Method and device for simulating suitable habitat of species in future climate

Abstract

The invention relates to a species suitable habitat simulation method and device under future climate, and the method comprises the steps: obtaining remote sensing data, foundation observation data and species distribution data in a target region, and carrying out the unified preprocessing of the data, and obtaining processed data; constructing environment characteristic variables based on the processed data, and performing screening to obtain a training data set meeting species suitability modeling input conditions; and constructing a species suitability model under the current climate condition by using the training data set, predicting species suitable habitat distribution under a future climate scene by using the species suitability model, and analyzing a spatial change trend according to the species suitable habitat distribution. Therefore, the problems that remote sensing and foundation observation data are isomerism in the aspects of time, space, scale and the like, an effective data fusion and index conversion mechanism is lacked, registration errors and feature inconsistency are easily caused, and the accuracy and reliability of ecological parameter estimation are affected are solved.

Description

Technical Field

[0001] The present application relates to the technical field of ecological environment simulation, and in particular to a method and device for simulating suitable habitats for species under future climate conditions. Background Art

[0002] As global ecological and environmental changes become increasingly severe, obtaining large-scale, high-precision, and time-series continuous ecological information has become a key requirement for ecological monitoring, resource management, and environmental decision-making. Remote sensing technology, due to its advantages of large-scale coverage and rapid updates, has been widely used in monitoring ecological parameters such as vegetation cover, carbon storage, and water status. At the same time, ground-based ecological observation systems (such as eddy covariance flux towers, ecological plot surveys, and sensor networks) provide high-precision, long-term, and stable field measurement data, which are an important foundation for supporting the accuracy verification of remote sensing inversion and the study of ecological process mechanisms.

[0003] Among related technologies, the integration of multi-source remote sensing data (such as optical, radar, lidar, multispectral / hyperspectral, etc.) with ground-based ecological observations has become a research hotspot to enhance the comprehensiveness and accuracy of ecological and environmental monitoring. Multi-source remote sensing provides complementary information at varying temporal and spatial resolutions and sensor sensitivities, while ground-based observations provide precise point-level data support for remote sensing inversion modeling and verification. Multi-source data fusion facilitates the construction of integrated ecological information maps that are continuous across time and space, and across scales, enhancing the ability to monitor, model, and predict complex ecological processes.

[0004] However, in related technologies, remote sensing and ground-based data are heterogeneous in terms of time, space, scale and observation mechanism. There is a lack of efficient data fusion algorithms and unified ecological indicator conversion mechanisms, making it difficult to achieve effective matching and conversion of heterogeneous features between the two types of data. This leads to error propagation and fusion instability in data alignment, feature extraction and model coupling, thereby affecting the accuracy and reliability of ecological parameter estimation. It is difficult to meet the actual needs of high-precision monitoring and modeling of ecological processes, and this problem needs to be solved urgently. Summary of the Invention

[0005] The present application provides a method and device for simulating suitable habitats for species under future climate conditions, in order to solve the problem in related technologies that remote sensing and ground-based data are heterogeneous in time, space, scale and observation mechanism, making fusion difficult, which can easily lead to error propagation in data alignment, feature extraction and model coupling, unstable fusion and difficulty in converting ecological indicators, thereby affecting the accuracy and reliability of ecological parameter estimation.

[0006] The first aspect of the present application provides a method for simulating suitable habitats for species under future climate conditions, comprising the following steps: acquiring remote sensing data, ground-based observation data, and species distribution data in a target area, and uniformly preprocessing the remote sensing data, the ground-based observation data, and the species distribution data to obtain processed data; constructing environmental characteristic variables based on the processed data, and screening them to obtain a training data set that meets the input conditions for species suitability modeling; using the training data set to construct a species suitability model under current climate conditions, so as to use the species suitability model to predict the distribution of suitable habitats for species under future climate scenarios, and analyzing spatial change trends based on the distribution of suitable habitats for species.

[0007] Through the above technical means, remote sensing data, ground-based observation data and species distribution data are integrated to construct a species suitability model under current climate conditions, and then the distribution of suitable habitats for species is predicted under future climate scenarios and spatial change trend analysis is carried out. This can fully reveal the response mechanism of species to climate change and the dynamic evolution characteristics of habitats, effectively improve the prediction accuracy and ecological representativeness of the model, help identify key habitat expansion or contraction areas, scientifically assess ecological risks, support the delineation of biodiversity conservation priority areas and the formulation of ecological management strategies, and thus provide a solid scientific basis and decision-making support for ecological protection and resource management.

[0008] Optionally, in one embodiment of the present application, the remote sensing data, ground-based observation data and species distribution data in the target area are acquired, and the remote sensing data, the ground-based observation data and the species distribution data are uniformly preprocessed to obtain processed data, including: acquiring the remote sensing data of the target area, and performing projection conversion, spatial resampling and time matching on the remote sensing data to construct a basic raster dataset; collecting the ground-based observation data, and performing outlier removal, time series interpolation and raster reconstruction on the ground-based observation data to obtain data with enhanced spatiotemporal continuity; extracting the occurrence point information of the target species, screening out records with abnormal geographic coordinates, missing collection time or unclear species identification to establish a positive sample set, and constructing a pseudo-negative sample point set according to the spatial uniformity or ecological distance principle in areas where the species is not observed; spatially aligning the basic raster dataset, the positive sample set and the pseudo-negative sample point set, and rasterizing them at the target spatial resolution, and cropping them to the target study area to form a data cube for modeling as the processed data.

[0009] Through the above technical means, by preprocessing and quality controlling the remote sensing data, ground-based observation data and species distribution data respectively, and combining them with a unified target spatial resolution for rasterization, a data cube with standardized structure, consistent time and space, and multi-variable fusion is constructed. This can effectively integrate multi-source heterogeneous data, improve data compatibility and integrity, and provide a high-quality, standardized input basis for species suitability modeling. It can improve the accuracy and stability of the model, enhance the model's response ability to complex ecological environments, support dynamic simulation and prediction under multiple time points and multiple climate scenarios, and thus promote the scientific nature and effectiveness of ecological protection decision-making.

