Biodiversity prediction method and device

Abstract

The invention provides a biodiversity prediction method and device, belongs to the technical field of biology, and aims to predict the biodiversity of an unobserved area in a research area and provide a spatial distribution pattern of the overall biodiversity of the research area, thereby facilitating protection planning of the research area. The method comprises the following steps: screening biodiversity data in a research area, training a random forest model by adopting the screened biodiversity data (biodiversity data of an observed area), predicting environment variable data of an unobserved area through the random forest model, and obtaining biodiversity prediction data of the unobserved area, and obtaining the biodiversity data of the unobserved area according to the biodiversity prediction data of the unobserved area and the corresponding biodiversity prediction residual error.

Description

Technical Field

[0001] The present invention relates to the field of biotechnology, and in particular to a method and device for predicting biodiversity. Background Art

[0002] The spatial distribution pattern of biodiversity is one of the core scientific issues in ecology, biogeography, and conservation biology; it is also the basis for conservation planning such as conservation priority areas, ecological corridors, and ecological restoration. At present, my country has carried out a large number of biodiversity monitoring surveys, using artificial sample lines, infrared cameras, voiceprints, or video collection equipment, and has accumulated a batch of detailed basic data on species distribution.

[0003] However, these survey data often present scattered, local point distributions, and are unable to provide the spatial distribution pattern of biodiversity in the study area as a whole, which is not conducive to conservation planning in the study area. Summary of the invention

[0004] The present invention proposes a biodiversity prediction method and device, which can predict the biodiversity of unobserved areas in a study area and provide the spatial distribution pattern of the biodiversity of the entire study area, thereby facilitating the protection planning of the study area.

[0005] In order to achieve the above object, the present invention adopts the following technical solution:

[0006] In the first aspect, the present invention provides a biodiversity prediction method, comprising: obtaining environmental variable data and biodiversity data of a study area within a period of time; the biodiversity data includes: species name, number of individuals and discovery location; the environmental variable data includes: climate factors, terrain factors, habitat factors and interference factors. Based on the species integrity and the end slope of the species accumulation curve, the biodiversity data is screened and a sample set is determined; in the study area, the part where the discovery location is located in the screened biodiversity data is the observed area, otherwise it is the unobserved area. The random forest model predicts the environmental variable data of the unobserved area within a period of time to obtain the biodiversity prediction data of the unobserved area within a period of time; the random forest model is obtained by training the sample set. Then, based on the biodiversity prediction data and the biodiversity prediction residuals corresponding to the biodiversity prediction data, the biodiversity data of the unobserved area within a period of time is determined; the biodiversity prediction residuals are obtained by interpolating the prediction residuals of the random forest model.

[0007] The biodiversity prediction method provided by the present invention screens the biodiversity data in the study area, and uses the screened biodiversity data (biodiversity data of the observed area) to train the random forest model, and predicts the environmental variable data of the unobserved area through the random forest model. After obtaining the biodiversity prediction data of the unobserved area, the biodiversity data of the unobserved area is obtained based on the biodiversity prediction data of the unobserved area and the corresponding biodiversity prediction residual. At this point, both the biodiversity data of the observed area and the biodiversity prediction data of the unobserved area are obtained. It can be seen that the present invention can predict the biodiversity of the unobserved area in the study area, and give the spatial distribution pattern of the biodiversity of the study area as a whole, thereby facilitating the protection planning of the study area.

[0008] In an implementation of the first aspect, based on species completeness and the end slope of the species accumulation curve, screening biodiversity data and determining a sample set of a random forest model includes:

[0009] Divide the study area into blocks;

[0010] For each of the plurality of blocks, biodiversity data of the discovery location falling within the geographic range of the block are associated with the block; and species completeness and a slope at the end of a species accumulation curve of the block are calculated using the biodiversity data associated with the block;

[0011] The biodiversity data in multiple blocks whose species integrity is greater than a first threshold and whose end slope of the species accumulation curve is less than a second threshold and the environmental variable data corresponding to the biodiversity data are used as a sample set;

[0012] The blocks associated with the biodiversity data in the sample set were classified as observed areas, and the remaining blocks in the study area were classified as unobserved areas.

[0013] In an implementation of the first aspect, the training method of the random forest model is:

[0014] The random forest model is trained through the training set; the input of the random forest model is environmental variable data, and the output of the random forest model is biodiversity prediction data; the sample set includes a training set and a test set.

[0015] In an implementation of the first aspect, a method for determining a biodiversity prediction residual is:

[0016] The trained random forest model is tested on the test set, and the difference between the actual biodiversity data in the test set and the predicted biodiversity data output by the trained random forest model is used as the prediction residual of the observed area; the sample set includes the training set and the test set;

[0017] Kriging interpolation is performed on the prediction residuals of the observed areas to obtain the biodiversity prediction residuals of the study area.

