Method for screening and planting aquatic plant combination in lakeside zone based on complementary niche

Abstract

The present application relates to the technical field of ecological environment restoration, solves the technical problems of lack of systematicness in aquatic plant configuration, insufficient optimization means and unstable community in the prior art, and particularly relates to a method for screening and planting of aquatic plant combination in lakeside zone based on complementary niches, which first constructs an ecological characteristic database, then calculates the niche complementarity index between aquatic plants and establishes a complementarity matrix. The non-dominated sorting genetic algorithm is used to optimize the aquatic plant combination to obtain an optimal aquatic plant combination. The selected aquatic plants are configured according to the suitable water depth, and feedback regulation is performed in combination with the ecological monitoring results to realize dynamic optimization of the community structure. The present application can fully consider the niche difference and functional complementarity of plants, and use quantitative indicators and intelligent algorithms for combination optimization to construct a more stable and functional plant community in the lakeside zone, and has significant ecological restoration and application promotion value.

Description

Technical Field

[0001] This invention relates to the field of ecological environment restoration technology, and in particular to a method for screening and planting aquatic plant combinations in lakeside areas based on complementary ecological niches. Background Technology

[0002] Lakeside zones, as crucial buffer zones in lake ecosystems, play a vital role in water purification, biodiversity maintenance, and habitat supply. The integrity of their vegetation structure and ecological function directly impacts the overall health of the lake. In recent years, with the increasing prominence of water pollution, lakeside zone reconstruction based on ecological principles has become one of the core technologies for aquatic ecological restoration. Among these, the "ecological buffer zone construction method," primarily using aquatic plant planting, has been widely applied in river and lake ecological management projects. However, existing methods for configuring aquatic plants in lakeside zones still suffer from the following prominent problems:

[0003] (1) The selection of plants lacks systematicity, and the combination and configuration are highly arbitrary.

[0004] Currently, the selection of plant species in engineering practice largely relies on experience or localized experiments, often guided by aesthetics or local adaptability, lacking in-depth analysis of ecological relationships between species (such as complementary resource utilization and synergistic habitat adaptation). This "experience-based" configuration results in poor plant community resilience, limited functionality, and difficulty in forming long-term stable composite systems.

[0005] (2) Community construction ignores niche differences and functional complementarity.

[0006] Aquatic plants exhibit significant niche differentiation in terms of water depth adaptation, root structure, and nutrient absorption methods. Different species have complementary relationships in resource utilization and functional contributions; scientific combination can improve the overall purification efficiency and ecological stability of the community. However, existing methods often overlook this complementary mechanism, leading to severe competition among plants, easy community degradation, and impaired restoration effects.

[0007] (3) Lack of quantitative optimization tools makes it difficult to achieve calculable and reproducible configuration schemes.

[0008] While existing studies have proposed some screening indicators based on functional traits or purification capabilities, they lack systematic methods for combinatorial optimization. In particular, they lack effective algorithmic support in dealing with multi-objective trade-offs (such as purification function and diversity), resulting in a lack of theoretical foundation for scheme optimization and unreproducible and difficult-to-generalize results.

[0009] (4) The planting material does not match the plant type, which affects growth and structural stability.

[0010] During the construction of the lakeside area, a large number of uniform substrate materials, such as coconut coir blankets and polyester fiber mats, were used. However, the dependence of plant root types on parameters such as porosity and degradability was ignored, resulting in low planting rates, slow growth, and even problems such as vegetation drift or substrate collapse in the early stages of planting.

[0011] In summary, there is an urgent need for a new method for configuring aquatic plants in the lakeside zone that integrates niche theory with intelligent optimization methods and takes into account both functionality and stability. This method should be able to systematically quantify the complementarity between species and dynamically adjust the community structure by combining water depth zoning and ecological monitoring feedback, thereby achieving the construction of a highly functional, highly stable, and highly adaptable plant community system. Summary of the Invention

[0012] To address the shortcomings of existing technologies, this invention provides a method for screening and planting aquatic plant combinations in lakeshore zones based on complementary ecological niches, falling within the technical scope of aquatic ecosystem restoration, lakeshore zone reconstruction, and water purification plant configuration. It is used to scientifically construct structurally stable and functionally complete aquatic plant communities in lakeshore zones, thereby enhancing the purification and restoration capabilities of aquatic ecosystems and overcoming problems such as lack of systematic configuration, insufficient optimization methods, and community instability in existing technologies.

