Smart forestry data management system and method based on data fusion

Abstract

The present invention relates to the technical field of data management, and in particular to a smart forestry data management system based on data fusion and a method thereof. The method comprises the following steps: obtaining forest area sensor data and forest area remote sensing data, and performing forest area resource trend analysis, thereby obtaining forest area resource trend data; constructing a seasonal forest area resource status model set based on forest area grid resource status data and forest area resource trend data; dividing forest suitable areas according to the seasonal forest area resource status model set, and obtaining forest suitable area data; calculating forest growth offset factors according to the seasonal forest area resource status model set, and obtaining forest growth offset factors; evaluating forest area tree growth health based on the forest growth offset factors, obtaining forest area tree growth health data, and optimizing forest area planting strategies, and obtaining forest area planting management strategies. The present invention can improve the automation level of forest resource management.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to a smart forestry data management system and method based on data fusion. Background Art

[0002] Smart forestry is to use advanced technologies such as the Internet of Things, big data, and artificial intelligence to monitor, manage, and support decision-making for forest resources through information technology, thereby improving the level of intelligent, precise, and scientific management of forestry resources. With the introduction of technologies such as the Internet of Things, remote sensing technology, and geographic information systems (GIS), the field of forestry management has begun to gradually collect more real-time data. These data include various types of information such as meteorology, soil, forest growth, and forest fire monitoring, involving a large amount of dynamic and static data. Traditional forestry management methods rely mostly on manual monitoring and simple manual recording methods, lacking effective automation, real-time data collection and processing capabilities, resulting in a lag in data acquisition and difficulty in providing accurate and timely decision support. In addition, most traditional forestry data management systems are localized and isolated systems, lacking effective data sharing and integration between systems, resulting in a waste of data resources and making it impossible to conduct comprehensive and timely monitoring and analysis of the status and changes of forest resources. Summary of the invention

[0003] Based on this, it is necessary for the present invention to provide a smart forestry data management system based on data fusion and a method thereof to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above purpose, a smart forestry data management method based on data fusion includes the following steps:

[0005] Step S1: Acquire forest area sensor data and forest area remote sensing data, and integrate forest area grid resource status according to the forest area sensor data and forest area remote sensing data, so as to obtain forest area grid resource status data; perform forest area resource trend analysis based on the forest area grid resource status data, so as to obtain forest area resource trend data;

[0006] Step S2: Based on the forest area grid resource status data and the forest area resource trend data, multi-source data physical layer fusion is performed to obtain a forest area resource status model; the forest area resource status model is divided into seasonal resource states to obtain a seasonal forest area resource status model set;

[0007] Step S3: integrating the seasonal environment of the forest area according to the seasonal forest area resource status model set, thereby obtaining seasonal environmental data of the forest area, and dividing the forest suitable area based on the seasonal environmental data of the forest area, thereby obtaining the forest suitable area data;

[0008] Step S4: Estimating the regional tree growth cycle according to the seasonal forest resource status model set, thereby obtaining regional tree growth cycle data, and calculating the tree growth offset factor based on the regional tree growth cycle data and the tree suitable area data, thereby obtaining the tree growth offset factor;

[0009] Step S5: Evaluate the health of forest tree growth in the forest area based on the forest growth deviation factor to obtain the data on forest tree growth health in the forest area, and optimize the forest planting strategy based on the forest tree growth health data and the data on suitable areas for trees to obtain the forest planting management strategy.

[0010] The present invention can obtain various data such as meteorology, soil, plant growth, etc. in the forest area in real time through sensors and remote sensing technology, which has greater timeliness and accuracy than the traditional manual recording method. Through grid management, resource status monitoring can be carried out in specific areas to provide high-precision forest resource data. This data integration can better capture changes in different geographical locations, seasons, and environments. Timely and accurate resource status information provides strong data support for forestry management personnel and improves the accuracy and scientificity of forest resource management. Multi-source data fusion technology can effectively integrate different types of data (such as meteorological data, soil data, forest growth data, etc.), eliminating the deviations and limitations that may be caused by a single data source, thereby providing more comprehensive and reliable data support for resource analysis and prediction. By performing trend analysis on resource status, the changing laws of forest resources can be discovered, and changes in resource supply and demand can be predicted in advance, providing a strong scientific basis for future forestry planning and resource management. The fusion of the physical layer unifies various resource data into a comprehensive platform for analysis, which improves the systematicness and integrity of forest resource monitoring. Seasonal changes have a direct impact on the growth of trees. The seasonal resource status division can provide forestry management with more predictions and assessments of dynamic factors such as climate change and soil moisture, helping forestry managers to better cope with environmental changes in different seasons. Through seasonal environmental integration, the tree planting area and forest resource distribution can be optimized according to different seasons and climate conditions, and forest resources can be used to the maximum extent to avoid waste or unreasonable distribution of resources. Through the division of suitable areas, it is possible to determine which areas are more suitable for specific tree species, help select the best tree planting area, and improve the success rate of planting. Based on the seasonal resource status model, the growth cycle of trees in the region can be estimated more scientifically to judge the growth laws of different types of trees, helping forestry managers to formulate more accurate planting and felling plans. The calculation of the growth deviation factor can effectively capture the impact of environmental changes on tree growth, especially the growth deviation caused by external factors such as climate change and soil quality changes, which helps to predict and adjust the normal state of tree growth. Through the analysis of the growth cycle and growth deviation factor, growth anomalies can be discovered in time and corresponding remedial measures can be taken to reduce the losses caused by poor growth. Through the assessment of forest growth health, we can fully understand the growth status of forests, discover potential pests and diseases, water shortages and other problems in advance, and provide data support for forest health management. Based on health data and suitable area information, we can optimize the planting strategy of forest areas, adjust planting density, select suitable tree species, improve soil conditions, etc., so as to improve the productivity and disaster resistance of forests. Through refined management and optimization of planting strategies, we can achieve sustainable utilization of forest resources and promote a win-win situation for ecological environmental protection and economic benefits.Overall, these technical steps can not only improve the automation and intelligence level of forest resource management, but also provide a scientific basis for forestry decision-making through accurate data analysis and prediction, and enhance the dynamic monitoring and flexible response capabilities of forest resources. This data-driven forestry management model has greatly improved the utilization efficiency of forest resources and the ecological protection effect, and also provided a more scientific solution to climate change and forest health issues.

[0011] Optionally, step S2 specifically includes:

[0012] Step S21: Acquire historical meteorological data of the forest area, and spatially align the forest area meteorological pattern data, forest area grid resource status data, and forest area resource trend data, so as to obtain aligned historical meteorological data, aligned forest area grid resource status data, and aligned forest area resource trend data;

[0013] Step S22: performing forest area meteorological pattern recognition on the aligned historical meteorological data, thereby obtaining forest area meteorological pattern data, and constructing a forest area meteorological change model based on the forest area meteorological pattern data;

[0014] Step S23: describing the spatial distribution of forest resources according to the aligned forest grid resource status data and the aligned forest resource trend data, thereby obtaining a forest resource distribution spatial model;

[0015] Step S24: Perform physical layer fusion on the forest area meteorological change model and the forest area resource distribution spatial model, so as to obtain a forest area resource status model;

[0016] Step S25: Divide the forest resource status model into seasonal resource statuses, thereby obtaining a seasonal forest resource status model set.

[0017] The present invention can ensure the spatial consistency of different types of data by acquiring and aligning historical meteorological data, grid resource status data and resource trend data, eliminate the differences between data sources, and improve the reliability and accuracy of data. This data alignment provides a basis for subsequent analysis and modeling, ensuring the integrity and accuracy of information. In the process of identifying and constructing a forest area meteorological change model, by deeply analyzing the changes in meteorological patterns, the influence trend of meteorological factors on forest area resources can be revealed, providing a scientific basis for predicting and responding to meteorological changes. Based on this meteorological model, future meteorological changes and their potential impact on forest area ecosystems can be more accurately predicted, which is helpful to formulate response measures and ensure the sustainable use of forest area resources. Further, by describing the spatial distribution of forest area resources, a more sophisticated resource distribution model can be constructed to reflect the spatial differences and trends of resources in different regions, which is helpful to accurately manage and optimize the allocation of resources. The establishment of this spatial model can not only provide guidance for forest area resource protection, but also support decision-making on ecological restoration and sustainable development. By integrating the forest area meteorological change model with the resource distribution spatial model at the physical layer, it is possible to fully integrate meteorological and resource information to form a more comprehensive forest area resource status model, thereby comprehensively evaluating the resource status and changes in the forest area and providing precise support for scientific management and decision-making. Finally, the forest area resource status model is seasonally divided so that the model can reflect the impact of different seasons on the resource status, and a seasonal resource status model set is obtained, which provides a specific basis for seasonal resource management and helps to formulate more reasonable seasonal resource protection and utilization strategies, thereby improving the effectiveness of forest area resource management and ensuring the stability and sustainability of the ecosystem in different seasons.

[0018] Optionally, step S25 is specifically:

[0019] Step S251: dividing the forest area meteorological pattern data into forest area seasonal meteorological patterns, thereby obtaining forest area seasonal meteorological pattern data;

[0020] Step S252: performing forest seasonal period identification on the periodic forest resource state change data, thereby obtaining forest seasonal resource state change data;

[0021] Step S253: integrating the seasonal patterns of the forest area based on the seasonal meteorological pattern data of the forest area and the seasonal resource status change data of the forest area, thereby obtaining the seasonal pattern data of the forest area;

[0022] Step S254: dividing the forest resource status model into seasonal resource status according to the forest seasonal pattern data, thereby obtaining a seasonal forest resource status model set.

