A method for predicting river aquatic biodiversity based on remote sensing and environmental DNA

Abstract

The present invention discloses a method for predicting river aquatic biodiversity based on remote sensing and environmental DNA. The method comprises the following steps: determining the abundance of a first species at each sampling point in the entire river basin based on the environmental DNA of each sampling point; determining the patch corresponding to each sampling point based on the sampling point and the species activity range, and determining the first remote sensing index based on the pre-acquired remote sensing vegetation index and the corresponding first species abundance in the patch; training a first preset model based on the first remote sensing index and the corresponding first species abundance in each patch, and determining the first preset model as a species abundance model when the first preset model reaches a preset standard; dividing the entire river basin into multiple patches, and determining a second remote sensing index in each patch, wherein the second remote sensing index is determined by the remote sensing vegetation index corresponding to the first remote sensing index; inputting the second remote sensing index in each patch into the species abundance model, and determining the second species abundance in the entire river basin, so as to determine the species diversity in the entire river basin.

Description

Technical Field

[0001] The present invention relates to the field of biological monitoring technology, and in particular to a method for predicting river aquatic biodiversity based on remote sensing and environmental DNA. Background Art

[0002] Aquatic biodiversity in a river basin is one of the important criteria for evaluating water ecological integrity. The biodiversity evaluation of the entire river basin and the spatial distribution of species within the river basin are issues that need to be urgently addressed in the evaluation of aquatic biodiversity in the river basin.

[0003] With the development of molecular sequencing technology, eDNA has facilitated efficient aquatic biodiversity monitoring. However, eDNA technology remains a point-based monitoring method, making it difficult to monitor difficult-to-reach areas with complex terrain, and it cannot meet the needs of high-resolution species diversity assessment and spatial distribution monitoring.

[0004] At present, in response to the above technical defects, eDNA technology is mainly regarded as particulate matter or sediment. Existing river transport models are used to combine the eDNA technology monitoring results in the entire river basin to predict the species distribution in the entire river basin. This does not take into account the relationship between organisms and the surrounding environment, and deviates from the basic theories of traditional ecology. Summary of the Invention

[0005] In view of this, the present invention provides a method for predicting river aquatic biodiversity based on remote sensing and environmental DNA. Based on remote sensing data and environmental DNA, the biodiversity and species distribution in the basin are predicted respectively, so as to solve the problem that the existing technology cannot evaluate the aquatic biodiversity and monitor the species distribution of the entire basin.

[0006] To achieve the above objectives, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for predicting river aquatic biodiversity based on remote sensing and environmental DNA, the method comprising:

[0008] Based on the environmental DNA of each sampling site, the abundance of the first species at each sampling site in the entire watershed was determined;

[0009] Determining patches corresponding to the sampling points based on the sampling points and the species activity range, obtaining remotely sensed vegetation indicators within the patches, and determining a first remotely sensed indicator based on the remotely sensed vegetation indicators and the corresponding abundance of the first species;

[0010] training a first preset model based on the first remote sensing indicator and the corresponding first species abundance in each patch, and determining the first preset model as a species abundance model when the first preset model meets a preset standard;

[0011] Dividing the entire watershed into a plurality of patches, and determining a second remote sensing indicator in each patch, wherein the second remote sensing indicator is determined by each of the remote sensing vegetation indicators corresponding to the first remote sensing indicator;

[0012] The second remote sensing indicator in each patch is input into the species abundance model to determine the second species abundance in the entire basin, so as to determine the species diversity in the entire basin.

[0013] In a possible implementation, the patches corresponding to the sampling points are determined based on the sampling points and the activity range of the species, specifically as follows:

[0014] Based on the activity ranges of different species, buffer zones of different distances were generated with the sampling points as the center;

[0015] Determine each remote sensing vegetation index within each buffer zone, calculate the correlation between each remote sensing vegetation index and the abundance of the first species, and determine the buffer zone corresponding to the highest total correlation as the patch.

[0016] In a possible implementation, determining the first remote sensing indicator based on each of the remotely sensed vegetation indicators and the corresponding first species abundance includes:

[0017] Determining the mean value and standard deviation of each remotely sensed vegetation indicator, and performing correlation analysis on each mean value and each standard deviation and the corresponding first species abundance;

[0018] If there is a significant correlation between the average value and the first species abundance, the average value is used as the first remote sensing indicator;

[0019] If the standard deviation is significant to the first species abundance, the standard deviation is also used as the first remote sensing indicator.

[0020] In a second aspect, the present invention provides a method for predicting the distribution of river aquatic species based on remote sensing and environmental DNA, the method comprising:

[0021] A second preset model is trained based on the first species abundance and the first remote sensing index obtained by the method of any embodiment of the first aspect, and pre-acquired 0-1 matrix data on the presence or absence of a species. When the second preset model meets a preset standard, the second preset model is determined to be a species distribution model, wherein the 0-1 matrix data on the presence or absence of the species is determined by the first species abundance, so as to determine the distribution of the species;

[0022] The second species abundance and the second remote sensing index obtained by the method of any embodiment of the first aspect are input into the species distribution model to determine the distribution of each species in the entire watershed.

