Optimization method of dynamic pattern of lake wetland habitat considering multiple ecological subjects

Abstract

The application discloses a kind of lake wetland habitat dynamic pattern optimization allocation method considering multiple ecological subjects, including S1, obtaining multiple basic data;S2, data processing and analysis are carried out to basic data;S3, target function is constructed;S4, the constraint condition of habitat pattern optimization allocation is set;S5, optimization allocation model is constructed;S6, multiple basic data are input into optimization allocation model;S7, multiple parameters in optimization allocation model are set, and optimal parameter combination is selected;S8, the optimization allocation result of lake wetland habitat dynamic pattern of target area is output;S9, optimization effect evaluation is carried out.The method considers the habitat dynamic demand of different ecological subjects life stage of lake wetland, realizes the dynamic optimization of lake wetland habitat pattern, establishes the dynamic effective association of pattern and ecological process, perfects lake wetland habitat pattern optimization allocation research theoretical system, and is helpful to realize the benefit maximization of ecological system.

Description

Technical Field

[0001] This invention belongs to the technical field of water ecological protection and restoration, specifically relating to a method for optimizing the dynamic pattern of lake wetland habitats that considers multiple ecological subjects. Background Technology

[0002] The habitat patterns of lakes and wetlands directly impact the energy flow, nutrient cycling, and population dynamics within the ecosystem, thus affecting its stability and health and hindering the effective functioning of ecosystem services. Optimizing habitat patterns is a crucial means of restoring and protecting ecosystems. Commonly used methods include optimization methods, system dynamics, the minimum cumulative resistance coefficient method, and swarm intelligence algorithms. Among these, swarm intelligence algorithms, combined with GIS technology, can simultaneously achieve comprehensive optimization of both quantity and space, and are widely used. Currently, the most commonly used swarm intelligence algorithms include ant colony optimization, particle swarm optimization, and genetic algorithms. Ant colony optimization, with its inherent parallel mechanism and global optimization characteristics, is particularly suitable for solving multi-objective optimization problems and is the most widely applied.

[0003] When using ant colony optimization to solve multi-objective habitat pattern optimization problems, the appropriate setting of the objective function is crucial. Currently, most landscape pattern optimization methods generally consider land use suitability, clustering, and minimum conversion costs in setting the objective function, lacking consideration from the perspective of the habitat needs of the ecological subjects. For lakes and wetlands, due to their unique hydrological rhythm characteristics, rich habitat types, and diverse ecological subjects, the coordination between different habitat types should be particularly considered when determining the optimization objective. Furthermore, in pattern optimization, land use suitability analysis is mostly based on landscape type maps interpreted from remote sensing images of a specific period within the year, failing to achieve a dynamic correlation with ecological processes. Summary of the Invention

[0004] The purpose of this invention is to address the aforementioned shortcomings in the prior art by providing a method for optimizing the dynamic pattern of lake wetland habitats that considers multiple ecological subjects, in order to solve or improve the problems mentioned above.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A method for optimizing the dynamic pattern of lake wetland habitats considering multiple ecological actors includes the following steps:

[0007] S1. Obtain multiple basic data on the dynamic pattern of lake wetland habitats;

[0008] S2. Perform data processing and analysis on the acquired basic data;

[0009] S3. Construct a habitat suitability objective function and a spatial clustering objective function for multiple ecological subjects;

[0010] S4. Set constraints for habitat pattern optimization based on area constraints and type conversion constraints;

[0011] S5. Based on the objective function constructed in step S3 and the constraints set in step S4, construct an optimization configuration model;

[0012] S6. Input the multiple basic data from step S2 into the optimization configuration model;

[0013] S7. Set multiple parameters in the optimization configuration model and select the optimal parameter combination;

[0014] S8. Based on the input of basic data and the selection of parameter combinations, output the optimized configuration results of the dynamic pattern of lake and wetland habitat in the target area;

[0015] S9. Compare the habitat type area, structural composition, and spatial pattern characteristics before and after the optimization of the dynamic pattern of lake wetland habitat, and evaluate the optimization effect.

[0016] Furthermore, the basic data in step S1 includes:

[0017] Vector boundaries of lakes and their respective administrative regions, lake wetland habitat type data, lake wetland land use data, lake topographic data, water level monitoring data, and regional or lake ecological protection plans;

[0018] Among them, the data on lake and wetland habitat types are determined by land use data, lake topographic data, water level monitoring data, and suitable habitat water depth thresholds.