[0010] Optionally, in one embodiment of the present application, the environmental characteristic variables are constructed based on the processed data, and are screened to obtain a training data set that meets the input conditions for species suitability modeling, including: extracting multiple characteristic variables and frequency characteristics of the environmental characteristic variables from the time series of the processed data; analyzing the linear correlation between the multiple characteristic variables according to the correlation coefficient matrix, calculating the variance inflation factor of each variable to determine collinearity, so as to eliminate redundant variables that meet preset conditions; and performing multiple rounds of screening based on the contribution to sample classification to determine the set of key environmental variables used for modeling and determine the training data set.

[0011] Through the above technical means, the variance inflation factor of each variable is used to detect and eliminate redundant variables with strong collinearity. At the same time, combined with multiple rounds of screening of the variable's contribution to sample classification, the key environmental variable set used for modeling is determined. This can effectively reduce multicollinearity between environmental variables, improve the independence and information content of variables, and screen out a set of key environmental variables with strong ecological representativeness and significant impact on model performance, optimize model input, improve the stability and prediction accuracy of the species suitability model, and enhance the model's generalization ability and interpretability, thereby providing solid data support for accurately simulating species distribution and guiding ecological protection.

[0012] Optionally, in one embodiment of the present application, the use of the training data set to construct a species suitability model under current climatic conditions includes: extracting the corresponding environmental variable values ​​for all positive and negative sample points in the training data set to construct a training data set with a feature-label structure; based on the training data set with the feature-label structure, using a preset model and a random forest model (a non-parametric classifier based on ensemble learning) to perform modeling respectively, and tuning the model parameters until the accuracy and stability of the model meet the preset training conditions to construct the species suitability model.

[0013] Through the above technical means, the preset model and the random forest model (a non-parametric classifier based on ensemble learning) are used to build models respectively, and the model parameters are tuned to construct the species suitability model. This can give full play to the advantages of different models, improve the accuracy and stability of the species suitability model, enhance the model's adaptability to complex ecological data, improve the accuracy of capturing species distribution patterns, reduce the risk of overfitting, and achieve more generalized prediction effects, thereby providing a reliable scientific basis and technical support for species protection planning and ecological environmental management.

[0014] Optionally, in one embodiment of the present application, the species suitability model is used to predict the distribution of species suitable habitats under future climate scenarios, and spatial change trends are analyzed based on the distribution of species suitable habitats, including: while keeping other environmental variables unchanged, replacing the climate variables used in the species suitability model with predicted climate variable data under future scenarios; inputting the predicted climate variable data under the future scenarios into the model to obtain species suitability probability maps under future scenarios at different time periods, and determining the distribution of species suitable habitats; comparing the current species suitability probability image output by the species suitability model with the species suitability probability maps under future scenarios at different time periods, and performing raster calculations to quantify the absolute change and relative change percentage of suitable habitat area, so as to analyze the spatial migration direction, concentration degree and fragmentation trend of the habitat, identify at least one of the potential expansion area, contraction area and future risk area of ​​the species, and determine the spatial change trend.

[0015] Through the above technical means, by comparing the current species suitability probability image output by the species suitability model and the species suitability probability map under different future scenarios, the spatial migration direction, concentration degree and fragmentation trend of the habitat are analyzed, and then the spatial change trend is determined. It can quantify the changes in the spatial structure of the habitat, help reveal the dynamic response characteristics of species to climate change, accurately identify potential habitat expansion or contraction areas, and assess habitat connectivity and ecological security risks, thereby supporting the scientific formulation of protection priority areas and ecological restoration strategies, and improving the foresight and pertinence of ecological management.

[0016] The second aspect of the present application provides a device for simulating suitable habitats for species under future climate conditions, including: a processing module for acquiring remote sensing data, ground-based observation data and species distribution data in a target area, and uniformly preprocessing the remote sensing data, the ground-based observation data and the species distribution data to obtain processed data; an acquisition module for constructing environmental characteristic variables based on the processed data, and screening them to obtain a training data set that meets the input conditions for species suitability modeling; a simulation module for using the training data set to construct a species suitability model under current climate conditions, so as to use the species suitability model to predict the distribution of suitable habitats for species under future climate scenarios, and analyze spatial change trends based on the distribution of suitable habitats for species.

[0017] Through the above technical means, remote sensing data, ground-based observation data and species distribution data are integrated to construct a species suitability model under current climate conditions, and then the distribution of suitable habitats for species is predicted under future climate scenarios and spatial change trend analysis is carried out. This can fully reveal the response mechanism of species to climate change and the dynamic evolution characteristics of habitats, effectively improve the prediction accuracy and ecological representativeness of the model, help identify key habitat expansion or contraction areas, scientifically assess ecological risks, support the delineation of biodiversity conservation priority areas and the formulation of ecological management strategies, and thus provide a solid scientific basis and decision-making support for ecological protection and resource management.

[0018] Optionally, in one embodiment of the present application, the processing module includes: a construction unit for acquiring remote sensing data of the target area, and performing projection conversion, spatial resampling and time matching on the remote sensing data to construct a basic raster dataset; an acquisition unit for collecting the ground-based observation data, and performing outlier removal, time series interpolation and raster reconstruction on the ground-based observation data to obtain data with enhanced spatiotemporal continuity; an establishment unit for extracting the occurrence point information of the target species, screening out records with abnormal geographic coordinates, missing collection time or unclear species identification to establish a positive sample set, and constructing a pseudo-negative sample point set according to the spatial uniformity or ecological distance principle in areas where the species is not observed; a processing unit for spatially aligning the basic raster dataset, the positive sample set, and the pseudo-negative sample point set, and performing rasterization processing at the target spatial resolution, and cropping to the target study area to form a data cube for modeling as the processed data.

