Method for predicting potential distribution of birds in urban center
By combining circuit theory and geographically weighted regression model with the PLUS model, the problem of predicting the impact of dynamic changes in urban landscape on bird distribution was solved, and high-precision bird distribution prediction was achieved, supporting urban ecological protection and green space planning.
Patent Information
- Application Number
- CN202510946157.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-09
- Publication Date
- 2025-10-17
AI Technical Summary
Existing technologies struggle to accurately predict the nonlinear and spatially heterogeneous impacts of rapid and dynamic changes in urban landscapes on bird distribution. Insufficient landscape connectivity indicators, weak integration of future scenarios, and neglect of spatial heterogeneity all contribute to inaccurate bird distribution predictions.
The functional connectivity index was calculated using a circuit theory model, and combined with geographically weighted regression and the PLUS land use simulation model, a spatial relationship model of bird distribution was constructed to predict the potential distribution of birds in urban central areas in the future.
It achieves high-precision, spatially explicit predictions of future bird distribution, improves the model's explanatory power and prediction accuracy, and supports urban ecological protection planning and green space system optimization.
Smart Images
Figure CN120806251A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of urban ecology, biogeography and urban planning, and particularly relates to a potential distribution prediction method for urban central city birds. BACKGROUND
[0002] Urban birds are important indicator species of urban ecosystem health, and their distribution is deeply affected by land use / cover change in the process of urbanization. Accurate prediction of their future distribution is crucial for urban biodiversity protection and green space planning.
[0003] However, the existing prediction methods have the following limitations:
[0004] 1. Static model: Traditional species distribution models (such as MaxEnt) are mostly based on static environmental variables, which cannot effectively capture the nonlinear, spatial heterogeneity effects of rapid changes in urban landscape (especially habitat fragmentation and connectivity changes) on bird distribution.
[0005] 2. Insufficient landscape connectivity indicators: Common landscape indices (such as patch area, proximity index) have limitations in describing the actual movement and diffusion ability of species (especially birds) in complex urban matrices. The functional connectivity of circuit theory can more realistically simulate the "flow" resistance of species in heterogeneous landscapes.
[0006] 3. Weak integration of future scenarios: Existing prediction methods rarely systematically incorporate high-precision, spatially explicit future land use change scenarios (such as urban expansion, green space layout changes) as core driving factors into bird distribution prediction models.
[0007] 4. Ignoring spatial heterogeneity: Global models (such as ordinary least squares regression) assume that variable relationships are constant in space, and cannot reflect the differences in the effects of environmental factors on bird distribution in different regions within the city. SUMMARY
[0008] The purpose of the present application is to overcome the shortcomings of existing methods and provide a potential distribution prediction method for urban central city birds, which more accurately depicts the spatial heterogeneity effects of urban landscape dynamics (especially green space pattern and connectivity changes) on bird distribution. Achieve high-precision, spatially explicit prediction of the potential distribution of target birds in urban central city at a specific future period (such as 2035). Provide scientific basis and decision support for urban ecological protection planning, green space system optimization and biodiversity improvement.
[0009] To solve the above technical problems, the technical solution adopted by the present application is:
[0010] A potential distribution prediction method for urban central city birds is provided, comprising the following steps:
[0011] S1. obtaining land use data of two historical periods of a target urban central city and target bird distribution data;
[0012] S2. extracting green space data based on the historical land use data, and calculating a functional connectivity index grid of the two historical periods by applying a circuit theory model;
[0013] S3. spatially matching the target bird distribution data of the two historical periods with the functional connectivity index of the corresponding period, and constructing a functional connectivity-bird distribution spatial relationship model by using a geographic weighted regression (GWR);
[0014] S4. predicting a land use scenario map of a target year by using a future land use change simulation model;
[0015] S5. extracting green space data based on the land use scenario map of the target year obtained in step S4, and calculating a functional connectivity index grid of the target year by applying the same circuit theory model parameters as in step S2;
[0016] S6. inputting the functional connectivity index grid of the target year obtained in step S5 into the functional connectivity-bird distribution spatial relationship model constructed in step S3, and inversely calculating a potential distribution map of the target bird in the urban central city of the target year.
