Site distribution analysis system based on natural update limitation of davidia involucrata population

Abstract

The invention relates to the technical field of species distribution analysis, and discloses a dove tree population natural update limitation-based species distribution analysis system, which comprises a habitat factor acquisition module, an update limitation evaluation module, a distribution prediction module, a spatial constraint correction module and an adaptability atlas optimization module. The habitat factor acquisition module acquires data in real time and constructs a multi-layer environment factor topology; the update limit evaluation module evaluates update limit in combination with genetic diversity and seed diffusion constraint; the distribution prediction module generates a distribution strategy through spatial modeling and ecological niche optimization; the spatial constraint correction module corrects a diffusion range and genetic parameters based on terrain and climate factors; and the adaptability graph optimization module feeds back an analysis result to carry out closed-loop optimization on the strategy. The system is provided with a multi-stage obstacle avoidance mechanism and a redundant link, so that the habitat change response capability is improved. According to the method, accurate analysis and dynamic optimization of davidia involucrata distribution are realized, and technical support is provided for species protection.

Description

Technical Field

[0001] The invention relates to the technical field of species distribution analysis, in particular to a species distribution analysis system based on natural regeneration restrictions of Davidia involucrata populations. Background Art

[0002] As a nationally protected wild plant, Davidia involucrata is known as a "living fossil." Research on its natural regeneration and distribution patterns is crucial for biodiversity conservation. However, Davidia involucrata populations currently face the severe challenge of limited natural regeneration, making scientific analysis of their distribution particularly urgent.

[0003] From a natural environmental perspective, Davidia involucrata requires stringent habitat conditions for growth. Soil physical and chemical properties, such as fertility and pH, directly influence seed germination and seedling growth. Light competition gradients are also crucial, as varying light conditions can alter the growth and regeneration capacity of populations. Furthermore, differences in altitude lead to significant variations in microhabitats. Climatic factors such as temperature, humidity, and wind speed vary at different altitudes, forming a complex habitat subnetwork that influences the spread and establishment of Davidia involucrata populations.

[0004] Davidia involucrata faces numerous limitations in its population reproduction and spread. Seed dispersal is hindered by geographical barriers such as mountains and rivers, limiting its breeding range. Insufficient genetic diversity can reduce the population's ability to adapt to environmental changes, making it susceptible to regeneration problems in the face of climate fluctuations. Furthermore, human interference, such as land development and tourism, can damage Davidia involucrata's habitat, increasing the risk of disrupting its dispersal pathways and further complicating its natural regeneration.

[0005] Existing species distribution analysis methods have obvious shortcomings when dealing with the particularity of Davidia involucrata populations. Traditional methods are often unable to comprehensively and in real time collect and analyze multi-dimensional habitat factors, and it is difficult to construct an accurate topological structure of environmental factors to reflect complex microhabitat changes. When evaluating population renewal restrictions, there is a lack of comprehensive consideration of genetic diversity and seed dispersal constraints, and it is impossible to dynamically evaluate the impact of reproductive barriers on distribution. In terms of distribution prediction, most existing models do not fully consider dynamic factors such as terrain isolation, climate fluctuations, and human interference, resulting in low accuracy and reliability of prediction results, and unable to provide scientific and effective guidance for the protection and management of Davidia involucrata populations.

[0006] Therefore, there is an urgent need for a system that can integrate multi-source data based on the characteristics of natural regeneration limitations of Davidia involucrata populations, realize real-time collection of habitat factors, comprehensive assessment of regeneration limitations, and dynamic prediction of distribution, so as to solve the problems existing in existing technologies and provide strong technical support for the protection and population restoration of Davidia involucrata. Summary of the Invention

[0007] The purpose of the present invention is to provide a species distribution analysis system based on the natural regeneration restriction of Davidia involucrata population, so as to solve the problems raised in the above background technology.

[0008] To achieve the above object, the present invention provides the following technical solution: a species distribution analysis system based on natural regeneration restrictions of Davidia involucrata populations, the system comprising: Habitat factor collection module, renewal restriction assessment module and distribution prediction module; The habitat factor acquisition module includes a population dynamics monitoring node and a microhabitat parameter network; the renewal restriction assessment module includes a genetic diversity analysis unit and a seed diffusion constraint unit; the distribution prediction module includes a spatial modeling node and an ecological niche optimization unit; the population dynamics monitoring node is used to collect the spatial coordinate data of the Davidia involucrata mother tree and the survival status parameters of the seedlings in real time; according to the soil physical and chemical indicators and the light competition gradient, a multi-layer environmental factor topology is constructed through the microhabitat parameter network; the microhabitat parameter network is composed of N habitat subnetworks based on altitude thresholds in parallel; based on the real-time ecological response characteristics of the N habitat subnetworks, a potential distribution benchmark strategy is generated through the spatial modeling node, and combined with the reproductive obstacle signal output by the seed diffusion constraint unit, the species diffusion path is dynamically probabilistically sorted through the ecological niche optimization unit.

[0009] Preferably, the system further comprises a spatial constraint correction module and an adaptive map optimization module; the spatial constraint correction module comprises a terrain isolation unit and a climate fluctuation compensation unit; According to the generated potential distribution benchmark strategy and dynamic diffusion probability ranking results, the geographical barrier of the Davidia involucrata population diffusion range is corrected through the terrain isolation unit, and the genetic parameters of the renewal restriction assessment module are dynamically corrected through the climate fluctuation compensation unit; at the same time, the seedling settlement obstacle rate is analyzed in real time through the reproduction success rate assessment network constructed by the adaptive map optimization module, and the analysis results are fed back to the spatial modeling node and the niche optimization unit to perform closed-loop correction on the distribution strategy.

[0010] Preferably, the terrain isolation unit is composed of diffusion path circuits corresponding to habitat subnetworks connected in parallel in the microhabitat parameter network; the steps of constructing the multi-layer environmental factor topology include: Collect the mother tree coordinate data and seedling survival rate fluctuation signal, input them into the maximum entropy model configured in each level of habitat subnet, generate the habitat suitability compensation coefficient and store it in the corresponding habitat subnet; Based on the suitability compensation coefficients stored in habitat subnetworks at all levels and combined with the instantaneous change rate of climatic factors, the population diffusion probability is distributed to N habitat subnetworks through the habitat suitability index algorithm to form a dynamic habitat matching topology.

[0011] Preferably, the habitat subnet construction conditions configured in the population dynamics monitoring node include: the mother tree coordinates are within a preset altitude range, the seed germination rate corresponding to the habitat subnet is greater than a threshold, and the light competition gradient is less than the allowable deviation range; The genetic diversity analysis unit is also configured with a multi-scale switching link; the multi-scale switching link includes a steady-state diffusion sub-link and an obstacle avoidance sub-link; the steady-state diffusion sub-link is generated based on the matching degree between the reproduction success rate signal of the update restriction evaluation module and the potential distribution benchmark strategy; the obstacle avoidance sub-link is constructed based on the correlation between the instantaneous rupture risk of the population diffusion path and human interference events.

[0012] Preferably, the spatial modeling node is configured with a multi-source data fusion model; the multi-source data fusion model includes a soil-climate mapping table corresponding to the habitat factor acquisition module and an obstacle avoidance parameter library of the update restriction assessment module; the step of generating the potential distribution benchmark strategy includes: Based on the real-time soil data from the habitat factor acquisition module and the diffusion probability ranking results, the theoretical diffusion range upper limit of each level of habitat subnetwork in the steady-state diffusion sub-link is calculated; The upper limit of the theoretical diffusion range and transient climate fluctuations are input into the multi-source data fusion model to generate a species distribution reference map, and the reference map is corrected for fracture risk through the obstacle avoidance sub-link to form a potential distribution benchmark strategy.

[0013] Preferably, the parameter updating step of the multi-source data fusion model includes: When it is detected that the coordinates of the mother tree are beyond the preset altitude range or the barrier rate of seedling settlement exceeds the threshold, the first obstacle avoidance instruction is triggered, that is, the alternative habitat unit is activated and the analysis granularity of the update restriction assessment module is adjusted; If the distribution range still does not return to the allowed range after the execution of the first obstacle avoidance instruction, the second obstacle avoidance instruction is triggered, that is, switching to the obstacle avoidance sub-link through the multi-scale switching link, and redistributing the topological weight of the population diffusion path based on the duration of the human interference event.