[0013] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for screening and planting aquatic plant combinations in lakeside areas based on complementary ecological niches, the method comprising the following steps:

[0014] Step 1: Construct a database of the ecological characteristics of aquatic plants in the lakeside zone.

[0015] The ecological characteristic database is used to provide a unified parameter standard for subsequent screening and calculation. The database includes at least the suitable water depth range (in centimeters), ecological niche width (the difference between the maximum and minimum suitable water depth), root type, propagation method, nutrient absorption capacity (e.g., TN / TP absorption rate per unit biomass per unit time), and purification capacity parameters (e.g., TN / TP removal rate or removal efficiency per unit biomass) for each aquatic plant.

[0016] Preferably, the suitable water depth is within the range [ d min , d max Characterized by the ecological niche width. d max –d min Root type was determined by anatomical observation and literature comparison; propagation method was determined by field survey and literature comparison; nutrient absorption and purification capacity were determined by indoor culture and small-scale water purification experiments.

[0017] Furthermore, database fields are stored with uniform units, uniform measurement periods, and uniform sampling standards to ensure comparability and calculability among various indicators.

[0018] Step 2: Calculate the niche complementarity index between aquatic plant pairs. .

[0019] The complementarity index is used to quantify the complementary relationship between two species in terms of deep-water niches and key functional traits. CI The calculation model is as follows:

[0020]

[0021] in, Aquatic plants with aquatic plants Niche complementarity index (0~1) between them; 、 and The weight coefficients for the corresponding features satisfy... .

[0022] The degree of overlap The crossover ratio of the suitable water depth ranges for two species is defined as:

[0023]

[0024] in, , Aquatic plants with aquatic plants Suitable water depth range;

[0025] The attribute difference 、 and Normalized to 0-1. Preferably:

[0026] Root type diversity : 0 for two plants with the same root system type, 1 for taproot type and fibrous root type, and 0.5 for mixed type and taproot / fibrous root type;

[0027] Differences in propagation methods : If they are the same, take 0; if they are different, take 1.

[0028] Differences in nutrient absorption Maximum difference normalization is used:

[0029]

[0030] in, Aquatic plants with aquatic plants The rate of total nitrogen uptake per unit biomass, M, in all pairwise combinations of the candidate species set. and Take the biggest one.

[0031] Furthermore, all candidate plants were categorized according to... The complementarity matrix is ​​calculated using the formula. .

[0032] Step 3: Construct a complementarity matrix and optimize plant combinations.

[0033] The combinatorial optimization employs a non-dominated sorting genetic algorithm (NSGA–II), with the dual objectives of maximizing complementarity and maximizing purification function. The objective function is:

[0034]

[0035]

[0036]

[0037] in, The number of compositions. For species The percentage of biomass; and For species k The normalized value of the biomass removal capacity per unit; The weights are adjustable (e.g., 0.5, 0.5).

[0038] The crowding distance is used to maintain the diversity of the solution set, and it is calculated as follows:

[0039]

[0040] in, For the first The objective function values ​​(corresponding to) F 1 and F 2) For individuals located at both ends of each non-dominated front, the crowding distance is taken to be infinite or recorded as the maximum value to preserve the boundary solution.

[0041] Preferably, the individual code is of length [length missing]. ,in The preferred value is 3. Indicates the first Aquatic plants in each water depth zone are numbered; the initial population size is not less than 50 and the number of evolutionary generations is not less than 100, and tournament selection and crowding distance sorting are used.

[0042] Furthermore, during optimization, penalty terms or constraints are used to ensure that the combination simultaneously includes three types of plants: submerged, emergent, and floating-leaved, and that the number of species in each zone meets the lower limit requirement, thereby enhancing the feasibility of the project.

[0043] Step 4: Implement the planting configuration of aquatic plants according to the water depth gradient.

[0044] Based on the perennial water depth distribution along the lake shore, the region is divided into a high water level zone (0–20 cm), a medium water level zone (20–50 cm), and a low water level zone (50–100 cm).

[0045] The optimized combinations are configured in the corresponding zone areas according to the suitable water depth range of each species.