[0019] Through the above technical means, the variance inflation factor of each variable is used to detect and eliminate redundant variables with strong collinearity. At the same time, combined with multiple rounds of screening of the variable's contribution to sample classification, the key environmental variable set used for modeling is determined. This can effectively reduce multicollinearity between environmental variables, improve the independence and information content of variables, and screen out a set of key environmental variables with strong ecological representativeness and significant impact on model performance, optimize model input, improve the stability and prediction accuracy of the species suitability model, and enhance the model's generalization ability and interpretability, thereby providing solid data support for accurately simulating species distribution and guiding ecological protection.

[0020] Optionally, in one embodiment of the present application, the acquisition module includes: an extraction unit for extracting multiple characteristic variables and frequency characteristics of the environmental characteristic variables from the time series of the processed data; a elimination unit for analyzing the linear correlation between the multiple characteristic variables according to the correlation coefficient matrix, and calculating the variance inflation factor of each variable to determine collinearity, so as to eliminate redundant variables that meet preset conditions; a first determination unit for performing multiple rounds of screening in combination with the contribution to sample classification, determining the set of key environmental variables for modeling, and determining the training data set.

[0021] Through the above technical means, the variance inflation factor of each variable is used to detect and eliminate redundant variables with strong collinearity. At the same time, combined with multiple rounds of screening of the variable's contribution to sample classification, the key environmental variable set used for modeling is determined. This can effectively reduce multicollinearity between environmental variables, improve the independence and information content of variables, and screen out a set of key environmental variables with strong ecological representativeness and significant impact on model performance, optimize model input, improve the stability and prediction accuracy of the species suitability model, and enhance the model's generalization ability and interpretability, thereby providing solid data support for accurately simulating species distribution and guiding ecological protection.

[0022] Optionally, in one embodiment of the present application, the simulation module includes: a first construction unit, used to extract the corresponding environmental variable values ​​of all positive and negative sample points in the training data set, and construct a training data set with a feature-label structure; a second construction unit, used to model the training data set based on the feature-label structure using a preset model and a random forest model (a non-parametric classifier based on ensemble learning), and to tune the model parameters until the accuracy and stability of the model meet the preset training conditions, thereby constructing the species suitability model.

[0023] Through the above technical means, the preset model and the random forest model (a non-parametric classifier based on ensemble learning) are used to build models respectively, and the model parameters are tuned to construct the species suitability model. This can give full play to the advantages of different models, improve the accuracy and stability of the species suitability model, enhance the model's adaptability to complex ecological data, improve the accuracy of capturing species distribution patterns, reduce the risk of overfitting, and achieve more generalized prediction effects, thereby providing a reliable scientific basis and technical support for species protection planning and ecological environmental management.

[0024] Optionally, in one embodiment of the present application, the simulation module includes: a replacement unit for replacing the climate variables used in the species suitability model with predicted climate variable data under future scenarios while keeping other environmental variables unchanged; a second determination unit for inputting the predicted climate variable data under the future scenarios into the model to obtain species suitability probability maps under future scenarios at different time periods and determine the distribution of suitable habitats for the species; a third determination unit for comparing the current species suitability probability image output by the species suitability model with the species suitability probability maps under future scenarios at different time periods, and performing raster calculations to quantify the absolute change and relative change percentage of the suitable habitat area, so as to analyze the spatial migration direction, concentration degree and fragmentation trend of the habitat, identify at least one of the potential expansion area, contraction area and future risk area of ​​the species, and determine the spatial change trend.

[0025] Through the above technical means, by comparing the current species suitability probability image output by the species suitability model and the species suitability probability map under different future scenarios, the spatial migration direction, concentration degree and fragmentation trend of the habitat are analyzed, and then the spatial change trend is determined. It can quantify the changes in the spatial structure of the habitat, help reveal the dynamic response characteristics of species to climate change, accurately identify potential habitat expansion or contraction areas, and assess habitat connectivity and ecological security risks, thereby supporting the scientific formulation of protection priority areas and ecological restoration strategies, and improving the foresight and pertinence of ecological management.

[0026] The third aspect of the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the program to implement the method for simulating suitable habitats for species under future climate as described in the above embodiment.

[0027] The fourth aspect of the present application provides a computer-readable storage medium, which stores a computer program. When the program is executed by a processor, it implements the above method for simulating suitable habitats for species under future climate.

[0028] The fifth embodiment of the present application provides a computer program product, including a computer program, which, when executed, is used to implement the above method for simulating suitable habitats for species under future climate.

[0029] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0031] Figure 1 This is a flow chart of a method for simulating suitable habitats for species under future climate conditions provided in accordance with an embodiment of the present application;

[0032] Figure 2 This is a flow chart of a method for simulating suitable habitats for species under future climate conditions according to one embodiment of the present application;

[0033] Figure 3 This is a block diagram of a device for simulating suitable habitats for species under future climate conditions according to an embodiment of the present application;

[0034] Figure 4 The figure is a schematic structural diagram of an electronic device provided according to an embodiment of the present application.

[0035] Reference numerals:

[0036] 10 - Device for simulating suitable habitats for species under future climate conditions; 100 - processing module, 200 - acquisition module, 300 - simulation module; 401 - memory, 402 - processor, 403 - communication interface. DETAILED DESCRIPTION

[0037] The following describes in detail embodiments of the present application, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0038] The following describes the method and device for simulating suitable habitats for species under future climate according to the embodiment of the present application with reference to the accompanying drawings. In view of the heterogeneity of remote sensing and ground-based data in terms of time, space, scale and observation mechanism mentioned in the above background technology, the lack of efficient data fusion algorithm and unified ecological indicator conversion mechanism, and the difficulty in achieving effective matching and conversion of heterogeneous features between the two types of data, resulting in error propagation and fusion instability in data registration, feature extraction and model coupling, thereby affecting the accuracy and reliability of ecological parameter estimation, and making it difficult to meet the actual needs of high-precision monitoring and modeling of ecological processes. The present application provides a method for simulating suitable habitats for species under future climate. In this method, a unified ecological system is constructed by processing and screening the acquired remote sensing data, ground-based observation data and species distribution data in the target area. A dataset is used to construct a species suitability model under current climate conditions, which is then replaced with future climate scenario data to predict the potential distribution of suitable habitats for species at different time points. This can effectively improve the ecological representativeness and spatiotemporal continuity of the variables used in the modeling, and enhance the model's ability to depict the dynamic response characteristics of species-environment relationships. At the same time, through multiple rounds of screening of key variables, the stability and generalization of the model can be improved, and suitability simulations for multiple climate scenarios and time points can be achieved. This can be used to support forward-looking analysis of habitat evolution trends and ecological risks, providing a scientific basis for national park planning, ecological red line delineation, and identification of key species protection priority areas, and has broad potential for promotion and application. This solves the problem of heterogeneity between remote sensing and ground-based observation data in terms of time, space, and scale, and the lack of effective data fusion and indicator conversion mechanisms, which can easily lead to registration errors and inconsistent features, thereby affecting the accuracy and reliability of ecological parameter estimates.