[0018] In a second aspect, the present invention provides a biodiversity prediction device, including an acquisition module, a screening module, a prediction module, and a determination module. The acquisition module is used to obtain environmental variable data and biodiversity data of a study area within a period of time; the biodiversity data includes: species name, number of individuals, and discovery location. The screening module is used to screen the biodiversity data and determine the sample set based on species integrity and the slope at the end of the species accumulation curve; in the study area, the part where the discovery location falls in the screened biodiversity data is the observed area, otherwise it is the unobserved area. The prediction module is used to predict the environmental variable data of the unobserved area within a period of time using a random forest model to obtain the biodiversity prediction data of the unobserved area within a period of time; the random forest model is obtained by training the sample set. The determination module is used to determine the biodiversity data of the unobserved area within a period of time based on the biodiversity prediction data and the biodiversity prediction residuals corresponding to the biodiversity prediction data; the biodiversity prediction residuals are obtained by interpolating the prediction residuals of the random forest model.

[0019] In an implementation of the second aspect, the screening module is specifically used to divide the study area into multiple blocks. For each of the multiple blocks, the biodiversity data whose location falls within the geographical scope of the block is associated with the block; and the species integrity of the block and the end slope of the species accumulation curve are calculated through the biodiversity data associated with the block. Blocks in the multiple blocks whose species integrity is greater than a first threshold and whose end slope of the species accumulation curve is less than a second threshold are classified as observed areas, and the remaining blocks in the multiple blocks are classified as unobserved areas. The biodiversity data associated with the blocks classified as observed areas and the environmental variable data corresponding to the biodiversity data are used as sample sets.

[0020] In an implementation of the first aspect and the second aspect, the living thing includes a bird.

[0021] In an implementation of the first aspect and the second aspect, the calculation formula for species integrity is as follows;

[0022]

[0023] Among them, SI represents species integrity, Sobs represents the number of species observed in the sample, Chao 1 represents the richness index, n 1 represents the number of species containing only one individual, n 2 represents the number of species containing only two individuals;

[0024] The slope at the end of the species accumulation curve is calculated by the cumulative curve of the number of species in the block over a period of time.

[0025] In a third aspect, the present invention provides an electronic device, comprising a processor and a memory coupled to the processor; the memory is used to store computer instructions, and when the electronic device is running, the processor executes the computer instructions stored in the memory, so that the electronic device executes the method as described in the first aspect or any one of its implementations.

[0026] In a fourth aspect, the present invention provides a computer-readable storage medium, comprising computer program instructions, which, when executed by a computer, enable the computer to execute the method in the first aspect or any one of its implementations.

[0027] In a fifth aspect, the present invention provides a computer program product, comprising computer program instructions, which, when executed on a computer, enable the computer to execute the method in the first aspect or any one of its implementations.

[0028] The technical effects corresponding to the above-mentioned second to fifth aspects and their possible implementations can refer to the above-mentioned description of the technical effects of the first aspect and its possible implementations, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is one of the schematic diagrams of the biodiversity prediction method provided in the embodiments of the present application;

[0030] Figure 2 This is the second schematic diagram of the biodiversity prediction method provided in the embodiment of the present application;

[0031] Figure 3 This is the third schematic diagram of the biodiversity prediction method provided in the embodiment of the present application;

[0032] Figure 4 This is the fourth schematic diagram of the biodiversity prediction method provided in the embodiment of the present application;

[0033] Figure 5 This is the fifth schematic diagram of the biodiversity prediction method provided in the embodiments of the present application;

[0034] Figure 6 It is a schematic diagram of the distribution grid of biodiversity for research provided in the embodiment of the present application;

[0035] Figure 7 This is one of the structural schematic diagrams of the biodiversity prediction device provided in the embodiments of the present application. DETAILED DESCRIPTION

[0036] The terms "first" and "second" etc. in the description and claims of the present invention are used to distinguish different objects rather than to describe a specific order of the objects.

[0037] The “and / or” in the embodiments of the present application indicates the relationship between objects. For example, A and / or B may indicate the following three situations: A exists alone, B exists alone, and A and B exist at the same time.

[0038] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0039] In the description of the present invention, unless otherwise specified, the meaning of "plurality" refers to two or more than two. For example, a plurality of blocks refers to two or more than two blocks.

[0040] The method and device provided in the embodiments of the present application relate to the spatial distribution pattern of biodiversity. The embodiments of the present application predict the unobserved biodiversity data in the study area by using the environmental variable data and biodiversity data that have been observed in the study area, thereby obtaining the spatial distribution pattern of the biodiversity of the entire study area.