[0023] The present invention can capture the impact of meteorological changes between different seasons on forest ecosystems by seasonal division of meteorological data, and provide accurate meteorological background data for subsequent resource management. Periodic identification of changes in forest resource status helps to identify the fluctuation characteristics of resource status in different seasons or cycles, thereby providing a basis for the prediction and management of resource changes. This process enhances the dynamic monitoring capability of forest resources, enabling managers to make targeted adjustments in different seasons and periodic changes. The seasonal data integration based on meteorology and resource status further improves the overall cognition of changes in forest ecological environment, ensures the comprehensive utilization of multi-dimensional data, and enhances the adaptive analysis of complex forest environment. The division of resource status model by seasonal pattern data can more accurately reflect the law of resource changes in forest areas in different seasons or specific cycles, thereby optimizing resource allocation and scheduling, and reducing environmental risks and management costs. These steps not only improve the accuracy of seasonal patterns, but also enable forest resource management to make more reasonable and scientific decisions based on seasonal changes, and ultimately promote the implementation of sustainable development in forest areas.

[0024] Optionally, step S3 specifically includes:

[0025] According to the seasonal forest resource status model set, the seasonal environment of the forest area is integrated to obtain the seasonal environmental data of the forest area, and the seasonal environmental characteristics of the forest area seasonal environmental data are divided to obtain the seasonal hydrological data and seasonal soil data of the forest area;

[0026] Obtaining forest growth rules, including forest growth cycle rules, forest suitable growth soil rules, forest suitable growth hydrological rules and forest suitable growth terrain rules;

[0027] According to the rules of forest suitable growth terrain, the forest terrain suitable area is selected for the remote sensing data to be analyzed, so as to obtain the forest terrain suitable area data;

[0028] According to the hydrological rules suitable for tree growth, seasonal hydrological data of forest areas are screened for suitable areas for tree hydrology, so as to obtain data on suitable areas for tree hydrology;

[0029] Based on the tree-suitable growth climate rules, seasonal soil data of forest areas are screened for tree-suitable soil areas, thereby obtaining tree-suitable soil area data;

[0030] The forest terrain suitable area data, forest hydrology suitable area data and forest soil suitable area data are regionally superimposed to obtain forest suitable area data.

[0031] The present invention integrates the seasonal environment of forest areas through a seasonal forest resource state model set, and can accurately extract environmental characteristic information of different seasons in forest areas, covering hydrology, soil, climate and other factors. This integration provides a comprehensive basis for subsequent data analysis and ensures the comprehensiveness and reliability of the data. Subsequently, by characterizing the seasonal environmental data, the hydrological and soil change laws of different seasons can be revealed, thereby providing a basis for the precise control of the tree growth cycle. Secondly, the tree growth rules, especially the soil, hydrology, terrain and climate conditions suitable for growth, are obtained, so that the model can analyze the suitable growth conditions of different regions in a targeted manner, and optimize the configuration of various resources in the forest area accordingly. According to the rules of terrain, hydrology and soil suitable for tree growth, the optimal area for tree growth is screened out, and the environment can be accurately evaluated from multiple dimensions such as terrain, climate, soil and hydrology, thereby avoiding the influence of unsuitable environmental conditions on tree growth. In addition, through the spatial superposition analysis of the data of each suitable area, the synergy of different environmental factors can be comprehensively considered to obtain a more scientific suitable growth area for trees, which provides strong support for precise planting. Overall, this series of steps achieves seamless connection from environmental data acquisition, rule extraction to regional screening, improves the accuracy of forest planting and management, can effectively promote the sustainable development of forest resources, optimize the overall health of forestry ecosystems, and provide reliable support for relevant decision-making.

[0032] Optionally, step S4 is specifically:

[0033] Step S41: extracting seasonal growth characteristics of regional trees from the seasonal forest resource status model set, thereby obtaining seasonal growth characteristic data of regional trees;

[0034] Step S42: Estimating the regional forest growth cycle according to the regional forest seasonal growth characteristic data, thereby obtaining regional forest growth cycle data;

[0035] Step S43: performing regional forest growth cycle error calculation on the forest growth cycle rule and the regional forest growth cycle data, thereby obtaining regional forest growth cycle error data;

[0036] Step S44: performing statistics on trees with high growth cycle errors based on the regional tree growth cycle error data, thereby obtaining tree data with high growth cycle errors;

[0037] Step S45: extracting suitable areas for trees with high growth cycle errors from the suitable area data for trees with high growth cycle errors according to the tree data with high growth cycle errors, thereby obtaining suitable area data for trees with high growth cycle errors;

[0038] Step S46: extracting high growth cycle error tree planting areas from the seasonal forest resource state model set according to the high growth cycle error tree data, thereby obtaining high growth cycle error tree planting area data;

[0039] Step S47: Calculate the tree growth offset factor for the high growth cycle error tree suitable area data and the high growth cycle error tree planting area data, so as to obtain the tree growth offset factor.

[0040] By extracting regional forest seasonal growth characteristic data, the present invention can accurately grasp the growth pattern of trees in different seasons, provide basic data for subsequent growth cycle estimation, and thus improve the understanding of the growth law of trees. The error calculation of the growth cycle can reveal the deviation between the model prediction and the actual growth, which helps to identify and correct areas or specific tree species with large errors, thereby improving the prediction accuracy. By counting trees with high growth cycle errors and extracting suitable areas, the selection of tree planting areas can be optimized, avoiding resource waste due to growth cycle errors, and improving the feasibility and benefits of planting. By calculating the tree growth offset factor, the tree growth area and planting strategy can be adjusted more scientifically, thereby further improving the management efficiency and economic benefits of forestry resources and ensuring the best match between the tree growth cycle and actual conditions.

[0041] Optionally, step S47 is specifically:

[0042] Step S471: extracting regional environmental features from the high growth cycle error forest tree suitable area data and the high growth cycle error forest tree planting area data, thereby obtaining forest tree suitable area environmental data and forest tree planting area environmental data;

[0043] Step S472: performing environmental error calculation on the forest suitable area environmental data and the forest planting area environmental data, thereby obtaining planting environment error data;

[0044] Step S473: extracting tree planting density features from the tree planting area data with high growth cycle errors, thereby obtaining tree planting density data;

[0045] Step S474: obtaining forest tree species characteristic data, and performing growth density deviation calculation on the forest tree species characteristic data and forest tree planting density data, thereby obtaining planted tree growth density deviation data;

[0046] Step S475: Perform growth cycle error multivariate regression analysis based on regional forest growth cycle error data, planted tree growth density deviation data, and planting environment error data to obtain a forest growth deviation factor.

[0047] The present invention can effectively identify and quantify environmental conditions suitable for growth by extracting regional environmental characteristics from suitable area data of high growth cycle error trees and planting area data of high growth cycle error trees, and provide an important basis for accurate tree planting decisions. The extracted environmental data helps to further evaluate the environmental differences between the planting area and the suitable area, thereby revealing potential planting environment errors and ensuring the healthy growth of trees. By calculating the planting environment error data, the deviation between the environment of the existing planting area and the ideal environment can be more clearly understood, providing a basis for optimizing planting strategies and improving forest resource utilization. In addition, by extracting the characteristics of tree planting density and calculating the growth density deviation, the influence of planting density on tree growth can be accurately grasped, providing data support for rationally planning planting density and optimizing tree species configuration. Combined with the analysis of tree species characteristic data, it is helpful to better understand the differences in tree species growth characteristics and their adaptability to the environment, thereby providing scientific guidance for the adjustment of planting plans. Finally, comprehensive multivariate regression analysis can reveal the intrinsic relationship between growth cycle errors and planting density and environmental differences, further accurately predict growth deviation factors, provide theoretical support for improving tree growth benefits and reducing adverse effects, and ultimately achieve efficient and sustainable tree planting management.

[0048] Optionally, the present invention further provides a smart forestry data management system based on data fusion, which is used to execute the smart forestry data management method based on data fusion as described above, and the smart forestry data management system based on data fusion includes:

[0049] The forest area resource status integration module is used to obtain forest area sensor data and forest area remote sensing data, and integrate forest area grid resource status according to forest area sensor data and forest area remote sensing data, so as to obtain forest area grid resource status data; perform forest area resource trend analysis based on forest area grid resource status data, so as to obtain forest area resource trend data;

[0050] The seasonal resource status classification module is used to perform physical layer fusion of multi-source data based on forest area grid resource status data and forest area resource trend data, so as to obtain a forest area resource status model; perform seasonal resource status classification on the forest area resource status model, so as to obtain a seasonal forest area resource status model set;

[0051] The forest suitable area division module integrates the seasonal environment of the forest area according to the seasonal forest area resource status model set, thereby obtaining the seasonal environmental data of the forest area, and divides the forest suitable area based on the seasonal environmental data of the forest area, thereby obtaining the forest suitable area data;

[0052] A growth offset factor calculation module is used to estimate the regional forest growth cycle according to the seasonal forest resource status model set, thereby obtaining regional forest growth cycle data, and calculate the forest growth offset factor based on the regional forest growth cycle data and forest suitable area data, thereby obtaining the forest growth offset factor;

[0053] The growth health assessment module is used to assess the growth health of trees in the forest area based on the tree growth deviation factor, so as to obtain the growth health data of trees in the forest area, and optimize the planting strategy of the forest area based on the growth health data of trees in the forest area and the data of suitable areas for trees, so as to obtain the planting management strategy of the forest area.