[0023] In a third aspect, the present invention provides a device for predicting river aquatic biodiversity based on remote sensing and environmental DNA, the device comprising:

[0024] The species abundance module is used to determine the abundance of the first species at each sampling point in the entire watershed based on the environmental DNA of each sampling point;

[0025] a first remote sensing indicator module, configured to determine a patch corresponding to each sampling point based on the sampling point and the species activity range, obtain each remote sensing vegetation indicator within the patch, and determine a first remote sensing indicator based on each remote sensing vegetation indicator and the corresponding first species abundance;

[0026] a species abundance model module, configured to train a first preset model based on the first remote sensing indicator and the corresponding first species abundance in each patch, and determine that the first preset model is a species abundance model when the first preset model meets a preset standard;

[0027] A second remote sensing indicator module is used to divide the entire watershed into a plurality of patches and determine a second remote sensing indicator in each patch, where the second remote sensing indicator is determined by each of the remote sensing vegetation indicators corresponding to the first remote sensing indicator;

[0028] The first prediction module is used to input the second remote sensing indicator in each patch into the species abundance model to determine the abundance of the second species in the entire basin, so as to determine the species diversity in the entire basin.

[0029] In a possible implementation, the first remote sensing indicator module is specifically configured to:

[0030] Based on the activity ranges of different species, buffer zones of different distances were generated with the sampling points as the center;

[0031] Determine each remote sensing vegetation index within each buffer zone, calculate the correlation between each remote sensing vegetation index and the abundance of the first species, and determine the buffer zone corresponding to the highest total correlation as the patch.

[0032] In a possible implementation, the first remote sensing indicator module is specifically configured to:

[0033] Determining the mean value and standard deviation of each remotely sensed vegetation indicator, and performing correlation analysis on each mean value and each standard deviation and the corresponding first species abundance;

[0034] If there is a significant correlation between the average value and the first species abundance, the average value is used as the first remote sensing indicator;

[0035] If the standard deviation is significant to the first species abundance, the standard deviation is also used as the first remote sensing indicator.

[0036] In a fourth aspect, the present invention provides a device for predicting the distribution of river aquatic species based on remote sensing and environmental DNA, the device comprising:

[0037] a species distribution model module, configured to train a second preset model based on the first species abundance and the first remote sensing index obtained according to any embodiment of the third aspect, and pre-acquired 0-1 matrix data on the presence or absence of a species, and determine, when the second preset model meets a preset standard, that the second preset model is a species distribution model, wherein the 0-1 matrix data on the presence or absence of a species is determined by the first species abundance to determine the distribution of the species;

[0038] The second prediction module is used to input the second species abundance and the second remote sensing index obtained in any embodiment of the third aspect into the species distribution model to determine the distribution of each species in the entire basin.

[0039] In a fifth aspect, the present invention provides an electronic device, comprising a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;

[0040] Memory for storing computer programs;

[0041] A processor configured to implement the steps of the method for predicting river aquatic biodiversity based on remote sensing and environmental DNA according to any one of the embodiments of the first aspect when executing the program stored in the memory;

[0042] Alternatively, when used to execute a program stored in the memory, the steps of the method for predicting the distribution of river aquatic species based on remote sensing and environmental DNA as described in the second embodiment are implemented.

[0043] In a sixth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for predicting river aquatic biodiversity based on remote sensing and environmental DNA as described in any one of the embodiments of the first aspect;

[0044] Alternatively, when the computer program is executed by a processor, the steps of the method for predicting the distribution of river aquatic biological species based on remote sensing and environmental DNA as described in the second embodiment are implemented.

[0045] The above technical solution provided by the embodiment of the present invention has the following advantages compared with the prior art:

[0046] An embodiment of the present invention provides a method for predicting river aquatic biodiversity based on remote sensing and environmental DNA. The method determines the abundance of a first species at each sampling point across the entire river basin based on the environmental DNA of each sampling point. The method then determines the corresponding patch for each sampling point based on the sampling point and the species' range. Remotely sensed vegetation indicators within the patch are obtained, and a first remotely sensed indicator is determined based on each remotely sensed vegetation indicator and the corresponding first species abundance. A first preset model is trained based on the remotely sensed indicator within each patch and the corresponding first species abundance. When the first preset model meets a preset standard, the first preset model is determined as a species abundance model. The entire river basin is divided into multiple patches, and a second remotely sensed indicator is determined within each patch. The second remotely sensed indicator within each patch is input into the species abundance model to determine the second species abundance across the entire river basin, thereby determining the species diversity within the entire river basin. Furthermore, a method for predicting the distribution of aquatic species in rivers is provided. This method addresses the existing problem of the inability to evaluate aquatic biodiversity and monitor species distribution across the entire river basin. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 A schematic flow chart of a method for predicting river aquatic biodiversity based on remote sensing and environmental DNA provided by an embodiment of the present invention;