[0046] Furthermore, each zone should be planted with no fewer than two species of aquatic plants, and the overall combination should include at least one species each of submerged, emergent, and floating-leaved plants; the planting density can be determined according to the plant size and substrate carrying capacity (e.g., 5–15 plants / m²). 2 ), and preferably a gentle slope strip layout to enhance hydrodynamic exchange and habitat connectivity.

[0047] Step 5: Select suitable ecological substrate materials for planting.

[0048] The matrix material is selected from natural fibers or biodegradable materials and meets the verifiable physical properties and environmental protection indicators: surface porosity 40% to 80%, initial dry tensile strength ≥15MPa, and quality retention rate ≥80% after immersion in water for 30 days.

[0049] Preferably, the residual mass after degradation 6–12 months after planting is ≤10%, the absolute change in pH of the water body during the degradation process is ≤0.5, and the leaching concentration of heavy metals meets the relevant national standard limits.

[0050] Furthermore, the substrate is matched with the root type: for fibrous root types, high-porosity lightweight fibers are preferred, while for taproot types, medium-density mats are preferred, to ensure initial fixation and mid-term ecological safety.

[0051] Step Six: Conduct ecological monitoring and community adjustment.

[0052] The ecological monitoring is conducted every 30 days for a period of no less than 6 months. The monitoring indicators include plant cover, Shannon-Wiener diversity index, and TN / TP concentrations in the influent and effluent. A combined adjustment will be initiated when any of the following conditions occur: plant cover at any depth zone decreases by more than 20% consecutively; the diversity index falls below 1.0; or the effluent TN / TP concentration is higher than the initial value by 20% for two consecutive monitoring periods.

[0053] The aforementioned adjustments include at least: replacing the worst-performing species in the affected zones, and prioritizing the introduction of species that are complementary to the existing communities. Higher or purification capabilities Stronger alternative species; further increase the weights related to purification in NSGA–II, and run several generations of rapid iterations to obtain new Pareto solutions.

[0054] Preferably, the weighting parameter 、 and Data-driven optimization using regression models: with community purification efficiency and stability as the dependent variables, and... 、 and Establish a linear or ridge regression model for the independent variables, i.e.:

[0055]

[0056] After normalizing the regression coefficients, they were respectively used as... 、 and The value is set; it is periodically updated when the newly added monitoring data reaches the set threshold, so as to enhance the method's adaptability to seasonal and interannual variations.

[0057] By employing the above technical solution, the present invention provides a method for screening and planting aquatic plant combinations in lakeside zones based on complementary ecological niches, which has at least the following beneficial effects:

[0058] 1. Based on a full consideration of plant niche differences and functional complementarity, this invention utilizes quantitative indicators and intelligent algorithms for combined optimization to construct a more stable and functional lakeside plant community, which has significant value for ecological restoration and application promotion.

[0059] 2. This invention utilizes a non-dominated sorting genetic algorithm to perform dual-objective optimization of aquatic plant combinations, balancing complementary structure and purification capacity to obtain optimal aquatic plant combinations with reasonable structure and high function. The selected aquatic plants are zoned according to suitable water depth, and feedback regulation is implemented based on ecological monitoring results to achieve dynamic optimization of the community structure.

[0060] 3. The method proposed in this invention has clear parameters and an extensible model, making it applicable to ecological restoration projects for various types of water bodies such as lakes, wetlands, and rivers. Attached Figure Description

[0061] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0062] Figure 1 This is a schematic diagram of the screening and planting results of aquatic plant combinations in the lakeside zone in Embodiment 1 of the present invention, where (a) is the screening process of aquatic plant combinations and (b) is the planting result of aquatic plant combinations.

[0063] Figure 2 This is a scatter plot of F1–F2 showing the distribution of all feasible species combinations in Embodiment 1 of the present invention in terms of complementarity and purification function.

[0064] Figure 3 This is a schematic diagram of the Pareto front solution set of the F1–F2 scatter points in Embodiment 1 of the present invention;

[0065] Figure 4 This is a heatmap of the symmetric complementarity matrix CI drawn in Embodiment 2 of the present invention;

[0066] Figure 5 This is a scatter plot of F1–F2 showing the distribution of all feasible species combinations in Embodiment 2 of the present invention in terms of complementarity and purification function.

