval index.

[0078] The specific steps for generating the species distribution density field are as follows:

[0079] 1) Extract species detection frequency and spatial coordinates from the species detection information database to establish a spatial point pattern dataset;

[0080] 2) The kernel density estimation method is used to process the spatial point pattern dataset, calculate the species density value in each spatial grid cell, and generate an initial density distribution map;

[0081] 3) Introduce the time dimension, divide the initial density distribution map into multiple time slices according to the release time series, perform Gaussian smoothing on each time slice, and output the time series density field;

[0082] 4) Combine the behavioral parameters of the released fish to perform spatial correction on the time series density field, and generate a species distribution density field that includes spatial grid density values, time slice data and confidence interval indicators; among which, the behavioral parameters of the released fish include the diurnal vertical movement range or preference for specific water temperatures.

[0083] (5) Calculation of release range: Analyze the changing trend of species distribution density field over time, extract distribution boundary and density gradient information, calculate the diffusion range and aggregation degree of released fish, and generate quantitative assessment results of release range; the quantitative assessment results of release range include spatial coverage index, distribution uniformity coefficient and diffusion rate parameter.

[0084] The specific steps for calculating the diffusion range and aggregation degree of the released fish are as follows:

[0085] 1) Based on the density gradient changes in the species distribution density field, detect the location of abrupt changes in density gradient, determine the coordinates of boundary points based on the first and second differences of density values, connect the boundary points to form a distribution contour line, and calculate the area enclosed by the contour line as the diffusion range.

[0086] 2) Calculate the coefficient of variation and spatial autocorrelation index of the density values ​​in the species distribution density field. Then, weight and fuse the reciprocal of the coefficient of variation with the spatial autocorrelation index to obtain the distribution uniformity coefficient, which is the aggregation index.

[0087] The coefficient of variation is the ratio of the standard deviation to the mean, and it is calculated using the following formula: In the formula, Represents the coefficient of variation. The standard deviation of the density value The mean of the density values; coefficient of variation The larger the value, the greater the relative fluctuation of the density value in space, that is, the more uneven the distribution.

[0088] Spatial autocorrelation index, using the global Moran index. The Moran index assesses whether the density value is spatially clustered, discrete, or random. The Moran index ranges from -1 to 1. A value greater than 0 indicates positive spatial autocorrelation, i.e., spatially clustered, while a value less than 0 indicates negative spatial autocorrelation, i.e., spatially discrete.

[0089] The distribution uniformity coefficient, also known as the clustering index, is determined by the coefficient of variation. reciprocal Spatial autocorrelation index, namely the Moran index The weighting coefficients are obtained through weighted fusion. and satisfy ,in The weights are the reciprocals of the coefficient of variation. The weights of the spatial autocorrelation index; the distribution uniformity coefficient. The higher the distribution evenness coefficient value, the more even the spatial distribution pattern of the species.

[0090] The flowchart for generating the quantitative assessment results of the release range is as follows: Figure 3 As shown, it includes the following steps:

[0091] 1) Extract the boundary coordinate sequence of the diffusion range, calculate the area of ​​the polygon enclosed by the sequence, and use it as a spatial coverage index;

[0092] 2) Calculate the coefficient of variation and spatial autocorrelation index of the density values ​​in the species distribution density field. Then, weight and fuse the reciprocal of the coefficient of variation with the spatial autocorrelation index to obtain the distribution uniformity coefficient, which is the index of the degree of aggregation.

[0093] 3) Obtain the timestamp of the release event and the current assessment time point, calculate the time interval, and combine it with the spatial coverage index to calculate the diffusion distance per unit time, which is used as the diffusion rate parameter;

[0094] 4) Combine the spatial coverage index, distribution uniformity coefficient, and diffusion rate parameter into a structured data record, add sampling point identifiers and evaluation timestamps, fill in the data according to the preset report template, and generate a quantitative evaluation result of the release range containing a quantitative index table and a spatial distribution diagram.

[0095] (6) Dynamically adjust the release strategy: compare the diffusion range in the quantitative assessment results of the release range with the preset target range, and calculate the release target adjustment factor; the preset target range is determined based on the ecological habit parameters of the released fish and the historical carrying capacity data of the water area.