[0048] Figure 2 This is a flowchart of the patch division method;

[0049] Figure 3 A flowchart of the method for determining remote sensing indicators;

[0050] Figure 4 This is a schematic diagram of the species abundance prediction results for the entire river basin;

[0051] Figure 5 A schematic flow chart of a method for predicting river aquatic species based on remote sensing and environmental DNA provided by an embodiment of the present invention;

[0052] Figure 6 This is a schematic diagram of the predicted species distribution structure of the whole basin for Ephemeroptera, Lysimachia, and Chironomidae;

[0053] Figure 7 A schematic diagram of the structure of a river aquatic biodiversity prediction device based on remote sensing and environmental DNA provided by an embodiment of the present invention;

[0054] Figure 8 This is a structural diagram of the first remote sensing indicator module;

[0055] Figure 9 A schematic diagram of the structure of a device for predicting river aquatic species based on remote sensing and environmental DNA provided by an embodiment of the present invention;

[0056] Figure 10 A schematic structural diagram of an electronic device is provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0057] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0058] To facilitate understanding of the embodiments of the present invention, specific embodiments will be further explained below with reference to the accompanying drawings. The embodiments do not limit the embodiments of the present invention.

[0059] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like to indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0060] In the present invention, the word "exemplary" or "in an example" is used to mean "used as an example, illustration or description". Any embodiment described in the present invention as "exemplary" is not necessarily to be construed as being preferred or advantageous over other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes are not described in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed herein.

[0061] Currently, the inability to assess aquatic biodiversity and monitor species distribution across the entire river basin is primarily addressed by treating eDNA technology like particulate matter or sediment. Existing river transport models are used to combine eDNA monitoring results across the entire river basin to predict species distribution across the entire river basin. This approach fails to consider the relationship between organisms and their surrounding environment and deviates from the fundamentals of traditional ecology. The development of remote sensing technology has addressed this shortcoming. Remote sensing utilizes the differences in the reflectivity of ground objects to electromagnetic waves to detect their properties. This technology leverages the ability to simultaneously observe large areas and provide information on the attributes of ground objects within the river basin, but it is unable to monitor species distribution within the basin. This addresses the current inability to assess aquatic biodiversity and monitor species distribution across the entire river basin. This invention uses remote sensing data and environmental DNA monitoring data to predict river basin biodiversity within the river basin and, based on these results, to predict the distribution of specific organisms within the river basin. This method establishes a highly accurate modeling approach for high-resolution river basin biodiversity assessment, potentially addressing the challenges of biodiversity monitoring in complex locations within the river basin. Specifically, the present invention provides a method for predicting biodiversity and species distribution within a watershed. Figures 1 to 4 As shown, Figure 1 A schematic flow chart of a method for predicting river aquatic biodiversity provided by an embodiment of the present invention includes the following steps:

[0062] Step 110 : determining the abundance of the first species at each sampling point in the entire watershed based on the environmental DNA at each sampling point.

[0063] First, multiple sampling points are set up upstream and downstream of the river, and surface water (such as water at a depth of 0-5 cm) is collected at each sampling point. 3L of water sample is collected at each sampling point, and then the water sample is filtered. At the same time, a blank sample is set as the sample. The processing method of the blank sample is the same as the above-mentioned water sample, specifically filtering the pure water through a filter membrane.

[0064] Secondly, DNA was extracted according to the DNA extraction kit (DNeasy Blood & Tissue Kits) process, lysis buffer was added, and after repeated centrifugation, liquid was added for purification to complete the DNA extraction process. Stable PCR extension conditions were explored based on the selected enzymes and primers for PCT amplification.

[0065] Finally, the PCR-amplified DNA is sent to a sequencing platform for high-throughput sequencing, and the DNA fragments are sequenced. Specifically, the process includes the following three steps:

[0066] (1) The same executable operational units (OTUs) at the same sampling point were screened and merged. If the frequency of the OTUs was less than 50% of the number of different samples at the same sampling point, it was considered that the OTUs did not exist at the sampling point. Otherwise, the same OTU data for different samples at the same sampling point were merged.

[0067] (2) The number of OTUs was processed and the frequency of each OTU in all sampling points was calculated. If the frequency was greater than 10% of the total number of sampling points, the OTU was considered to exist;

[0068] (3) Merge different OTUs representing the same species. Perform a database search to obtain the specific species name and calculate the total number of species at each sampling point.

[0069] Step 120 , determining the patch corresponding to each sampling point based on the sampling point and the species activity range, obtaining each remote sensing vegetation index within the patch, and determining the first remote sensing index based on each remote sensing vegetation index and the corresponding first species abundance.