[0067] Figure 6 This is a schematic diagram of the Pareto front solution set of the F1–F2 scatter points in Embodiment 2 of the present invention;

[0068] Figure 7 This is a graph showing the continuous monitoring of plant coverage for 6 months in Example 3 of the present invention;

[0069] Figure 8 This is a graph showing the diversity index monitored continuously for 6 months in Example 3 of the present invention;

[0070] Figure 9 This is a graph showing the TN removal rate monitored continuously for 6 months in Example 3 of the present invention;

[0071] Figure 10 This is a graph showing the TP removal rate monitored continuously for 6 months in Example 3 of the present invention. Detailed Implementation

[0072] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. This will allow for a full understanding of how the present application uses technical means to solve technical problems and achieve technical effects, and to facilitate its implementation.

[0073] This embodiment proposes a method for screening and planting aquatic plant combinations in lakeshore areas based on complementary ecological niches. This falls under the technical scope of aquatic ecosystem restoration, lakeshore reconstruction, and water purification plant configuration, aiming to enhance the ecological stability and water purification function of aquatic plant communities. Figure 1As shown in (a), a plant ecological characteristic database was first constructed, including parameters such as suitable water depth range, root type, propagation method, and nutrient absorption capacity. Then, the niche complementarity index between aquatic plant pairs was calculated based on attribute differences. A complementarity matrix was constructed. The non-dominated sorting genetic algorithm (NSGA-II) was used to perform bi-objective optimization of the aquatic plant combination, balancing complementary structure and purification capacity, to obtain an optimal aquatic plant combination with reasonable structure and high function, such as... Figure 1 As shown in (b), the selected aquatic plants are zoned according to their suitable water depth, and feedback regulation is implemented based on ecological monitoring results to achieve dynamic optimization of the community structure. This method has clearly defined parameters and a scalable model, making it suitable for ecological restoration projects of various water types, such as lakes, wetlands, and rivers.

[0074] Example 1: An ecological characteristic database was constructed based on measured ecological parameters, and three groups of lakeside plant communities were selected for the construction of a freshwater lakeside zone.

[0075] The research subject is Lake Jia, a typical freshwater lake within a provincial wetland park. It has a water area of ​​approximately 9.4 hectares and a shoreline length of approximately 960 meters. The lake depth gradually increases from the shoreline towards the inner lake, with a year-round depth range of 0–100 cm, meeting the typical requirements for lakeshore restoration and ecological reconstruction. The research period was from April to October 2023, lasting six consecutive months.

[0076] Phase 1 (Parameter Collection and Database Construction): Twenty common aquatic plants in the region were selected, including submerged plants (such as *Ceratophyllum demersum* and *Vallisneria natans*), emergent plants (such as reeds and cattails), and floating-leaved plants (such as water lilies and water chestnuts). As shown in Table 1, the ecological characteristic database includes:

[0077] (1) The suitable water depth range for each aquatic plant was determined by a combination of manual measurement and buoy-assisted method. d min , d max (Unit: cm) and niche width ( d max –d min );

[0078] (2) Determine the root system type (tap root type, fibrous root type, mixed type) and propagation method (seed propagation, vegetative propagation, etc.) through literature review and field investigation.

[0079] (3) Collect plant samples, determine the total nitrogen / total phosphorus uptake rate per unit biomass, and determine the TN / TP removal capacity per unit area in a small-scale water purification experiment; all indicators are stored in the database according to a unified dimension and measurement period.

[0080] Table 1 Ecological Characteristics Database

[0081]

[0082] Phase Two (Complementarity Index Calculation): Based on the database, the niche complementarity index is calculated for each pair of all aquatic plants. The following formula is used:

[0083]

[0084] in, The degree of overlap between the suitable water depth ranges of two species is represented by the intersection-union ratio of the intervals:

[0085]

[0086] Attribute Difference D r , D f The value is assigned as 0 / 1 based on whether they are of the same type or have the same propagation method (the difference between mixed type and main / fibrous root type can be taken as 0.5); D n Normalization of the maximum difference using the nitrogen uptake rate difference:

[0087]

[0088] In this embodiment, α=0.3, β=0.3, and γ=0.4 are set. A calculation script is written using Python to obtain a 20×20 symmetric complementarity matrix CI=[ CI ij ],diagonal CI ij =0, the number of unique pairs M = (20 × 19) ÷ 2 = 190.