[0019] Furthermore, lake wetland habitat types are classified as follows: cultivated land, forest and grassland, reed terraces, shallow water areas, shallow water areas, deep water areas, construction land, and bare beaches.

[0020] Furthermore, the objective function for habitat suitability for multiple ecological subjects is constructed in step S3 as follows:

[0021]

[0022] α1+α2+α3+α4+α5=1

[0023] Where F represents the sum of habitat suitability of each cell network within the study area, and α1, α2, α3, α4, and α5 represent the weights assigned to different ecological subjects regarding habitat suitability. Let I represent the habitat suitability of different ecological subjects in the y-th month when the habitat type is k at grid (i, j), where I is the number of rows in the cell grid and J is the number of columns in the cell grid.

[0024] Furthermore, the objective function for spatial clustering in step S3 is constructed as follows:

[0025] The degree of agglomeration is represented by a domain identity index. A binary function is defined based on whether the land use type at grid cell (i, j) is k.

[0026]

[0027] Among them, u ij Let (i, j) be the habitat type at grid (i, j), then the neighborhood identity index of habitat type k at (i, j) is... for:

[0028]

[0029] Where (i, j) ≠ (s, t), s is the row number of the neighborhood grid, and t is the column number of the neighborhood grid;

[0030] Using an eight-neighborhood as the window, the spatial clustering objective function F U Defined as:

[0031]

[0032] Furthermore, in step S4, based on area constraints, the constraints for optimizing habitat pattern configuration are set as follows:

[0033] The constraints for the wetland conservation area and the total area of ​​the region are as follows:

[0034]

[0035] in, and Let be the minimum and maximum areas of habitat type k, respectively. For wetland area counting units, when the habitat type at cell (i, j) is k, It is 1 if it is true, otherwise it is 0.

[0036] Furthermore, in step S4, based on type conversion constraints, the constraints for optimizing habitat pattern configuration are set as follows:

[0037]

[0038] in, For the conversion parameters, k is the original habitat type of the unit, and p is the converted habitat type.

[0039] And set the following type conversion constraints:

[0040] The conversion of shallow water areas, shallow water areas, and deep water areas to other habitat types is prohibited. The three types of shallow water areas, shallow water areas, and deep water areas can be converted into each other. The habitat type of deep water areas is a fixed unit.

[0041] Other habitat types must not be converted into bare beaches;

[0042] The conversion between construction land and cultivated land is prohibited.

[0043] Furthermore, the optimized configuration model is constructed in step S5 as follows:

[0044] S5.1. A greedy algorithm is used to transform the multi-objective optimization problem into a single-objective maximization problem, and a heuristic information function is used to change the habitat cell at cell (i, j) from category k to p.

[0045]

[0046] Where m is the target number and N is the ant colony size;

[0047] S5.2, Set the pheromone update function and pheromone release function;

[0048] The pheromone update function is:

[0049]

[0050] in, Q(t) represents the pheromone released by ants during the process of transforming the grid (i,j) from habitat type k to p when the number of iterations is t. Q(t) is the pheromone intensity factor.

[0051] At this point, the pheromone update function of the algorithm is defined as:

[0052]

[0053] Where ρ is the pheromone volatile factor and N is the ant colony size;

[0054] S5.3 The probability of the current cell changing from its current type to another type is given by the probability transfer function:

[0055]

[0056] in, Let t be the probability that the a-th ant in grid (i, j) changes from land class k to p. a Let α be the habitat type unit that the a-th ant has already visited, β be the expected heuristic factor, and α be the heuristic factor.

[0057] The method for optimizing the dynamic pattern of lake wetland habitats considering multiple ecological subjects provided by this invention has the following beneficial effects:

[0058] This invention analyzes the demand for suitable habitats from the perspective of multiple ecological subjects, selects water depth as a key dynamic factor in ecological processes, and classifies habitat types by integrating land use data, measured topographic data, water level monitoring data, and suitable habitat water depth thresholds, thus endowing the static pattern with dynamic attributes.

[0059] Based on the habitat type classification results with composite dynamic attributes, this invention conducts research on the optimal allocation of lake wetland habitat patterns in conjunction with current status analysis, and presents a multi-objective optimal allocation scheme for lake wetland habitat patterns.