[0070] Let’s first introduce the identification of plaques:

[0071] Specifically, Figure 2 Flowchart for the plaque determination method, e.g. Figure 2 As shown in Figure 2, the patch division steps are:

[0072] Step 210 : Based on the activity ranges of different species, buffer zones of different distances are generated with the sampling point as the center.

[0073] Step 220 , determining each remote sensing vegetation index within each buffer zone, and calculating the correlation between each remote sensing vegetation index and the abundance of the first species, and determining the buffer zone corresponding to the highest total correlation as the patch.

[0074] The following is an introduction to remote sensing vegetation indicators:

[0075] Collect Sentinel-2 remote sensing images from the year before the sampling (cloud cover less than 10%) and perform the following operations:

[0076] (1) Radiation calibration Absolute radiation calibration is to establish a mathematical relationship between the pixel value and the actual radiation value, with the purpose of obtaining the absolute radiation value of the target. Absolute radiation calibration can be performed on the basis of relative radiation calibration, or a mathematical calibration model can be directly established through the original DN value and the actual radiation value to obtain the target surface radiation. Absolute radiation calibration obtains the radiation brightness (or reflectivity) of the top layer of the atmosphere. The radiation brightness calculation of the top layer of the atmosphere is to convert the initial DN value into radiation brightness, and its expression is as follows:

[0077] L λ =K*DN+C

[0078] Reflectivity and radiance can be converted into each other.

[0079] (2) Due to the influence of the atmosphere on electromagnetic waves, the top atmosphere irradiance is converted into surface reflectivity to eliminate the atmospheric influence on electromagnetic waves. The purpose of atmospheric correction is to eliminate the influence of atmospheric and light factors on the reflection of ground objects and obtain formal physical model parameters such as ground object reflectivity, emissivity, and surface temperature. This includes eliminating the influence of atmospheric water vapor, oxygen, carbon dioxide, methane, and ozone on ground object reflection; and eliminating the influence of atmospheric molecules and aerosol scattering. The FLAASH model (ENVI5.2) is selected.

[0080] (3) Calculate the following remote sensing vegetation indices (also known as candidate indices). B1, B2, B3, B4, B5, B6, B7, B8, and B11 refer to bands 1, 2, 3, 4, 5, 6, 7, 8, and 11 of Sentinel-2. The sum and average are used to generate the corresponding annual average vegetation index data, as shown in Table 1.

[0081] In summary, alternative patch sizes are proposed based on the activity ranges of different species. For example, aquatic insects have activity ranges generally between 500 and 5000 meters. Buffer zones of 500, 1000, and 5000 meters are generated around the sampling point. Correlations between various remotely sensed vegetation indicators within the buffer zones and the abundance of the first species in step 110 are calculated, and the distance with the highest overall correlation is selected as the patch size. For example, Table 2 shows the correlations between remotely sensed vegetation indicators and species abundance within different patch sizes.

[0082] Table 1 Remote sensing vegetation indicators

[0083]

[0084] Table 2 Correlation between remote sensing vegetation indices and species abundance in different patch areas

[0085]

[0086] The determination of the first remote sensing index is actually the screening of remote sensing vegetation. For details, see Figure 3 As shown, Figure 3 As shown, the first remote sensing indicator is determined by the following steps:

[0087] Step 310 : determining the mean value and standard deviation of each remote sensing vegetation index, and performing correlation analysis on each mean value and each standard deviation and the corresponding first species abundance.

[0088] In step 320, if there is a significant correlation between the average value and the abundance of the first species, the average value is used as the first remote sensing indicator.

[0089] In step 330 , if there is a significant correlation between the standard deviation and the abundance of the first species, the standard deviation is used as the first remote sensing indicator.

[0090] Table 3 Correlation between species abundance and remote sensing vegetation indices and GIS indicators (bold parts indicate significant correlation)

[0091]

[0092]

[0093] Specifically, the mean and standard deviation of various remote sensing vegetation indices within the patch were calculated respectively, combined with the latitude, longitude and elevation data of the sampling points, and correlation analysis was performed with the abundance of the first species. The criterion for screening was whether the mean of each remote sensing vegetation index was significantly correlated with species diversity (P < 0.5), and the first remote sensing index was determined. Specifically, the mean value of a certain remote sensing vegetation index was used as the first screening criterion. If there was a significant correlation, the mean value of the remote sensing vegetation index was used as the first remote sensing index, and then whether its standard deviation was significantly correlated was considered. If so, the standard deviation of the vegetation index was also used as the first remote sensing index.

[0094] Step 130 : training the first preset model based on the remote sensing index and the corresponding first species abundance in each patch, and determining the first preset model as a species abundance model when the first preset model meets a preset standard.