[0096] If the release target adjustment factor is lower than the expected threshold, the release location is re-determined based on the species detection information database in step (3), the species distribution density field in step (4), and the quantitative assessment results of the release range in step (5), combined with the diffusion pattern of rare fish and the spatial characteristics of the release water area; after the release operation is carried out at the re-determined release location, steps (2)-(5) are repeated to update the species distribution density field and the quantitative assessment results of the release range until the release target adjustment factor reaches the expected threshold; if the release target adjustment factor reaches or exceeds the expected threshold, the location of the current release location is recorded and the current release location is determined as the target release location;

[0097] The specific steps for re-determining the release location are as follows:

[0098] 1) Based on the species detection information database in step (3), extract the detection frequency and geographical location association information of each sampling point of rare fish, and identify the blank distribution area where the fish detection frequency is zero and the core aggregation area with a high detection frequency.

[0099] 2) Retrieve the species distribution density field from step (4), analyze the spatial gradient change characteristics of density values, and clarify the direction of fish diffusion, the short-spot area and the low-density extension area.

[0100] 3) Extract the spatial coverage and diffusion rate parameters from the quantitative assessment results of the release range in step (5) to determine the spatial orientation of fish diffusion restriction and the area with weak diffusion capacity.

[0101] 4) Combining the dispersal patterns of rare fish species, matching suitable dispersal paths and spatial habitat requirements for fish, and considering the spatial characteristics of the release area, identifying hydrologically connected areas and habitat-suitable areas in the water that are suitable for fish survival and conducive to their extension to blank distribution areas and areas with dispersal shortcomings, and combining the results of the coupling analysis of the above multi-dimensional information, selecting areas that can fill the gaps in fish distribution and promote the dispersal of fish to the target range as new release sites, while taking into account the hydrological connectivity with the original core aggregation area to ensure the habitat continuity between the new release site and the existing distribution area.

[0102] The formula for calculating the release target adjustment factor is as follows: ,in Represents the target's diffusion range. Represents the actual spread range. It is a dimensionless numerical value, ranging from 0 to 1; when When, it indicates that the release target adjustment factor is lower than the expected threshold; when This indicates that the release target adjustment factor has reached or exceeded the expected threshold; the target diffusion range is determined by statistically analyzing the diffusion range quantiles of historical successful release cases.

[0103] Example: This example uses the release range assessment of the rare fish species *Leptochloa longfin* in a certain watershed as an example. Based on the present invention's quantitative assessment method for the release range of rare fish species using eDNA macrobarcoding, the release range of *Leptochloa longfin* is assessed. The release area is a medium-sized reservoir with an area of ​​75 km², a water depth of 2-20 m, a semi-enclosed water body type, a water flow velocity of 0.1-0.3 m / s, and a dissolved oxygen content of 7-9 mg / L. Based on the diurnal vertical migration range of *Leptochloa longfin* (2-15 m), its preferred water temperature of 18-24℃, and the historical release locations and the number of rare fish released in this area, the initial release location is determined to be the central part of the reservoir (latitude and longitude coordinates 118.45°E, 29.28°N). The release scale is 10,000 healthy juvenile *Leptochloa longfin*, with a body length of 5-7 cm. Then, the quantitative assessment of the release range is carried out according to the following steps:

[0104] S1. Data Acquisition and Processing: In this embodiment, water samples are collected from the release water area by systematically deploying environmental DNA sampling devices. The environmental DNA sampling devices typically include sterile water samplers, filtration equipment, and sample preservation tubes. During operation, a certain volume of water is collected at preset points according to the gridded sampling plan. The water is immediately filtered using a filter membrane to capture environmental DNA particles. The filtered membrane is then transported to the laboratory under low-temperature conditions for subsequent analysis. In a laboratory setting, total DNA was extracted from the filter membrane using a commercial DNA extraction kit, following standard operating procedures including lysis, binding, washing, and elution. The obtained total DNA sample was tested for concentration and purity using a micro-nucleic acid quantification instrument to ensure that the DNA quality met amplification requirements. Specific primers were designed and synthesized targeting fish-specific macrobarcode regions, with primer targets selected from hypervariable regions such as the mitochondrial COI gene or 12S rRNA gene. The target fragment was amplified using polymerase chain reaction (PCR). The reaction system included DNA template, primers, polymerase, and buffer. Cyclic conditions were optimized to minimize non-specific amplification. After verification by agarose gel electrophoresis, the amplified products were used to construct sequencing libraries using a library construction kit, adding sequencing adapters and indexes. Finally, paired-end sequencing was performed using a high-throughput sequencing platform such as Illumina MiSeq to obtain raw sequence data in FASTQ format.