[0089] Phase 3 (NSGA–II combinatorial optimization): Construct a complementarity matrix and introduce NSGA–II for optimization screening.

[0090] (1) Simulation optimization was performed using Python and the DEAP genetic algorithm library (v1.3.1);

[0091] (2) The population size was set to 60 and the number of generations was set to 150. Tournament selection and crowding distance sorting were used to maintain solution set diversity.

[0092] (3) The individual code is preferably an integer vector of length 3 [C1, C2, C3], corresponding to one "core species" in each of the high / medium / low water level zones;

[0093] (4) The biobjective function is:

[0094]

[0095]

[0096]

[0097] TN k TP k This is the normalized value of the removal capacity per unit of biomass; in this embodiment, a=b=0.5 is used in the optimization stage, starting from... w k Equal weighting is applied, and updates are made based on actual measured values ​​of biomass (or coverage) during on-site implementation. w k .

[0098] To visually demonstrate the distribution of all feasible species combinations in terms of complementarity and purification function, an F1–F2 scatter plot was created, as shown below. Figure 2 As shown. Each point represents a core species assemblage of a (high / medium / low) zone (zone adaptation threshold ≥10cm, a=b=0.5, w Equal rights).

[0099] Based on non-dominated sorting and crowding distance, the Pareto front solution set is obtained as follows: Figure 3 As shown in Table 2, the frontier solution is located in the upper right region of the F1–F2 plane, representing a trade-off between complementarity and purification function. Three optimal Pareto "core species combinations" were obtained after optimization.

[0100] Combination A: Goldfish algae + Yellow calamus + Water chestnut;

[0101] Combination B: Vallisneria natans + Sagittaria sagittifolia + Water hyacinth;

[0102] Combination C: Hydrilla verticillata + Alisma plantago-aquatica + Nymphaea rubra.

[0103] Table 2 Objective function values ​​of representative combinations

[0104]

[0105] Phase 4 (Segregated Planting and Monitoring Design): Each group (core + supplement) was divided into zones according to suitable water depth: high water level zone (0–20 cm), medium water level zone (20–50 cm), and low water level zone (50–100 cm). Three independent replicate bank sections were set up for each group (each section 30 m long, randomly arranged with an interval ≥20 m to reduce mutual influence), and three empirical control sections and three blank control sections were also set up. The planting substrate used was coconut fiber mats with an initial dry tensile strength ≥15 MPa, surface porosity 65%, and a quality retention rate ≥80% after 30 days of immersion in water.

[0106] The monitoring frequency is once every 30 days; the monitoring content includes:

[0107] Coverage (Drone + Parallel calibration);

[0108] Shannon–Wiener Diversity Index ,in Calculated based on the percentage of species coverage within the zone;

[0109] Removal rate by η =( Cin C out ) / C in Calculate by multiplying by 100% (ensuring consistent sampling diameter and stable flow rate).

[0110] Preferably, TN / TP is tested using national or industry standard methods; after the data are tested for normality and homogeneity of variance, ANOVA / Tukey or Kruskal–Wallis / Dunn is used for inter-group comparisons.

[0111] Phase 5 (Results and Evaluation): Monitoring results over six consecutive months showed that combination A performed best in terms of purification capacity and community stability: the average coverage increased to 55.6%, the Shannon-Wiener index was 1.27, the effluent TN removal rate was 71.3%, the TP removal rate was 58.2%, the plants grew well, the community structure was stable, and no degradation or substrate loosening occurred.

[0112] Comparative verification showed that the configuration of "optimized combination + supplementary species" was superior to the empirical combination and the blank control, indicating that the technical path of "ecological characteristic database - CI index - NSGA-II optimization - zonal configuration - monitoring feedback" proposed in this invention has good feasibility and stable effect under engineering conditions.

[0113] This embodiment verifies the effectiveness of the ecological feature database construction, CI index model and genetic optimization algorithm in the screening of aquatic plant combinations in the lakeside zone. The results have practical engineering feasibility and ecological application value.

[0114] Example 2: The NSGA-II algorithm was used for multi-objective optimization screening to construct a combination of five aquatic plants that maximized functionality and balanced stability.