[0041] It is understandable that biodiversity refers to the diversity of all biological species, genes and ecosystems on the earth. Protecting biodiversity is of great significance to maintaining the ecological balance of the earth. Biodiversity includes genetic diversity, species diversity and ecosystem diversity. In the embodiments of this application, biodiversity refers to species diversity.

[0042] Since in the prior art, the acquisition of biodiversity data is often obtained through artificial observation, and artificial observation has the characteristics of randomness, high uncertainty and large errors, so that the observed biodiversity data presents scattered and local point distribution, and it is impossible to give the spatial distribution pattern of the biodiversity of the entire study area, which is not conducive to the protection planning of the study area. In order to solve the above problems, the embodiment of the present application provides a biodiversity prediction method and device, the biodiversity data in the study area are screened, and the screened biodiversity data (biodiversity data of the observed area) are used to train a random forest model, and the environmental variable data of the unobserved area are predicted by the random forest model. After obtaining the biodiversity prediction data of the unobserved area, the biodiversity data of the unobserved area is obtained according to the biodiversity prediction data of the unobserved area and the corresponding biodiversity prediction residual. So far, both the biodiversity data of the observed area and the biodiversity prediction data of the unobserved area are obtained. It can be seen that the embodiment of the present application can predict the biodiversity of the unobserved area in the study area, and give the spatial distribution pattern of the biodiversity of the entire study area, which is conducive to the protection planning of the study area.

[0043] Exemplarily, the biodiversity prediction method provided in the embodiment of the present invention may be executed by an electronic device having a processing function, for example, the electronic device may be a computer, a server, etc. Taking the electronic device as a computer as an example, the hardware part of the computer may include: a processor, a memory, a network interface, a user interface, a communication bus, etc.

[0044] The processor is used to control the electronic device to perform related processing and computing tasks. The processor may include a central processing unit (CPU) or other processors. The processor may be single-core or multi-core. For example, the processor may include multiple CPUs.

[0045] The memory is used to store computer instructions and related data, for example, to store environmental variable data, biodiversity data, sample sets, biodiversity prediction data, and biodiversity prediction residuals. The memory can be a random access memory (RAM), a read only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, or an optical memory, a disk storage medium or other magnetic storage device, or any other medium that can be used to store program code or data that can be accessed by a computer. Optionally, the memory can be integrated into the processor, or the memory can be independent of the processor.

[0046] The network interface is used for the computer to communicate with other devices or communication networks. The network interface can be a transceiver with transceiver functions. Optionally, the network interface can include a standard wired interface, a wireless interface (such as a WI-FI interface, a Bluetooth interface, a 5G interface).

[0047] The communication bus is used to realize the connection and communication between various components. For example, the processor, memory, network interface and user interface mentioned above can be interconnected through the communication bus.

[0048] The user interface may include a display screen and an input unit (such as a keyboard). Optionally, the user interface may also include a standard wired interface and a wireless interface.

[0049] Those skilled in the art will appreciate that the above-mentioned computer may also include more or fewer components, or a combination of certain components, or different arrangements of components, which is not limited in the embodiments of the present application.

[0050] like Figure 1 As shown, the biodiversity prediction method provided in the embodiment of the present application includes S101-S104.

[0051] S101. Obtain environmental variable data and biodiversity data of the study area over a period of time.

[0052] Among them, biodiversity data include: species name, number of individuals and discovery location; environmental variable data include: climate factors, terrain factors, habitat factors and interference factors.

[0053] The above-mentioned organisms may be birds, mammals, reptiles or amphibians, and the embodiments of the present application are not limited thereto.

[0054] Optionally, specific parameters and acquisition methods of the above environmental variable data are shown in Table 1.

[0055] Table 1

[0056]

[0057]

[0058] In one application scenario, taking birds as an example, biodiversity data spontaneously observed by bird enthusiasts / workers can be obtained from the China Bird Watching Record Center.

[0059] S102. Based on species integrity and the end slope of the species accumulation curve, screen the biodiversity data and determine the sample set; in the study area, the part where the discovery location is located in the screened biodiversity data is the observed area, otherwise it is the unobserved area.

[0060] Optionally, combined Figure 1 ,like Figure 2 As shown, S102 includes S1021-S1024.

[0061] S1021. Divide the study area into multiple blocks.

[0062] S1022. For each block among the multiple blocks, the biodiversity data whose location falls within the geographical scope of the block is associated with the block; and the species completeness of the block and the end slope of the species accumulation curve are calculated through the biodiversity data associated with the block.