[0014] Preferably, the parameter updating step of the multi-source data fusion model further includes: When the seed germination rate of a certain level of habitat subnetwork in the habitat factor acquisition module is lower than the threshold, the third obstacle avoidance instruction is triggered, that is, the diffusion output channel of the subnetwork is blocked, and the corresponding population is transferred to other habitat subnetworks through the terrain isolation unit; During the execution of the third obstacle avoidance instruction, if other habitat subnetworks are detected to be overloaded, the fourth obstacle avoidance instruction is triggered, that is, the adaptive graph optimization module is called to downgrade the output strategy of the niche optimization unit and limit the diffusion demand of the edge habitat.

[0015] Preferably, the degradation processing logic of the adaptive graph optimization module includes: Based on the obstacle rate data and population genetic diversity indicators output by the reproductive success rate assessment network, a habitat diffusion safety level table is constructed; When the fourth obstacle avoidance instruction is triggered, the diffusion path priority is dynamically downgraded according to the safety level table, and the analysis frequency of the update restriction evaluation module is synchronously adjusted to match the downgraded distribution strategy.

[0016] Preferably, the step of constructing the habitat subnet further includes: Based on the spatial clustering results of the mother tree coordinate data and the historical fluctuation range of seedling survival rates, an altitude gradient adaptability index table was generated; According to the differences in suitability compensation coefficients of each habitat subnetwork in the index table, the diffusion path circuits of N habitat subnetworks are topologically weighted using the geographically weighted regression algorithm, and redundant connection links are established between habitat subnetworks.

[0017] Preferably, the process of generating the species distribution reference map includes: After inputting the theoretical diffusion range upper limit and transient climate fluctuations into the multi-source data fusion model, the preset ecological corridor database and human activity interference layer were simultaneously loaded; Based on the allele frequency data output by the genetic diversity analysis unit, the break threshold of the population diffusion path in the reference map was corrected, and the terrain roughness parameter was superimposed on the obstacle avoidance sub-link for secondary calibration.

[0018] Compared with the prior art, the present invention has the following beneficial effects: The system implements multi-dimensional dynamic monitoring of Davidia involucrata habitats through a habitat factor collection module. Population dynamics monitoring nodes collect real-time spatial coordinates of mother trees and seedling survival status. A microhabitat parameter network constructs a multi-layered environmental factor topology based on altitude thresholds, enabling detailed characterization of the interactions of soil, light, and other factors across different altitude gradients. This multi-node, multi-subnetwork monitoring architecture captures instantaneous changes in habitat factors in real time, providing high-density, high-precision data support for subsequent analysis and addressing the limitations of traditional monitoring methods, which suffer from single and delayed data.

[0019] The Regeneration Constraint Assessment Module provides a three-dimensional diagnosis of regeneration barriers in Davidia involucrata populations. The Genetic Diversity Analysis Unit, equipped with a multi-scale switching link, dynamically switches analysis modes based on reproductive success and diffusion risk. This allows both assessment of genetic stability under steady-state conditions and rapid identification of risk sources when barriers emerge. The Seed Dispersal Constraint Unit, linked to the Spatial Modeling Node, converts reproductive barrier signals into probabilistic rankings of diffusion paths. This allows the system to quantify regeneration constraints from both genetic and physical diffusion perspectives, transcending the limitations of traditional assessment methods that focus on a single factor.

[0020] The distribution prediction module achieves dynamic and accurate predictions through spatial modeling and niche optimization. A multi-source data fusion model integrates soil-climate mapping tables and a barrier avoidance parameter library, generating a reference map based on theoretical diffusion limits and climate fluctuations. It then uses barrier avoidance sub-links to correct for disruption risks, ensuring that predictions are both consistent with ecological laws and resilient to sudden disturbances. The niche optimization unit dynamically ranks diffusion paths and adjusts distribution strategies based on real-time data, significantly improving the timeliness and adaptability of predictions and providing a more forward-looking basis for protected area planning.

[0021] The Spatial Constraint Correction Module and the Adaptive Spectrum Optimization Module form a closed-loop feedback mechanism. The Terrain Isolation Unit corrects geographic barriers based on the diffusion path circuits of the habitat subnetwork, while the Climate Fluctuation Compensation Unit dynamically adjusts genetic parameters. Together, these two enable the system to adapt to the dual constraints of terrain and climate. The Adaptive Spectrum Optimization Module analyzes settlement barriers in real time through a reproductive success assessment network and feeds the results back to the modeling and optimization units, forming a closed "monitoring-analysis-correction" loop. This ensures that the distribution strategy continues to meet the actual needs of the population, improving the system's robustness and long-term reliability.

[0022] The system's multi-level obstacle avoidance command mechanism enhances its ability to respond to unexpected risks. Four levels of obstacle avoidance commands cover a variety of abnormal scenarios, including exceeding altitude limits, exceeding settlement obstacle rates, low seed germination rates, and habitat subnet overload. Through tiered response measures, including activating alternative habitat units, switching analysis links, relocating populations, and degrading diffusion strategies, a multi-layered risk prevention and control system is formed. This mechanism enables the system to rapidly adjust to uncertainties such as human interference and natural disasters, maximizing the survival and spread of Davidia involucrata populations and demonstrating the technology's intelligent and user-friendly design.

[0023] Furthermore, the system's innovative design in topology construction and parameter updating further enhances analytical performance. The habitat subnetwork utilizes spatial clustering and a geographically weighted regression algorithm to generate an adaptive index table and assign topological weights, establishing redundant connections and enhancing the system's modeling capabilities and fault tolerance for complex habitats. The dynamic parameter update mechanism of the multi-source data fusion model automatically adjusts the analysis granularity and topological weights based on real-time monitoring data, ensuring the system maintains optimal operation and laying a solid foundation for long-term population dynamics research. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a working principle diagram of the species distribution analysis system based on natural regeneration restriction of Davidia involucrata populations according to the present invention; Figure 2 This is the design diagram of the topological structure of the microhabitat parameter network; Figure 3 A diagram showing how spatial modeling nodes work; Figure 4Flowchart for multi-scale handover link control; Figure 5 Logic diagram for degradation processing of adaptive graph optimization module. DETAILED DESCRIPTION

[0025] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0026] See also Figure 1-Figure 5 The present invention relates to a species distribution analysis system based on natural regeneration restrictions of Davidia involucrata populations, and the specific implementation steps are as follows: The system includes a habitat factor collection module, a regeneration constraint assessment module, and a distribution prediction module. Within the habitat factor collection module, a population dynamics monitoring node collects real-time spatial coordinate data of Davidia involucrata mother trees and seedling survival parameters. A microhabitat parameter network constructs a multi-layered environmental factor topology based on soil physical and chemical parameters (such as pH and nutrient content) and light competition gradients (such as canopy density and solar radiation intensity). This microhabitat parameter network is composed of N parallel habitat subnetworks based on altitude thresholds (e.g., each 100-meter interval represents an altitude interval), each corresponding to microhabitat characteristics within a specific altitude range.

[0027] In the renewal restriction assessment module, the genetic diversity analysis unit analyzes the genetic parameters such as the allele frequency and heterozygosity of the Davidia involucrata population through gene sequencing technology to evaluate the level of genetic diversity; the seed diffusion constraint unit monitors the activity range and terrain obstacles of seed propagation media (such as birds and wind) and outputs reproductive obstacle signals.

[0028] In the distribution prediction module, the spatial modeling node uses a spatial analysis algorithm to generate potential distribution benchmark strategies based on the real-time ecological response characteristics of N habitat subnetworks (such as the impact of temperature and humidity changes at different altitudes on seedling survival). The niche optimization unit, combined with the reproductive barrier signals output by the seed dispersal constraint unit, prioritizes species dispersal pathways using a dynamic probabilistic ranking algorithm. For example, the probability of dispersal success for each pathway is calculated based on seed dispersal distance and terrain barrier probability.

[0029] The present invention will be further described below in conjunction with Examples 1 to 5: Example 1: The system sets up a spatial constraint correction module and an adaptive map optimization module. The spatial constraint correction module includes a terrain isolation unit and a climate fluctuation compensation unit. The adaptive map optimization module is used to construct a breeding success rate evaluation network and realize feedback correction of the analysis results.