[0039] Specifically, Figure 1 A schematic flow chart of a method for simulating suitable habitats for species under future climate conditions provided in an embodiment of the present application.

[0040] like Figure 1 As shown in Figure 2, the method for simulating suitable habitats for species under future climate conditions includes the following steps:

[0041] In step S101 , remote sensing data, ground-based observation data and species distribution data in a target area are acquired, and the remote sensing data, ground-based observation data and species distribution data are uniformly preprocessed to obtain processed data.

[0042] As a possible implementation method, the embodiment of the present application obtains remote sensing image data of the target area, which may include but is not limited to terrain data (such as DEM (Digital Elevation Model)), current climate elements (such as annual average temperature, precipitation), future climate prediction data (such as CMIP6 (Coupled Model Intercomparison Project Phase 6, Phase 6 Coupled Model Intercomparison Program) / SSP2 (Shared Socioeconomic Pathway, Shared Socioeconomic Pathway) -4.5, SSP5-8.5 simulation results), vegetation indices (such as NDVI (Normalized Difference Vegetation Index, Normalized Vegetation Index, EVI (Enhanced Vegetation Index)), night light remote sensing images (such as VIIRS (Visible Infrared Imaging Radiometer Suite, Visible Infrared Imaging Radiometer Suite), DMSP / OLS (Defense Meteorological Satellite Program / Operational Linescan System, National Defense Meteorological Satellite Program / Operational Line Scan System); ground-based observation data, including but not limited to long-term monitoring data from meteorological stations, flux towers, national ecological stations and phenological observation points, such as temperature, precipitation, wind speed, humidity, NPP (Net Primary Productivity), SIF (Solar-Induced Chlorophyll Fluorescence), soil moisture, etc.; species distribution data, including but not limited to the occurrence point information of target species extracted from global or national species observation databases (such as GBIF (Global Biodiversity Information Facility), national biodiversity monitoring platforms, etc.).

[0043] It can be understood that the embodiment of the present application performs unified preprocessing of data to ensure the consistency of various types of data in spatial scale, time scale and coordinate projection system, thereby laying an accurate and reliable foundation for subsequent data fusion, model construction and result analysis.

[0044] Optionally, in one embodiment of the present application, remote sensing data, ground-based observation data and species distribution data in the target area are obtained, and the remote sensing data, ground-based observation data and species distribution data are uniformly preprocessed to obtain processed data, including: obtaining remote sensing data of the target area, and performing projection conversion, spatial resampling and time matching on the remote sensing data to construct a basic raster dataset; collecting ground-based observation data, and performing outlier removal, time series interpolation and raster reconstruction on the ground-based observation data to obtain data with enhanced spatiotemporal continuity; extracting the occurrence point information of the target species, screening out records with abnormal geographic coordinates, missing collection time or unclear species identification to establish a positive sample set, and constructing a pseudo-negative sample point set according to the spatial uniformity or ecological distance principle in areas where the species is not observed; spatially aligning the basic raster dataset, positive sample set and pseudo-negative sample point set, and rasterizing them at the target spatial resolution, and cropping them to the target study area to form a data cube for modeling as processed data.

[0045] Specifically, in the embodiments of the present application, for remote sensing data, a spatial resampling operation may be first performed to adjust data of different resolutions to a unified target spatial resolution (e.g., 250 m or 500 m). Common resampling methods include, but are not limited to, bilinear interpolation or nearest neighbor. The process may be as follows:

[0046]

[0047] Where Resampled Pixel Value represents the resampled pixel value; neighboring pixels represents the sum of the neighboring pixel values ​​used for calculation; and n represents the number of pixels in the neighborhood.

[0048] Subsequently, the embodiment of the present application can use a unified geographic reference system (such as WGS84 (World Geodetic System 1984)) to perform projection conversion on all image data to ensure that the remote sensing data is strictly aligned with the ground-based observation data in space.

[0049] Furthermore, for ground-based observation data, the present application embodiment can perform missing value filling and spatial interpolation. Among them, the commonly used method is Kriging interpolation, which is based on the principle of geographic proximity and spatial autocorrelation and can more accurately estimate the value of unobserved locations. Its mathematical expression can be shown as follows:

[0050] Z(x)=μ(x)+∑λ i [Z(x i )-μ(xi )],

[0051] Where Z(x) represents the estimated value at position x; λ i Represents the weight coefficient; Z(x i ) represents the value of a known sample point; μ(x i ) represents the spatial smoothing term; x represents the spatial coordinates of the position to be estimated; i represents the position coordinates of the known sample points.

[0052] During the actual implementation process, after the embodiment of the present application completes the fusion of the above-mentioned remote sensing and ground-based data, the two can also be spatially aligned and rasterized, that is, all data are rasterized into a unified grid system and cropped according to the boundaries of the study area, thereby forming a standardized data cube that can be used for subsequent analysis.

[0053] In step S102, environmental characteristic variables are constructed based on the processed data and screened to obtain a training data set that meets the input conditions for species suitability modeling.

[0054] It can be explained that, based on the above data collection and preprocessing results, the embodiment of the present application can extract environmental characteristic variables from the preprocessed time series, which may include but are not limited to annual maximum value, minimum value, mean, standard deviation, CV (Coefficient of Variation), interannual variation range (ΔY), seasonal range, length of green growth period, growth peak time, etc., and frequency characteristics can be extracted at the same time, including but not limited to high temperature frequency, drought frequency, vegetation fluctuation frequency, etc.