[0063] In one implementation, the calculation formula for species completeness is as follows;

[0064]

[0065] Among them, SI represents species integrity, Sobs represents the number of species observed in the sample, Chao 1 represents the richness index, n 1 represents the number of species containing only one individual, n 2 Represents the number of species containing only two individuals.

[0066] The end slope of the above-mentioned species accumulation curve is calculated by calculating the cumulative curve of the number of species in the block within a period of time. The specific calculation method is a common method in the technical field and will not be described in detail in the embodiments of the present application.

[0067] S1023. Take the biodiversity data in multiple blocks whose species integrity is greater than a first threshold and whose end slope of the species accumulation curve is less than a second threshold, and the environmental variable data corresponding to the biodiversity data as a sample set.

[0068] Optionally, the first threshold may be 0.85, the second threshold may be 1, and the values ​​of the first threshold and the second threshold may be selected within a reasonable range, which is not further limited in the embodiment of the present application.

[0069] S1024. Classify the blocks associated with the biodiversity data in the sample set into the observed area, and classify the remaining blocks in the study area into the unobserved area.

[0070] In one application scenario, taking birds as an example, in the biodiversity data spontaneously observed by the above-mentioned bird enthusiasts / workers, due to the small amount of observation data in some areas, the number of recorded bird species is far less than the actual number of species in the area. It is necessary to exclude the above-mentioned insufficiently observed biodiversity data through the above-mentioned screening method to ensure that the data in the sample set for training the random forest model is closer to the actual situation, thereby improving the prediction ability of the random forest model.

[0071] In the above application scenario, the study area is divided into multiple grids (i.e., the above blocks) to obtain a grid map of the study area. The collected environmental variable data are preprocessed to generate auxiliary data (projected into the same coordinate system, resampled to the same resolution, and cropped to the size of the study area). The processed environmental variable data are extracted into the grid at the corresponding position to obtain the environmental variable data corresponding to the location of each grid. The biodiversity data whose location falls within the geographical range of the block is associated with the grid. The sample set is then screened through the above S1022-S1024.

[0072] In one implementation, combining Figure 2 ,like Figure 3 As shown, before S103, the above method also includes: S105, training a random forest model.

[0073] Optionally, the training method of the random forest model is: training the random forest model through a training set; the input of the random forest model is environmental variable data, and the output of the random forest model is biodiversity prediction data; the sample set includes a training set and a test set. Exemplarily, 20% of the data in the above sample set can be used as a test set, and 80% of the data in the above sample set can be used as a training set. The sample set can also be divided according to other proportions, which is not limited in the embodiments of the present application.

[0074] In an application scenario, taking birds as an example, the training set includes biodiversity data and environmental variable data of an observed area over a period of time, and the biodiversity data and environmental variable data correspond to grids one by one.

[0075] In the above application scenario, the training process of the random forest model is as follows.

[0076] Step 1: Randomly select a subsample set.

[0077] A portion of samples is randomly selected with replacement from the training set (80% of the grid samples) multiple times to form multiple new sub-sample sets, each of which includes biodiversity data and environmental variable data of the observed area over a period of time.

[0078] Step 2: Randomly select a subset of features.

[0079] During the training of the decision tree, a feature subset is randomly selected with replacement from the total number of features in the subsample set. The size of the subset is smaller than the total number of features (for example, eight features such as temperature, precipitation, and vegetation index are selected as a feature subset from a total of nine environmental variables).

[0080] For example, for one of the feature subsets, multiple environmental variables (e.g., temperature, precipitation) are selected, and then data corresponding to multiple environmental variables in a subsample set are selected as samples (e.g., sample 1 has 100 species, 20 temperature, and 50 precipitation; sample 2 has 200 species, 10 temperature, and 30 precipitation). A decision tree is constructed and trained based on the samples.

[0081] Step 3: Train the decision tree.

[0082] Use the subsample set and feature subset selected in step 1 and step 2 to train a decision tree model. The decision tree model includes a root node and multiple internal nodes (also called decision nodes or split nodes) connected to the root node, and multiple child nodes connected to the multiple internal nodes. Among them, the root node includes all the data in the subsample set, the child nodes include the subset data selected based on the root node, and the leaf nodes include the prediction results of the decision tree. In the subsequent growth process of the decision tree, each child node will continue to select the best features from the remaining features for division in addition to the features already selected by its parent node (such as the second split: select seven out of eight features as a new feature subset) until the stopping condition is reached (such as all samples belong to the same category or the number of samples contained in the node is less than a certain threshold).