[0017] Further, the functional connectivity index is at least one of current density, cumulative current or point-to-point connectivity calculated by the circuit theory model.
[0018] Further, one or more environmental covariates are added when constructing the functional connectivity-bird distribution spatial relationship model in step S3, and the environmental covariates include one or more of distance from road, distance from water body, building density, elevation and slope.
[0019] Further, the future land use change simulation model in step S4 is a PLUS model.
[0020] Further, when applying the circuit theory model in steps S2 and S5, the constructed resistance surface is given a low resistance value for green space and a high resistance value for construction land.
[0021] Further, when constructing the model by using GWR in step S3, the AICc criterion is used to optimize the bandwidth.
[0022] Further, the potential distribution map obtained in step S6 is a bird habitat suitability map.
[0023] Compared with the prior art, the present application has the following beneficial effects:
[0024] 1. Innovative Integration: The innovative fusion of circuit theory, functional connectivity, geographically weighted regression (GWR), and the PLUS land use simulation model addresses the challenges of modeling urban landscape dynamics and spatial heterogeneity in bird distribution.
[0025] 2. Clearer physical mechanism: Using circuit theory to simulate the "flow" of birds in urban landscapes can better reflect the actual diffusion process and habitat connectivity effectiveness than traditional landscape indices.
[0026] 3. Capturing spatial heterogeneity: The GWR model effectively reveals the spatial non-stationarity of the impact of factors such as functional connectivity on bird distribution (i.e., the degree of influence varies in different regions), significantly improving the model's explanatory power and prediction accuracy.
[0027] 4. Future scenario-driven: Closely integrated with high-precision, spatially explicit future land use simulation results, the prediction results are made more forward-looking and practical, directly serving the future urban ecological network planning.
[0028] 5. Strong operability: The method has a clear process and can be implemented based on remote sensing, GIS and ecological model software, making it easy to promote and apply to different cities and species.
[0029] 6. High decision-making support value: The prediction results can be directly used to identify key areas for urban biodiversity conservation, optimize green space layout and ecological network construction, and enhance urban ecological resilience. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] Figure 1 The figure is an overall technical flow chart of the method of the present invention.
[0031] Figure 2 This is an example map of land use in the historical period (t1, t2).
[0032] Figure 3 An example diagram of the calculation results of functional connectivity (current density) for the historical period (t1, t2).
[0033] Figure 4 This is an example diagram of GWR model results.
[0034] Figure 5 This is an example of the land use scenario in 2035 simulated and predicted by the PLUS model.
[0035] Figure 6 This is an example diagram of the calculation results of functional connectivity (current density) in 2035.
[0036] Figure 7 This is an example map of the predicted potential distribution (suitability) of target birds in 2035. DETAILED DESCRIPTION
[0037] The application will be further described below in conjunction with the specific embodiments. Among them, the drawings are only used for exemplary description, and the representation is only a schematic diagram, not a physical diagram, and cannot be understood as a limitation on the patent; in order to better illustrate the embodiments of the application, some components of the drawings will be omitted, enlarged or reduced, and do not represent the size of the actual product; for those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings can be omitted.
[0038] The same or similar reference numerals in the drawings of the embodiments of the application correspond to the same or similar components; in the description of the application, it should be understood that if the orientation or positional relationship indicated by the terms "upper", "lower", "left", "right" and the like is based on the orientation or positional relationship shown in the drawings, it is only for the convenience of describing the application and simplifying the description, and does not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, therefore the terms describing the positional relationship in the drawings are only used for exemplary description, and cannot be understood as a limitation on the patent, for those skilled in the art, the specific meaning of the above terms can be understood according to the specific situation.
[0039] The embodiments of the application disclose a potential distribution prediction method for birds in urban central city areas. Referring to Figures 1-7 A potential distribution prediction method for birds in urban central city areas, comprising the following steps:
[0040] S1. Data collection and preprocessing.