[0060] The method of this invention systematically considers the dynamic habitat needs of different ecological subjects at different life stages in lake wetlands, realizes the dynamic optimization of lake wetland habitat patterns, establishes a dynamic and effective correlation between patterns and ecological processes, improves the theoretical system for the research on the optimal allocation of lake wetland habitat patterns, and helps to maximize the benefits of the ecosystem. Attached Figure Description

[0061] Figure 1 A flowchart for optimizing the dynamic pattern of lake wetland habitats considering multiple ecological entities.

[0062] Figure 2 A coding framework for a habitat pattern optimization model based on ant colony algorithm. Detailed Implementation

[0063] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0064] Example 1, Reference Figure 1 This scheme considers the dynamic pattern optimization configuration method of lake wetland habitats with multiple ecological subjects. This method can optimize the configuration method of habitat pattern by taking into account the dynamic habitat needs of different ecological subjects and the synergistic effect of different ecological service functions for lake wetlands that are strongly affected by human activities. The specific steps are as follows:

[0065] Step S1: Obtain multiple basic data on the dynamic pattern of lake wetland habitats;

[0066] Step S2: Perform data processing and analysis on the acquired basic data.

[0067] Step S3: Construct the habitat suitability objective function and spatial clustering objective function for multiple ecological subjects;

[0068] Step S4: Set constraints for habitat pattern optimization based on area constraints and type conversion constraints;

[0069] Step S5: Based on the objective function constructed in step S3 and the constraints set in step S4, construct an optimization configuration model;

[0070] Step S6: Input the multiple basic data from step S2 into the optimization configuration model;

[0071] Step S7: Set multiple parameters in the optimization configuration model and select the optimal parameter combination;

[0072] Step S8: Based on the input of basic data and the selection of parameter combinations, output the optimized configuration results of the dynamic pattern of lake and wetland habitats in the target area;

[0073] Step S9: Compare the habitat type area, structural composition, and spatial pattern characteristics before and after the optimization of the dynamic pattern of lake wetland habitat, and evaluate the optimization effect.

[0074] Based on the habitat type classification results with composite dynamic attributes, this embodiment conducts a study on the optimal allocation of lake wetland habitat patterns in conjunction with the current situation analysis, and presents a multi-objective optimal allocation scheme for lake wetland habitat patterns.

[0075] Example 2, this example specifically includes the following steps:

[0076] Step S1: Obtain multiple basic data;

[0077] It mainly includes the vector boundaries of lakes and their respective administrative regions, lake wetland habitat type data, lake wetland land use data, lake topographic data, water level monitoring data, and regional or lake ecological protection plans.

[0078] The data on lake wetland habitat types were determined using land use data, lake topographic data, water level monitoring data, and suitable habitat water depth thresholds. The specific classification of lake wetland habitat types is shown in Table 1.

[0079] Table 1 Classification of Lake Wetland Habitat Types

[0080]

[0081]

[0082] Step S2: Perform data processing and analysis on the acquired basic data.

[0083] This mainly involves water level data and habitat type data. Water level data processing primarily involves analyzing the annual water level variation patterns across different decades, selecting water level periods less affected by human activities as healthy eco-hydrological rhythms, and using these as a reference to determine the reasonable annual water level variation process under the current annual average water level conditions. Habitat type data processing mainly involves obtaining the spatial distribution of habitat types for each month of the current year based on the reasonable annual water level variation process and the lake / wetland habitat type classification standards.

[0084] Step S3: Construct the habitat suitability objective function and spatial clustering objective function for multiple ecological subjects;

[0085] Construct a habitat suitability objective function for multiple ecological subjects, including:

[0086] Considering the different life stages of multiple ecological subjects in lake wetlands, this study assesses the suitability of different habitat types for different ecological subjects at different life stages, specifically for different months. This yields monthly spatial distribution maps of suitability for the study area from the perspective of different ecological subjects. For a single ecological subject, the cumulative multiplication method is used to aggregate the 12-month grid cell suitability distribution maps into an annual-scale grid cell suitability distribution map. This distribution map represents the distribution of grid cell habitat type suitability information under a healthy eco-hydrological rhythm scenario, and also covers the suitability information of habitat types throughout the monthly life cycle of the ecological subject. Based on this, a multi-ecological-subject habitat suitability objective function is constructed as follows:

[0087]

[0088] α1+α2+α3+α4+α5=1

[0089] Where F represents the sum of habitat suitability of each cell network within the study area, and α represents the weighting of habitat suitability for different ecological subjects. Let y represent the habitat suitability of different ecological subjects in month y when the habitat type is k at grid (i, j).