[0083] Specifically, during the construction of the decision tree, the algorithm will recursively select the optimal features for splitting, starting from the root node, and divide the data set into different subsets based on the selected features and the corresponding division criteria (such as information gain, Gini index, etc.), and correspond to different child nodes respectively. This process will continue, and new child nodes will continue to be generated until the stopping condition is met (such as the number of samples in the node is less than a certain threshold, the node purity reaches a certain standard, the depth of the tree reaches a preset maximum value, etc.). Finally, this node is regarded as a leaf node, and the node will determine a prediction result based on the sample data it contains. The prediction result is determined based on the statistics of all sample data in the node (such as the mean, median, etc.).

[0084] Step 4: Repeat steps 2 and 3.

[0085] Repeat steps 2 and 3 multiple times to generate multiple decision trees and obtain a random forest model.

[0086] Step 5: Regression prediction.

[0087] The random forest model averages the predictions of each tree to get the final prediction result.

[0088] For example, the above prediction results The following formula is satisfied.

[0089]

[0090] Where N is the number of decision trees in the random forest model, i is the number of decision trees, and h is the i (x) represents the output result of the decision tree numbered i.

[0091] In another implementation, combining Figure 3 ,like Figure 4 As shown, after S105 and before S103, the method further includes: S106, determining the biodiversity prediction residual of the study area.

[0092] Exemplary, combined Figure 4 ,like Figure 5 As shown, S106 includes S1061-S1062.

[0093] S1061. Testing the trained random forest model with a test set, and taking the difference between the actual biodiversity data in the test set and the biodiversity prediction data output by the trained random forest model as the prediction residual of the observed area.

[0094] S1062. Perform Kriging interpolation on the prediction residuals of the observed area to obtain the biodiversity prediction residuals of the study area.

[0095] It should be noted that interpolation refers to the use of the value of a function at a certain point to estimate the approximate value of a function at other points. There are various interpolation methods, including linear interpolation, polynomial interpolation, spline interpolation, etc. Each method has its specific application scenarios and advantages and disadvantages. Kriging interpolation is an interpolation method that takes into account the geographical location and spatial correlation of the data. Since the embodiment of the present application needs to take spatial correlation into account during the interpolation of the prediction residuals of the observed area, the biodiversity prediction residuals of the study area obtained by Kriging interpolation can be more in line with the actual situation of the study area, thereby improving the accuracy of the biodiversity prediction residuals of the study area.

[0096] It can be understood that Kriging interpolation is a spatial interpolation method that performs a weighted summation of a certain attribute of several discrete points in space (the prediction residual of the random forest model in the embodiment of the present application), thereby interpolating the prediction residual of an unobserved location at the location where the attribute is located. 0 ) satisfies the following formula.

[0097]

[0098] Among them, i represents the index variable, which is used to traverse all sampling points, n represents the number of sampling points, and λ i is the weight coefficient, which can satisfy the point (x0 ,y 0 ) is the residual estimate Z(x) 0 ) and the residual observation Z(x i ) is the optimal coefficient with the smallest difference.

[0099] The above point (x 0 ,y 0 ) is a coordinate point in a two-dimensional coordinate system set based on the grid map of the above-mentioned study area; the two-dimensional coordinate system includes x and y axes, and the origin of the two-dimensional coordinate system can be the center position of the grid map of the study area, or it can be other positions in the grid map of the study area. The embodiment of the present application does not make further limitations on the two-dimensional coordinate system.

[0100] S103. The random forest model predicts the environmental variable data of the area that has not been observed for a period of time, and obtains the biodiversity prediction data of the area that has not been observed for a period of time.

[0101] Among them, the random forest model is trained through the sample set.

[0102] In one application scenario, taking birds as an example, environmental variable data of an area that has not been observed for a period of time is input into multiple decision trees in a random forest model. Leaf nodes in multiple decision trees are matched according to the input environmental variable data to obtain multiple decision tree output results. The output results of multiple decision trees are averaged to finally obtain the prediction result of the random forest model, that is, the biodiversity prediction data of the area that has not been observed for a period of time.

[0103] S104. Determine the biodiversity data of an area that has not been observed for a period of time based on the biodiversity prediction data and the biodiversity prediction residuals corresponding to the biodiversity prediction data.

[0104] Among them, the biodiversity prediction residuals are obtained by interpolating the prediction residuals of the random forest model.

[0105] In the embodiment of the present application, the biodiversity data of the region not observed within a period of time is the sum of the biodiversity prediction data of the region not observed within a period of time and the biodiversity prediction residual.

[0106] In one application scenario, taking birds as an example, Figure 6 This is the final grid map of the spatial distribution pattern of the overall biodiversity in the study area. Figure 6 As shown, each grid corresponds to a color, and each color represents the number of bird species found in the grid.