[0041] Among them, the historical period data: collect the data of the target urban central city area in two historical periods (such as 2013, 2023).
[0042] Bird distribution data: obtain the occurrence point data or abundance data of the target bird through field investigation, citizen science database (such as eBird) or literature data. Spatialization processing (such as generating point layer or density grid) is carried out.
[0043] Land use / cover data (LUC_t1, LUC_t2): obtain high-resolution (such as ≤30 meters) land use / cover classification map corresponding to the period. Extract green land (such as park, woodland, grassland, wetland) as core habitat patch.
[0044] Environmental variable data: collect auxiliary environmental variables (optional, but recommended to include, such as distance from road, distance from water body, building density, elevation, slope, etc.) that may affect the distribution of birds, with the same spatial resolution as the LUC data.
[0045] Future driving factor data: collect scenario data (such as urban development plan, traffic planning map, DEM, slope, restricted development area, etc.) for driving PLUS model simulation to predict land use in 2035.
[0046] S2. Historical period functional connectivity calculation.
[0047] Based on the green space data of the two historical periods extracted in S1.
[0048] Apply circuit theory model (using software Circuitscape or Linkage Mapper) to calculate the functional connectivity of target bird species in t1 and t2 periods.
[0049] Core definition: Define eligible green space patches as "source" patches (e.g., area greater than minimum threshold).
[0050] Resistance surface construction: Construct landscape resistance surface based on LUC_t1 and LUC_t2. Typically, green space is assigned low resistance (easy to pass), construction land is assigned high resistance (difficult to pass), and other land types are assigned intermediate resistance values. Resistance values can be set according to literature or expert knowledge, which is a potential innovation point / adjustable parameter.
[0051] Functional connectivity index calculation: Calculate the circuit theory connectivity index of each grid cell or key area, such as:
[0052] Current density: Measures the importance of the pixel in maintaining overall landscape connectivity.
[0053] Point-to-point connectivity: Identifies important corridors or bottleneck areas connecting key source patches.
[0054] Cumulative current: Similar to current density, reflects connectivity contribution. Select one of the core indicators (e.g., current density) as the independent variable (X_FC) for subsequent modeling.
[0055] Output functional connectivity index raster maps for t1 and t2 periods (FC_t1, FC_t2).
[0056] S3. Build functional connectivity-bird distribution relationship model.
[0057] Data matching: Match bird distribution data in S1 with functional connectivity index in corresponding periods in S2 in spatial location. If bird data is point data, extract the functional connectivity value at the location of the point; if it is raster data, perform spatial overlay.
[0058] Geographically weighted regression modeling:
[0059] Model construction: where (u, v) is the spatial coordinate, and are local regression coefficients that vary with geographic location, and ε is the error term. This is the core innovation point, which uses GWR to capture spatial heterogeneity.
[0060] Optional: Important environmental variables can be added as covariates to the GWR model to improve the model's explanatory power and prediction accuracy.
[0061] Model fitting and validation: The data from period t1 and period t2 were used to fit the model. The accuracy of the model was evaluated using cross-validation and other methods (e.g. ). A single-period model with higher accuracy is preferred, or two-period data are considered to be combined (if the trend is stable) to construct the final relationship model (Model_FC_Bird). The core output of this model is the regression coefficient surface of spatial variation ( , ...).
[0062] S4. Predict the land use scenario in 2035 (PLUS model).
[0063] The PLUS model is used to predict the land use scenario of the central urban area of the target city in 2035.
[0064] Drivers: Input the future driver data collected in S1.
[0065] Conversion rules and restricted areas: Based on historical land use change patterns, development plans, and policies, conversion rules between land use types and restricted development areas are set. This is a key parameter setting step.
[0066] Development probability calculation: Use the land expansion analysis strategy module to extract the driving factors of expansion of each land use type.
[0067] CA model based on multi-class random patch seeds: combining development probability, conversion rules, neighborhood effects and random factors to simulate and generate the spatial distribution map of land use in 2035.