[0095] Specifically, the first remote sensing indicator selected in step 120 is modeled with the first species abundance at the corresponding sampling point, and a gradient boosting regression model is selected. The following is a brief introduction to the gradient boosting regression model: The gradient boosting regression tree is a technique that learns from its errors. It essentially pools ideas and integrates a number of inferior learning algorithms for learning. The general idea is:

[0096] Multiple layers of weak classifiers are trained on the same training set. Each layer uses the training set to train a weak classification model. They obtain prediction results from the trained model, and then determine the weights to be redistributed on each sample based on whether the sample classification in the training set is correct and the accuracy of the overall classification. The new data set with modified weights is used to train a lower-layer classifier.

[0097] Specifically, for example, in this application, there are a total of 51 sampling points, of which two-thirds of the sampling points are used as training sets for model training, and the remaining one-third of the sampling points are used for verification and parameter optimization until the difference between the accuracy of the training set and the test set is less than 10% (the process is to adjust the learning rate, repeatedly try from 0.5-0.7, and repeatedly experiment with the depth of the tree from 1000-2000), and the model is considered to be built.

[0098] Step 140: Divide the entire watershed into a plurality of patches, and determine a second remote sensing index in each patch. The second remote sensing index is determined by each remote sensing vegetation index corresponding to the first remote sensing index.

[0099] Step 150: Input the second remote sensing indicator in each patch into the species abundance model to determine the second species abundance in the entire watershed, so as to determine the species diversity in the entire watershed.

[0100] The entire watershed is divided into patches with the area selected in sub-step 220 of step 120 as the area, and the remote sensing indicators selected in sub-steps 320 and 330 of step 120 are calculated as the input set and input into the species abundance model in step 130 to determine the species abundance of the entire watershed, so as to predict the species diversity of the entire watershed, that is, the community level, as shown in FIG. Figure 4 shown.

[0101] The present invention provides a method for predicting river aquatic biodiversity based on remote sensing and environmental DNA. The method uses environmental DNA from each sampling point to determine the abundance of a first species at each sampling point throughout the entire river basin. The method then determines the patch corresponding to each sampling point based on the sampling point and the species' range. Remotely sensed vegetation indicators within the patch are obtained, and a first remotely sensed indicator is determined based on each remotely sensed vegetation indicator and the corresponding first species abundance. A first preset model is trained based on the remotely sensed indicator within each patch and the corresponding first species abundance. When the first preset model meets a preset standard, the first preset model is determined as a species abundance model. The entire river basin is divided into multiple patches, and a second remotely sensed indicator is determined within each patch. The second remotely sensed indicator within each patch is input into the species abundance model to determine the second species abundance across the entire river basin, thereby determining the species diversity within the entire river basin. Furthermore, the method provides a method for predicting the distribution of aquatic species in rivers. This method addresses the existing problem of the inability to evaluate aquatic biodiversity across the entire river basin.

[0102] The above is a method for predicting river aquatic biodiversity based on remote sensing and environmental DNA provided by an embodiment of the present invention. The following describes an embodiment of a method for predicting the distribution of river aquatic organisms based on remote sensing and environmental DNA provided by the present invention. Figure 5 , Figure 5 A flow chart of a method for predicting river aquatic species based on remote sensing and environmental DNA is provided in an embodiment of the present invention. Figure 5 As shown in Figure 2, the method for predicting river aquatic species based on remote sensing and environmental DNA includes the following steps:

[0103] In step 510, the second preset model is trained based on the first species abundance obtained in step 110 and the first remote sensing index obtained in step 120, as well as the pre-acquired 0-1 matrix data of whether a certain species exists. When the second preset model meets the preset standard, the second preset model is determined to be a species distribution model, wherein the 0-1 matrix data of whether a certain species exists is determined by the first species abundance to determine the distribution of the certain species.

[0104] First, we introduce whether a certain species exists in 0-1 matrix data: OTUs is translated as the smallest operational unit that can perform classification in taxonomy. More simply, the existence of this OTU is a DNA fragment specific to a certain organism. According to OTUs, we can directly determine whether the corresponding species exists at that point (if a certain OTU exists, the species exists, and the value here is 1; if it does not exist, it is 0, and finally a 0-1 matrix data is formed). Then, the distribution status of a species is determined based on the distribution status of its OTUs in the watershed.

[0105] In step 520 , the second species abundance obtained in step 150 and the second remote sensing index obtained in step 140 are input into a species distribution model to determine the distribution of each species in the entire watershed.

[0106] Specifically, the species abundance in step 110 is used as the basis to determine whether the species exists based on 0-1 data, and the species abundance of the sampling point and the parameters of the training sample in 130 are used as input parameters. The training set and test set samples in step 130 are used as the basis to select the random forest model. The general idea is: as the name suggests, a forest is established in a random way. There are many decision trees in the forest, and there is no correlation between each decision tree of the random forest. After obtaining the forest, when a new input sample enters, each decision tree in the forest is allowed to make a judgment to see which category the sample should belong to (for the classification algorithm), and then see which category is selected the most, and predict the sample to be that category. The random forest is composed of decision trees. The decision tree is actually a method of dividing the space with a hyperplane. Each time it is divided, the current space is divided into two to generate a model that can predict whether a species exists. Combined with the species abundance predicted in step 150, the species distribution is predicted. Taking the prediction of the four-season mayfly, the limp beetle, and the midge in the entire basin as an example, the final model predicts the distribution of each species as follows Figure 6 shown.