[0115] Based on the ecological characteristic database established in Example 1, eight aquatic plants with different ecological functions and morphological structures were selected, including: goldfish algae, yellow calamus, water chestnut, eelgrass, wild arrowhead, water hyacinth, whorled hydrangea, and water lily.

[0116] The niche complementarity index uses the same model as in Example 1, with weights set at α=0.3, β=0.3, and γ=0.4 in this example.

[0117] Based on the above eight plant species, an 8×8 symmetric complementarity matrix CI (with pairwise combinations M=28) was calculated and used as the optimization input. To visually demonstrate the differences in complementarity among species, a heatmap of the symmetric complementarity matrix CI was plotted, as shown below. Figure 4 As shown.

[0118] The non-dominated sorting genetic algorithm (NSGA–II) was used to screen the three-zone combinations. The encoding method was an integer vector of length 3, C=[C1,C2,C3], representing the plant numbers of the high / medium / low water level zones, respectively; the species and the zone depth overlap length of not less than 10cm were considered a fit (consistent with Example 1).

[0119] The optimized basic parameters are as follows:

[0120] (1) Initial population size: 80;

[0121] (2) Evolutionary algebra: 200;

[0122] (3) Crossover probability: 0.9; Mutation probability: 0.1;

[0123] (4) Encoding method: a vector of length 3 These represent the numbers of aquatic plants in the low, medium, and high water level zones, respectively.

[0124] (5) Objective function: Niche complementarity index of all aquatic plant pairs in the combination. value The weighted average TN / TP purification capacity per unit biomass of aquatic plants within the combination. .

[0125] In this embodiment, only TN data is available, so we temporarily take... a =1, b =0, and normalize TN to 0 to 1 to obtain E k To visually demonstrate the distribution of all feasible combinations across the dimensions of complementarity and purification, an F1–F2 scatter plot (all combinations) is drawn, as shown below. Figure 5 As shown.

[0126] Based on non-dominated ordering and crowding distance, 22 Pareto front solutions were obtained, located in the upper right region of the F1–F2 plane, reflecting a trade-off between complementarity and purification function. Figure 6 As shown. From Figure 6 Among the frontier solutions shown, five typical combinations that balance function and diversity were selected, and the results are listed in Table 3.

[0127] Table 3 Optimal Combinations and Their Model Output Results

[0128]

[0129] As can be seen from the results in Table 2, combination 3 exhibits superior niche complementarity ( Combination 1 is optimal in terms of purification ability. This combination performs best. Different combinations can be flexibly selected based on the actual engineering objectives.

[0130] This example demonstrates the process from a database— CI The quantitative process of quantification—NSGA-II dual-objective optimization—frontier screening—combination selection verifies that the method of this invention can achieve optimized configuration of water purification capacity while ensuring community stability. Furthermore, if the requirement of "no less than two species per zone" needs to be met during the engineering implementation phase, the "supplementary species selection principle" of Example 1 can be referred to (prioritizing species that are similar to the core species within the zone). CI Higher and E k For species whose value is not lower than the mean value within the band, one supplementary species is added to each band without changing the optimized coding complexity and basic conclusions of this embodiment.

[0131] Example 3: Practice of community structure adjustment and combinatorial optimization based on dynamic monitoring results.

[0132] This embodiment is based on combination 1 (yellow iris + water chestnut + goldfish algae) selected in embodiment 2, and is applied in an engineering manner on the west bank of a lake in a city park. The total length of the planting strip is 120m, covering a water depth of 0–100cm. It is divided into high-level zone (0–20cm), mid-level zone (20–50cm), and low-level zone (50–100cm) according to the normal water level. The planting density of each zone is uniformly 10 plants / m². 2 (The density is calculated as equivalent plant number / coverage based on seedling size). The planting material is a natural coconut husk fiber base mat with an initial tensile strength of 16.5 MPa, a surface porosity of 65%, a quality retention rate of 83.2% after 30 days of immersion in water, and can be completely degraded within 10 months, with the degradation process meeting the pH and heavy metal leaching limits of the water body.

[0133] Monitoring design: Continuous monitoring for 6 months (once every 30 days), with indicators including: (1) vegetation cover; (2) Shannon–Wiener diversity index. (3) Effluent TN / TP concentration and removal rate η = ( C in C out ) / C in×100%. Monitoring results are shown in Table 4. The time variation curves of coverage, diversity index, and TN / TP are shown in Table 4. Figures 7-10 .