[0090] Construct the spatial clustering objective function, including:

[0091] According to the agglomeration effect in economic geography, clustering land of similar types together can reduce costs and create higher economic benefits. Similarly, when optimizing habitat type allocation, unit clustering is beneficial for expanding ecological effects and should therefore be considered. The degree of habitat type clustering is related not only to the habitat type of the current grid unit but also to the habitat types of nearby grid units; therefore, neighborhood search should be considered during the optimization process.

[0092] The degree of clustering is represented by a domain identity index. A binary function is defined based on whether the land use type at grid cell (i, j) is k.

[0093]

[0094] Among them, u ij Let (i, j) be the habitat type at grid (i, j). Then, the neighborhood identity index of habitat type k at (i, j) is:

[0095]

[0096] Where (i, j) ≠ (s, t), and using an eight-neighborhood as the window, the spatial clustering objective function is defined as:

[0097]

[0098] Step S4: Set constraints for habitat pattern optimization based on area constraints and type conversion constraints;

[0099] Constraints are divided into two main categories: area constraints and type conversion constraints. Area constraints mainly involve wetland area and total area of ​​the region, and are specifically expressed as follows:

[0100]

[0101] in, and Let be the minimum and maximum areas for habitat type k, respectively. When the habitat type at cell (i, j) is k, ... It is 1 if it is true, otherwise it is 0.

[0102] During the optimization of habitat type allocation, the types of habitat units change due to reconfiguration, resulting in conversion costs. Lower costs mean easier implementation of the plan, and achieving the highest overall benefits at the lowest cost aligns with the planning objective. Since the conversion costs between different habitat types are difficult to obtain, certain restrictions are imposed on the type conversion process based on actual circumstances, namely:

[0103]

[0104] This step sets the following type conversion constraints:

[0105] Transformation from shallow water areas, shallow water areas, and deep water areas to other habitat types is prohibited, but these three types can be converted into each other, with the deep water area habitat type being a fixed unit.

[0106] Other habitat types must not be converted into bare beaches;

[0107] The conversion between construction land and cultivated land is prohibited.

[0108] Step S5, Model Building;

[0109] refer to Figure 2 Based on the establishment of a multi-objective system and a constraint system, and combined with core function processing, an optimization configuration model is constructed using MATLAB, which specifically includes:

[0110] Step S5.1, Heuristic Information Function;

[0111] In multi-objective optimization configuration problems, each ant represents a habitat type optimization configuration scheme. A common approach is to transform each individual objective into a maximization problem. When calculating the heuristic information function for different habitat types in each unit, a greedy algorithm can be used to transform the multi-objective optimization problem into a single-objective maximization problem, ensuring that all objective functions contribute to different habitat types.

[0112] Therefore, the heuristic function for converting the habitat cell at cell (i, j) from category k to p is expressed as:

[0113]

[0114] Where m is the target number and N is the ant colony size.

[0115] Step S5.2, pheromone update function;

[0116] Multi-objective optimization simulates the positive feedback mechanism achieved by ants leaving pheromones during foraging, and sets key functions such as pheromone update function and pheromone release function;

[0117] The algorithm updates the pheromone level in each iteration based on the result of the previous iteration. The pheromone release function is shown below:

[0118]

[0119] in, Let Q(t) be the pheromone released by ants during the process of transforming the grid (i,j) from habitat type k to p when the number of iterations is t. Q(t) is the pheromone intensity factor.

[0120] At this point, the pheromone update function of the algorithm is defined as:

[0121]

[0122] Where ρ is the pheromone evaporation factor and N is the ant colony size.

[0123] Step S5.3, probability transition function;

[0124] During habitat type optimization, when the current unit type is configured to another type, the change only occurs within the current unit and does not cause type conversion in other units in the neighborhood. Therefore, we only need to consider the probability of the current unit changing from its current type to another type, and its probability transfer function is as follows:

[0125]

[0126] in, Let t be the probability that the a-th ant in grid (i, j) changes from land class k to p. a Let α be the habitat type unit that the a-th ant has already visited, β be the expected heuristic factor, and α be the heuristic factor.

[0127] Step S6: Prepare the data and input it into the model from step S5;

[0128] The habitat type vector data is converted into raster data to obtain grid data. There are eight habitat types: cultivated land, forest and grassland, reed beds, shallow water areas, shallow water areas, deep water areas, construction land, and bare tidal flats. The layer attributes that need to be defined are shown in the table below.