[0107] The present invention provides a method for predicting the distribution of river aquatic species based on remote sensing and environmental DNA. A second preset model is trained based on the first species abundance and first remote sensing index obtained by the method according to any embodiment of the first aspect, as well as pre-acquired 0-1 matrix data indicating the presence or absence of a species. When the second preset model meets a preset standard, the second preset model is determined to be a species distribution model, wherein the 0-1 matrix data indicating the presence or absence of a species is determined by the first species abundance to determine the distribution of the species. The second species abundance and second remote sensing index obtained by the method according to any embodiment of the first aspect are input into the species distribution model to determine the distribution of each species within the entire river basin. This method solves the problem in existing technologies that it is impossible to monitor the distribution of aquatic species throughout the entire river basin.

[0108] The above is an embodiment of the method for predicting river aquatic biodiversity and species distribution based on remote sensing and environmental DNA provided by the present invention. The following describes other embodiments of the method for predicting river aquatic biodiversity and species distribution based on remote sensing and environmental DNA provided by the present invention. Please refer to the following for details.

[0109] Figure 7 A schematic diagram of a device for predicting river aquatic biodiversity according to an embodiment of the present invention is shown in FIG. Figure 7 As shown, the river aquatic biodiversity prediction device includes: a species abundance model 71 , a first remote sensing indicator module 72 , a species abundance model module 73 , a second remote sensing indicator module 74 and a first prediction module 75 .

[0110] The species abundance module 71 is used to determine the abundance of the first species at each sampling point in the entire watershed based on the environmental DNA of each sampling point.

[0111] The first remote sensing index module 72 is used to determine the patch corresponding to each sampling point based on the sampling point and the species activity range, obtain each remote sensing vegetation index within the patch, and determine the first remote sensing index based on each remote sensing vegetation index and the corresponding first species abundance.

[0112] The species abundance model module 73 is used to train the first preset model based on the first remote sensing indicator and the corresponding first species abundance in each patch, and determine the first preset model as the species abundance model when the first preset model meets the preset standard.

[0113] The second remote sensing index module 74 is used to divide the entire watershed into multiple patches and determine a second remote sensing index in each patch. The second remote sensing index is determined by each remote sensing vegetation index corresponding to the first remote sensing index.

[0114] The first prediction module 75 is used to input the second remote sensing indicator in each patch into the species abundance model to determine the second species abundance in the entire watershed, so as to determine the species diversity in the entire watershed.

[0115] In one example, if Figure 8 As shown, the first remote sensing indicator module 72 includes a buffer unit 721 and a patch unit 722, wherein:

[0116] The buffer unit 721 is used to generate buffer zones of different distances with the sampling point as the center based on the activity ranges of different species;

[0117] The patch unit 722 is used to determine each remote sensing vegetation index within each buffer zone, calculate the correlation between each remote sensing vegetation index and the abundance of the first species, and determine the buffer zone corresponding to the highest total correlation as the patch.

[0118] In one example, if Figure 8 As shown, the first remote sensing indicator module 72 further includes an analysis unit 723, a first judgment unit 724 and a second judgment unit 725, which are specifically used to:

[0119] An analysis unit 723 is configured to determine the average value and standard deviation of each remotely sensed vegetation indicator, and perform correlation analysis on each average value and each standard deviation with the corresponding first species abundance;

[0120] A first judgment unit 724 is configured to use the average value as a first remote sensing indicator if there is a significance between the average value and the abundance of the first species;

[0121] The second judgment unit 725 is configured to use the standard deviation as the first remote sensing indicator if there is a significance between the standard deviation and the first species abundance.

[0122] The functions performed by the various components in the river aquatic biodiversity prediction device provided by the embodiment of the present invention have been described in detail in any of the above method embodiments, and therefore will not be repeated here.

[0123] An embodiment of the present invention provides a river aquatic biodiversity prediction device based on remote sensing and environmental DNA. The device determines the abundance of a first species at each sampling point across the entire river basin based on the environmental DNA of each sampling point. The device then determines the patch corresponding to each sampling point based on the sampling point and the species' range. Remotely sensed vegetation indicators within the patch are obtained, and a first remotely sensed indicator is determined based on each remotely sensed vegetation indicator and the corresponding first species abundance. A first preset model is trained based on the remotely sensed indicator within each patch and the corresponding first species abundance. When the first preset model meets a preset standard, the model is determined as a species abundance model. The entire river basin is divided into multiple patches, and a second remotely sensed indicator is determined within each patch. The second remotely sensed indicator within each patch is input into the species abundance model to determine the second species abundance across the entire river basin, thereby determining the species diversity within the entire river basin. Furthermore, a method for predicting the distribution of aquatic species in rivers is provided. This method addresses the existing problem of the inability to evaluate aquatic biodiversity across the entire river basin.