[0105] Then, the DADA2 denoising program was used to process the raw sequence data, setting the maximum expected error rate to 0.05. Sequencing errors were identified and corrected, and chimeric sequences were removed, outputting an error-corrected sequence set. This set was then length-normalized, removing sequences shorter than 200 bp or longer than 300 bp. Abundance normalization was performed using a resampling method, converting the sequence count of each sample into a proportion per million sequences. Finally, a clean sequence dataset containing a high-quality read length set, a sequence quality score vector, and sampling point identifiers was generated.

[0106] S2. Species identification and database construction: The sequences in the clean sequence dataset are locally compared with the reference macro barcode database for rare fish species. The reference database contains the COI gene standard sequences of 30 rare fish species, such as the longfin brevicornu. The sequence similarity matching threshold is set to ≥97%. After calculating the sequence similarity, the species sequences that reach the matching threshold are identified, and preliminary species identification results are generated.

[0107] The frequency of occurrence of *Leptochloa longfini* at each sampling point was statistically analyzed based on the preliminary identification results. Combined with the latitude and longitude coordinates of each sampling point (e.g., sampling point S01 has coordinates of 118.5°E, 29.3°N), a species detection frequency distribution map was generated. This map was then integrated with the timestamp data of the release events, and the time interval and spatial location correlation of each detection event were marked. A species detection information database containing species classification codes, a detection frequency matrix, and geographic location labels was constructed. The freshwater fish classification codes adopt a hierarchical coding structure, typically containing four levels of taxonomic unit information: order, family, genus, and species. For example, the classification code for *Leptochloa longfini* is CYPR-2025-LF01.

[0108] S3. Generation of Distribution Density Field: The detection frequency and corresponding spatial coordinates of the sampling points for *Leptochloa longfini* are extracted from the species detection database to establish a spatial point pattern dataset. This dataset is processed using kernel density estimation, with a spatial grid cell size of 500m × 500m. The species density value within each grid cell is calculated to generate an initial density distribution map.

[0109] Introducing a time dimension, the initial density distribution map was divided into six time slices at 1, 2, 3, 4, 5, and 6 months post-release. Gaussian smoothing was applied to each time slice to smooth the density, outputting a time-series density field. Combining the behavioral parameters of the longfin brevicornu (a type of fish), including its diurnal vertical migration range of 2-15m and preferred water temperature of 18-24℃, spatial correction was performed on the time-series density field, generating a species distribution density field containing spatial grid density values, time slice data, and confidence interval indices.

[0110] S4. Release Range Assessment: Analyze the density gradient changes in the species distribution density field, detect abrupt changes in the density gradient, determine the boundary point coordinates based on the first and second differences of the density values, use the Alpha shape algorithm to connect the boundary points to form a distribution contour line, and use the shoelace formula to calculate the area enclosed by the contour line as the diffusion range.

[0111] Calculate the coefficient of variation of density values ​​within a species distribution density field. The spatial autocorrelation index, also known as the global Moran index. The formula for calculating the coefficient of variation is: In the formula, where, The standard deviation of the density value The mean of the density values ​​is represented by the coefficients w1=0.6 and w2=0.4 based on historical data (satisfying w1+w2=1). The reciprocal of the coefficient of variation is weighted and fused with the spatial autocorrelation index to obtain the distribution uniformity coefficient. This coefficient is an indicator of the degree of aggregation. The dynamic change of the distribution evenness coefficient of rare fish species at different stages after release is shown in the figure below. Figure 4 As shown;

[0112] S5. Dynamic Adjustment of Release Strategy: Compare the diffusion range in the quantitative assessment results of the release area with the preset target range, and calculate the release target adjustment factor; based on the ecological habit parameters of the longfin lanceolate and the historical carrying capacity data of the water area, determine that the diffusion range of the longfin lanceolate in the release area is ≥50 km² 6 months after release; use the formula... Calculate the release target adjustment factor, where Represents the target range. This represents the actual diffusion range. The dynamic relationship between the release target adjustment factor and the diffusion range after release is shown in the following figure. Figure 5 As shown.