[0030] The terrain isolation unit in the spatial constraint correction module implements its functions based on the diffusion path circuits corresponding to the parallel habitat subnetworks in the microhabitat parameter network. Each habitat subnetwork's diffusion path circuit corresponds to a digital model of the terrain characteristics for a specific elevation range. This model is constructed using terrain data collected by a geographic information system (GIS), including vector data for physical barrier elements such as slope, aspect, river distribution, and mountain orientation. The terrain isolation unit's workflow is linked to the processing of the potential distribution benchmark strategy and dynamic diffusion probability ranking results. After obtaining the potential distribution benchmark strategy output by the distribution prediction module and the dynamic diffusion probability ranking results generated by the niche optimization unit, the terrain isolation unit first traverses the diffusion path circuits corresponding to each habitat subnetwork and extracts the terrain barrier parameters within each path circuit. For example, for a habitat subnetwork within a certain elevation range, if its diffusion path circuit contains mountains with a slope greater than 30 degrees, the area is marked as a high terrain barrier area; if a river or canyon is present, the area is marked as an absolute terrain barrier area. Based on these markers, the terrain isolation unit makes geographical barrier corrections to the diffusion range of Davidia involucrata populations: for absolute barrier areas, they are directly removed from the diffusion path; for high barrier areas, the inhibitory effect of terrain on seed diffusion is reflected by reducing the diffusion probability weight of the path (such as multiplying it by a correction coefficient of 0.3).

[0031] The Climate Fluctuation Compensation Unit relies on the collection and analysis of real-time climate data. Connecting to the meteorological monitoring network, the unit acquires real-time data on climate factors, including temperature, precipitation, sunshine duration, and extreme weather event warnings. The Climate Fluctuation Compensation Unit establishes a data interface with the Genetic Diversity Analysis Unit within the Regeneration Limitation Assessment Module, receiving real-time monitoring results of genetic parameters, such as changes in allele frequency and genetic drift rates. Its dynamic correction mechanism is as follows: When temperature fluctuations exceeding 20% ​​of the historical average within a specific time period are detected, the Climate Fluctuation Compensation Unit triggers a genetic parameter correction program. Based on a pre-defined climate-genetic response model (e.g., a positive correlation between temperature increase and mutation rate), the model parameters used to calculate genetic parameters in the Genetic Diversity Analysis Unit are adjusted. For example, if the model presumes a 0.05% increase in mutation rate for every 1°C increase in temperature, then if the actual temperature rises by 2°C compared to the historical period, the mutation rate parameter will be automatically adjusted to increase by 0.1%. This dynamic correction ensures that genetic diversity analysis results more accurately reflect the immediate impact of climate fluctuations on the genetic structure of the population.

[0032] The reproductive success rate assessment network constructed by the adaptive graph optimization module consists of sensor nodes distributed across each habitat subnetwork. These sensors include soil moisture sensors, soil temperature sensors, light intensity sensors, and pest and disease monitoring cameras. These sensor nodes collect real-time data on key environmental parameters and biotic factors during the seedling establishment phase, such as soil moisture content, soil nutrient content, frequency of seedling pest and disease infestation, and the intensity of competition between seedlings and surrounding plants. The reproductive success rate assessment network preprocesses the raw data using edge computing nodes to extract characteristic parameters related to seedling establishment barriers. For example, soil moisture content below 40% of field capacity is defined as water stress, and pest and disease infestation frequency greater than once per week is defined as biotic stress. The assessment network uses machine learning models (such as the random forest algorithm) to comprehensively analyze these barrier characteristics and output a real-time assessment of the seedling establishment barrier rate. This value is expressed as a probability value between 0 and 1, with higher values ​​indicating greater barriers.

[0033] A bidirectional data transmission link is established between the Adaptability Graph Optimization module, the Spatial Modeling Node, and the Niche Optimization Unit. When the reproductive success assessment network calculates that the seedling establishment barrier rate for a habitat subnetwork exceeds a preset threshold (e.g., 0.6), the module packages the barrier rate data and the corresponding habitat subnetwork identifier into a feedback packet and transmits it to the Spatial Modeling Node and the Niche Optimization Unit. Upon receiving this feedback data, the Spatial Modeling Node first locates the corresponding habitat subnetwork and its diffusion pathway within the potential distribution baseline strategy. It then adjusts the habitat suitability compensation coefficient for the habitat subnetwork based on the barrier rate (e.g., for every 0.1 increase in barrier rate, the compensation coefficient decreases by 0.05) and recalculates the theoretical diffusion range upper limit for the subnetwork. The Niche Optimization Unit then adjusts the weight factor for the habitat subnetwork in the dynamic probability ranking based on the feedback barrier rate data, for example, by using an exponential decay function to reduce the diffusion pathway priority of subnetworks with high barrier rates. This closed-loop correction mechanism enables the distribution prediction module to dynamically optimize species distribution strategies based on real-time monitoring of seedling establishment, improving the accuracy of predictions.

[0034] In terms of data interaction and processing, the Spatial Constraint Correction Module and the Adaptability Map Optimization Module communicate with other modules through standardized data interfaces. For example, the Terrain Isolation Unit retrieves terrain data for each habitat subnetwork from the Microhabitat Parameter Network via a RESTful API. The Climate Fluctuation Compensation Unit transmits correction parameters to the Genetic Diversity Analysis Unit using JSON data encapsulation. Computational tasks within each module are handled by a distributed computing cluster, ensuring efficient and real-time large-scale data processing.

[0035] It should be noted that during the construction of the diffusion path circuit for the terrain isolation unit, the resolution of the terrain data for habitat subnets with different altitude thresholds can be adjusted according to actual needs. For example, in mountainous areas with drastic altitude changes, the altitude interval of the habitat subnet can be set to 50 meters, corresponding to a terrain data resolution of 10 meters; while in areas with flat terrain, the altitude interval can be expanded to 200 meters, with a terrain data resolution of 50 meters. This differentiated setting ensures the accuracy of terrain correction while avoiding data redundancy and improving system operation efficiency.

[0036] Furthermore, the climate fluctuation compensation unit imposes time constraints on the dynamic correction of genetic parameters. For short-term climate fluctuations (e.g., high temperatures lasting less than three days), the correction parameters are valid only during the fluctuation and automatically return to the original parameters after the fluctuation ends. For long-term climate trends (e.g., precipitation decreases lasting more than one month), the genetic parameters are adjusted based on the trend prediction model and continuously monitored until the trend reverses. This design avoids overreaction to short-term climate anomalies while adapting to the impact of long-term climate change on the population's genetic structure.

[0037] The adaptive graph optimization module's reproductive success rate assessment network also has self-learning capabilities. By accumulating historical data, the system continuously updates the training set of the machine learning model, improving the accuracy of barrier rate assessments. For example, if the outbreak probability of a certain pest or disease under specific soil moisture and temperature conditions has been repeatedly verified, the model will automatically adjust the barrier feature weights for that combination, making subsequent assessments more accurate.

[0038] In terms of system hardware deployment, the population dynamics monitoring nodes, the sensor nodes of the microhabitat parameter network, and the sensor nodes of the reproductive success rate assessment network all utilize low-power IoT devices. These devices are powered by solar energy and utilize wireless communication modules (such as LoRa and 5G) for data transmission, ensuring long-term stable operation in complex field environments. The algorithm logic for the spatial constraint correction module and the adaptive map optimization module is implemented in software programs deployed on cloud servers, facilitating remote maintenance and algorithm upgrades.

[0039] In summary, the Spatial Constraint Correction Module, through its Terrain Isolation Unit and Climate Fluctuation Compensation Unit, adjusts distribution strategies based on geographic barriers and climate dynamics, respectively. The Adaptive Map Optimization Module monitors seedling settlement barriers in real time and provides feedback to the Distribution Prediction Module, forming a closed-loop optimization mechanism. These three modules work in tandem with other system components to accurately analyze and dynamically predict species distribution under the constraints of natural regeneration in Davidia involucrata populations.

[0040] Example 2: The diffusion path circuit of the system's topographically isolated units is constructed based on a multi-layered environmental factor topology, and its construction process is closely related to this topology. This multi-layered environmental factor topology is a data structure generated by the microhabitat parameter network within the habitat factor acquisition module. It integrates multiple environmental factors, such as altitude, soil physical and chemical parameters, and light competition gradients, to form a hierarchical network topology. The construction of the diffusion path circuit is based on this topology, converting the spatial distribution characteristics of environmental factors into nodes and connections in the circuit model. Each node represents a habitat region under specific environmental conditions, and the strength of the connections between nodes indicates the ease with which seeds can spread between different regions.

[0041] The first step in constructing the diffusion path circuit involves collecting parent tree coordinate data and seedling survival rate fluctuation signals. Parent tree coordinate data is accurately acquired using a Global Positioning System (GPS) device, with each parent tree location corresponding to a geographic coordinate point. Seedling survival rate fluctuation signals are monitored in real time by a network of sensors distributed throughout each habitat. These sensors, including infrared counters and image recognition cameras, automatically identify and count changes in seedling populations. The collected data is transmitted to the system's data processing center via a wireless communication network.