[0054] The smart forestry data management system based on data fusion of the present invention can realize any one of the smart forestry data management methods based on data fusion of the present invention, and is used to combine the medium of operation and signal transmission between various modules to complete the smart forestry data management method based on data fusion. The internal modules of the system cooperate with each other, thereby improving the automation and intelligence level of forest resource management. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0056] Figure 1 It is a schematic diagram of the steps of the smart forestry data management method based on data fusion of the present invention;

[0057] Figure 2 Detailed step flow diagram of step S1 in the present invention;

[0058] Figure 3 Detailed step flow diagram of step S2 in the present invention;

[0059] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0060] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are 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 technicians in this field without creative work are within the scope of protection of the present invention.

[0061] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0062] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0063] To achieve this, please refer to Figures 1 to 3 The present invention provides a smart forestry data management method based on data fusion, the method comprising the following steps:

[0064] Step S1: Acquire forest area sensor data and forest area remote sensing data, and integrate forest area grid resource status according to the forest area sensor data and forest area remote sensing data, so as to obtain forest area grid resource status data; perform forest area resource trend analysis based on the forest area grid resource status data, so as to obtain forest area resource trend data;

[0065] In this embodiment, multiple sensors (such as temperature and humidity sensors, soil moisture sensors, and CO2 concentration sensors) are installed in the forest area, and the NDVI (normalized difference vegetation index) data and ground elevation data of the forest area are obtained through remote sensing satellites. First, the sensor records data once an hour, and the remote sensing data is updated once a month. The sensor data and remote sensing data are divided into grids using a GIS system and integrated into resource status data for each grid (such as soil moisture content, temperature changes, and vegetation coverage). Through data analysis methods (such as time series analysis), trend analysis of grid resource status is performed to predict resource change trends in the next few seasons, and obtain forest resource trend data (such as the probability of resource growth or decline, predicted forest coverage changes, etc.).

[0066] Step S2: Based on the forest area grid resource status data and the forest area resource trend data, multi-source data physical layer fusion is performed to obtain a forest area resource status model; the forest area resource status model is divided into seasonal resource states to obtain a seasonal forest area resource status model set;

[0067] In this embodiment, a machine learning method (such as a support vector machine or a random forest) is used to fuse data from different sources. First, the forest area sensor data and remote sensing data are fused by weighted averaging or a Kalman filter algorithm. Then, a forest area resource status model is established based on the fused data, and the model can describe the resource distribution under different climatic conditions, soil properties, and vegetation conditions. Furthermore, the model is divided according to seasonal climate changes (such as low temperature in winter, high temperature season, rainy season, etc.) to obtain multiple seasonal models. These seasonal resource status model sets help to accurately describe the impact of different seasons on forest area resources, thereby optimizing the management of forest area resources.

[0068] Step S3: integrating the seasonal environment of the forest area according to the seasonal forest area resource status model set, thereby obtaining seasonal environmental data of the forest area, and dividing the forest suitable area based on the seasonal environmental data of the forest area, thereby obtaining the forest suitable area data;

[0069] In this embodiment, a seasonal resource status model set is used to perform spatial distribution analysis, and combined with climate data (such as temperature, precipitation, etc.), an environmental suitability index for each season is established. Based on the index, a spatial interpolation algorithm (such as Kriging interpolation) is used to evaluate the environmental suitability of each grid. For example, factors such as temperature, humidity, and sunshine time have different effects on the growth of trees in different seasons. Therefore, these factors are taken into consideration and areas suitable for tree growth are divided. The final tree suitable area data can provide a basis for tree planting decisions. For example, some areas are suitable for specific tree species, while other areas may require highly adaptable tree species.

[0070] Step S4: Estimating the regional tree growth cycle according to the seasonal forest resource status model set, thereby obtaining regional tree growth cycle data, and calculating the tree growth offset factor based on the regional tree growth cycle data and the tree suitable area data, thereby obtaining the tree growth offset factor;

[0071] In this embodiment, the growth cycle of trees in different regions is estimated through a growth model (such as a tree ring model) based on historical forest growth data and a set of seasonal resource status models. In this model, the growth cycle involves stages such as seed germination, growth, and maturity, which are affected by multiple factors such as climate and soil. For example, the growth cycle is predicted by factors such as seasonal climate and soil moisture to obtain the growth cycle data for each grid. Combined with the data on suitable areas for trees, regression analysis or neural networks are used to calculate the offset factor, indicating that under specific climatic conditions, the growth cycle of trees in certain areas may be advanced or delayed. The offset factor reflects the impact of climate change on forest growth and can be used to adjust forest management strategies.

[0072] Step S5: Evaluate the health of forest tree growth in the forest area based on the forest growth deviation factor to obtain the data on forest tree growth health in the forest area, and optimize the forest planting strategy based on the forest tree growth health data and the data on suitable areas for trees to obtain the forest planting management strategy.

[0073] In this embodiment, tree growth health assessment indicators (such as tree height, diameter at breast height, leaf area, etc.) are combined with growth offset factors to evaluate the health status of trees in different areas of the forest area. Through remote sensing images and ground field surveys, the growth health indicators of trees such as chlorophyll content and crown width changes are evaluated. If the health of trees in a certain area is low, and through the analysis of growth offset factors, it can be determined that certain environmental factors (such as poor soil quality and unsuitable climatic conditions) have led to a decrease in tree health. Combined with data on suitable areas for trees, planting strategies are optimized. For example, in some areas, it is recommended to use tree species with strong drought resistance, or adjust irrigation and fertilization plans to improve the growth health of trees. Ultimately, the overall forest resource management level of the forest area is improved through the optimization of planting management strategies.

[0074] Optionally, step S1 specifically includes:

[0075] Step S11: Acquire forest area sensor data and forest area remote sensing data, and perform data preprocessing on the forest area sensor data and forest area remote sensing data, respectively, so as to obtain sensor data to be analyzed and remote sensing data to be analyzed;

[0076] In this embodiment, multiple sensors (such as temperature, humidity, light intensity, etc.) and remote sensing equipment (such as satellite remote sensing, drone aerial photography, etc.) are deployed in the forest area. The sensor collects forest environment data at regular intervals (for example, every hour), and the remote sensing equipment collects image data regularly (such as every month). For sensor data, the original data of the sensor is firstly de-noised, and a filtering algorithm (such as median filtering) is used to remove sensor errors. For remote sensing data, image enhancement techniques (such as histogram equalization) are used to improve image quality, eliminate the influence of clouds and fog, and perform radiation correction and geometric correction to ensure the consistency of data collected at different time points. For example, assuming that the humidity data collected by a certain sensor may have errors, the humidity data is smoothed by the Kalman filter algorithm to obtain an accurate humidity value.

[0077] Step S12: Calculating the regional vegetation coverage rate of the remote sensing data to be analyzed, thereby obtaining regional vegetation coverage rate data, and dividing the remote sensing data to be analyzed into forest area remote sensing grids according to the regional vegetation coverage rate data, thereby obtaining forest area grid remote sensing data;

[0078] In this embodiment, the regional vegetation coverage is calculated based on the processed remote sensing image data by using the NDVI (normalized difference vegetation index) algorithm. The specific method is to use the red light band and near infrared band data in the remote sensing image to calculate the NDVI value. For example, using the formula: ; Among them, NIR is the reflectance of the near-infrared band, and RED is the reflectance of the red light band. The calculated NDVI value can reflect the vegetation coverage of the area. The closer the value is to 1, the more lush the vegetation is, and the closer the value is to -1, the less vegetation is present. Then, according to the calculated vegetation coverage, the spatial analysis method is used to divide the area into grid cells with different vegetation coverage rates. For example, the size of each grid is 1km², and the vegetation coverage rate of each grid is in the interval of (0,1) for subsequent analysis.

[0079] Step S13: extracting sensor location features from the sensor data to be analyzed, thereby obtaining sensor location data, and fusing forest area grid data with the forest area grid remote sensing data and the sensor data to be analyzed according to the sensor location data, thereby obtaining forest area grid fusion data;

[0080] In this embodiment, during the positioning process of the sensor, the geographic coordinates (longitude, latitude) of the sensor are obtained through the GPS positioning module. The sensor's altitude, azimuth and other information are further extracted as the location features of the sensor. For example, three temperature and humidity sensors are deployed in forest area A, and their locations are (longitude 1, latitude 1), (longitude 2, latitude 2), and (longitude 3, latitude 3). These location information are combined with the grid information in the remote sensing data, and the corresponding remote sensing data is fused with the sensor data according to the grid of the area where the sensor is located. The data fusion method can use a weighted average method or a multidimensional interpolation method to fuse the data collected by the sensor with the remote sensing grid data, thereby improving the temporal and spatial consistency of the data.