[0113] Six months after the release, when the release range was assessed according to the above steps, the actual diffusion range was 40 km², which was 10 km² lower than the expected threshold. At this point, the release target adjustment factor was adjusted. Therefore, it is necessary to redetermine the release location, extract the detection frequency and geographical location association information of each sampling point of *Leptochloa lanceolata* based on the species detection information database in step (3), identify the blank distribution area with zero detection frequency of *Leptochloa lanceolata* and the core aggregation area with high detection frequency; retrieve the species distribution density field in step (4), analyze the spatial gradient change characteristics of density values, and clarify the short-spot areas and low-density extension areas of *Leptochloa lanceolata* diffusion direction; extract the spatial coverage and diffusion rate parameters from the quantitative assessment results of the release range in step (5), determine the spatial orientation of *Leptochloa lanceolata* diffusion restriction and the weak diffusion capacity area; combine the diffusion pattern of *Leptochloa lanceolata*, match the suitable diffusion path and spatial habitat requirements of *Leptochloa lanceolata*. To determine the optimal release locations, the following steps were taken: First, considering the spatial characteristics of the release area, hydrological connectivity and habitat suitability areas suitable for the survival of *Sclerodermus longfinnis* and conducive to its expansion into undeveloped distribution areas and areas with limited dispersal potential were identified. Second, a combined analysis of these multi-dimensional information was conducted to select areas that could fill gaps in fish distribution and promote their spread to the target area as new release sites. Third, hydrological connectivity with existing core aggregation areas was considered to ensure habitat continuity between the new release sites and existing distribution areas. Based on these steps, the newly determined release sites were located in the northeast (118.52°E, 29.33°N) and southwest (118.38°E, 29.23°N) waters of the reservoir. A second release was then conducted based on these newly determined sites, repeating steps S1-S4 to update the species distribution density field and the quantitative assessment results of the release range.

[0114] The final evaluation results six months after the second release showed that the spatial coverage of the longfin brook reached 45 km², the distribution uniformity coefficient was 0.75, and the diffusion rate was 7 km² / month. All indicators met the evaluation requirements. Based on the evaluation results, the location of the current release site was recorded and determined as the target release site.

[0115] The data structure of the quantitative assessment report on release range is shown in Table 1, which defines the fields and descriptions of the data records.

[0116] Table 1: Data Structure Table for Quantitative Assessment Report of Release Range

[0117] Spatial coverage index floating-point numbers The diffusion range is a polygonal area, usually measured in square meters or square kilometers. Distribution uniformity coefficient floating-point numbers The weighted fusion value is dimensionless and its range depends on the input metrics. Diffusion rate parameters floating-point numbers The rate of change of area per unit time, such as square meters per day. Sampling point identifier String A unique code that identifies a sampling point, such as "S001". Evaluation timestamp Date and Time The exact time the report was generated, in the format YYYY-MM-DDHH:MM:SS