[0042] The collected data is input into a maximum entropy model configured for each habitat subnetwork. The maximum entropy model is a machine learning-based ecological niche modeling method that predicts the suitability of species in unknown areas by analyzing the environmental characteristics of known distribution points. In this system, each habitat subnetwork corresponds to an independent maximum entropy model, which is pre-configured with environmental variable parameters relevant to that altitude range, such as soil type, temperature, and precipitation. After inputting the parent tree coordinate data and seedling survival rate fluctuation signals, the maximum entropy model iteratively determines the contribution of each environmental variable to seedling survival and generates a habitat suitability compensation coefficient. This coefficient is a value between 0 and 1 that reflects the relative likelihood of Davidia involucrata seedling survival under specific habitat conditions, with higher values ​​indicating more suitable habitats. The generated habitat suitability compensation coefficient is stored in the corresponding habitat subnetwork database and serves as an important parameter for subsequent diffusion path calculations.

[0043] Based on the suitability compensation coefficients stored at each level of habitat subnetwork and the instantaneous rate of change of climatic factors, the Habitat Suitability Index (HSI) algorithm distributes population dispersal probabilities to N habitat subnetworks, forming a dynamic habitat matching topology. The instantaneous rate of change of climatic factors includes the short-term rate of change of meteorological elements such as temperature, precipitation, and wind speed, which are collected in real time by meteorological monitoring stations. The Habitat Suitability Index algorithm comprehensively considers the habitat suitability compensation coefficients and the rate of change of climatic factors to calculate a comprehensive suitability score for each habitat subnetwork. The algorithm uses the location of the mother tree as the dispersal source and allocates population dispersal probabilities to each habitat subnetwork based on the comprehensive suitability score and the physical characteristics of seed dispersal (such as dispersal distance and direction). This allocation process forms a dynamic habitat matching topology, in which the node weights of each habitat subnetwork adjust in real time as environmental conditions change.

[0044] The conditions for constructing the habitat subnet configured by the population dynamics monitoring node include that the coordinates of the mother tree are within the preset altitude range, the seed germination rate corresponding to the habitat subnet is greater than the threshold, and the light competition gradient is less than the allowable deviation range. The preset altitude range is determined based on the ecological habits and geographical distribution characteristics of Davidia involucrata, usually between 1500-2000 meters. The seed germination rate is obtained in real time through seed monitors buried in each habitat area. These monitors can record information such as the germination time and number of seeds. The light competition gradient is measured by light sensors installed in the canopy layer, reflecting the light competition pressure faced by Davidia involucrata seedlings during their growth. Only areas that meet these three conditions at the same time will be included in the construction scope of the habitat subnet.

[0045] The multi-scale switching link configured in the Genetic Diversity Analysis Unit includes a steady-state diffusion sublink and an obstacle avoidance sublink. The steady-state diffusion sublink is generated based on the matching degree between the reproductive success rate signal from the Update Constraint Assessment Module and the underlying distribution benchmark strategy. When the reproductive success rate is stable and consistent with the prediction of the underlying distribution benchmark strategy, the system automatically activates the steady-state diffusion sublink for diffusion simulation. This sublink uses a relatively simple diffusion model, primarily considering the natural dispersal capacity of seeds and habitat suitability, while ignoring temporary interference factors.

[0046] The obstacle avoidance sub-link is constructed based on the correlation between the instantaneous disruption risk of population diffusion paths and human disturbance events. The instantaneous disruption risk of population diffusion paths is assessed by analyzing terrain data, human activity records and other information. For example, when a road is about to be built or large-scale land development is taking place on a diffusion path, the disruption risk of the path will increase significantly. Monitoring of human disturbance events is achieved by integrating satellite remote sensing data, government planning information and social media data, which can capture the impact of human activities on the habitat of Davidia involucrata in real time. When a disturbance event is detected, the system automatically switches to the obstacle avoidance sub-link, which uses a more complex diffusion model that not only takes into account habitat suitability and seed dispersal capacity, but also actively avoids high-risk areas and replans diffusion paths.

[0047] During operation, the system dynamically adjusts the construction of habitat subnetworks and the selection of dispersal pathways based on real-time monitoring data. For example, if the seed germination rate in a habitat area that previously met the construction criteria suddenly drops below a threshold, the system temporarily excludes that area from the habitat subnetwork and redistributes dispersal probabilities. Similarly, after a human disturbance event ends, the system assesses the recovery of the affected area and decides whether to reintegrate it into the dispersal pathway network.

[0048] The construction and management of habitat subnetworks also involves data storage and maintenance. Relevant data for each habitat subnetwork, including environmental parameters, suitability compensation coefficients, and dispersal probabilities, is stored in a distributed database. This data is regularly updated and verified to ensure the system uses the latest and most accurate information. The system also backs up historical data for trend analysis and model validation.

[0049] The construction and maintenance of diffusion path circuits consumes significant computing resources. To improve system efficiency, distributed computing and parallel processing technologies are employed. Data processing tasks are distributed to multiple computing nodes for simultaneous execution, significantly reducing processing time. Furthermore, the system employs a caching mechanism to cache frequently used calculation results and intermediate data, avoiding repeated calculations and further improving system response speed.

[0050] In terms of the system's hardware deployment, the sensor devices for the population dynamics monitoring nodes and microhabitat parameter network utilize a low-power, high-reliability design. These devices are typically equipped with solar panels and high-capacity batteries to ensure long-term stable operation in remote mountainous areas. Data transmission utilizes a combination of communication methods, using 4G / 5G networks in areas with network coverage and satellite communications or low-power wide-area network technologies such as LoRa in areas with weaker network signals.

[0051] The system also possesses self-learning and adaptive capabilities. By analyzing and mining historical data, it continuously optimizes the parameters of the maximum entropy model and habitat suitability index algorithm, improving the accuracy of habitat suitability predictions. Furthermore, the system can adjust the construction conditions of habitat subnetworks and the selection strategy of diffusion pathways based on actual operational conditions, making the entire system more adaptable to complex and changing natural environments.

[0052] In terms of the system's interface design, the Terrain Isolation Unit communicates with other modules via standardized data interfaces. These interfaces utilize a RESTful API architecture and support multiple data formats, such as JSON and XML, facilitating integration and expansion with other systems. For example, the Terrain Isolation Unit can feed generated diffusion path data into a Geographic Information System (GIS) for intuitive visualization. Furthermore, it can also obtain additional environmental data from other environmental monitoring systems to enrich its analytical capabilities.

[0053] The process of constructing diffusion path circuits for terrain-isolated cells also involves data quality control and exception handling. During the data acquisition phase, the system verifies sensor data in real time to check its integrity and plausibility. If any abnormal data is detected, the system automatically flags it and attempts to supplement or correct it using other data sources. During the data processing phase, the system checks the plausibility of the calculated results to ensure that the generated diffusion paths conform to biological and ecological principles. If any abnormal results are detected, the system automatically triggers recalculation or manual intervention.

[0054] Multi-layered security measures are implemented to safeguard the system. SSL / TLS encryption protocols are used during data transmission to prevent data theft or tampering. Data storage utilizes encryption technology to encrypt sensitive data and prevent data leaks. Furthermore, a strict access control mechanism is implemented to ensure that only authorized personnel can access and operate the system's core functions and data.

[0055] The deployment and implementation of the system must take into account the actual geographical environment and ecological conditions. In areas with concentrated Davidia involucrata distribution, sensor density can be appropriately increased to improve monitoring accuracy and reliability. In areas with complex terrain and inaccessible transportation, data collection using drones or remote monitoring equipment can be used to reduce labor costs and workload. Furthermore, system deployment requires collaboration with local ecological protection authorities and research institutions, leveraging their expertise and resources to ensure that implementation meets ecological protection requirements and objectives.

[0056] The construction and application of diffusion pathway circuits also provide important decision-making support for the conservation and management of Davidia involucrata. By analyzing diffusion pathways and habitat suitability, conservationists can identify key ecological corridors and protected areas, enabling targeted conservation efforts. For example, at key nodes along diffusion pathways, measures such as enhanced monitoring and reduced human interference can be implemented to ensure smooth seed dispersal and seedling survival. For areas with high habitat suitability but not yet occupied by Davidia involucrata, artificial introduction or ecological restoration can be considered to expand its distribution range.