[0107] In summary, in a biodiversity prediction method provided by an embodiment of the present application, the biodiversity data in the study area is screened, and the screened biodiversity data (biodiversity data in the observed area) is used to train a random forest model, and the environmental variable data of the unobserved area is predicted by the random forest model. After obtaining the biodiversity prediction data of the unobserved area, the biodiversity data of the unobserved area is obtained based on the biodiversity prediction data of the unobserved area and the corresponding biodiversity prediction residual. At this point, both the biodiversity data of the observed area and the biodiversity prediction data of the unobserved area are obtained. It can be seen that the embodiment of the present application can predict the biodiversity of the unobserved area in the study area, and provide the spatial distribution pattern of the biodiversity of the study area as a whole, which is beneficial to the protection planning of the study area.

[0108] In one application scenario, the random forest model was trained and tested in three grid maps of the study area with three resolutions (grid sizes of 10km×10km, 50km×50km, and 100km×100km, respectively). In each case, the training and testing process of the random forest model is as follows.

[0109] The sample set was divided into a training set and a test set in an 8:2 ratio. After the random forest model was obtained by training the training set, the difference between the biodiversity data (i.e., the number of species) in the training set and the prediction results of the random forest model was used as the prediction residual of the observed area, and the prediction residual of the unobserved area was obtained by Kriging interpolation, thereby obtaining the prediction residual of the entire study area.

[0110] Afterwards, the environmental variable data in the test set were input into the random forest model, and the sum of the prediction results of the random forest model and the corresponding prediction residuals were subjected to Pearson correlation analysis with the biodiversity data (i.e., the number of species) corresponding to the environmental variable data in the test set.

[0111] At the same time, corresponding to each of the above three situations, the Maxent distribution model and the traditional expert distribution surface overlay method were used to perform biodiversity deduction at the same resolution and in the same area (i.e., predict the number of species in the unobserved area in the test set based on the data in the training set), and perform Pearson correlation analysis with the corresponding biodiversity data (i.e., the number of species) in the test set.

[0112] It should be noted that the Maxent species distribution model is a species distribution prediction model based on the maximum entropy principle. It analyzes the known distribution data and environmental variables of species and uses the maximum entropy principle to infer the distribution probability of species in unknown areas. In this case, the Maxent species distribution model is run through Java software. Based on the biodiversity data of a single organism (for example, a certain bird) in a certain grid in the training set, combined with the environmental variable information contained in its location, the distribution of the organism in the unknown area of ​​the study area is predicted, and finally a simulated distribution grid map of the species is obtained. Then, by traversing the biodiversity data of each organism in the training set, the number of species in each grid in the grid map of the study area can be predicted. The traditional expert distribution surface is based on the existing bird distribution range delineated by experts, made into a grid, and the number of species is directly superimposed.

[0113] The results of the biodiversity data obtained by combining the random forest model provided in the embodiment of the present application with Kriging interpolation, the Maxent distribution model and the traditional expert distribution surface overlay method and the actual situation are shown in the following Tables 1, 2 and 3.

[0114] Table 1

[0115]

[0116] Table 2

[0117]

[0118] Table 3

[0119]

[0120] It can be understood that the higher the correlation obtained by Pearson correlation analysis, that is, the closer the predicted value is to the true value, the higher the accuracy of the model. Generally speaking, correlation |r|>0.7 is a strong correlation, 0.4<|r|≤0.7 is a moderate correlation, and |r|≤0.4 is a weak correlation.

[0121] Tables 1, 2 and 3 above show the correlation results between the biodiversity data and the actual situation obtained by combining the random forest model provided in the embodiment of the present application with Kriging interpolation, the Maxent distribution model and the traditional expert distribution surface overlay method when the resolution of the grid map of the study area is 10km, 50km and 100km (i.e., the grid sizes are 10km×10km, 50km×50km and 100km×100km, respectively).

[0122] It can be seen from Tables 1, 2 and 3 that when the grid map of the study area is of the above three resolutions, the correlation between the final prediction results of the random forest model combined with the Kriging interpolation method provided in this application and the true value is higher than that of the traditional expert distribution surface overlay method. At a resolution of 10km×10km, the correlation of the Maxent distribution model is higher than that of the random forest model combined with the Kriging interpolation method. At a resolution of 50km×50km and 100km×100km, the correlation of the Maxent distribution model is lower than that of the random forest model combined with the Kriging interpolation method. It can be seen that at a resolution of 50km×50km and 100km×100km, the final prediction results of the random forest model combined with the Kriging interpolation method provided in this application are closer to the actual situation, which proves that the random forest model provided in this application combined with the Kriging interpolation method has the highest accuracy.