Claims

1. A method for quantitatively assessing the release range of rare fish species based on eDNA macrobarcoding, characterized in that, Includes the following steps: (1) Determine the initial release location of rare fish: Collect hydrological characteristics of the release area, ecological habits of rare fish and historical release data of the area, and then determine the initial release location of rare fish based on the hydrological characteristics of the release area, ecological habits of rare fish and historical release data of the area; the hydrological characteristics include water flow velocity and dissolved oxygen content; the ecological habits of rare fish include diurnal vertical movement range and preferred water temperature; the historical release data of the area includes at least the release location and the number of rare fish released; the release area is a lake, reservoir and slow-flowing river section closed or semi-closed water body with an area range of 50–1000 km² and a water depth range of 2–30 m; (2) Data acquisition and processing: Environmental DNA samples were collected from the release water area and subjected to macrobarcode sequencing to obtain raw sequence data. Quality control operations were performed on the raw sequence data to remove low-quality reads and contaminated sequences, generating a clean sequence dataset. The clean sequence dataset includes a set of high-quality reads, a sequence quality score vector, and sampling point identifiers. (3) Species identification and database construction: The clean sequence dataset is compared with the rare fish reference macro barcode database to identify matching species sequences, record the detection frequency and sampling point spatial coordinates of each species, and construct a species detection information database; the species detection information database includes species classification code, detection frequency matrix and geographic location label; the rare fish reference macro barcode region is the hypervariable region of mitochondrial COI gene or 12SrRNA gene; (4) Generation of distribution density field: Based on the detection frequency and spatial coordinates in the species detection information database, the distribution density of species in the release water area is estimated by spatial interpolation method. Combined with the release time series data, the species distribution density field is generated. The species distribution density field includes spatial grid density values, time slice data and confidence interval index. (5) Calculation of release range: Analyze the changing trend of species distribution density field over time, extract distribution boundary and density gradient information, calculate the diffusion range and aggregation degree of released fish, and generate quantitative assessment results of release range; the quantitative assessment results of release range include spatial coverage index, distribution uniformity coefficient and diffusion rate parameter. (6) Dynamically adjust the release strategy: compare the diffusion range in the quantitative assessment results of the release range with the preset target range, and calculate the release target adjustment factor; If the release target adjustment factor is lower than the expected threshold, the release location is re-determined based on the species detection information database in step (3), the species distribution density field in step (4), and the quantitative assessment results of the release range in step (5), combined with the diffusion pattern of rare fish and the spatial characteristics of the release water area; after the release operation is carried out at the re-determined release location, steps (2)-(5) are repeated to update the species distribution density field and the quantitative assessment results of the release range until the release target adjustment factor reaches the expected threshold; if the release target adjustment factor reaches or exceeds the expected threshold, the location of the current release location is recorded and the current release location is determined as the target release location; The specific steps for re-determining the release location are as follows: 1) Based on the species detection information database in step (3), extract the detection frequency and geographical location association information of each sampling point of rare fish, and identify the blank distribution area where the fish detection frequency is zero and the core aggregation area with a high detection frequency. 2) Retrieve the species distribution density field from step (4), analyze the spatial gradient change characteristics of density values, and clarify the direction of fish diffusion, the short-spot area and the low-density extension area. 3) Extract the spatial coverage and diffusion rate parameters from the quantitative assessment results of the release range in step (5) to determine the spatial orientation of fish diffusion restriction and the area with weak diffusion capacity. 4) Combining the dispersal patterns of rare fish species, matching suitable dispersal paths and spatial habitat requirements for fish, and considering the spatial characteristics of the release area, identifying hydrologically connected areas and habitat-suitable areas in the water that are suitable for fish survival and conducive to their extension to blank distribution areas and areas with dispersal shortcomings, and combining the results of the coupling analysis of the above multi-dimensional information, selecting areas that can fill the gaps in fish distribution and promote the dispersal of fish to the target range as new release sites, while taking into account the hydrological connectivity with the original core aggregation area to ensure the habitat continuity between the new release site and the existing distribution area.

2. The method for quantitatively assessing the release range of rare fish species based on eDNA macrobarcoding according to claim 1, characterized in that, The rare fish mentioned in step (1) must cover at least one of the following categories: 1) Fish species under first- or second-class protection; 2) Threatened fish species; 3) Critically endangered, endangered or vulnerable fish species.

3. The method for quantitatively assessing the release range of rare fish species based on eDNA macrobarcoding according to claim 1, characterized in that, The specific steps for generating the clean sequence dataset in step (2) are as follows: 1) Collect water samples using an environmental DNA sampling device, extract total DNA and amplify fish-specific macrobarcode regions, prepare sequencing libraries, and obtain raw sequence data; 2) Use a noise reduction program to process the raw sequence data, correct sequencing errors and remove chimeras, and output the error-corrected sequence set; 3) The error-corrected sequence set is length-normalized and abundance-normalized to generate a clean sequence dataset containing a high-quality read length set, a sequence quality score vector, and sampling point identifiers.

4. The method for quantitatively assessing the release range of rare fish species based on eDNA macrobarcoding according to claim 1, characterized in that, The specific steps for constructing the species detection information database in step (3) are as follows: 1) Perform local alignment between the sequences in the clean sequence dataset and the reference macro barcode database, calculate the sequence similarity, identify species sequences that meet the matching threshold, and generate preliminary species identification results; the matching threshold is defined as a sequence similarity ≥ 97% and an alignment coverage ≥ 90%; 2) Count the number of times each species appears at different sampling points in the preliminary species identification results, and combine the spatial coordinates of the sampling points to form a species detection frequency distribution map; 3) Integrate the species detection frequency distribution map with the timestamp data of release events, mark the detection time intervals and spatial location associations, and construct a species detection information database containing species classification codes, detection frequency matrices and geographic location labels.