[0057] Constructing diffusion pathway circuits for topographically isolated units is a complex and systematic process. It comprehensively considers multiple factors, including environmental factors, population dynamics, and human activities. Using advanced algorithms and techniques, it generates scientifically sound diffusion pathway predictions. This process not only provides a powerful tool for studying the natural regeneration of Davidia involucrata populations but also provides an important reference for biodiversity conservation and ecosystem management.

[0058] Example 3: The multi-source data fusion model configured within the system's spatial modeling nodes integrates the soil-climate mapping table corresponding to the habitat factor acquisition module and the obstacle avoidance parameter library from the updated constraint assessment module to form a composite analysis framework. The soil-climate mapping table, constructed based on long-term ecological monitoring data, records the corresponding relationship matrix between soil types and climate factors at different altitudes. This includes the spatiotemporal coupling characteristics of physical and chemical indicators such as soil texture, pH, and organic matter content, and climate factors such as temperature, precipitation, and daylight duration. The obstacle avoidance parameter library stores quantitative parameters for obstacle factors such as terrain slope, human activity intensity, and vegetation cover, each associated with a specific spatial distribution layer.

[0059] The potential distribution benchmark strategy generation step first calculates the theoretical diffusion range upper limit for each habitat subnetwork in the steady-state diffusion sublink based on the real-time soil data and diffusion probability ranking results from the habitat factor acquisition module. Real-time soil data is acquired through a sensor network distributed across each habitat area, including dynamic parameters such as soil moisture, temperature, and nutrient content. Diffusion probability ranking results are generated by the niche optimization unit based on factors such as seed dispersal distance and terrain suitability. The theoretical diffusion range upper limit is calculated using an improved random walk model that considers the directionality of seed dispersal and environmental heterogeneity, and is expressed as follows:

[0060] in, represents the upper limit of the theoretical diffusion range, Indicates the The seed dispersal distance in each direction, represents the propagation probability in this direction, Indicates the area of ​​suitable habitat in that direction, Represents the total habitat area.

[0061] The upper limit of the theoretical diffusion range and transient climate fluctuations are input into a multi-source data fusion model to generate a species distribution reference map. Transient climate fluctuation data comes from real-time data from the meteorological monitoring network, including changes in short-term climate factors such as temperature variability and precipitation anomaly index. The multi-source data fusion model uses a Bayesian network structure to fuse information from the soil-climate mapping table and the barrier avoidance parameter library through a conditional probability table. The model first makes a climate suitability correction to the upper limit of the theoretical diffusion range and adjusts the suitability score of each habitat area according to transient climate fluctuations; it then overlays the terrain and human interference layers in the barrier avoidance parameter library to identify potential diffusion barrier areas. Through this process, a species distribution reference map containing spatial distribution probabilities is generated, in which each pixel value represents the probability of Davidia involucrata appearing at that location.

[0062] The obstacle avoidance sublink corrects the reference map for disruption risk, forming a potential distribution benchmark strategy. The obstacle avoidance sublink is based on an instantaneous disruption risk assessment of population dispersal pathways, which is provided in real time by a terrain isolation unit and a human activity monitoring system. The terrain isolation unit analyzes terrain complexity using a digital elevation model to identify terrain features that could disrupt dispersal pathways, such as steep canyons and rivers. The human activity monitoring system integrates satellite remote sensing data and ground survey information to track human disturbance events such as road construction and urban expansion in real time. For identified high-risk areas, the obstacle avoidance sublink uses a path replanning algorithm to prioritize alternative pathways with lower disruption risk. For example, if a highway under construction is detected on a dispersal pathway, the system automatically searches for alternative pathways that bypass the area and calculates a new dispersal probability based on factors such as the path's length and habitat suitability. In this way, the species distribution reference map is corrected, ultimately forming a potential distribution benchmark strategy.

[0063] The parameter update mechanism of the multi-source data fusion model is implemented based on multi-level trigger conditions. When it is detected that the coordinates of the mother tree exceed the preset altitude range, the first obstacle avoidance instruction is triggered. The preset altitude range is determined according to the ecological adaptability of Davidia involucrata, usually 1500-2500 meters. The mother tree coordinate data is obtained in real time through the Global Positioning System (GPS). When the system finds that the position of a mother tree exceeds this range, the alternative habitat unit is immediately activated. The alternative habitat unit pre-stores suitable habitat data of the surrounding area, including soil type, climate conditions, terrain characteristics and other information. Based on this data, the system will find suitable migration target areas for mother trees that exceed the altitude range, and adjust the analysis granularity of the update restriction assessment module, refining it from the population level to the individual gene level, in order to more accurately assess the adaptability of individuals to the new environment.

[0064] If the distribution range has not returned to the allowed range after the execution of the first instruction, the second obstacle avoidance instruction will be triggered. The allowed range refers to the minimum distribution area threshold for the Davidia involucrata population to maintain stable survival and reproduction. This threshold is determined based on population ecology theory and historical monitoring data. After the second obstacle avoidance instruction is activated, the system switches to the obstacle avoidance sub-link through the multi-scale switching link, and reallocates the topological weights of the population diffusion paths based on the duration of the human interference event. The duration of the human interference event is obtained by integrating multi-source data such as government planning data and project progress information. For interference events with a longer duration, such as large-scale infrastructure construction projects, the system will significantly reduce the topological weights of the affected paths and give priority to paths with short interference durations to reduce risks during the population diffusion process.

[0065] During actual operation, the system monitors changes in parent tree coordinates, population distribution, and human disturbance events in real time, dynamically adjusting parameter update strategies. For example, after a human disturbance event in a particular area ends, the system reassesses the habitat suitability of that area and gradually restores its topological weight within the diffusion pathway network. Simultaneously, the system verifies and optimizes the updated potential distribution baseline strategy, continuously adjusting model parameters to improve prediction accuracy by comparing predicted results with actual monitoring data.

[0066] The parameter update process of the multi-source data fusion model also involves data quality control and uncertainty handling. When acquiring multi-source data such as parent tree coordinates, climate data, and information on human disturbance, the system conducts a quality assessment and removes outliers and erroneous data. For data with uncertainty, such as the predicted duration of a human disturbance event, the system represents it using a probability distribution and considers the propagation of this uncertainty during the model calculation process. For example, for a road construction project expected to last 6-12 months, the system assigns corresponding probability weights to different durations and calculates the topological weights of the diffusion paths under various possible scenarios, ultimately producing a comprehensive result.

[0067] The system also possesses adaptive learning capabilities, continuously optimizing its parameter update mechanism based on historical data and actual feedback. By analyzing past cases in which obstacle avoidance commands were triggered, the system can determine how different types of disturbances affect the distribution of Davidia involucrata populations and adjust triggering conditions and response strategies accordingly. For example, if a particular type of human disturbance is found to have a relatively minor impact on the Davidia involucrata population despite its prolonged duration, the system will appropriately relax the triggering threshold for that type of event, minimizing unnecessary policy adjustments.

[0068] In terms of hardware deployment, computational tasks for spatial modeling nodes are handled by a cluster of high-performance servers. These servers are equipped with powerful computing capabilities and large memory capacities, capable of handling large-scale spatial data and complex model calculations. Data storage utilizes a distributed file system to ensure data security and scalability. Furthermore, the system has established a data backup and recovery mechanism, regularly backing up critical data to prevent data loss and system failures.

[0069] The system's software architecture adopts a microservices design concept, splitting functional modules such as the multi-source data fusion model, obstacle avoidance algorithm, and parameter update mechanism into independent microservices. Each microservice is responsible for specific business logic and communicates through standardized interfaces. This architectural design makes the system highly scalable and maintainable, facilitating the subsequent addition and optimization of features.

[0070] In terms of user interaction, the system provides an intuitive visual interface, allowing users to view the generation process and results of potential distribution benchmark strategies. The interface supports a variety of spatial analysis functions, such as buffer analysis and overlay analysis, allowing users to customize their analysis based on their needs. The system also provides a data export function, allowing users to export analysis results to common Geographic Information System (GIS) formats, facilitating integration with other software and further analysis.

[0071] The implementation of the spatial modeling node also involves collaboration with other modules. For example, when working with the terrain isolation unit, the spatial modeling node can obtain accurate terrain data and diffusion barrier information to generate more precise diffusion paths. When working with the climate fluctuation compensation unit, it can obtain real-time changes in climate factors and dynamically adjust distribution strategies. This collaborative mechanism forms an organic whole, capable of comprehensively and accurately reflecting the distribution dynamics and influencing factors of Davidia involucrata populations.