[0068] Output a high-resolution land use / cover raster map for 2035.
[0069] S5. Calculate functional connectivity in 2035.
[0070] Extract the green space data for 2035 from LUC_2035 output from S4.
[0071] Use the same circuit theory model parameter settings as S2 (core definition standard, resistance surface assignment scheme, connectivity index calculation method).
[0072] Calculate the functional connectivity index raster map (FC_2035) for the year 2035. Ensuring comparability between historical and future connectivity calculations is key.
[0073] S6. Inversely predict the potential distribution of birds in 2035.
[0074] Input the FC_2035 raster map obtained in S5 into the final functional connectivity-bird distribution relationship model (Model_FC_Bird) constructed in S3.
[0075] Spatial inversion calculation: using the local coefficients of the GWR model and the value of FC_2035 at each spatial location, the suitability of the target bird in 2035 (Y_pred_2035) at this location is calculated.
[0076] Formula: Y_pred_2035(u, v) = +....
[0077] Output the potential distribution map of the target bird in 2035 (suitability raster map).
[0078] S7. Result analysis and application.
[0079] Analyze the spatial pattern, hotspot area and change trend (compared with the historical period) of the potential distribution of birds in 2035.
[0080] Identify the key protection area (high suitability area), potential corridor (based on functional connectivity and predicted distribution) and restoration area (low connectivity and low suitability but potential area).
[0081] Apply the prediction results to urban green space system planning, ecological network construction, biodiversity protection strategy formulation, etc.
[0082] Obviously, the above embodiments of the present application are only examples for clearly illustrating the present application, and are not intended to limit the embodiments of the present application. Based on the above description, other different forms of changes or variations can be made by those skilled in the art. It is not necessary or possible to exhaust all the embodiments. Any modification, equivalent replacement and improvement made within the spirit and principle of the present application shall be included in the protection scope of the claims of the present application.
Claims
1. A method for predicting the potential distribution of birds in a city center, characterized by: The following steps are involved: S1. Obtain land use data and target bird distribution data for two historical periods in the central urban area of the target city; S2. extracting green space data based on the historical land use data, and applying a circuit theory model to calculate a functional connectivity index grid map for two historical periods; S3. Spatially match the target bird distribution data of the two historical periods with the functional connectivity indicators of the corresponding periods, and construct a functional connectivity-bird distribution spatial relationship model using geographically weighted regression (GWR); S4. Use the future land use change simulation model to predict the land use scenario map for the target year; S5. Extracting green space data based on the target year land use scenario map obtained in step S4, and calculating the functional connectivity index grid map for the target year using the same circuit theory model parameters as in step S2; S6. Input the functional connectivity index grid map of the target year obtained in step S5 into the functional connectivity-bird distribution spatial relationship model constructed in step S3, and perform inverse calculation to obtain the potential distribution map of the target birds in the city center in the target year.
2. The method for predicting the potential distribution of birds in a city center according to claim 1, characterized in that: The functional connectivity indicator is at least one of current density, cumulative current or point-to-point connectivity calculated by a circuit theory model.
3. A method for predicting the potential distribution of birds in a city center according to claim 1 or 2, characterized in that: When constructing the functional connectivity-bird distribution spatial relationship model in step S3, one or more environmental covariates are also added, and the environmental covariates include one or more of distance to roads, distance to water bodies, building density, altitude, and slope.
4. The method for predicting the potential distribution of birds in a city center according to claim 1, characterized in that: In step S4, the future land use change simulation model is the PLUS model.
5. The method for predicting the potential distribution of birds in a city center according to claim 1, characterized in that: When the circuit theory model is applied in step S2 and step S5, the resistance surface constructed assigns a low resistance value to the green land and a high resistance value to the construction land.
6. The method for predicting the potential distribution of birds in a city center according to claim 1, characterized in that: When constructing the model using GWR in step S3, the bandwidth is optimized using the AICc criterion.
7. The method for predicting the potential distribution of birds in a city center according to claim 1, characterized in that: The potential distribution map obtained in step S6 is a bird habitat suitability map.