[0055] Optionally, in one embodiment of the present application, environmental characteristic variables are constructed based on the processed data and screened to obtain a training data set that meets the input conditions for species suitability modeling, including: extracting multiple characteristic variables and frequency characteristics of environmental characteristic variables from the time series of the processed data; analyzing the linear correlation between multiple characteristic variables according to the correlation coefficient matrix, calculating the variance inflation factor of each variable to determine collinearity, so as to eliminate redundant variables that meet the preset conditions; performing multiple rounds of screening based on the contribution to sample classification, determining the set of key environmental variables for modeling, and determining the training data set.

[0056] The preset condition may be that VIF (Variance Inflation Factor) exceeds 5, or that VIF exceeds 10, and may be set by those skilled in the art according to actual conditions, and is not specifically limited here.

[0057] It can be explained that the multi-round screening method may include but is not limited to random forest variable importance ranking, PCA (Principal Component Analysis, principal component analysis), etc.

[0058] Specifically, after completing the preprocessing of remote sensing and ground-based data, the embodiment of the present application can extract environmental characteristic variables from multi-source time series and screen out key variable sets through correlation analysis, which can be used as input for subsequent species suitability modeling.

[0059] As a specific example, first, for each time series variable, the embodiment of the present application can calculate its extreme values ​​(maximum value, minimum value), coefficient of variation (CV), inter-annual variation and other statistical characteristics. The calculation formula of the coefficient of variation is as follows:

[0060]

[0061] Among them, σ is the standard deviation, μ is the mean, and CV reflects the relative volatility of the variable on the time scale.

[0062] Furthermore, the interannual variation can be estimated by the absolute value of the mean difference between adjacent years, as shown below:

[0063]

[0064] in, It is the coefficient of determination of the regression model established by taking the jth variable as the dependent variable and the remaining variables as independent variables, where j represents the variable.

[0065] In the actual implementation process, if VIF j >10, it means that the variable has strong collinearity with other variables and should be eliminated.

[0066] In step S103, a species suitability model under current climate conditions is constructed using the training data set, so as to predict the distribution of species suitable habitats under future climate scenarios using the species suitability model, and analyze the spatial change trend based on the distribution of species suitable habitats.

[0067] Optionally, in one embodiment of the present application, a species suitability model under current climatic conditions is constructed using a training data set, including: extracting the corresponding environmental variable values ​​for all positive and negative sample points in the training data set to construct a training data set with a feature-label structure; using a preset model and a random forest model to model the training data set based on the feature-label structure, and tuning the model parameters until the accuracy and stability of the model meet the preset training conditions to construct a species suitability model.

[0068] The default model can be MaxEnt (Maximum Entropy Model). Based on the principle of maximum entropy, MaxEnt is applicable to situations where only records of species presence are available. It can infer the potential suitable distribution of species based on limited species observation data. The random forest model is a nonparametric classification method based on ensemble learning, with strong nonlinear modeling capabilities and variable interpretation capabilities.

[0069] During the modeling process of the embodiment of the present application, methods such as cross validation can be used to tune the model parameters to improve the generalization ability and stability of the model.

[0070] Furthermore, embodiments of the present application may employ machine learning algorithms to construct species suitability models, and may verify performance using commonly used evaluation indicators.

[0071] Specifically, the embodiment of the present application can use the maximum entropy model (MaxEnt) and random forest (RandomForest) algorithms for modeling. Among them, the maximum entropy model is based on the maximum entropy principle, and its construction should meet the probability distribution with the maximum entropy under the constraint conditions, and its form can be expressed as:

[0072]

[0073] Where P(x) is the probability of a species appearing at location x; w i is the variable weight; f i (x) is the environmental variable; e represents the base of the natural logarithm (Euler constant, approximately 2.718), which is used to construct an exponential function to improve the sensitivity of the model; i represents the index of the environmental variable, which is used to enumerate multiple environmental variables (such as temperature, precipitation, soil type, etc.); x represents the location coordinates in geographic space.

[0074] Furthermore, the embodiment of the present application can construct multiple decision trees for prediction through the random forest model, which can be shown as follows:

[0075]

[0076] Among them, y represents the prediction result; f(x i ) is the prediction result of the i-th tree; N is the number of trees in the forest; i represents the i-th tree.

[0077] In some cases, performance evaluation indicators may include but are not limited to the receiver operating characteristic AUC (Area Under the Curve), Kappa coefficient and TSS (True Skill Statistic), which can comprehensively evaluate the performance of each model in terms of accuracy, reliability and discrimination ability, and select the model with the best performance, so as to output the probability distribution image of species suitability under current climate conditions, and provide a high-quality model foundation for subsequent prediction of potential species habitats under future climate scenarios.

[0078] As a specific example, the embodiment of the present application can use AUC and TSS to evaluate the performance of the model. The specific calculation of the embodiment of the present application can be as follows:

[0079]

[0080] TSS=Sensitivity+Specificity-1,

[0081] Among them, TruePositive Rate (x) represents the proportion of samples that are correctly identified as positive among all actually positive samples; FalsePositive Rate (x) represents the proportion of samples that are incorrectly identified as positive among all actually negative samples; Sensitivity represents the ability of the model to identify positive samples; Specificity represents the ratio of samples that are correctly identified as negative among all actually negative samples; x represents the false positive rate.

[0082] It can be explained that the closer the AUC value is to 1, the better the model performance is, and the higher the TSS value is, the more accurate the model is.

[0083] Optionally, in one embodiment of the present application, a species suitability model is used to predict the distribution of suitable habitats for species under future climate scenarios, and spatial change trends are analyzed based on the distribution of suitable habitats for species, including: while keeping other environmental variables unchanged, replacing the climate variables used in the species suitability model with predicted climate variable data under future scenarios; inputting the predicted climate variable data under future scenarios into the model to obtain species suitability probability maps under future scenarios at different time periods, and determining the distribution of suitable habitats for species; comparing the current species suitability probability image output by the species suitability model with the species suitability probability map under future scenarios at different time periods, and performing raster calculations to quantify the absolute change and relative change percentage of suitable habitat area, so as to analyze the spatial migration direction, concentration degree and fragmentation trend of the habitat, identify at least one of the potential expansion area, contraction area and future risk area of ​​the species, and determine the spatial change trend.