[0123] On this basis, since the training process of the random forest model is to optimize the model performance by building multiple decision trees and integrating their prediction results. The construction process of each decision tree is independent, and the diversity of the model is increased by randomly selecting features and sample data. During the training process, the random forest will constantly try different feature combinations and splitting methods to find the best prediction model. This training method enables the random forest to capture the complex nonlinear relationships in the data. In contrast, the linear regression method describes the relationship between features and targets by fitting a straight line (or plane, hyperplane). If the relationship between the distribution of biodiversity and the independent variables is nonlinear, or there are complex interactions between the independent variables, the linear regression method may not be able to accurately fit this relationship, resulting in inaccurate prediction results.

[0124] At the same time, the random forest model effectively improves the generalization ability of the random forest model by constructing multiple decision trees and integrating their prediction results. In particular, compared with the existing linear regression method, the random forest model provided by this application combined with the Kriging interpolation method is more sensitive to outliers, and thus can provide more accurate predictions even when faced with complex natural phenomena such as biodiversity distribution.

[0125] Accordingly, the present application embodiment provides a biodiversity prediction device, such as Figure 7 As shown, it includes an acquisition module 501 , a screening module 502 , a prediction module 503 and a determination module 504 .

[0126] The acquisition module 501 is used to acquire environmental variable data and biodiversity data of the research area within a period of time; the biodiversity data includes: species name, number of individuals and discovery location. For example, the acquisition module 501 is used to implement S101 of the above biodiversity prediction method.

[0127] The screening module 502 is used to screen the biodiversity data and determine the sample set based on species integrity and the end slope of the species accumulation curve; in the study area, the part where the discovery location in the screened biodiversity data falls is the observed area, otherwise it is the unobserved area. For example, the screening module 502 is used to implement S102 of the above-mentioned biodiversity prediction method.

[0128] The prediction module 503 is used to predict the environmental variable data of the unobserved area for a period of time using a random forest model to obtain the biodiversity prediction data of the unobserved area for a period of time; the random forest model is obtained by training the sample set. For example, the prediction module 503 is used to implement S103 of the above biodiversity prediction method.

[0129] The determination module 504 is used to determine the biodiversity data of the unobserved area within a period of time based on the biodiversity prediction data and the biodiversity prediction residuals corresponding to the biodiversity prediction data; the biodiversity prediction residuals are obtained by interpolating the prediction residuals of the random forest model. For example, the determination module 504 is used to implement S104 of the above-mentioned biodiversity prediction method.

[0130] Optionally, the screening module 502 is specifically used to: divide the study area into multiple blocks. For each of the multiple blocks, the biodiversity data whose location falls within the geographical scope of the block is associated with the block; and the species integrity of the block and the end slope of the species accumulation curve are calculated through the biodiversity data associated with the block. Blocks in the multiple blocks whose species integrity is greater than a first threshold and whose end slope of the species accumulation curve is less than a second threshold are classified as observed areas, and the remaining blocks in the multiple blocks are classified as unobserved areas. The biodiversity data associated with the blocks classified as observed areas and the environmental variable data corresponding to the biodiversity data are used as a sample set. For example, the screening module 502 is specifically used to implement S1021-S1024 of the above-mentioned biodiversity prediction method.

[0131] Optionally, the biodiversity prediction device further includes a training module 505. The training module 505 is used to train a random forest model. For example, the training module 505 is used to implement S105 of the biodiversity prediction method.

[0132] In one application scenario, the training module 505 is specifically used to: train a random forest model through a training set; the input of the random forest model is environmental variable data, and the output of the random forest model is biodiversity prediction data; the sample set includes a training set and a test set.

[0133] Optionally, the biodiversity prediction device further comprises a residual determination module 506. The residual determination module 506 is used to determine the biodiversity prediction residual of the study area. For example, the residual determination module 506 is used to implement S106 of the biodiversity prediction method.

[0134] In one application scenario, the residual determination module 506 is specifically used to: test the trained random forest model through a test set, and use the difference between the actual biodiversity data in the test set and the biodiversity prediction data output by the trained random forest model as the prediction residual of the observed area; the sample set includes the training set and the test set. The prediction residual of the observed area is interpolated to obtain the biodiversity prediction residual of the study area. For example, the residual determination module 506 is specifically used to implement S1061-S1062 of the above-mentioned biodiversity prediction method.

[0135] Each module of the above-mentioned biodiversity prediction device can also be used to execute other steps in the above-mentioned method embodiment. All relevant contents involved in the above-mentioned method embodiment can be referred to the functional description of the corresponding functional module, which will not be repeated here.