[0129] Table 2 Grid Layer Attribute Structure Table

[0130]

[0131] Step S7: Parameter setting and comparison;

[0132] The important parameters of the ant colony algorithm in the model are basically the same as those of the basic ant colony algorithm. They mainly involve heuristic factor α, expected heuristic factor β, and pheromone evaporation factor ρ. Different parameter combinations will not only affect the convergence speed and convergence result of the objective function, but the parameter effect will also be directly reflected in the optimization configuration result.

[0133] The heuristic factor α and the expected heuristic factor β, which have a significant impact on ant colony search, were selected as the main influencing parameters for parameter combination settings, and the effects were compared.

[0134] Step S8: Model running and result output;

[0135] By running the model, the optimized configuration results of the habitat pattern in the study area are obtained. Since the output of the MATLAB model is a text matrix, ArcGIS software is also needed to convert the text file into a raster file to complete the output of the results.

[0136] Step S9: Evaluation of optimization effect;

[0137] The optimization effect is evaluated by comparing the habitat type area, structural composition, and spatial pattern characteristics before and after the habitat pattern optimization.

[0138] The spatial pattern characteristics of habitats are mainly analyzed using landscape pattern indices, focusing on aspects such as habitat structure rationality, habitat fragmentation, habitat shape rationality, habitat connectivity, and habitat diversity. Specific indices can be calculated using Fragstats 4.2 software.

[0139] Example 3: This example describes the construction of an objective function for habitat suitability for multiple ecological subjects, specifically including:

[0140] 1. Analysis of the suitability of the main ecological habitats:

[0141] Lake wetland ecosystems are mainly composed of humans, aquatic biological communities, and the abiotic environment. The aquatic biological community includes biotic factors such as wetland, mesophytic, and aquatic plants, animals, and microorganisms. The abiotic environment includes sunlight, water, and soil. Generally, the ecological subject refers to the living organisms within the ecosystem. Considering the ecological status of populations and their impact on the ecological environment, "humans," "birds," "fish," "emergent plants," and "submerged plants" are selected as the ecological subjects of lake wetlands, and habitat suitability analysis is conducted.

[0142] 1.1. Human habitat suitability analysis;

[0143] Human survival and social development are closely related to lakes and wetlands, which provide multiple ecological services, including material supply, flood control and irrigation, climate regulation, habitat maintenance, and recreation. In terms of habitat type demand, human activities primarily involve arable land, reed beds, and construction land. Arable land and reed beds mainly provide crops and cash crops, while construction land serves as the primary residential area. The demand for these three habitat types remains relatively constant throughout the season. Furthermore, shallow, shallow, and deep water areas provide material supply, as well as climate regulation, water purification, flood control and irrigation, and recreation. However, from November to March of the following year, the weather conditions are relatively poor, resulting in lower service functions and suitability. From April to October, water purification and recreation functions are enhanced. After July, the flood control and irrigation functions of shallow and deep water areas are significantly strengthened, with deep water areas exhibiting the strongest water storage capacity and the highest suitability. Forests and grasslands primarily serve the functions of climate regulation and soil and water conservation, with their seasonal demand intensity remaining constant. The demand for bare beach land is almost zero. Based on the above analysis, the suitability of different habitat types for human populations in different months is shown in Table 3.

[0144] Table 3. Habitat type suitability for human subjects in different months.

[0145]

[0146] 1.2 Analysis of bird habitat suitability;

[0147] Birds are the most representative group of wildlife in lake wetlands and play a vital role in maintaining the stability of wetland ecosystems. Wetland birds can be mainly divided into wading birds, waterfowl, and shore birds, each utilizing different habitats for foraging, resting, and roosting throughout the day. Water depth is the most important factor limiting waterbird habitat utilization. It not only determines whether a habitat can be used by waterbirds but also affects their feeding behavior and energy consumption, which is crucial for wintering waterbirds in their foraging grounds. Generally, large and medium-sized plovers and sandpipers typically forage in wetlands with a depth of no more than 15 cm, while wading birds such as mallards and herons forage in wetlands with a depth of no more than 30 cm. Some waterfowl also need to utilize deeper open water for swimming and playing. Therefore, a combination of habitats consisting of shallow water, mudflats, and grassy areas is often required.