[0084] Specifically, the embodiments of the present application can replace current climate variables with future climate forecast data to predict species suitability and quantify habitat changes.

[0085] For example, the embodiment of the present application can use climate data under future climate scenarios such as SSP2-4.5 and SSP5-8.5 to replace the climate variables in the current climate model, which can be specifically expressed as:

[0086] T future =T current +ΔT,

[0087] Among them, T future represents the predicted value of future climate; T current represents the current climate value; ΔT is the amount of climate change; T represents the climate value; future represents the future; current represents the present.

[0088] Furthermore, the embodiments of the present application can input future climate scenario data into the trained species suitability model to predict the distribution of suitable habitats for species at time points such as 2050 and 2100. By comparing the suitability images at different time points, the embodiments of the present application can calculate the area change of suitable habitats, which can be specifically expressed as:

[0089] ΔA=A future -A current ,

[0090] Where ΔA represents the change in the area of ​​suitable habitat; A future Indicates the area of ​​suitable habitat at a future time point, A current Indicates the area of ​​suitable habitat at the current time node.

[0091] like Figure 2 As shown, taking a specific example, the process of the method for simulating suitable habitats for species under future climate is explained in detail. The embodiment of the present application may specifically include:

[0092] (1) Fusion and preprocessing of remote sensing data and ground-based data.

[0093] The embodiment of the present application first collects multi-source ecological and environmental data to support species suitability modeling, including but not limited to remote sensing data from satellites such as MODIS (Moderate Resolution Imaging Spectroradiometer), Landsat (Landsat Satellite Program), and Sentinel-2 (Sentinel-2 Multispectral Instrument) (such as vegetation indices, night lights, digital elevation models, etc.); species occurrence information recorded by ecological monitoring stations (including but not limited to birds, mammals, amphibians, reptiles and plants); and ground-based observation data (such as temperature and precipitation data from meteorological stations and flux data from FLUXNET (The Global Network of Eddy Covariance Flux Towers)).

[0094] Furthermore, in the data preprocessing stage of the embodiment of the present application, remote sensing data can be resampled by methods such as bilinear interpolation or nearest neighbor interpolation to achieve unified spatial resolution and scale alignment; ground-based observation data can use the Kriging interpolation method to construct a continuous spatial layer.

[0095] The embodiments of the present application can establish an environmental variable dataset with good spatial consistency and ecological representativeness through the fusion and standardization of multi-source heterogeneous data, providing high-quality data support for subsequent species suitability modeling and habitat prediction under future climate scenarios.

[0096] (2) Construction and screening of environmental characteristic variables

[0097] In an embodiment of the present application, after completing the standardization processing of multi-source ecological and environmental data, environmental characteristic variables covering the target area are further extracted, including but not limited to multiple dimensions such as climate, hydrology, topography, vegetation, and human activity interference. The collinearity between variables is evaluated and eliminated through correlation analysis (such as VIF, etc.), and a set of key variables with low information redundancy and strong ecological representativeness is screened out as input for subsequent species suitability modeling.

[0098] (3) Construction and training of species suitability models

[0099] As a possible implementation method, the embodiment of the present application uses the maximum entropy model (MaxEnt) or the random forest model (Random Forest) to construct a species suitability prediction model based on the key environmental characteristic variables and species sample data obtained by screening, and uses the training set data for parameter fitting and model training. At the same time, the model structure and parameters are optimized by combining cross-validation and other methods to improve the stability and generalization ability of the model.

[0100] In the embodiments of the present application, after the model training is completed, a species suitability probability distribution map under current climate conditions can be output, providing key support for subsequent suitability simulation under different climate scenarios and time scales, and providing a scientific basis for ecological protection planning and habitat management.

[0101] (4) Prediction of species suitability and habitat quantification

[0102] Specifically, the embodiment of the present application is based on the trained species suitability model, and can use indicators such as AUC and TSS to comprehensively evaluate the model performance to ensure that it has high prediction accuracy and discrimination ability.

[0103] Furthermore, the embodiment of the present application extracts model parameters and structural information, and replaces environmental variables under current climate conditions with predicted climate data under different future climate scenarios (such as CMIP6 / SSP2-4.5, SSP5-8.5), driving the model to perform spatial simulation of species suitability at different time nodes in the future (such as 2030, 2050, 2100, etc.).

[0104] By comparing the suitability distribution patterns at different times and scenarios, the embodiments of the present application can predict the evolution trend of potential habitats of species, and provide a scientific basis for dynamic monitoring of ecosystems, identification of priority areas for habitat protection, and early warning of ecological risks.

[0105] According to the method for simulating suitable habitats for species under future climates proposed in the embodiments of the present application, remote sensing data, ground-based observation data and species distribution data of the target area are obtained. After processing and screening, a species suitability model under current climate conditions is constructed. On this basis, future climate scenario data are replaced to predict the potential suitable habitat distribution of species at different time nodes. This can effectively improve the ecological representativeness and spatiotemporal continuity of the variables used in modeling, and enhance the model's ability to depict the dynamic response characteristics in species-environment relationships. At the same time, through multiple rounds of screening of key variables, the stability and generalization ability of the model can be improved, and suitability simulation for multiple climate scenarios and time nodes can be achieved. It can be used to support forward-looking analysis of habitat evolution trends and ecological risks, provide a scientific basis for national park planning, ecological red line demarcation and identification of priority areas for key species protection, and has broad potential for promotion and application.

[0106] Next, a device for simulating suitable habitats for species under future climate conditions proposed in accordance with an embodiment of the present application will be described with reference to the accompanying drawings.

[0107] Figure 3 Schematic diagram of a device for simulating suitable habitats for species under future climate according to an embodiment of the present application.