[0081] Step S14: Integrate the forest area grid resource status according to the forest area grid fusion data, so as to obtain the forest area grid resource status data;

[0082] In this embodiment, the resource status of the forest area is further integrated based on the obtained forest area grid fusion data. Specifically, the remote sensing data (such as NDVI) and the sensor data (such as temperature, humidity, etc.) are matched according to the grid unit, and the resource status indicators in each grid unit are calculated. For example, the average temperature, humidity, vegetation coverage, etc. in each grid are calculated. For example, in a certain grid, the NDVI value is 0.75, the temperature is 22°C, and the humidity is 65%. Combining these data, a comprehensive resource status report for the grid is generated to reflect the ecological status of the area. Classification algorithms such as decision trees or support vector machines can be used to classify the resource status of the forest area based on multiple feature data, such as determining whether the area belongs to the category of "ecological health", "drought" or "forest degradation".

[0083] Step S15: Perform forest resource trend analysis based on the forest grid resource status data to obtain forest resource trend data.

[0084] In this embodiment, trend analysis is performed based on the obtained forest grid resource status data. The analysis content includes the temporal change trends of resources such as temperature, humidity, and vegetation coverage. Time series analysis methods (such as ARIMA models) can be used to predict the changing trends of forest resources in the future. For example, the vegetation coverage rate of a certain forest area has a significant increasing trend every year. Based on this trend, the vegetation coverage rate changes in the next five years can be predicted, and corresponding forestry management recommendations can be made. In addition, by comparing forest resource data in different years and seasons, potential trends in resource decline or recovery can be identified to help decision makers take corresponding protection measures or management strategies.

[0085] Optionally, step S14 is specifically:

[0086] Step S141: performing grid camera extraction, grid soil feature extraction and grid water feature extraction on the forest area grid fusion data, thereby obtaining grid camera data, grid soil sensing data, grid remote sensing water index data and grid water sensing data;

[0087] In this embodiment, the grid camera data, grid soil sensor data, grid remote sensing water index data and grid water sensor data contained in the forest area grid fusion data are extracted. These data are gridded through high-definition camera images taken by drones or satellites to extract the surface image data of each grid in the forest area. These image data are processed by image processing technology (such as edge detection and image segmentation) to identify different surface features, such as forests, grasslands, bare land, etc. The soil sensor (such as conductivity sensor or spectral sensor) measures the soil in the area in real time, and extracts the soil characteristic data such as humidity, pH, temperature, etc. of each grid. The water sensor (water quality sensor, water temperature sensor, etc.) measures the water in the area in real time, and extracts the water characteristic data such as pH, temperature, water quality, etc. of each grid. Remote sensing technology (such as using NDWI index) monitors the water in the area and extracts the water index data of each grid. This data will provide a basis for subsequent soil and water analysis.

[0088] Step S142: Calculate the grid tree planting density according to the regional vegetation coverage data, thereby obtaining grid tree planting density data, and integrate the grid soil resource characteristics according to the grid soil sensor data and the grid tree spatial planting density data, thereby obtaining grid soil resource characteristic data;

[0089] In this embodiment, high-resolution remote sensing images (such as Landsat or MODIS data) are used to analyze the coverage of vegetation in the region and calculate the density of greening or trees in the region. Combined with the regional vegetation coverage data, classification is performed using vegetation indices (such as NDVI), and the planting density of trees in each grid is further determined. For example, areas with high vegetation indexes indicate that the density of greening or tree planting is large, while low areas indicate that the trees are sparse. Afterwards, spatial analysis algorithms (such as K-means clustering or nearest neighbor analysis) are used to calculate the tree planting density data in each grid. In terms of soil resource feature integration, combined with soil sensor data in the grid, spatial feature analysis is performed using soil types (such as sand, clay, etc.) and physical properties (such as soil moisture, pH). Specifically, data is collected through ground sensors, a soil resource evaluation model based on a physical model is established, each grid is classified and weighted, and grid soil resource feature data is obtained.

[0090] Step S143: performing grid tree growth feature recognition based on the grid camera data to obtain grid tree growth feature data, and performing grid tree resource feature integration based on the grid tree growth feature data and the grid tree planting density data to obtain grid tree resource feature data;

[0091] In this embodiment, the identification of grid tree growth characteristics relies on the analysis of high-resolution images, and uses computer vision technology, such as convolutional neural networks (CNN) in deep learning, to identify the growth status of trees. By identifying the height, crown width, leaf area and other characteristics of trees, the forest growth characteristic data of the grid is generated. At the same time, combined with the tree planting density data, the impact of the growth status of trees in each grid on growth is analyzed to integrate resources. Using phenological data (such as spring budding, summer growth, etc.) and growth models, the growth trend of trees is predicted through time series data analysis, and combined with the tree planting density, the forest resource status of the grid is evaluated.

[0092] Step S144: Calculating the grid water body coverage rate of the grid remote sensing water body index data to obtain water body grid coverage rate data, and performing grid water body spatial distribution statistics based on the water body grid coverage rate data to obtain grid water body spatial distribution data;

[0093] In this embodiment, the calculation of grid water coverage mainly relies on remote sensing images and water body index (such as NDWI). The water area in each grid is extracted through remote sensing images taken by satellites or drones. Image processing technology is used, combined with the reflective characteristics of the water body, to classify water body and non-water body pixels, thereby calculating the water body coverage. Spatial distribution statistics are performed through GIS technology, and the water body coverage data of each grid is correlated with the surrounding grid data to draw a spatial distribution map of the water body. Specifically, sample detection can be carried out in a typical area, and the algorithm model can be corrected by comparing with the actual data on the ground.

[0094] Step S145: integrating grid water resource characteristics of grid water spatial distribution data according to grid water sensor data, thereby obtaining grid water resource characteristic data;

[0095] In this embodiment, the spatial distribution data of the grid water body is integrated according to the data collected by the water body sensor. The water body sensor can measure water quality indicators (such as dissolved oxygen, pH value, conductivity, etc.), and combine the remote sensing water body index to perform water resource characteristic analysis. Through spatial data fusion technology (such as Kalman filtering or Bayesian method), the sensor data of the water body is combined with the spatial distribution data to generate comprehensive characteristic data of the grid water body resources. These data reflect the spatial changes, stock and other characteristics of water resources.

[0096] Step S146: Performing grid resource spatial integration on the grid forest resource characteristic data, the grid soil resource characteristic data and the grid water resource characteristic data, thereby obtaining forest area grid resource status data.

[0097] In this embodiment, a geographic information system (GIS) is used to spatially integrate grid forest resource characteristic data, grid soil resource characteristic data, and grid water resource characteristic data to form a comprehensive grid resource status data. The resource status of each grid is integrated through multi-dimensional data fusion methods (such as weighted average method, principal component analysis method, etc.). This process requires standardization of different data sources (such as soil quality, forest planting density, water body coverage, etc.) to ensure that they are synthesized under the same dimension. During implementation, a GIS platform can be used to overlay and analyze different resource data, draw a comprehensive resource distribution map, and evaluate the overall resource status of the forest area and its spatial distribution.

[0098] Optionally, step S15 is specifically:

[0099] Perform regional water body coverage time series analysis on water body grid coverage data to obtain regional water body coverage time series change data;

[0100] In this embodiment, satellite remote sensing image data (such as Landsat or Sentinel images) and ground observation data are collected to ensure that the temporal and spatial resolution of the data is suitable for time series analysis. For example, the 30m spatial resolution of Landsat images is used to obtain image data once a year. Water body information is extracted from remote sensing images using a water body identification algorithm (such as NDWI or MNDWI index), and then the area is divided into grids (such as 1km×1km grids) using GIS software. The water body coverage is extracted from each grid unit, and the water body coverage rate in each grid is calculated. The water body coverage rate of different grid units is analyzed in time series using statistical analysis methods (such as trend analysis or sliding average method) to obtain the water body coverage rate time series data of each grid. These data can reveal trends such as interannual changes and seasonal fluctuations in water body coverage. For example, the rasterio library in Python is used for image reading and writing, and the numpy and pandas libraries are used for calculation and statistics of time series data, and finally the water body coverage rate change curve can be drawn through visualization tools such as Matplotlib.

[0101] Conduct time series analysis on regional vegetation coverage data to obtain time series change data on regional vegetation coverage;

[0102] In this embodiment, remote sensing images (such as MODIS NDVI data or Landsat images) and ground observation data are obtained. The spatiotemporal resolution of MODIS data is suitable for the temporal change analysis of vegetation coverage (such as 16-day cycle, 250m spatial resolution). NDVI (normalized difference vegetation index) value is used as a metric for vegetation coverage. The range of NDVI values ​​is usually between -1 and +1, and the higher the value, the better the vegetation coverage. The area is divided into grids, the average NDVI value in each grid cell is calculated, and the vegetation coverage is determined according to the set threshold. Time series analysis methods such as autoregressive model (ARIMA) or seasonal decomposition are used to analyze the changing trend of vegetation coverage in grids in different regions, and explore the seasonal and long-term change characteristics of vegetation changes. Image data processing is performed using the Google Earth Engine (GEE) platform or the geemap and rasterio libraries in Python, and the temporal data is modeled and predicted through the statsmodels library, and finally a temporal change diagram of vegetation coverage is drawn.