[0148] The main habitat types for wetland birds include cultivated land, forest and grassland, reed beds, shallow water areas, shallow water areas, and deep water areas. The Great Bustard, Oriental White Stork, Siberian Crane, and Red-crowned Crane, all Class I protected birds in China, were selected as indicator species for analysis. During the wintering season, the Great Bustard and other birds primarily forage in wheat fields and fallow lands, showing a preference for lakes and wetlands when choosing habitats. Oriental White Storks tend to forage in open water, reed marshes, and mudflats during the breeding season, with foraging depths ranging from 5 to 40 cm. Siberian Cranes are wading birds and cannot forage in deep water; they also face difficulties finding food in areas with insufficient moisture and excessively dry substrate. Siberian Cranes primarily inhabit water depths of 5-45 cm, with the vast majority in areas no deeper than 30 cm. Red-crowned Cranes mainly forage in marshy wetlands dominated by sedges in spring, with foraging depths mostly between 5 and 15 cm, requiring a human disturbance distance of more than 1500 m. Based on the above analysis, from November to March of the following year, the suitability of cultivated land is relatively high. Shallow areas are relatively more suitable than shallow and deep water areas, but the suitability of shallow water areas decreases due to winter freezing. From April to October, the suitability of shallow, shallow, and deep water areas all improves. Forest grasslands and reed beds become habitats and nesting sites for some birds, possessing a certain degree of suitability. Bare beaches can only serve as short-term stopovers for birds, with lower suitability. The suitability of different habitat types for the main bird species in different months is shown in Table 4.

[0149] Table 4. Habitat type suitability for bird species in different months

[0150]

[0151]

[0152] 1.3. Fish habitat requirements;

[0153] Fish are a crucial link in the food chain of shallow lake wetlands, impacting the lake ecosystem primarily through feeding behavior, nutrient excretion, and sediment disturbance. Different fish species have different diets, resulting in vertical variability in their distribution within the water. Silver carp primarily feed on algae, rotifers, and water fleas, thus mainly inhabiting the upper and middle layers of the water; grass carp and crucian carp primarily feed on plant roots and shoots, and *Ceratophyllum demersum*, thus mainly inhabiting the middle and lower layers; common carp primarily feed on benthic animals, thus mainly inhabiting the bottom layer; snakehead mainly feeds on common carp and crucian carp, primarily living in the middle and lower layers; yellow catfish primarily feeds on benthic animals and fish larvae, with a similar habitat to snakehead. Different fish species have specific needs for habitats in shallow, shallow, and deep water areas. Shallow water areas serve as important spawning grounds and foraging grounds, while deep water areas mainly function as overwintering grounds. Furthermore, flooded areas in reed beds also become feeding and resting places for some fish species. Based on the above analysis, the suitability of different habitat types for fish populations in different months is shown in Table 5.

[0154] Table 5. Habitat type suitability for fish populations in different months.

[0155]

[0156] 1.4. Habitat requirements for reeds;

[0157] Emergent plants, as the dominant group in lake wetlands, mainly include reeds, lotus, and narrow-leaved cattails. Reeds, as a key emergent plant in lake wetland areas, play an irreplaceable role in the functioning of wetlands. The dormancy period for reeds is generally from January to February, during which water depth requirements are not high, with an optimal depth of about 15 cm. The germination period is generally from March to April, during which a water depth of no more than 30 cm is required. The vegetative growth stage is mainly from May to August, during which water depth requirements are higher, with an optimal range of 40-60 cm. September and October are the reproductive growth stage, with the optimal water depth range similar to the vegetative growth stage. November and December are the seed dispersal stage, with the optimal water depth requirements similar to the seed dormancy period. Reed growth mainly involves reed beds, shallow water areas, and shallow water zones. Based on the above analysis, the suitability of different habitat types for reeds in different months is shown in Table 6.

[0158] Table 6. Habitat type suitability for reeds in different months

[0159]

[0160] 1.5. Habitat requirements of Ceratophyllum demersum;

[0161] Submerged plants are the most basic and crucial link in the ecosystem of "grass-type lakes." A healthy submerged plant community structure can cope with changes in the aquatic environment, and its restoration is key to the restoration of the aquatic ecological environment and an important guarantee against the transformation of "grass-type lakes" into "algae-type lakes." Dominant species of submerged plants include *Ceratophyllum demersum*, *Potamogeton crispus*, and *Potamogeton pectinata*. Among them, *Ceratophyllum demersum* not only has a strong capacity for removing total nitrogen and total phosphorus but also a strong capacity for accumulating heavy metals, playing an important role in the water purification function of shallow lakes. Water depth is a decisive factor for the growth and reproduction of submerged plants. Increased water depth leads to changes in factors such as light, dissolved oxygen, and phytoplankton attachment, thereby affecting the morphological and physiological indicators of submerged plants. *Ceratophyllum demersum* is a perennial herbaceous plant, generally flowering from June to July and fruiting from August to October. It mainly grows in still water areas of lakes, easily forming dense underwater communities at depths of 1-3 meters. Based on the above analysis, the suitability of different habitat types for *Ceratophyllum demersum* in different months is shown in Table 7.