[0124] Figure 9 A schematic diagram of a device for predicting river aquatic species based on remote sensing and environmental DNA is provided in an embodiment of the present invention. Figure 9 As shown, the river aquatic organism species prediction device based on remote sensing and environmental DNA includes: a species distribution model module 91 and a second prediction module 92.

[0125] The species distribution model module 91 is used to train the second preset model based on the first species abundance and the first remote sensing index obtained according to any embodiment of the third aspect, and the pre-acquired 0-1 matrix data of whether a certain species exists. When the second preset model meets the preset standard, the second preset model is determined to be the species distribution model, wherein the 0-1 matrix data of whether a certain species exists is determined by the first species abundance to determine the distribution of a certain species.

[0126] The second prediction module 92 is used to input the second species abundance and the second remote sensing index obtained in any embodiment of the third aspect into the species distribution model to determine the distribution of each species in the entire basin.

[0127] The functions performed by each component in the river aquatic species distribution prediction device based on remote sensing and environmental DNA provided by the embodiment of the present invention have been described in detail in any of the above method embodiments, and therefore will not be repeated here.

[0128] The present invention provides a device for predicting the distribution of river aquatic species based on remote sensing and environmental DNA. A second preset model is trained based on the first species abundance and first remote sensing index obtained in the above embodiment, as well as pre-acquired 0-1 matrix data indicating the presence of a species. When the second preset model meets a preset standard, the second preset model is determined as a species distribution model. The 0-1 matrix data indicating the presence of a species is determined by the first species abundance, which is used to determine the distribution of the species. The second species abundance and second remote sensing index obtained in the above embodiment are input into the species distribution model to determine the distribution of each species within the entire river basin. This solves the problem of the existing technology in being unable to monitor the distribution of aquatic species throughout the entire river basin.

[0129] like Figure 10 As shown, an embodiment of the present invention provides an electronic device, including a processor 111 , a communication interface 112 , a memory 113 and a communication bus 114 , wherein the processor 111 , the communication interface 112 , and the memory 113 communicate with each other via the communication bus 114 .

[0130] Memory 113, for storing computer programs;

[0131] In one embodiment of the present invention, the processor 111 is configured to execute the program stored in the memory 113 to implement the steps of the method for predicting river aquatic biodiversity based on remote sensing and environmental DNA as provided in any of the aforementioned method embodiments;

[0132] Alternatively, when used to execute a program stored in the memory, the steps of the method for predicting the distribution of river aquatic species based on remote sensing and environmental DNA as provided in any of the aforementioned method embodiments are implemented.

[0133] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for predicting river aquatic biodiversity based on remote sensing and environmental DNA as provided in any of the aforementioned method embodiments;

[0134] Alternatively, when the computer program is executed by a processor, the steps of the method for predicting the distribution of river aquatic organism species based on remote sensing and environmental DNA as provided in any of the aforementioned method embodiments are implemented.

[0135] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0136] The entire river basin is divided into patches of different sizes, fully considering the relationship between species diversity and patch environment, and establishing a species diversity prediction model based on the relationship between species diversity and patch vegetation and geographical attributes, which meets the requirements of species diversity prediction in the entire river basin. On this basis, the species abundance (total number of species) in each patch is combined with vegetation, geographical attributes and whether species exist to establish corresponding relationships, thereby meeting the needs of predicting species distribution in the entire river basin.

[0137] It should be noted that, in this article, relational terms such as "first" and "second" are merely used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or equipment. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or equipment comprising the elements. In addition, in the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0138] Professionals should also be further aware that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0139] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein may be implemented using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0140] The foregoing is merely a detailed description of the present invention, intended to enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein, but is to be construed in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. A method for predicting river aquatic biodiversity based on remote sensing and environmental DNA, characterized by: The method comprises: Based on the environmental DNA of each sampling site, the abundance of the first species at each sampling site in the entire watershed was determined; Determining patches corresponding to the sampling points based on the sampling points and the species activity range, obtaining remotely sensed vegetation indicators within the patches, and determining a first remotely sensed indicator based on the remotely sensed vegetation indicators and the corresponding abundance of the first species; training a first preset model based on the first remote sensing indicator and the corresponding first species abundance in each patch, and determining the first preset model as a species abundance model when the first preset model meets a preset standard; Dividing the entire watershed into a plurality of patches, and determining a second remote sensing indicator in each patch, wherein the second remote sensing indicator is determined by each of the remote sensing vegetation indicators corresponding to the first remote sensing indicator; Inputting the second remote sensing indicator in each patch into the species abundance model to determine the abundance of the second species in the entire watershed, so as to determine the species diversity in the entire watershed; The patch corresponding to each sampling point is determined by the sampling point and the activity range of the species, specifically: based on the activity range of different species, buffer zones of different distances are generated with the sampling point as the center; each remote sensing vegetation indicator within each buffer zone is determined, and the correlation between each remote sensing vegetation indicator and the abundance of the first species is calculated, and the buffer zone corresponding to the highest total correlation is determined as the patch.