[0108] like Figure 3 As shown, the device 10 for simulating suitable habitats for species under future climate conditions includes: a processing module 100 , an acquisition module 200 , and a simulation module 300 .

[0109] The processing module 100 is used to obtain remote sensing data, ground-based observation data, and species distribution data within the target area, and to pre-process the remote sensing data, ground-based observation data, and species distribution data to obtain processed data.

[0110] Acquisition module 200 is used to construct environmental characteristic variables based on the processed data and screen them to obtain a training data set that meets the input conditions for species suitability modeling.

[0111] The simulation module 300 is used to construct a species suitability model under current climate conditions using the training data set, so as to use the species suitability model to predict the distribution of species suitable habitats under future climate scenarios, and to analyze spatial change trends based on the distribution of species suitable habitats.

[0112] Optionally, in one embodiment of the present application, the processing module 100 includes: a construction unit, an acquisition unit, an establishment unit and a processing unit.

[0113] Among them, the construction unit is used to obtain the remote sensing data of the target area, and perform projection conversion, spatial resampling and time matching on the remote sensing data to construct a basic raster dataset.

[0114] The acquisition unit is used to collect ground-based observation data and perform outlier removal, time series interpolation and grid reconstruction on the ground-based observation data to obtain data with enhanced spatiotemporal continuity.

[0115] Establish units to extract the occurrence point information of target species, screen out records with abnormal geographic coordinates, missing collection time or unclear species identification to establish a positive sample set, and construct a pseudo-negative sample point set according to the spatial uniformity or ecological distance principle in areas where the species is not observed.

[0116] The processing unit is used to spatially align the basic raster dataset, positive sample set, and pseudo negative sample point set, perform rasterization processing at the target spatial resolution, and crop them to the target study area to form a data cube for modeling as processing data.

[0117] Optionally, in one embodiment of the present application, the acquisition module 200 includes: an extraction unit, a elimination unit, and a first determination unit.

[0118] The extraction unit is used to extract multiple characteristic variables and frequency characteristics of environmental characteristic variables from the time series of the processed data.

[0119] The elimination unit is used to analyze the linear correlation between multiple feature variables based on the correlation coefficient matrix, calculate the variance inflation factor of each variable to determine collinearity, and eliminate redundant variables that meet the preset conditions.

[0120] The first determination unit is used to perform multiple rounds of screening based on the contribution to sample classification, determine the set of key environmental variables used for modeling, and determine the training data set.

[0121] Optionally, in one embodiment of the present application, the simulation module 300 includes: a first construction unit and a second construction unit.

[0122] The first construction unit is used to extract the corresponding environmental variable values ​​of all positive and negative sample points in the training data set to construct a training data set with a feature-label structure.

[0123] The second construction unit is used to model the training data set based on the feature-label structure using a preset model and a random forest model (a non-parametric classifier based on ensemble learning), and to tune the model parameters until the accuracy and stability of the model meet the preset training conditions to construct a species suitability model.

[0124] Optionally, in one embodiment of the present application, the simulation module 300 includes: a replacement unit, a second determination unit and a third determination unit.

[0125] Among them, the replacement unit is used to replace the climate variables used in the species suitability model with the predicted climate variable data under future scenarios while keeping other environmental variables unchanged.

[0126] The second determination unit is used to input the predicted climate variable data under future scenarios into the model to obtain species suitability probability maps under future scenarios at different time periods and determine the distribution of suitable habitats for species.

[0127] The third determination unit is used to compare the current species suitability probability image output by the species suitability model with the species suitability probability map under future scenarios in different time periods, and perform raster calculations to quantify the absolute change and relative change percentage of suitable habitat area, so as to analyze the spatial migration direction, concentration and fragmentation trend of the habitat, identify at least one of the potential expansion area, contraction area and future risk area of ​​the species, and determine the spatial change trend.

[0128] It should be noted that the above explanations of the embodiment of the method for simulating suitable habitats for species under future climate are also applicable to the device for simulating suitable habitats for species under future climate of this embodiment, and will not be repeated here.

[0129] According to the device for simulating suitable habitats for species under future climates proposed in the embodiment of the present application, remote sensing data, ground-based observation data and species distribution data of the target area are obtained. After processing and screening, a species suitability model under current climate conditions is constructed. On this basis, future climate scenario data are replaced to predict the potential suitable habitat distribution of species at different time nodes. This can effectively improve the ecological representativeness and spatiotemporal continuity of the variables used in modeling, and enhance the model's ability to depict the dynamic response characteristics in species-environment relationships. At the same time, through multiple rounds of screening of key variables, the stability and generalization ability of the model can be improved, and suitability simulation for multiple climate scenarios and time nodes can be achieved. It can be used to support forward-looking analysis of habitat evolution trends and ecological risks, provide a scientific basis for national park planning, ecological red line demarcation and identification of priority areas for key species protection, and has broad potential for promotion and application.

[0130] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include:

[0131] Memory 401 , processor 402 , and computer programs stored in the memory 401 and executable on the processor 402 .

[0132] When the processor 402 executes the program, the method for simulating suitable habitats for species under future climate conditions provided in the above embodiment is implemented.

[0133] Furthermore, the electronic device further includes:

[0134] The communication interface 403 is used for communication between the memory 401 and the processor 402 .

[0135] The memory 401 is used to store computer programs that can be run on the processor 402 .

[0136] The memory 401 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0137] If the memory 401, the processor 402, and the communication interface 403 are implemented independently, the communication interface 403, the memory 401, and the processor 402 can be connected to each other via a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0138] Optionally, in a specific implementation, if the memory 401 , the processor 402 and the communication interface 403 are integrated on a chip, the memory 401 , the processor 402 and the communication interface 403 can communicate with each other through an internal interface.

[0139] The processor 402 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0140] This embodiment also provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the method for simulating suitable habitats for species under future climate conditions as described above is implemented.

[0141] An embodiment of the present application also provides a computer program product, including a computer program, which can run computer instructions. When the computer instructions are executed by a processor, the method for simulating suitable habitats for species under future climate provided in an embodiment of the present application is implemented.