[0162] Table 7. Habitat type suitability for Ceratophyllum demersum in different months

[0163]

[0164] 2. Construction of the suitability objective function;

[0165] Based on the above habitat type suitability analysis for different ecological subjects in different months, and combined with the habitat type classification results for different months of the year, the spatial distribution of habitat type suitability for different ecological subjects during the monthly process is obtained. An averaging method is used to aggregate the habitat suitability indices for different months, ultimately yielding the spatial distribution of habitat suitability for different ecological subjects at the annual scale. Weighting coefficients are then used to optimize the combination of different ecological subjects. Finally, a suitability objective function considering the life stages of different ecological subjects is obtained, which is used to construct a dynamic habitat pattern optimization model, specifically including:

[0166] 2.1 Aggregation of habitat suitability for individual ecological subjects;

[0167] Based on the spatial distribution of current habitats from January to December, and combined with the suitability analysis results of habitat types for different ecological subjects in different months, the spatial distribution of current habitat suitability from January to December was obtained by matching habitat types and suitability for different months, with suitability values ​​ranging from 0 to 1. Spatial operations were used to average and aggregate the habitat suitability from January to December to obtain the annual-scale spatial distribution of habitat suitability. This result aggregates suitability information from different months and better represents the dynamic needs of ecological subjects throughout their life cycle. The habitat suitability information for a single grid is as follows:

[0168]

[0169] Among them, F s Let y represent the average habitat suitability for habitat type k at grid (i,j), and y represent the month. Let y be the habitat suitability in month y when the habitat type is k at grid (i, j).

[0170] The suitability information aggregated in this step comprehensively considers the degree of demand for different habitats during the life cycle of the ecological subject. By combining the suitability of different habitat types in different months, the habitat suitability information can be reflected on a dynamic scale of monthly processes, achieving matching and integration with life cycle information.

[0171] 2.2 Weighted aggregation of habitat suitability for multiple ecological subjects;

[0172] By considering the weights of different ecological subjects, a weighted aggregation of habitat suitability for multiple ecological subjects is performed to obtain the aggregated habitat suitability information for a single grid:

[0173]

[0174] Among them, F m Let α be the average habitat suitability of multiple ecological subjects with habitat type k at grid (i,j), and let α be the weight assigned to the habitat suitability of different ecological subjects. The habitat suitability for different ecological subjects in month y when the habitat type is k at grid (i, j).

[0175] In this step, the suitability information after the aggregation of multiple ecological subjects comprehensively considers the dynamic needs of multiple subjects' life cycles, rather than being limited to reflecting a single subject. This helps to fully reflect the habitat demand information of the main bodies of the ecosystem and improve the overall function of the ecosystem.

[0176] 2.3 Construction of the suitability objective function for multiple ecological subjects;

[0177] Considering that the optimization objective is a maximum optimization problem, the habitat suitability information of the grid cells in the study area is summed to obtain the multi-ecological-agent suitability objective function:

[0178]

[0179] Where F represents the sum of habitat suitability of the cell grid within the study area.

[0180] In this step, the suitability objective function for multiple ecological subjects is a maximum optimization problem. Constructing a suitability objective function for the entire region helps to improve the overall suitability of the ecosystem and increase land use efficiency.

[0181] Although specific embodiments of the invention have been described in detail with reference to the accompanying drawings, this should not be construed as limiting the scope of protection of this patent. Various modifications and variations that can be made by a person skilled in the art without inventive effort within the scope described in the claims still fall within the scope of protection of this patent.