[0136] The embodiment of the present application also provides an electronic device, including: a processor and a memory coupled to the processor; the memory is used to store computer instructions, and when the electronic device is running, the processor executes the computer instructions stored in the memory, so that the electronic device executes the method in the above embodiment. The processor can implement the above acquisition module 501, screening module 502, prediction module 503 and determination module 504; the above memory can also be used to store environmental variable data, biodiversity data, sample sets, biodiversity prediction data, and biodiversity prediction residuals.

[0137] An embodiment of the present application further provides a computer-readable storage medium, which includes a computer program. When the computer program runs on a computer, the method in the above embodiment is executed.

[0138] An embodiment of the present application further provides a computer program product, which includes computer program instructions. When the computer program instructions are run on a computer, the method in the above embodiment is executed.

[0139] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0140] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A biodiversity prediction method, characterized in that: include: Obtain data on environmental variables and biodiversity in the study area over a period of time; The biodiversity data include: species name, number of individuals and discovery location; the environmental variable data include: climate factors, terrain factors, habitat factors and interference factors; Based on species integrity and the end slope of the species accumulation curve, the biodiversity data is screened and a sample set is determined; in the research area, the part where the discovery location in the screened biodiversity data is located is the observed area, otherwise it is the unobserved area; The random forest model predicts the environmental variable data of the unobserved area within the period of time to obtain the biodiversity prediction data of the unobserved area within the period of time; the random forest model is obtained by training the sample set; The biodiversity data of the unobserved area within the period of time is determined according to the biodiversity prediction data and the biodiversity prediction residuals corresponding to the biodiversity prediction data; the biodiversity prediction residuals are obtained by interpolating the prediction residuals of the random forest model.

2. The method according to claim 1, characterized in that The method of screening the biodiversity data and determining a sample set of a random forest model based on species integrity and the end slope of the species accumulation curve includes: Dividing the study area into a plurality of blocks; For each of the plurality of blocks, associating the biodiversity data whose location falls within the geographical range of the block with the block; and calculating the species completeness and the end slope of the species accumulation curve of the block through the biodiversity data associated with the block; The biodiversity data in the plurality of blocks, in which the species integrity is greater than a first threshold and the slope of the end of the species accumulation curve is less than a second threshold, and the environmental variable data corresponding to the biodiversity data are used as the sample set; The blocks associated with the biodiversity data in the sample set are classified into the observed area, and the remaining blocks in the study area are classified into the unobserved area.

3. The method according to claim 1 or 2, characterized in that The calculation formula for the species integrity is as follows; Among them, SI represents species completeness, Sobs represents the number of species observed in the sample, Chao1 represents the richness index, n1 represents the number of species containing only one individual, and n2 represents the number of species containing only two individuals; The end slope of the species accumulation curve is calculated by using the species quantity accumulation curve in the block within the period of time.

4. The method according to claim 1, characterized in that The training method of the random forest model is: The random forest model is trained by the training set; the input of the random forest model is environmental variable data, and the output of the random forest model is biodiversity prediction data; the sample set includes the training set and the test set.

5. The method according to claim 1, characterized in that The method for determining the biodiversity prediction residual is: The trained random forest model is tested by the test set, and the difference between the actual biodiversity data in the test set and the biodiversity prediction data output by the trained random forest model is used as the prediction residual of the observed area; the sample set includes the training set and the test set; Kriging interpolation is performed on the prediction residuals of the observed area to obtain the biodiversity prediction residuals of the study area.

6. The method according to claim 1, characterized in that The organisms include birds.

7. A biodiversity prediction device, characterized in that: It includes an acquisition module, a screening module, a prediction module and a determination module; The acquisition module is used to acquire environmental variable data and biodiversity data of a research area within a period of time; the biodiversity data includes: species name, number of individuals and discovery location; The screening module is used to screen the biodiversity data and determine the sample set based on species integrity and the end slope of the species accumulation curve; in the research area, the part where the discovery position in the screened biodiversity data falls is the observed area, otherwise it is the unobserved area; The prediction module is used for predicting the environmental variable data of the unobserved area within the period of time using the random forest model to obtain the biodiversity prediction data of the unobserved area within the period of time; the random forest model is obtained by training the sample set; The determination module is used to determine the biodiversity data of the unobserved area within the period of time according to the biodiversity prediction data and the biodiversity prediction residuals corresponding to the biodiversity prediction data; the biodiversity prediction residuals are obtained by interpolating the prediction residuals of the random forest model.

8. An electronic device, characterized in that: It comprises a processor and a memory coupled to the processor; the memory is used to store computer instructions, and when the electronic device is running, the processor executes the computer instructions stored in the memory, so that the electronic device executes the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The method comprises computer program instructions, which, when executed by a computer, cause the computer to perform the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The method comprises computer program instructions, and when the computer program instructions are executed on a computer, the computer is caused to execute the method according to any one of claims 1 to 6.

Images