[0103] Based on the regional water coverage time series change data and regional vegetation coverage time series change data, the forest area macro resource trend analysis is carried out to obtain the forest area macro resource trend data;

[0104] In this embodiment, the water coverage rate and vegetation coverage rate data are combined, and the macro-resource characteristics of the region are extracted through multivariate statistical analysis or machine learning methods (such as principal component analysis PCA or cluster analysis). For example, the water body and vegetation coverage rate data are standardized and then PCA dimension reduction is performed to obtain a principal component as a macro trend indicator of forest resources. Time series analysis (such as trend test, regression analysis) is used to identify the long-term and short-term trends of changes in water body and vegetation coverage. Changes in these macro resources can provide decision support for forest management. For example, analyzing whether the vegetation coverage rate in a certain area shows a continuous downward trend, indicating the existence of forest degradation problems. The scikit-learn library can be used for data standardization and PCA analysis, and statsmodels can be used for regression analysis, and finally the analysis results can be visualized, such as drawing a trend chart.

[0105] According to the forest area grid resource status data, the forest area resource status fluctuation statistics are carried out to obtain the non-periodic forest area resource status change data and the periodic forest area resource status change data;

[0106] In this embodiment, the degree of fluctuation of the resource status is calculated by analyzing the volatility of the time series data in the forest grid resource status data (such as standard deviation, root mean square error, etc.). Then, based on the frequency and amplitude of the fluctuation, it is determined whether there is a periodic change. Periodic changes can be extracted by Fourier transform or wavelet transform, while non-periodic changes can be identified by anomaly detection methods (such as algorithms based on change point detection). Fourier transform is performed using scipy in Python, and change point detection is performed using the ruptures library.

[0107] Conduct forest micro-resource trend analysis based on non-periodic forest resource status change data, thereby obtaining forest micro-resource trend data;

[0108] In this embodiment, a fine-grained trend analysis method, such as local weighted regression (LOWESS) or support vector machine (SVM) regression model, is used to fit the resource change trend in the region. The support vector machine regression model in scikit-learn can be used for trend modeling, and seaborn can be used for data visualization to show the resource trend changes in the micro region.

[0109] Multi-layer resource trend coupling of forest area macro-resource trend data and forest area micro-resource trend data is carried out to obtain forest area resource trend data.

[0110] In this embodiment, the macro resource trend is coupled with the micro resource trend to analyze and establish a multi-level coupling model (such as a multi-layer perceptron, a neural network, etc.). Through the interaction of data at different levels, the key factors affecting the resource status of the forest area are extracted. By establishing a neural network model or a hierarchical clustering method, combined with time series data and regional characteristics, the coupling trend of the macro and micro levels is extracted. For example, the correlation between water changes and vegetation changes, and their impact on the dynamics of forest resources are analyzed. Keras or tensorflow can be used to build a multi-layer perceptron neural network for coupling model training, and matplotlib can be used to display the coupling trend of resource changes at each level.

[0111] Optionally, step S2 specifically includes:

[0112] Step S21: Acquire historical meteorological data of the forest area, and spatially align the forest area meteorological pattern data, forest area grid resource status data, and forest area resource trend data, so as to obtain aligned historical meteorological data, aligned forest area grid resource status data, and aligned forest area resource trend data;

[0113] In this embodiment, historical meteorological data of the forest area are obtained from a meteorological station or remote sensing platform, including parameters such as temperature, precipitation, humidity, and wind speed. These data are usually recorded in units of days or months. Then, the historical meteorological data, the aligned forest area resource trend data, and the grid resource status data of the forest area are spatially aligned through a geographic information system (GIS) to ensure that these data correspond to the same geographic grid position. Specifically, data from different sources can be spatially matched through interpolation algorithms (such as Kriging interpolation or inverse distance weighted method) to obtain a consistent spatiotemporal data set. These aligned data provide a reliable basis for subsequent meteorological pattern recognition and resource distribution analysis.

[0114] Step S22: performing forest area meteorological pattern recognition on the aligned historical meteorological data, thereby obtaining forest area meteorological pattern data, and constructing a forest area meteorological change model based on the forest area meteorological pattern data;

[0115] In this embodiment, pattern recognition is performed on the aligned historical meteorological data. The main patterns of forest meteorology are identified through data mining algorithms, such as K-means clustering or principal component analysis (PCA). For example, climate features such as "dry season pattern" or "wet season pattern" can be identified. Then, based on these pattern data, a forest meteorological change model is constructed. Specifically, regression analysis or time series prediction methods (such as ARIMA model) are used to predict future meteorological change trends. In addition, machine learning methods, such as support vector machines (SVM) or random forest models, can be introduced for nonlinear prediction. This model will help evaluate the long-term change trend of forest meteorology and provide decision support for forest resource management.

[0116] Step S23: describing the spatial distribution of forest resources according to the aligned forest grid resource status data and the aligned forest resource trend data, thereby obtaining a forest resource distribution spatial model;

[0117] In this embodiment, the aligned forest grid resource status data and resource trend data are used to describe the resource distribution of the forest area through a spatial analysis method. For example, raster data (such as land use type, forest coverage, species distribution, etc.) can be used to generate a resource distribution map of the forest area. The spatial distribution description can use a spatial interpolation method (such as IDW interpolation) to combine resource data from different grids to draw a high-precision resource distribution map. On this basis, a spatial statistical model (such as a spatial autoregressive model SAR) is used to establish a spatial model of forest resource distribution, which can reveal the distribution law and change trend of resources in different spatial locations.

[0118] Step S24: Perform physical layer fusion on the forest area meteorological change model and the forest area resource distribution spatial model, so as to obtain a forest area resource status model;

[0119] In this embodiment, the meteorological change model and the resource distribution spatial model are integrated to form a forest resource status model that can reflect the relationship between meteorology and resources. The specific method can use a multiple linear regression model or a more complex machine learning model (such as a neural network) to introduce the interaction terms of meteorological pattern data and resource distribution data into the model to simulate the impact of meteorological changes on resource status. For example, in the case of drought, the model can predict the decline in forest coverage or the reduction in soil moisture. Through this fusion process, a dynamic and comprehensive model that can reflect resource changes under different climatic conditions is finally constructed.

[0120] Step S25: Divide the forest resource status model into seasonal resource statuses, thereby obtaining a seasonal forest resource status model set.

[0121] In this embodiment, seasonal division is performed based on the obtained forest resource status model. The forest resource status is classified according to the characteristics of seasonal climate change (such as spring, summer, autumn, and winter). Seasonally changing meteorological data (such as temperature and precipitation) and the output of the forest resource model can be used to identify the resource status characteristics of different seasons through time series analysis. For example, spring is the peak season for forest growth, while winter is the resource consumption period. Based on these characteristics, a seasonal resource status model set is constructed, which will help forest managers make targeted resource management decisions in different seasons. In specific implementation, machine learning algorithms (such as decision trees) or rule-based models can be used for seasonal division.

[0122] Optionally, step S25 is specifically:

[0123] Step S251: dividing the forest area meteorological pattern data into forest area seasonal meteorological patterns, thereby obtaining forest area seasonal meteorological pattern data;

[0124] In this embodiment, a clustering algorithm (such as K-means or DBSCAN) is used to divide the patterns of these data. Different seasonal meteorological patterns are divided according to the seasonal fluctuation characteristics of meteorological data, for example, the data is divided into spring pattern, summer pattern, autumn pattern and winter pattern. In the specific implementation, the scikit-learn library in Python is used for clustering, and an appropriate number of clusters (for example, 4 categories) is selected, and each type of data is verified and optimized according to the changing law of meteorological factors to ensure that the seasonal pattern has good discrimination. Finally, a forest seasonal meteorological pattern data set is generated, which contains the meteorological parameter characteristics corresponding to each seasonal pattern.

[0125] Step S252: performing forest seasonal period identification on the periodic forest resource state change data, thereby obtaining forest seasonal resource state change data;

[0126] In this embodiment, a time series analysis is performed on the resource status of the forest area (such as forest coverage, vegetation growth status, soil moisture, etc.). Fourier transform (FFT) or wavelet transform is used to identify seasonal cycles in resource status data. For example, assuming that a period of 12 months of forest resource change data is collected, the periodic component can be extracted by Fourier transform to determine the seasonal fluctuation period of the resource status. These resource status data are processed using a period identification algorithm (such as FFT) to extract seasonal change cycles (for example, vegetation recovery in spring and autumn each year, and resource consumption in summer). Based on this analysis, a seasonal resource status change data set for the forest area is generated to describe the trend of resource status changes in each cycle.

[0127] Step S253: integrating the seasonal patterns of the forest area based on the seasonal meteorological pattern data of the forest area and the seasonal resource status change data of the forest area, thereby obtaining the seasonal pattern data of the forest area;

[0128] In this embodiment, the seasonal meteorological pattern data of the forest area is combined with the seasonal resource state change data, and a multivariate analysis method (such as principal component analysis PCA or multidimensional scaling analysis MDS) is used for data integration. Specifically, a multidimensional data set can be constructed, in which each dimension represents a seasonal meteorological pattern (such as temperature, precipitation, etc.), and the other dimension represents the seasonal resource state change related thereto (such as vegetation growth, soil moisture change, etc.). Through principal component analysis (PCA), the multidimensional data is reduced into several main components, and the main correlation between meteorological patterns and resource state changes is identified. For example, the main connection between temperature and precipitation changes and vegetation growth can be identified, and then a seasonal pattern data set containing integrated meteorological and resource state information can be constructed.

[0129] Step S254: dividing the forest resource status model into seasonal resource status according to the forest seasonal pattern data, thereby obtaining a seasonal forest resource status model set.