Claims

1. A method for optimizing the dynamic pattern of lake wetland habitats considering multiple ecological subjects, characterized in that, Includes the following steps: S1. Obtain multiple basic data on the dynamic pattern of lake wetland habitats; S2. Perform data processing and analysis on the acquired basic data; S3. Construct a habitat suitability objective function and a spatial clustering objective function for multiple ecological subjects; S4. Set constraints for habitat pattern optimization based on area constraints and type conversion constraints; S5. Based on the objective function constructed in step S3 and the constraints set in step S4, construct an optimization configuration model; S6. Input the multiple basic data from step S2 into the optimization configuration model; S7. Set multiple parameters in the optimization configuration model and select the optimal parameter combination; S8. Based on the input of basic data and the selection of parameter combinations, output the optimized configuration results of the dynamic pattern of lake and wetland habitat in the target area; S9. Compare the habitat type area, structural composition, and spatial pattern characteristics before and after the optimization of the dynamic pattern of lake wetland habitat, and evaluate the optimization effect. The basic data in step S1 includes: Vector boundaries of lakes and their respective administrative regions, lake wetland habitat type data, lake wetland land use data, lake topographic data, water level monitoring data, and regional or lake ecological protection plans; Among them, the data on lake and wetland habitat types were determined through land use data, lake topographic data, water level monitoring data, and suitable habitat water depth thresholds; The lake wetland habitat types are classified as follows: cultivated land, forest and grassland, reed terraces, shallow water area, shallow water area, deep water area, construction land and bare beach; The objective function for constructing the multi-ecological-subject habitat suitability in step S3 is as follows: in, F The sum of habitat suitability of each cell network within the study area is represented by α1, α2, α3, α4, and α5, which are weights assigned to the habitat suitability of different ecological subjects. Let y represent the habitat suitability of different ecological subjects in month y when the habitat type is k at grid (i, j). I For the number of rows in the cell, J This represents the number of columns in the cell.

2. The method for optimizing the dynamic pattern of lake wetland habitat considering multiple ecological subjects as described in claim 1, characterized in that, The objective function for constructing spatial clustering in step S3 is as follows: The degree of agglomeration is represented by a domain identity index, based on whether the land use type at grid cell (i, j) is... k Define a binary function : in, For grid ( i , j The habitat type at ) then ( i , j Habitat type k Neighborhood identity index for: in,( i , j ) ( s , t ), s For the row number of the neighboring grid, t The column number for the neighboring grid; Using an eight-neighborhood as the window, the spatial clustering objective function is... Defined as:

3. The method for optimizing the dynamic pattern of lake wetland habitat considering multiple ecological subjects as described in claim 2, characterized in that, In step S4, based on area constraints, the constraints for optimizing habitat layout are set as follows: The constraints for the wetland conservation area and the total area of ​​the region are as follows: in, and Habitat type k The minimum and maximum areas, For wetland area counting units, when the cell grid ( i , j The habitat type at this location is k hour, It is 1 if it is true, otherwise it is 0.

4. The method for optimizing the dynamic pattern of lake wetland habitat considering multiple ecological subjects according to claim 3, characterized in that, In step S4, based on type conversion constraints, the constraints for optimizing habitat pattern configuration are set as follows: in, For conversion parameters, k This refers to the original habitat type of the unit. To change habitat type; And set the following type conversion constraints: The conversion of shallow water areas, shallow water areas, and deep water areas to other habitat types is prohibited. The three types of shallow water areas, shallow water areas, and deep water areas can be converted into each other. The habitat type of deep water areas is a fixed unit. Other habitat types must not be converted into bare beaches; The conversion between construction land and cultivated land is prohibited.

5. The method for optimizing the dynamic pattern of lake wetland habitat considering multiple ecological subjects according to claim 4, characterized in that, The optimized configuration model constructed in step S5 is as follows: S5.

1. A greedy algorithm is used to transform the multi-objective optimization problem into a single-objective maximization problem, and the habitat unit at cell (i, j) is divided into categories. k Turn to p heuristic information function : in, m For the target number, N The size of the ant colony; S5.2, Set the pheromone update function and pheromone release function; The pheromone update function is: in, For iteration number t, the grid (i, j) changes from habitat type k Convert to p The pheromones released by the ants during the process For pheromone intensity factor; At this point, the pheromone update function of the algorithm is defined as: in, As a pheromone volatile factor, N The size of the ant colony; S5.3 The probability of the current cell changing from its current type to another type is given by the probability transfer function: in, For the number of iterations t Time a An ant is on the grid (i, j) from the land class k Convert to p The probability of tabu a For the first a The habitat type units that only ants have visited. β As the expected heuristic factor, α It is a heuristic factor.

Images