[0072] In practical applications, this system can provide important decision-making support for the conservation and management of Davidia involucrata. By generating a potential distribution benchmark strategy, conservationists can understand the likely spread direction and range of Davidia involucrata populations and formulate targeted conservation measures. For example, nature reserves or ecological corridors could be established at key nodes along the dispersal pathway to ensure smooth seed dispersal and seedling survival. In areas severely disturbed by human activity, ecological restoration measures could be implemented to improve habitat quality and enhance the adaptability and resilience of the population.

[0073] System implementation also requires consideration of actual operating and maintenance costs. To reduce costs, the system utilizes a modular design and standardized interfaces to facilitate device replacement and upgrades. Furthermore, the system provides remote monitoring and management capabilities, allowing technicians to monitor system status remotely over the internet, identify and resolve issues promptly, and reduce on-site maintenance workload and costs.

[0074] The multi-source data fusion model and obstacle avoidance mechanism of the spatial modeling node provide a comprehensive and dynamic approach for analyzing the distribution of Davidia involucrata populations. By integrating information on multiple environmental factors and human activities, the system can more accurately predict the potential distribution range of Davidia involucrata and develop appropriate response strategies for different disturbance scenarios. This not only provides a deeper understanding of the ecological needs and distribution patterns of Davidia involucrata populations, but also provides scientific evidence and technical support for biodiversity conservation and ecosystem management.

[0075] Example 4: At a high altitude (2200-2300 meters) in a Davidia involucrata nature reserve, the system detected that the seed germination rate in habitat subnetwork A consistently fell below the 15% threshold, triggering a third obstacle avoidance command. Sensor network feedback from this subnetwork indicated that the soil pH had gradually shifted from neutral to acidic to 5.2, and the organic matter content in the surface soil had dropped to 1.8%, deteriorating the seed germination environment. The system immediately blocked the diffusion output channel of subnetwork A and, through topographic isolation units, activated a redundant link to the adjacent, lower-altitude habitat subnetwork B (2000-2100 meters). Subnetwork B has similar soil types to subnetwork A, but a more humid microclimate, and a historical average seedling survival rate of 42%. The system rerouted the seed flow originally intended for diffusion through subnetwork A to subnetwork B and simultaneously activated subnetwork A's ecological restoration plan, including the artificial application of organic fertilizers and soil amendments to adjust soil pH.

[0076] Thirty days after the third obstacle avoidance directive was executed, the diffusion load in subnet B reached 85% of its carrying capacity, triggering the fourth obstacle avoidance directive. The Adaptive Graph Optimization module immediately constructed a habitat diffusion safety level table, assessing that subnet B had a "medium" safety level. While the adjacent habitat subnet C (1,900-2,000 meters above sea level) was not overloaded, it had a "low" safety level due to its proximity to areas of high human activity. Following the downgrade logic, the system lowered the priority of the diffusion paths in subnet C by 50% and simultaneously adjusted the analysis frequency of the update constraint assessment module from once daily to once every three days. Simultaneously, the system called the Adaptive Graph Optimization module to adjust the output strategy of the niche optimization unit, reallocating diffusion paths originally planned to pass through subnet C to the other, more secure subnets D and E.

[0077] During system operation, the habitat subnetwork construction is continuously and dynamically adjusted. Based on spatial clustering of parent tree coordinate data, the system identified three major Davidia involucrata population clusters: the core cluster (1800-2000 meters above sea level), the buffer cluster (2000-2200 meters above sea level), and the edge cluster (2200-2400 meters above sea level). Analysis of historical seedling survival rate fluctuations showed that the core cluster had the smallest fluctuation (±8%), followed by the buffer cluster (±12%), and the edge cluster had the largest fluctuation (±18%). Based on this, the system generated an altitude gradient adaptability index table. The habitat subnetwork suitability compensation coefficients corresponding to the core cluster were 0.92, 0.78 for the buffer cluster, and 0.65 for the edge cluster.

[0078] Using a geographically weighted regression algorithm, the system assigns topological weights to the diffusion paths of each habitat subnetwork. The core cluster's habitat subnetwork, due to its high suitability compensation coefficient, receives the highest diffusion weight (0.45) and establishes multiple redundant links with the buffer cluster subnetwork. For example, three backup diffusion paths are established between the core subnetwork F and the buffer subnetwork G, respectively, through valleys, ridges, and streams to enhance system robustness. If a path is blocked by a sudden landslide, the system can immediately switch to the backup path to ensure continuous seed dispersal.

[0079] In actual operation, when the system detected that the seed germination rate in the H habitat subnet of the edge cluster had dropped to 12%, it first triggered the third obstacle avoidance instruction, transferring the diffusion output of this subnet to the adjacent I subnet. However, the carrying capacity of the I subnet was limited and quickly reached a load level of 75%. At this time, the system triggered the fourth obstacle avoidance instruction, downgrading the diffusion path priority of the H subnet and initiating ecological restoration measures. The restoration team artificially planted nitrogen-fixing plants in the H subnet area to increase the soil organic matter content and set up a sunshade to adjust the light intensity and improve the seed germination environment.

[0080] The system's multi-scale switching links play a key role in different scenarios. In the steady-state diffusion mode, the system primarily relies on the habitat subnetworks in the core and buffer zones for seed dispersal. These areas have relatively stable environmental conditions and high seed germination and seedling survival rates. However, when large-scale human activity interference is detected, such as a highway expansion project surrounding the protected area, the system automatically switches to the obstacle avoidance subnetwork. In this case, the system avoids habitat subnetworks near the project area and prioritizes diffusion paths away from the interference source. For example, when the highway expansion on the east side of the protected area began, the system rerouted the diffusion path originally passing through the eastern J subnetwork to the western K subnetwork. Although the K subnetwork has a slightly lower suitability compensation coefficient, it effectively avoids human interference with seed dispersal and seedling growth.

[0081] Over its long-term operation, the system continuously accumulates data and optimizes its performance. Analysis of historical data revealed periodic fluctuations in sensor data from certain habitat subnetworks. Further investigation revealed this correlation with seasonal animal migration. For example, large numbers of migratory birds gather in certain areas each autumn, leading to increased soil disturbance and impacting seed germination. The system adjusted its monitoring strategy accordingly, increasing monitoring frequency during the bird migration season and optimizing its habitat suitability assessment model to account for animal activity.

[0082] The system's hardware deployment fully considers the complex terrain and harsh environment of the reserve. In high-altitude, inaccessible areas, low-power, solar-powered sensor nodes were deployed. These nodes utilize LoRa wireless communication technology, providing an effective coverage radius of up to 5 kilometers and ensuring reliable data transmission to the monitoring center. More powerful edge computing nodes were deployed in the core and buffer zones. These nodes not only process local sensor data in real time but also perform preliminary data analysis and decision-making, reducing the burden on cloud servers.

[0083] At the software level, the system utilizes a microservices architecture, with each functional module deployed and operated independently, improving scalability and fault tolerance. For example, if a spatial modeling node fails, the system automatically transfers some computing tasks to a backup node, ensuring the smooth operation of the entire system. The system also implements an automated version update mechanism, enabling seamless updates without disrupting system operations when new algorithms or features are released.

[0084] The system's user interface is intuitive and easy to use, providing convenient tools for protected area managers. Through a map-based visualization interface, managers can view the status of each habitat subnetwork, seed dispersal paths, and the implementation of obstacle avoidance strategies in real time. The interface also supports data query and statistical analysis, allowing managers to query monitoring data for specific time periods and generate various statistical reports to support decision-making.

[0085] The system is deeply integrated with the daily management of protected areas. For example, if the system detects an abnormally low survival rate of seedlings in a particular area, it automatically generates an early warning and sends it to relevant managers. Based on this warning, managers can promptly conduct field investigations and implement targeted conservation measures. Furthermore, the potential distribution benchmark strategies generated by the system provide an important basis for the planning and development of protected areas, such as identifying new protected areas and planning ecological corridors.

[0086] The system also demonstrates strong adaptability in responding to sudden natural disasters. For example, when a rainstorm triggered a mudslide in the protected area, the system quickly identified the affected habitat subnetworks and adjusted dispersal routes. During a mudslide in 2024, the system quickly shifted dispersal tasks from the affected L subnetwork to the adjacent M and N subnetworks, ensuring continuous seed dispersal and minimizing the impact of the disaster on the Davidia involucrata population.