[0130] In this embodiment, the seasonal pattern data of the forest area is used to seasonally divide the forest area resource status model. According to the characteristics of seasonal meteorological patterns and resource status changes, different seasonal resource status categories are set (for example, spring vegetation recovery, summer drought, high temperature season resource depletion, etc.). Then, the forest area resource status model is trained using a decision tree algorithm (such as CART) or a support vector machine (SVM) to divide the resource status model into different seasonal categories. In a specific implementation, historical data (such as meteorological data and resource change data from the past 20 years) can be used to train the model to ensure that each seasonal status category can accurately reflect the resource status changes in the forest area. For example, the DecisionTreeClassifier in Python's scikit-learn library is used to implement the division of seasonal resource status. Finally, a seasonal forest area resource status model set is generated, which contains the resource status classification results in different seasons for subsequent forest management and resource monitoring.

[0131] Optionally, step S3 specifically includes:

[0132] According to the seasonal forest resource status model set, the seasonal environment of the forest area is integrated to obtain the seasonal environmental data of the forest area, and the seasonal environmental characteristics of the forest area seasonal environmental data are divided to obtain the seasonal hydrological data and seasonal soil data of the forest area;

[0133] In this embodiment, a seasonal forest resource state model is adopted to integrate resource state models of different seasons according to a geographic information system (GIS) to form a unified spatiotemporal framework. The characteristics of the resource state model of each season are characterized to obtain environmental characteristics such as temperature, humidity, precipitation, soil moisture, and light in each season. Based on the integrated data, K-means clustering analysis or a fuzzy clustering algorithm based on adaptive segmentation is used to classify and divide the data to identify environmental characteristics of different seasons (spring, summer, autumn, and winter). According to the seasonal environmental characteristics divided, relevant hydrological data (such as water level, flow, soil moisture changes) and soil data (such as pH value, soil organic matter content, and soil structure) are extracted for seasonal analysis to obtain seasonal variation data of hydrology and soil. Remote sensing data and ground monitoring data (such as soil moisture sensor data) are used to process data through a data analysis platform (such as Python's Pandas and Numpy), and hydrological and soil data of each region are output by season.

[0134] Obtaining forest growth rules, including forest growth cycle rules, forest suitable growth soil rules, forest suitable growth hydrological rules and forest suitable growth terrain rules;

[0135] In this embodiment, based on the existing research literature on tree growth, the growth patterns of trees are extracted, including the growth cycle (such as from germination to maturity), suitable hydrological and soil environment (such as humidity, pH value, etc.), and terrain requirements (such as slope, altitude, etc.).

[0136] According to the rules of forest suitable growth terrain, the forest terrain suitable area is selected for the remote sensing data to be analyzed, so as to obtain the forest terrain suitable area data;

[0137] In this embodiment, according to the terrain rules for tree growth (for example, the slope is required to be less than 15° and the altitude is between 500 and 1500 meters), terrain data (such as DEM data) is obtained through remote sensing images to perform terrain analysis, and areas that meet the terrain requirements are screened out. GIS tools (such as the spatial analysis function of ArcGIS) are used to process the terrain data, mark the areas that meet the conditions, and form forest terrain suitable area data. The "slope analysis" tool in GIS is used to calculate the slope of the area based on the DEM data obtained by remote sensing, and screen out terrain areas suitable for specific tree species (such as Chinese fir).

[0138] According to the hydrological rules suitable for tree growth, seasonal hydrological data of forest areas are screened for suitable areas for tree hydrology, so as to obtain data on suitable areas for tree hydrology;

[0139] In this embodiment, according to the requirements of the hydrological conditions for the growth of trees (such as suitable annual precipitation, soil moisture retention capacity, etc.), the seasonal hydrological data of the forest area are screened to identify the hydrologically suitable areas that meet the conditions. A hydrological model (such as the SWAT model) or an algorithm based on statistical analysis can be used to analyze the annual precipitation, soil moisture changes and their effects on the growth of trees in each region, and finally obtain a suitable hydrological area. For example, the SWAT model is used to simulate the hydrological changes in different seasons in the forest area, and the hydrological area suitable for the growth of Chinese fir is identified, such as the area with an annual precipitation between 800 and 1200 mm. For example, for tree species (such as poplar) in humid areas, a higher annual precipitation and good soil permeability are required, while for drought-resistant tree species (such as sea buckthorn), they are more adapted to dry climates. ArcGIS is used to perform spatiotemporal analysis of hydrological data, and the area with suitable hydrological conditions is finally screened out by superimposing the matching degree of hydrological conditions and the suitable hydrological requirements of trees.

[0140] Based on the tree-suitable growth climate rules, seasonal soil data of forest areas are screened for tree-suitable soil areas, thereby obtaining tree-suitable soil area data;

[0141] In this embodiment, remote sensing technology and ground monitoring data are used to classify soil types and qualities based on the requirements of forest trees for soil conditions (such as pH value, soil fertility, organic matter content, etc.), and soil suitability is evaluated in combination with soil analysis models (such as soil erosion models). GIS tools are used to perform spatial distribution analysis on soil data to screen out soil areas that meet the requirements for forest growth. For example, if certain tree species (such as oaks) prefer deep, fertile, acidic soils, spatial analysis is performed through soil type and pH value data to screen out areas that meet the conditions. Soil data sets (such as USDA soil data) are superimposed with regional environmental data to determine soil areas that meet the requirements for tree growth.

[0142] The forest terrain suitable area data, forest hydrology suitable area data and forest soil suitable area data are regionally superimposed to obtain forest suitable area data.

[0143] In this embodiment, the spatial data of terrain suitable areas, hydrological suitable areas and soil suitable areas are superimposed, and the spatial superposition analysis method (such as "raster superposition") is used to identify the areas with the best environmental conditions. The final result is a comprehensive suitability area map, marking the specific spatial areas suitable for tree growth. After obtaining the forest terrain suitable areas, forest hydrological suitable areas and forest soil suitable areas, a GIS platform (such as ArcGIS or QGIS) is used for spatial superposition analysis. The spatial data layers of each suitable area are superimposed, and a comprehensive forest suitable growth area map is finally obtained through weighted analysis and weighted averaging. For example, the hydrological conditions are weighted 40%, the soil conditions are weighted 30%, and the terrain conditions are weighted 30%, and a comprehensive suitable area is obtained.

[0144] Optionally, step S4 is specifically:

[0145] Step S41: extracting seasonal growth characteristics of regional trees from the seasonal forest resource status model set, thereby obtaining seasonal growth characteristic data of regional trees;

[0146] In this embodiment, the growth characteristics of different tree species in the region are extracted by using remote sensing data and ground observation data and using the seasonal forest resource status model. Specifically, the changes in the forest area at different time points in each season are obtained through satellite remote sensing images, such as chlorophyll content (NDVI), tree height, canopy density and other growth characteristics, and combined with meteorological data (temperature, precipitation, etc.), a regression model is established to extract the growth characteristics of trees in a specific area in different seasons. Finally, the generated seasonal growth characteristic data includes parameters such as the tree growth index at each time point.

[0147] Step S42: Estimating the regional forest growth cycle according to the regional forest seasonal growth characteristic data, thereby obtaining regional forest growth cycle data;

[0148] In this embodiment, based on the extracted seasonal tree growth characteristic data, a growth model (such as a LOGISTIC model or an empirical formula based on the growing season) is used to periodically estimate the tree growth process. For example, for a specific forest area, the time nodes of the start and end growth of the trees are first determined by the seasonally changing vegetation index (such as the peak value of NDVI), and then the complete growth cycle of the trees is estimated. The length of the growth cycle (such as the number of days from germination to dormancy) and the characteristics of each stage (such as accelerated growth period, slow growth period, etc.) can be calibrated through historical growth data and meteorological data models to ultimately generate regional forest growth cycle data.

[0149] Step S43: performing regional forest growth cycle error calculation on the forest growth cycle rule and the regional forest growth cycle data, thereby obtaining regional forest growth cycle error data;

[0150] In this embodiment, the growth cycle error is calculated based on the tree growth cycle data estimated in the previous stage and combined with the actual observation data in the region. In specific implementation, the estimated growth cycle is first compared with the actually measured tree growth cycle (for example, data obtained through ground surveys, sample collection, etc.) to calculate the error. The error calculation can use statistical analysis methods, such as mean square error (MSE) or mean absolute error (MAE), to further analyze the distribution characteristics of the error. Regional growth cycle error data can provide error data for different types of trees (for example, the error may vary in different regions) and provide a basis for subsequent model optimization.

[0151] Step S44: performing statistics on trees with high growth cycle errors based on the regional tree growth cycle error data, thereby obtaining tree data with high growth cycle errors;

[0152] In this embodiment, trees with large growth cycle errors are screened out, that is, those tree populations whose errors in growth cycle estimation exceed a certain set threshold (such as more than 10%). Then, these trees are statistically analyzed to record the distribution of their growth cycle errors, tree species, and growth environment characteristics. This process can be achieved through cluster analysis, for example, using the K-means algorithm to classify trees with large errors, thereby obtaining group characteristic data of trees with high growth cycle errors. These data can provide support for subsequent tree suitable areas and growth optimization.