2. The method according to claim 1, characterized in that The determining of the first remote sensing indicator based on each of the remote sensing vegetation indicators and the corresponding first species abundance includes: Determining the mean value and standard deviation of each remotely sensed vegetation indicator, and performing correlation analysis on each mean value and each standard deviation and the corresponding first species abundance; If there is a significant correlation between the average value and the first species abundance, the average value is used as the first remote sensing indicator; If the standard deviation is significant to the first species abundance, the standard deviation is also used as the first remote sensing indicator.

3. A method for predicting the distribution of river aquatic species based on remote sensing and environmental DNA, characterized in that: The method comprises: A second preset model is trained based on the first species abundance and the first remote sensing indicator obtained by the method according to any one of claims 1 to 2, and pre-acquired 0-1 matrix data on the presence or absence of a species. When the second preset model meets a preset standard, the second preset model is determined to be a species distribution model, wherein the 0-1 matrix data on the presence or absence of a species is determined by the first species abundance, so as to determine the distribution of the species. The second species abundance and the second remote sensing index obtained by the method according to any one of claims 1-2 are input into the species distribution model to determine the distribution of each species in the entire watershed.

4. A river aquatic biodiversity prediction device based on remote sensing and environmental DNA, characterized by: The device comprises: The species abundance module is used to determine the abundance of the first species at each sampling point in the entire watershed based on the environmental DNA of each sampling point; a first remote sensing indicator module, configured to determine a patch corresponding to each sampling point based on the sampling point and the species activity range, obtain each remote sensing vegetation indicator within the patch, and determine a first remote sensing indicator based on each remote sensing vegetation indicator and the corresponding first species abundance; a species abundance model module, configured to train a first preset model based on the first remote sensing indicator and the corresponding first species abundance in each patch, and determine that the first preset model is a species abundance model when the first preset model meets a preset standard; A second remote sensing indicator module is used to divide the entire watershed into a plurality of patches and determine a second remote sensing indicator in each patch, where the second remote sensing indicator is determined by each of the remote sensing vegetation indicators corresponding to the first remote sensing indicator; a first prediction module, configured to input the second remote sensing indicator in each patch into the species abundance model to determine the abundance of the second species in the entire watershed, so as to determine the species diversity in the entire watershed; The first remote sensing indicator module is specifically used to generate buffer zones of different distances with the sampling point as the center based on the activity range of different species; determine each remote sensing vegetation indicator within each buffer zone, and calculate the correlation between each remote sensing vegetation indicator and the abundance of the first species, and determine the buffer zone corresponding to the highest total correlation as the patch.

5. The device according to claim 4, characterized in that The first remote sensing indicator module is used to: Determining the mean value and standard deviation of each remotely sensed vegetation indicator, and performing correlation analysis on each mean value and each standard deviation and the corresponding first species abundance; If there is a significant correlation between the average value and the first species abundance, the average value is used as the first remote sensing indicator; If the standard deviation is significant to the first species abundance, the standard deviation is also used as the first remote sensing indicator.

6. A device for predicting the distribution of river aquatic species based on remote sensing and environmental DNA, characterized in that: The device comprises: a species distribution model module, configured to train a second preset model based on the first species abundance and the first remote sensing index obtained by the apparatus according to any one of claims 4-5, and pre-acquired 0-1 matrix data on the presence or absence of a species, and determine, when the second preset model meets a preset standard, that the second preset model is a species distribution model, wherein the 0-1 matrix data on the presence or absence of a species is determined by the first species abundance, for determining the distribution of the species; The second prediction module is used to input the second species abundance and the second remote sensing index obtained by the device according to any one of claims 4-5 into the species distribution model to determine the distribution of each species in the entire watershed.

7. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory for storing computer programs; A processor, configured to implement the steps of the method for predicting river aquatic biodiversity based on remote sensing and environmental DNA as described in any one of claims 1 to 2 when executing the program stored in the memory; Alternatively, when used to execute the program stored in the memory, the steps of the method for predicting the distribution of river aquatic species based on remote sensing and environmental DNA as claimed in claim 3 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for predicting river aquatic biodiversity based on remote sensing and environmental DNA according to any one of claims 1 to 2 are implemented; Alternatively, when the computer program is executed by a processor, the steps of the method for predicting the distribution of river aquatic species based on remote sensing and environmental DNA as claimed in claim 3 are implemented.

Images