[0142] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0143] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of technical features indicated. Thus, a feature specified as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of this application, "N" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0144] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing a custom logical function or process step, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may be performed in a different order than shown or discussed, including performing functions in a substantially simultaneous manner or in a reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present application pertain.

[0145] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or N wires (electronic devices), a portable computer disk cartridge (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program can be obtained electronically by optically scanning the paper or other medium and then editing, interpreting or processing it in other suitable ways as necessary, and then storing it in a computer memory.

[0146] It should be understood that various parts of the present application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiment, the N steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. If implemented using hardware, as in another embodiment, it can be implemented using any one or a combination of the following technologies known in the art: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0147] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0148] In addition, the functional units in the various embodiments of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into a module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0149] The storage medium mentioned above may be a read-only memory, a magnetic disk, or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application. Persons skilled in the art may make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A method for simulating suitable habitats for species under future climate conditions, characterized in that: The following steps are involved: Acquiring remote sensing data, ground-based observation data, and species distribution data within a target area, and uniformly preprocessing the remote sensing data, the ground-based observation data, and the species distribution data to obtain processed data; constructing environmental characteristic variables based on the processed data and screening them to obtain a training data set that meets the input conditions for species suitability modeling; The training dataset is used to construct a species suitability model under current climate conditions, so as to use the species suitability model to predict the distribution of species suitable habitats under future climate scenarios, and to analyze spatial change trends based on the distribution of species suitable habitats.

2. The method according to claim 1, characterized in that The acquiring of remote sensing data, ground-based observation data, and species distribution data within the target area, and uniformly preprocessing the remote sensing data, the ground-based observation data, and the species distribution data to obtain processed data, includes: Acquire remote sensing data of the target area, and perform projection conversion, spatial resampling and time matching on the remote sensing data to construct a basic raster dataset; Collecting the ground-based observation data, and performing outlier removal, time series interpolation, and grid reconstruction on the ground-based observation data to obtain data with enhanced spatiotemporal continuity; Extract the occurrence point information of the target species, screen out records with abnormal geographic coordinates, missing collection time, or unclear species identification to establish a positive sample set, and construct a pseudo-negative sample point set based on spatial uniformity or ecological distance principles in areas where the species is not observed; The basic raster dataset, the positive sample set, and the pseudo-negative sample point set are spatially aligned, rasterized using a target spatial resolution, and cropped to the extent of a target study area to form a data cube for modeling as the processed data.

3. The method according to claim 1, characterized in that The environmental characteristic variables are constructed based on the processed data, and are screened to obtain a training data set that meets the input conditions for species suitability modeling, including: extracting a plurality of characteristic variables and frequency characteristics of the environmental characteristic variables from the time series of the processed data; Analyze the linear correlation between the multiple feature variables according to the correlation coefficient matrix, calculate the variance inflation factor of each variable to determine collinearity, and eliminate redundant variables that meet the preset conditions; Multiple rounds of screening are performed based on the contribution to sample classification to determine a set of key environmental variables for modeling and to determine the training data set.

4. The method according to claim 1, wherein The method of constructing a species suitability model under current climate conditions using the training data set includes: Extracting the corresponding environmental variable values ​​for all positive and negative sample points in the training data set to construct a training data set with a feature-label structure; The training data set based on the feature-label structure is modeled using a preset model and a random forest model respectively, and the model parameters are tuned until the accuracy and stability of the model meet the preset training conditions to construct the species suitability model.

5. The method according to claim 1, wherein The method of using the species suitability model to predict the distribution of suitable habitats for species under future climate scenarios and analyzing spatial variation trends based on the distribution of suitable habitats for species includes: Under the premise of keeping other environmental variables unchanged, the climate variables used in the species suitability model are replaced with the predicted climate variable data under the future scenario; Inputting the predicted climate variable data under the future scenario into the model to obtain species suitability probability maps under future scenarios at different time periods and determine the distribution of suitable habitats for the species; Compare the current species suitability probability image output by the species suitability model with the species suitability probability map under different future scenarios, and perform raster calculations to quantify the absolute change and relative change percentage of suitable habitat area, so as to analyze the spatial migration direction, concentration and fragmentation trend of the habitat, identify at least one of the potential expansion area, contraction area and future risk area of ​​the species, and determine the spatial change trend.

6. A device for simulating suitable habitats for species under future climate conditions, characterized in that: include: a processing module for acquiring remote sensing data, ground-based observation data, and species distribution data within a target area, and performing unified preprocessing on the remote sensing data, the ground-based observation data, and the species distribution data to obtain processed data; An acquisition module is used to construct environmental characteristic variables based on the processed data and screen them to obtain a training data set that meets the input conditions for species suitability modeling; A simulation module is used to use the training data set to construct a species suitability model under current climate conditions, to use the species suitability model to predict the distribution of species suitable habitats under future climate scenarios, and to analyze spatial change trends based on the distribution of species suitable habitats.

7. The device according to claim 6, characterized in that The processing module includes: A construction unit is used to obtain remote sensing data of the target area, and perform projection conversion, spatial resampling and time matching on the remote sensing data to construct a basic raster dataset; An acquisition unit, configured to collect the ground-based observation data, and perform outlier removal, time series interpolation, and grid reconstruction on the ground-based observation data to obtain data with enhanced spatiotemporal continuity; Establish units to extract the occurrence point information of target species, screen out records with abnormal geographic coordinates, missing collection time, or unclear species identification to establish a positive sample set, and construct a pseudo-negative sample point set based on spatial uniformity or ecological distance principles in areas where the species is not observed; A processing unit is used to spatially align the basic raster dataset, the positive sample set, and the pseudo-negative sample point set, perform rasterization processing using a target spatial resolution, and crop to the target study area to form a data cube for modeling as the processed data.

8. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for simulating suitable habitats for species under future climate as described in any one of claims 1 to 5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the method for simulating suitable habitats for species under future climate as described in any one of claims 1 to 5.

10. A computer program product comprising a computer program, characterized in that The computer program is executed to implement the method for simulating suitable habitats for species under future climate according to any one of claims 1 to 5.

Images