[0153] Step S45: extracting suitable areas for trees with high growth cycle errors from the suitable area data for trees with high growth cycle errors according to the tree data with high growth cycle errors, thereby obtaining suitable area data for trees with high growth cycle errors;

[0154] In this embodiment, the tree data with high growth cycle error is used to extract the suitable area for the tree. Specifically, by analyzing the relationship between the growth cycle error of the tree and the environmental factors such as climate, soil, and topography in different regions, the most suitable area for the high growth cycle error tree is extracted from the tree suitable area data using geographic information system (GIS) technology. For example, a spatial interpolation method (such as Kriging or inverse distance weighted method) can be used to determine under what environmental conditions a group of trees with large errors is more suitable for growth, thereby obtaining the suitable area for the group with high growth cycle error. The extracted suitable area data can provide a scientific basis for tree planting.

[0155] Step S46: extracting high growth cycle error tree planting areas from the seasonal forest resource state model set according to the high growth cycle error tree data, thereby obtaining high growth cycle error tree planting area data;

[0156] In this embodiment, based on the data of suitable areas for trees with high growth cycle errors, combined with the seasonal forest resource status model set, the tree planting area is further extracted. Specifically, the planting area of ​​trees with high growth cycle errors is extracted from the seasonal forest resource status model set using geographic information system (GIS) technology to obtain the actual planting area of ​​trees with high growth cycle errors.

[0157] Step S47: Calculate the tree growth offset factor for the high growth cycle error tree suitable area data and the high growth cycle error tree planting area data, so as to obtain the tree growth offset factor.

[0158] In this embodiment, the data of suitable areas for trees with high growth cycle errors and the data of planting areas for trees with high growth cycle errors in the region are used, combined with environmental factors and climate change models, to calculate the tree growth offset factor. Specifically, the tree growth offset factor refers to the offset of the actual growth cycle of the tree relative to the estimated cycle due to changes in factors such as climate, soil, and management measures. By establishing a multivariate regression model or a machine learning algorithm (such as a random forest), the relationship between various environmental parameters and the tree growth cycle error is considered, and the growth offset factor is finally calculated and applied to further adjustment and optimization of the tree growth cycle.

[0159] Optionally, step S47 is specifically:

[0160] Step S471: extracting regional environmental features from the high growth cycle error forest tree suitable area data and the high growth cycle error forest tree planting area data, thereby obtaining forest tree suitable area environmental data and forest tree planting area environmental data;

[0161] In this embodiment, the climate, precipitation, sunshine duration, soil moisture and other environmental characteristics in the region are extracted from the data of suitable areas for trees with high growth cycle errors and the data of tree planting areas with high growth cycle errors. In the data of suitable areas, special attention is paid to factors that may affect the growth cycle of trees, such as temperature, humidity, and soil pH; in the data of tree planting areas, analysis is performed based on factors such as land use type, vegetation coverage, slope, and planting density in the planting area. By using machine learning methods (such as random forests or support vector machines), the environmental characteristics of these areas are further extracted to generate environmental data of suitable areas for trees and environmental data of tree planting areas. Remote sensing technology (such as satellite images and drone data) can also be used to obtain climate data (temperature, precipitation), soil properties (pH value, organic matter content) and topographic features (slope, altitude, etc.) of suitable areas for trees and tree planting areas.

[0162] Step S472: performing environmental error calculation on the forest suitable area environmental data and the forest planting area environmental data, thereby obtaining planting environment error data;

[0163] In this embodiment, the error analysis method is used to calculate the environmental data of the forest suitable area and the forest planting area. First, the historical meteorological data and soil data are used as a benchmark to calculate the difference between the actual measurement value and the ideal value. Common error calculation methods include root mean square error (RMSE) or absolute error (MAE). For example, in the environmental data of the forest suitable area, if the temperature data deviates greatly from the temperature value measured in the environmental data of the forest planting area, the planting environment error data is generated by calculating the difference. In addition, the covariance between environmental factors can also be considered to identify the main factors affecting the source of error, so as to more accurately evaluate the error data.

[0164] Step S473: extracting tree planting density features from the tree planting area data with high growth cycle errors, thereby obtaining tree planting density data;

[0165] In this embodiment, tree planting density characteristics are extracted from the tree planting area data with high growth cycle errors. The specific operation includes obtaining the distribution density data of trees in the planting area through remote sensing images or ground surveys. Image processing techniques, such as segmentation algorithms (such as the Otsu method), are used to identify and classify the trees in the image and calculate the number of trees per unit area. The area of ​​each planting area is divided according to the geographic coordinate system to obtain tree planting density data. This data can help analyze the growth status of trees at different densities and their relationship with environmental factors.

[0166] Step S474: obtaining forest tree species characteristic data, and performing growth density deviation calculation on the forest tree species characteristic data and forest tree planting density data, thereby obtaining planted tree growth density deviation data;

[0167] In this embodiment, data on tree species characteristics are collected based on existing forest growth research literature, including tree growth rate, root expansion, photosynthesis efficiency, etc. These characteristics are then combined with the tree planting density data extracted in the previous step, and a growth density deviation model is established for calculation. If the tree density in a certain area is high, but the growth characteristics of the tree species are not suitable for this density, a growth density deviation will occur. These data are analyzed using a biostatistical model (such as linear regression or nonlinear regression) to calculate the growth density deviation of each tree species at different densities.

[0168] Step S475: Perform growth cycle error multivariate regression analysis based on regional forest growth cycle error data, planted tree growth density deviation data, and planting environment error data to obtain a forest growth deviation factor.

[0169] In this embodiment, a multiple regression analysis method is used to combine the regional growth cycle error data, the planted tree growth density deviation data, and the planting environment error data for comprehensive analysis. The multiple regression model can help find the relationship between these variables and the forest growth deviation and calculate the growth deviation factor. In specific implementation, a regression model containing multiple independent variables (environmental error, density deviation, etc.) and a dependent variable (growth deviation factor) is first constructed. The least squares method (OLS) is used to estimate the regression coefficients to obtain the degree of influence of each variable on the forest growth deviation. Finally, through the regression model, the deviation factor related to forest growth is obtained, thereby providing data support for further forestry management and planting optimization.

[0170] Optionally, the present invention further provides a smart forestry data management system based on data fusion, which is used to execute the smart forestry data management method based on data fusion as described above, and the smart forestry data management system based on data fusion includes:

[0171] The forest area resource status integration module is used to obtain forest area sensor data and forest area remote sensing data, and integrate forest area grid resource status according to forest area sensor data and forest area remote sensing data, so as to obtain forest area grid resource status data; perform forest area resource trend analysis based on forest area grid resource status data, so as to obtain forest area resource trend data;

[0172] The seasonal resource status classification module is used to perform physical layer fusion of multi-source data based on forest area grid resource status data and forest area resource trend data, so as to obtain a forest area resource status model; perform seasonal resource status classification on the forest area resource status model, so as to obtain a seasonal forest area resource status model set;

[0173] The forest suitable area division module integrates the seasonal environment of the forest area according to the seasonal forest area resource status model set, thereby obtaining the seasonal environmental data of the forest area, and divides the forest suitable area based on the seasonal environmental data of the forest area, thereby obtaining the forest suitable area data;

[0174] A growth offset factor calculation module is used to estimate the regional forest growth cycle according to the seasonal forest resource status model set, thereby obtaining regional forest growth cycle data, and calculate the forest growth offset factor based on the regional forest growth cycle data and forest suitable area data, thereby obtaining the forest growth offset factor;

[0175] The growth health assessment module is used to assess the growth health of trees in the forest area based on the tree growth deviation factor, so as to obtain the growth health data of trees in the forest area, and optimize the planting strategy of the forest area based on the growth health data of trees in the forest area and the data of suitable areas for trees, so as to obtain the planting management strategy of the forest area.