[0087] The implementation of the system has also facilitated scientific research. By accumulating extensive monitoring data over a long period of time, researchers have been able to delve deeper into the ecological characteristics, distribution patterns, and interrelationships between Davidia involucrata populations and environmental factors. For example, using system data, researchers have discovered that the dispersal distance of Davidia involucrata seeds is significantly negatively correlated with altitude. This finding provides important evidence for further optimizing seed dispersal models.

[0088] The system also provides strong support for the reserve's publicity and education efforts. By visualizing the distribution and dynamics of Davidia involucrata populations, the public can gain a more intuitive understanding of its survival and conservation significance, raising public awareness of its importance. The reserve can also leverage the scientific information generated by the system to carry out various activities, such as exhibitions and lectures, to educate the public about biodiversity conservation.

[0089] The system will continue to be upgraded and improved in the future. Plans are underway to incorporate artificial intelligence technologies to further enhance its predictive capabilities and decision-making capabilities. For example, deep learning algorithms will analyze multi-source data to predict potential habitat degradation or population crises in advance and automatically generate response plans. Furthermore, the system will strengthen data sharing and collaboration with other protected areas and research institutions to achieve broader conservation and management of Davidia involucrata populations.

[0090] This system provides a comprehensive and effective solution for the conservation of Davidia involucrata populations by dynamically adjusting habitat subnetworks, optimizing diffusion pathways, addressing various disturbances and disasters, and deeply integrating it with scientific research, management, and public education. This advanced technology-based conservation model is not only applicable to Davidia involucrata but also offers a valuable reference for the conservation of other endangered species.

[0091] Example 5: Taking a mountainous Davidia involucrata distribution area as an example, when generating a species distribution reference map, the system first inputs the theoretical upper limit of the diffusion range calculated using a random walk model (e.g., the potential area radiating outward from a core mother tree group within 5 kilometers) and transient climate fluctuation data (e.g., a 15% increase in precipitation over the historical average during a certain period) into a multi-source data fusion model. A simultaneously loaded ecological corridor database contains the region's three main natural vegetation corridors: an east-west canyon corridor, a north-south ridge corridor, and a linear corridor along streams. A human disturbance layer identifies the specific locations of mining areas, farmland, and roads under construction. The northeastern mining area, covering 2 square kilometers, is designated a high-disturbance area within 500 meters.

[0092] The allele frequency data output by the genetic diversity analysis unit showed that the heterozygosity of the Davidia involucrata population in this area was 0.38, lower than the average level of similar habitats (0.45), indicating that the genetic diversity of the population was low and its buffering capacity to environmental changes was weak. Based on this, the system revised the fracture threshold of the population diffusion path in the reference map, raising the original minimum connectivity area allowed from 0.5 hectares to 1 hectare. In other words, when a habitat patch with an area of ​​less than 1 hectare appears in the diffusion path, it is determined to be an unsustainable path and marked as a fracture risk area. For example, in the hilly areas of the southwest, there are multiple isolated woodlands with an area of ​​0.3-0.8 hectares in the original theoretical diffusion path. The system raised the path fracture risk level of these areas to "high" and removed them from the reference map.

[0093] In the obstacle avoidance sub-link, the system overlays the terrain roughness parameter for secondary calibration. Using a digital elevation model (DEM), the system extracts terrain relief data for the region, calculating an average roughness index of 0.6 (ranging from 0 to 1, with higher values ​​indicating more complex terrain) for the northeastern mountainous region and 0.3 for the southwestern hilly region. For areas with a roughness index greater than 0.5, the system automatically increases the resistance coefficient of the diffusion path by 40%. For example, a diffusion path in the northeastern mountainous region that crosses a steep ridge has an original diffusion probability of 35%, which is adjusted to 21% after the roughness parameter is overlaid. Furthermore, for flat areas with a roughness index less than 0.2 (such as river valley plains), the diffusion probability is increased by 15% to reflect the actual impact of terrain on seed dispersal.

[0094] During implementation, when the system processed a habitat subnetwork at an altitude of 1,800-2,000 meters, the theoretical upper limit of the diffusion range covered an area of ​​approximately 50 hectares of mixed coniferous and broad-leaved forest. However, after superimposing the ecological corridor database, it was found that this area was connected to the core population only by a stream corridor approximately 80 meters wide, and the middle of the corridor was divided into two sections by farmland. Through obstacle avoidance sub-link analysis, the system concluded that the effective connectivity of this corridor was only 60% of the theoretical value. Therefore, the distribution probability of this area in the reference map was adjusted from 0.7 to 0.42, and it was marked as an "edge habitat."

[0095] While processing the human disturbance layer, the system identified a southeastern highway under construction (planned length: 10 kilometers, width: 20 meters) that crossed the overlapping area of ​​two habitat subnetworks. The risk of diffusion path disruption in the highway construction area was assessed as "very high." The system automatically blocked the diffusion path in this area and initiated a path replanning mechanism. The newly planned path bypassed the highway construction area and continued along the western ridge corridor. Although the path length increased by 3 kilometers, it avoided the high-disturbance area. The average habitat suitability compensation coefficient along the route was 0.82, higher than the original path's 0.75.

[0096] When generating the reference map, the system also integrates data from the soil-climate mapping table in real time. For example, in areas above 2,200 meters above sea level, where mountain yellow soil is the dominant soil type, if transient climate fluctuations indicate a 2°C increase in temperature compared to the same period last year, the soil-climate mapping table indicates that this soil type's water retention capacity decreases by 25% under high-temperature conditions. Based on this information, the system adjusts the habitat suitability score for this area, revising the seedling survival probability from 45% to 34%, and lowering the color gradient (i.e., distribution probability) for this area in the reference map.

[0097] In another scenario, the coordinate data for mother trees in a habitat subnetwork showed that some mother trees had migrated to lower altitudes (1,600-1,700 meters). This triggered a parameter update in the multi-source data fusion model. Analysis of soil and climate data for the migrated mother trees revealed that the soil organic matter content in the lower-altitude region was 1.2% higher than in the original habitat, and the accumulated temperature during the growing season increased by 150°C per day, making it more suitable for Davidia involucrata growth. Based on this information, the system updated the parameters for the lower-altitude region in the soil-climate mapping table and expanded the potential distribution range of this region in the reference map, adding approximately 80 hectares of predicted suitable habitat.

[0098] The system's obstacle avoidance sub-link also features dynamic learning capabilities. For example, after repeatedly handling temporary flooding incidents caused by heavy rain, the system identified a 70% probability of flooding in river valleys below 1,500 meters above sea level during the rainy season. It then presets the diffusion path resistance coefficient for this type of terrain to 0.9 (meaning the diffusion probability is only 10% of normal). When the subsequent rainy season arrives, the system can quickly adjust the diffusion path in that area without real-time calculations, improving emergency response efficiency.

[0099] At the hardware level, the system uses high-resolution remote sensing equipment aboard drones to capture real-time image data of ecological corridors and human activities. Drones regularly patrol the protected areas (once a week). Image recognition algorithms automatically extract ground features from these images, such as road construction progress and farmland expansion, and update them in real time to the interference layer. This high-frequency data collection ensures the timeliness of the reference map, ensuring that the prediction results reflect the latest environmental changes.

[0100] At the software level, the engine generating the species distribution reference map utilizes parallel computing technology, dividing the entire study area into 100-by-100-meter grid cells, independently calculating distribution probabilities for each cell. For key elements such as ecological corridors and high-disturbance areas, a higher precision (50-by-50-meter) is used to ensure detailed accuracy. The generated map supports online query functionality, allowing users to enter coordinates or define an area to obtain detailed information such as habitat suitability scores and diffusion path priorities within that area.

[0101] In terms of collaboration with other modules, the reference map generated by the spatial modeling node provides basic data support for the niche optimization unit. For example, the niche optimization unit dynamically ranks the diffusion paths of each habitat subnetwork based on the distribution probabilities in the map, prioritizing paths in high-probability areas to the top three and allocating more seed dispersal resources. Simultaneously, information on fracture risk areas in the reference map is fed back to the terrain isolation unit, which updates the topology of the diffusion path circuit, ensuring data consistency and connectivity across modules.

[0102] In practice, the system helped the reserve management identify an overlooked potential suitable habitat: a 200-hectare natural secondary forest in the northwest of the reserve. Although remote from existing populations, this area is connected to the core area via an unmarked stream valley corridor, and the soil and climatic conditions are suitable for the growth of Davidia involucrata. Based on the system's recommendations, management established artificial assisted dispersal sites in the area, guiding seed dispersal through bird feeding and other methods. Within six months, the first seedlings were observed to germinate, confirming the accuracy of the system's predictions.