5. The method for quantitatively assessing the release range of rare fish species based on eDNA macrobarcoding according to claim 1, characterized in that, The specific steps for generating the species distribution density field in step (4) are as follows: 1) Extract species detection frequency and spatial coordinates from the species detection information database to establish a spatial point pattern dataset; 2) The kernel density estimation method is used to process the spatial point pattern dataset, calculate the species density value in each spatial grid cell, and generate an initial density distribution map; 3) Introduce the time dimension, divide the initial density distribution map into multiple time slices according to the release time series, perform Gaussian smoothing on each time slice, and output the time series density field; 4) Combine the behavioral parameters of the released fish to perform spatial correction on the time series density field, and generate a species distribution density field that includes spatial grid density values, time slice data and confidence interval indicators; among which, the behavioral parameters of the released fish include the diurnal vertical movement range or preference for specific water temperatures.

6. The method for quantitatively assessing the release range of rare fish species based on eDNA macrobarcoding according to claim 1, wherein the specific steps for calculating the diffusion range and aggregation degree of the released fish species in step (5) are as follows: 1) Based on the density gradient changes in the species distribution density field, detect the location of abrupt changes in density gradient, determine the coordinates of boundary points based on the first and second differences of density values, connect the boundary points to form a distribution contour line, and calculate the area enclosed by the contour line as the diffusion range. 2) Calculate the coefficient of variation and spatial autocorrelation index of the density values ​​in the species distribution density field. Then, weight and fuse the reciprocal of the coefficient of variation with the spatial autocorrelation index to obtain the distribution uniformity coefficient, which is the aggregation index.

7. The method for quantitatively assessing the release range of rare fish species based on eDNA macrobarcoding according to claim 1, characterized in that, The specific steps for generating the quantitative assessment results of the release range in step (5) are as follows: 1) Extract the boundary coordinate sequence of the diffusion range, calculate the area of ​​the polygon enclosed by the sequence, and use it as a spatial coverage index; 2) Calculate the coefficient of variation and spatial autocorrelation index of the density values ​​in the species distribution density field. Then, weight and fuse the reciprocal of the coefficient of variation with the spatial autocorrelation index to obtain the distribution uniformity coefficient, which is the index of the degree of aggregation. 3) Obtain the timestamp of the release event and the current assessment time point, calculate the time interval, and combine it with the spatial coverage index to calculate the diffusion distance per unit time, which is used as the diffusion rate parameter; 4) Combine the spatial coverage index, distribution uniformity coefficient, and diffusion rate parameter into a structured data record, add sampling point identifiers and evaluation timestamps, and generate a quantitative evaluation result of the release range containing a quantitative index table and a spatial distribution diagram.

8. The method for quantitatively assessing the release range of rare fish species based on eDNA macrobarcoding according to claim 1, characterized in that: The preset target range mentioned in step (6) is determined based on the ecological habit parameters of the released fish and the historical carrying capacity data of the water area.

9. The method for quantitatively assessing the release range of rare fish species based on eDNA macrobarcoding according to claim 1, characterized in that, The formula for calculating the release target adjustment factor mentioned in step (6) is as follows: ,in Represents the target's diffusion range. Represents the actual spread range. It is a dimensionless numerical value, ranging from 0 to 1; when When, it indicates that the release target adjustment factor is lower than the expected threshold; when This indicates that the release target adjustment factor has reached or exceeded the expected threshold; the target diffusion range is determined by statistically analyzing the diffusion range quantiles of historical successful release cases.

10. The method for quantitatively assessing the release range of rare fish species based on eDNA macrobarcoding according to claim 6, characterized in that: The coefficient of variation is the ratio of the standard deviation to the mean, and it is calculated using the following formula: In the formula, Represents the coefficient of variation. The standard deviation of the density value The mean of the density values; coefficient of variation The larger the value, the greater the relative fluctuation of the density value in space, that is, the more uneven the distribution. Spatial autocorrelation index, using the global Moran index. The Moran index assesses whether the density value is spatially clustered, discrete, or random. The Moran index ranges from -1 to 1. A value greater than 0 indicates positive spatial autocorrelation, i.e., spatially clustered, while a value less than 0 indicates negative spatial autocorrelation, i.e., spatially discrete. The distribution uniformity coefficient, also known as the clustering index, is determined by the coefficient of variation. reciprocal Spatial autocorrelation index, namely the Moran index The weighting coefficients are obtained through weighted fusion. and satisfy ,in The weights are the reciprocals of the coefficient of variation. The weights of the spatial autocorrelation index; the distribution uniformity coefficient. The higher the distribution evenness coefficient value, the more even the spatial distribution pattern of the species.

Images