[0176] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0177] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A smart forestry data management method based on data fusion, characterized in that: The following steps are involved: Step S1: Acquire forest area sensor data and forest area remote sensing data, and integrate forest area grid resource status according to the forest area sensor data and forest area remote sensing data, so as to obtain forest area grid resource status data; Conduct forest resource trend analysis based on forest grid resource status data to obtain forest resource trend data; Step S2: Based on the forest area grid resource status data and forest area resource trend data, multi-source data physical layer fusion is performed to obtain a forest area resource status model; the forest area resource status model is divided into seasonal resource states to obtain a seasonal forest area resource status model set; Step S2 is specifically as follows: Step S21: Acquire historical meteorological data of the forest area, and spatially align the forest area meteorological pattern data, forest area grid resource status data, and forest area resource trend data, so as to obtain aligned historical meteorological data, aligned forest area grid resource status data, and aligned forest area resource trend data; Step S22: performing forest area meteorological pattern recognition on the aligned historical meteorological data, thereby obtaining forest area meteorological pattern data, and constructing a forest area meteorological change model based on the forest area meteorological pattern data; Step S23: describing the spatial distribution of forest resources according to the aligned forest grid resource status data and the aligned forest resource trend data, thereby obtaining a forest resource distribution spatial model; Step S24: Perform physical layer fusion on the forest area meteorological change model and the forest area resource distribution spatial model, so as to obtain a forest area resource status model; Step S25: dividing the forest resource state model into seasonal resource states, thereby obtaining a seasonal forest resource state model set; Step S25 is specifically as follows: Step S251: dividing the forest area meteorological pattern data into forest area seasonal meteorological patterns, thereby obtaining forest area seasonal meteorological pattern data; Step S252: performing forest seasonal period identification on the periodic forest resource state change data, thereby obtaining forest seasonal resource state change data; Step S253: integrating the seasonal patterns of the forest area based on the seasonal meteorological pattern data of the forest area and the seasonal resource status change data of the forest area, thereby obtaining the seasonal pattern data of the forest area; Step S254: dividing the forest resource state model into seasonal resource states according to the forest seasonal pattern data, thereby obtaining a seasonal forest resource state model set; Step S3: Performing forest area seasonal environment integration according to the seasonal forest area resource status model set, thereby obtaining forest area seasonal environment data, and performing tree suitable area division based on the forest area seasonal environment data, thereby obtaining tree suitable area data; Step S3 is specifically as follows: According to the seasonal forest resource status model set, the seasonal environment of the forest area is integrated to obtain the seasonal environmental data of the forest area, and the seasonal environmental characteristics of the forest area seasonal environmental data are divided to obtain the seasonal hydrological data and seasonal soil data of the forest area; Obtaining forest growth rules, including forest growth cycle rules, forest suitable growth soil rules, forest suitable growth hydrological rules and forest suitable growth terrain rules; According to the rules of forest suitable growth terrain, the forest terrain suitable area is selected for the remote sensing data to be analyzed, so as to obtain the forest terrain suitable area data; According to the hydrological rules suitable for tree growth, seasonal hydrological data of forest areas are screened for suitable areas for tree hydrology, so as to obtain data on suitable areas for tree hydrology; Based on the tree-suitable growth climate rules, seasonal soil data of forest areas are screened for tree-suitable soil areas, thereby obtaining tree-suitable soil area data; Perform regional spatial superposition on forest topography suitable area data, forest hydrology suitable area data and forest soil suitable area data to obtain forest suitable area data; Step S4: Estimating the regional tree growth cycle according to the seasonal forest resource status model set, thereby obtaining regional tree growth cycle data, and calculating the tree growth offset factor based on the regional tree growth cycle data and the tree suitable area data, thereby obtaining the tree growth offset factor; Step S5: Evaluate the health of forest tree growth in the forest area based on the forest growth deviation factor to obtain the data on forest tree growth health in the forest area, and optimize the forest planting strategy based on the forest tree growth health data and the data on suitable areas for trees to obtain the forest planting management strategy.

2. The smart forestry data management method based on data fusion according to claim 1 is characterized in that: Step S1 is specifically as follows: Step S11: Acquire forest area sensor data and forest area remote sensing data, and perform data preprocessing on the forest area sensor data and forest area remote sensing data respectively, so as to obtain sensor data to be analyzed and remote sensing data to be analyzed; Step S12: Calculating the regional vegetation coverage rate of the remote sensing data to be analyzed, thereby obtaining regional vegetation coverage rate data, and dividing the remote sensing data to be analyzed into forest area remote sensing grids according to the regional vegetation coverage rate data, thereby obtaining forest area grid remote sensing data; Step S13: extracting sensor location features from the sensor data to be analyzed, thereby obtaining sensor location data, and fusing forest area grid data with the forest area grid remote sensing data and the sensor data to be analyzed according to the sensor location data, thereby obtaining forest area grid fusion data; Step S14: Integrate the forest area grid resource status according to the forest area grid fusion data, so as to obtain the forest area grid resource status data; Step S15: Perform forest resource trend analysis based on the forest grid resource status data to obtain forest resource trend data.

3. The smart forestry data management method based on data fusion according to claim 2 is characterized in that: Step S14 is specifically as follows: Step S141: performing grid camera extraction, grid soil feature extraction and grid water feature extraction on the forest area grid fusion data, thereby obtaining grid camera data, grid soil sensing data, grid remote sensing water index data and grid water sensing data; Step S142: Calculate the grid tree planting density according to the regional vegetation coverage data, thereby obtaining grid tree planting density data, and integrate the grid soil resource characteristics according to the grid soil sensor data and the grid tree spatial planting density data, thereby obtaining grid soil resource characteristic data; Step S143: performing grid tree growth feature recognition based on the grid camera data to obtain grid tree growth feature data, and performing grid tree resource feature integration based on the grid tree growth feature data and the grid tree planting density data to obtain grid tree resource feature data; Step S144: Calculating the grid water body coverage rate of the grid remote sensing water body index data to obtain water body grid coverage rate data, and performing grid water body spatial distribution statistics based on the water body grid coverage rate data to obtain grid water body spatial distribution data; Step S145: integrating grid water resource characteristics of grid water spatial distribution data according to grid water sensor data, thereby obtaining grid water resource characteristic data; Step S146: Performing grid resource spatial integration on the grid forest resource characteristic data, the grid soil resource characteristic data and the grid water resource characteristic data, thereby obtaining forest area grid resource status data.

4. The smart forestry data management method based on data fusion according to claim 2 is characterized in that: Step S15 is specifically as follows: Perform regional water body coverage time series analysis on water body grid coverage data to obtain regional water body coverage time series change data; Conduct time series analysis on regional vegetation coverage data to obtain time series change data on regional vegetation coverage; Based on the regional water coverage time series change data and regional vegetation coverage time series change data, the forest area macro resource trend analysis is carried out to obtain the forest area macro resource trend data; According to the forest area grid resource status data, the forest area resource status fluctuation statistics are carried out to obtain the non-periodic forest area resource status change data and the periodic forest area resource status change data; Conduct forest micro-resource trend analysis based on non-periodic forest resource status change data, thereby obtaining forest micro-resource trend data; Multi-layer resource trend coupling of forest area macro-resource trend data and forest area micro-resource trend data is carried out to obtain forest area resource trend data.

5. The smart forestry data management method based on data fusion according to claim 1 is characterized in that: Step S4 is specifically as follows: Step S41: extracting seasonal growth characteristics of regional trees from the seasonal forest resource status model set, thereby obtaining seasonal growth characteristic data of regional trees; Step S42: Estimating the regional forest growth cycle according to the regional forest seasonal growth characteristic data, thereby obtaining regional forest growth cycle data; Step S43: performing regional forest growth cycle error calculation on the forest growth cycle rule and the regional forest growth cycle data, thereby obtaining regional forest growth cycle error data; Step S44: performing statistics on trees with high growth cycle errors based on the regional tree growth cycle error data, thereby obtaining tree data with high growth cycle errors; Step S45: extracting suitable areas for trees with high growth cycle errors from the suitable area data for trees with high growth cycle errors according to the tree data with high growth cycle errors, thereby obtaining suitable area data for trees with high growth cycle errors; Step S46: extracting high growth cycle error tree planting areas from the seasonal forest resource state model set according to the high growth cycle error tree data, thereby obtaining high growth cycle error tree planting area data; Step S47: Calculate the tree growth offset factor for the high growth cycle error tree suitable area data and the high growth cycle error tree planting area data, so as to obtain the tree growth offset factor.

6. The smart forestry data management method based on data fusion according to claim 5 is characterized in that: Step S47 is specifically as follows: Step S471: extracting regional environmental features from the high growth cycle error forest tree suitable area data and the high growth cycle error forest tree planting area data, thereby obtaining forest tree suitable area environmental data and forest tree planting area environmental data; Step S472: performing environmental error calculation on the forest suitable area environmental data and the forest planting area environmental data, thereby obtaining planting environment error data; Step S473: extracting tree planting density features from the tree planting area data with high growth cycle errors, thereby obtaining tree planting density data; Step S474: obtaining forest tree species characteristic data, and performing growth density deviation calculation on the forest tree species characteristic data and forest tree planting density data, thereby obtaining planted tree growth density deviation data; Step S475: Perform growth cycle error multivariate regression analysis based on regional forest growth cycle error data, planted tree growth density deviation data, and planting environment error data to obtain a forest growth deviation factor.

7. A smart forestry data management system based on data fusion, characterized in that: Used to execute the smart forestry data management method based on data fusion as claimed in claim 1, the smart forestry data management system based on data fusion includes: The forest area resource status integration module is used to obtain forest area sensor data and forest area remote sensing data, and integrate forest area grid resource status according to forest area sensor data and forest area remote sensing data, so as to obtain forest area grid resource status data; perform forest area resource trend analysis based on forest area grid resource status data, so as to obtain forest area resource trend data; The seasonal resource status classification module is used to perform physical layer fusion of multi-source data based on forest area grid resource status data and forest area resource trend data, so as to obtain a forest area resource status model; perform seasonal resource status classification on the forest area resource status model, so as to obtain a seasonal forest area resource status model set; The forest suitable area division module integrates the seasonal environment of the forest area according to the seasonal forest area resource status model set, thereby obtaining the seasonal environmental data of the forest area, and divides the forest suitable area based on the seasonal environmental data of the forest area, thereby obtaining the forest suitable area data; A growth offset factor calculation module is used to estimate the regional forest growth cycle according to the seasonal forest resource status model set, thereby obtaining regional forest growth cycle data, and calculate the forest growth offset factor based on the regional forest growth cycle data and forest suitable area data, thereby obtaining the forest growth offset factor; The growth health assessment module is used to assess the growth health of trees in the forest area based on the tree growth deviation factor, so as to obtain the growth health data of trees in the forest area, and optimize the planting strategy of the forest area based on the growth health data of trees in the forest area and the data of suitable areas for trees, so as to obtain the planting management strategy of the forest area.

Images