[0103] The system also plays a crucial role in assessing the cumulative impact of human activities. For example, long-term monitoring of two primary dust diffusion pathways around a mining area revealed that dust deposition caused soil pH along these pathways to drop by 0.5 units, reducing seedling survival to 22%. Based on this, the system issued an early warning to management, prompting the mining area to implement dust control measures and plant dust-absorbing plants along the pathways. Three months later, soil pH returned to normal, and seedling survival rebounded to 38%.

[0104] The implementation of the system has also provided the research team with a new perspective. By analyzing the spatiotemporal variations in distribution probabilities within the reference map, researchers discovered an "altitude drift" phenomenon in Davidia involucrata populations. Under climate warming, the probability of distribution in high-altitude areas has increased by an average of 3.2% annually, while it has decreased by 2.5% in low-altitude areas. This finding provides critical data for predicting future distribution trends of Davidia involucrata, and the relevant research results have been applied to the long-term planning of the nature reserve.

[0105] In the future, the system plans to integrate more data sources, such as soil microbial community data and pollinator distribution data, to more comprehensively assess habitat suitability. Furthermore, a mobile application will be developed to facilitate field workers to upload observation data in real time and receive the latest distribution forecast information pushed by the system, further improving the relevance and efficiency of conservation efforts.

[0106] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0107] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A species distribution analysis system based on natural regeneration restrictions of Davidia involucrata populations, characterized in that: include: Habitat factor collection module, renewal restriction assessment module and distribution prediction module; The habitat factor acquisition module includes a population dynamics monitoring node and a microhabitat parameter network; the regeneration restriction assessment module includes a genetic diversity analysis unit and a seed dispersal constraint unit; the distribution prediction module includes a spatial modeling node and an ecological niche optimization unit; the population dynamics monitoring node is used to collect spatial coordinate data of Davidia involucrata mother trees and seedling survival status parameters in real time; based on soil physical and chemical indicators and light competition gradients, a multi-layer environmental factor topology is constructed through the microhabitat parameter network; The microhabitat parameter network is composed of N habitat subnetworks based on altitude thresholds in parallel. Based on the real-time ecological response characteristics of the N habitat subnetworks, a potential distribution benchmark strategy is generated through a spatial modeling node. Combined with the reproductive barrier signal output by the seed dispersal constraint unit, the species dispersal paths are dynamically probabilistically sorted through the niche optimization unit.

2. The species distribution analysis system based on natural regeneration restriction of Davidia involucrata population according to claim 1, characterized in that: The system also includes a spatial constraint correction module and an adaptive map optimization module; the spatial constraint correction module includes a terrain isolation unit and a climate fluctuation compensation unit; According to the generated potential distribution benchmark strategy and dynamic diffusion probability ranking results, the geographical barrier of the Davidia involucrata population diffusion range is corrected through the terrain isolation unit, and the genetic parameters of the renewal restriction assessment module are dynamically corrected through the climate fluctuation compensation unit; at the same time, the seedling settlement obstacle rate is analyzed in real time through the reproduction success rate assessment network constructed by the adaptive map optimization module, and the analysis results are fed back to the spatial modeling node and the niche optimization unit to perform closed-loop correction on the distribution strategy.

3. The species distribution analysis system based on natural regeneration restriction of Davidia involucrata population according to claim 2, characterized in that: The terrain isolation unit is composed of diffusion path circuits corresponding to habitat subnets connected in parallel in the microhabitat parameter network; the steps of constructing the multi-layer environmental factor topology include: Collect the mother tree coordinate data and seedling survival rate fluctuation signal, input them into the maximum entropy model configured in each level of habitat subnet, generate the habitat suitability compensation coefficient and store it in the corresponding habitat subnet; Based on the suitability compensation coefficients stored in habitat subnetworks at all levels and combined with the instantaneous change rate of climatic factors, the population diffusion probability is distributed to N habitat subnetworks through the habitat suitability index algorithm to form a dynamic habitat matching topology.

4. The species distribution analysis system based on natural regeneration restriction of Davidia involucrata population according to claim 3, characterized in that: The habitat subnet construction conditions configured for the population dynamics monitoring node include: the mother tree coordinates are within a preset altitude range, the seed germination rate corresponding to the habitat subnet is greater than a threshold, and the light competition gradient is less than the allowable deviation range; The genetic diversity analysis unit is also configured with a multi-scale switching link; the multi-scale switching link includes a steady-state diffusion sub-link and an obstacle avoidance sub-link; the steady-state diffusion sub-link is generated based on the matching degree between the reproduction success rate signal of the update restriction evaluation module and the potential distribution benchmark strategy; the obstacle avoidance sub-link is constructed based on the correlation between the instantaneous rupture risk of the population diffusion path and human interference events.

5. The species distribution analysis system based on natural regeneration restriction of Davidia involucrata population according to claim 4, characterized in that: The spatial modeling node is configured with a multi-source data fusion model; the multi-source data fusion model includes a soil-climate mapping table corresponding to the habitat factor acquisition module and an obstacle avoidance parameter library of the update restriction assessment module; The steps of generating the potential distribution benchmark strategy include: Based on the real-time soil data from the habitat factor acquisition module and the diffusion probability ranking results, the theoretical diffusion range upper limit of each level of habitat subnetwork in the steady-state diffusion sub-link is calculated; The upper limit of the theoretical diffusion range and transient climate fluctuations are input into the multi-source data fusion model to generate a species distribution reference map, and the reference map is corrected for fracture risk through the obstacle avoidance sub-link to form a potential distribution benchmark strategy.

6. The species distribution analysis system based on natural regeneration restriction of Davidia involucrata population according to claim 5, characterized in that: The parameter updating step of the multi-source data fusion model includes: When it is detected that the coordinates of the mother tree are beyond the preset altitude range or the barrier rate of seedling settlement exceeds the threshold, the first obstacle avoidance instruction is triggered, that is, the alternative habitat unit is activated and the analysis granularity of the update restriction assessment module is adjusted; If the distribution range still does not return to the allowed range after the execution of the first obstacle avoidance instruction, the second obstacle avoidance instruction is triggered, that is, switching to the obstacle avoidance sub-link through the multi-scale switching link, and redistributing the topological weight of the population diffusion path based on the duration of the human interference event.

7. The species distribution analysis system based on natural regeneration restriction of Davidia involucrata population according to claim 6, characterized in that: The parameter updating step of the multi-source data fusion model further includes: When the seed germination rate of a certain level of habitat subnetwork in the habitat factor acquisition module is lower than the threshold, the third obstacle avoidance instruction is triggered, that is, the diffusion output channel of the subnetwork is blocked and the corresponding population is transferred to other habitat subnetworks through the terrain isolation unit; During the execution of the third obstacle avoidance instruction, if other habitat subnetworks are detected to be overloaded, the fourth obstacle avoidance instruction is triggered, that is, the adaptive graph optimization module is called to downgrade the output strategy of the niche optimization unit and limit the diffusion demand of the edge habitat.

8. The species distribution analysis system based on natural regeneration restriction of Davidia involucrata population according to claim 7, characterized in that: The degradation processing logic of the adaptive graph optimization module includes: Based on the obstacle rate data and population genetic diversity indicators output by the reproductive success rate assessment network, a habitat diffusion safety level table is constructed; When the fourth obstacle avoidance instruction is triggered, the diffusion path priority is dynamically downgraded according to the safety level table, and the analysis frequency of the update restriction evaluation module is synchronously adjusted to match the downgraded distribution strategy.

9. The species distribution analysis system based on natural regeneration restriction of Davidia involucrata population according to claim 3, characterized in that: The steps of constructing the habitat subnetwork also include: Based on the spatial clustering results of the mother tree coordinate data and the historical fluctuation range of seedling survival rates, an altitude gradient adaptability index table was generated; According to the differences in suitability compensation coefficients of each habitat subnetwork in the index table, the diffusion path circuits of N habitat subnetworks are topologically weighted using the geographically weighted regression algorithm, and redundant connection links are established between habitat subnetworks.

10. The species distribution analysis system based on natural regeneration restriction of Davidia involucrata population according to claim 5, characterized in that: The generation process of the species distribution reference map includes: After inputting the theoretical diffusion range upper limit and transient climate fluctuations into the multi-source data fusion model, the preset ecological corridor database and human activity interference layer were simultaneously loaded; Based on the allele frequency data output by the genetic diversity analysis unit, the break threshold of the population diffusion path in the reference map was corrected, and the terrain roughness parameter was superimposed on the obstacle avoidance sub-link for secondary calibration.

Images