[0134] Table 4. Community ecological performance of Assortment 1 during the monitoring period

[0135]

[0136] Monitoring, interpretation, and proactive early warning triggering: such as Figure 7 As shown, coverage generally increased from January to May, but declined slightly in June (-0.8 percentage points). Figure 8 The diversity index showed only a slight decline in June (1.27 → 1.25), remaining above the safety threshold of 1.0. Meanwhile, Figure 9 and Figure 10 The data shows that the removal rates of TN and TP decreased by -1.5 and -0.9 percentage points respectively in month 6 compared to the previous month. Combined with the growth rate information: the average monthly growth rate of coverage from month 1 to month 5 was approximately +4.88 percentage points, while it turned -0.8 percentage points in months 5 and 6. Preferably, to enhance the robustness of the project, this embodiment adopts an "active early warning mechanism" (as a soft trigger in addition to the rigid trigger threshold): when the monthly growth rate of any key indicator turns from positive to negative and two consecutive indicators show a decline, a small-scale combination fine-tuning and parameter re-optimization process is initiated.

[0137] Combination adjustments and parameter updates:

[0138] (1) Species fine-tuning: *Nelumbo nucifera*, which had a weaker contribution to cover, was replaced with *Hydrilla verticillata*; the average niche complementarity index between *Nelumbo nucifera* and other species in the assemblage increased by 7.6% compared to the original pairing. Based on this, the objective function of the assemblage's average complementarity was changed from...

[0139]

[0140] Updated to:

[0141]

[0142] Under three pairs of approximations with similar CI magnitudes, F 1 Relative improvement .

[0143] (2) Weight coefficient correction: To enhance the guidance of purification function, γ in the complementarity model is adjusted from 0.4 to 0.5, and at the same time, the following is set:

[0144]

[0145] To meet α + β + γThe constraint =1. This adjustment improves the "variability in nutrient absorption capacity". D n "exist CI The significance of the pairings in the study of nutrient absorption gradients leads to a preference for pairings with high nutrient absorption gradients.

[0146] (3) Rapid algorithm iteration: Under the existing NSGA-II framework, a rapid re-optimization (tournament selection + crowding distance sorting) with a population of 80 and 20 iterations is adopted, retaining the top 10% of Pareto bodies, and updating the core combination to: Iris tectorum + Hydrilla verticillata + Ceratophyllum demersum. This step is completed without changing the engineering layout boundary and zonal structure.

[0147] Parameters and decision criteria (repeatability constraints):

[0148] The parameters and caliber are as follows: α + β + γ =1 (after optimization) α = β =0.25, γ =0.5); the zone adaptation threshold is ≥10 cm (the overlap length between the species' suitable depth range and the zone); the monitoring method and frequency are consistent before and after; the complementarity improvement of species replacement is based on the two new pairings and the original pairings. CI The values ​​are compared, and the relative percentage increase is given (7.6% in this example).

[0149] Evaluation of the optimized effect:

[0150] After two consecutive months of monitoring following optimization, the following results were observed: coverage rate 56.4%, H′ 1.30, TN removal rate 63.1%, and TP removal rate 58.0%. A comparison with the data from the previous month (before optimization) is shown in Table 5.

[0151] Table 5 Comparison of key indicators before and after combinatorial optimization (2 months)

[0152]

[0153] Without limiting the scope of this invention, the improvement and model F2=∑ in this embodiment w k E k Weighted guidance consistency: When γ=0.5 (α=β=0.25), the model prefers... D n Larger species pairings; two new pairings resulting from superimposed species replacement. CI The improvement (relative to the original pairing mean +7.6%) resulted in a simultaneous improvement in the F1 combination and the engineering explicit removal rate (%). Note: The algorithm internally calculates F2 at the species level. E kThe normalized value is included, and the engineering output uses the effluent and influent water removal rate (%). The two have different dimensions but the same trend.

[0154] This embodiment demonstrates that, before the rigid threshold of the combined adjustment process is triggered, a closed-loop strategy of "proactive early warning + minor fine-tuning + rapid re-optimization" is preferably adopted. This strategy can resolve mild community degradation and fluctuations in purification efficiency before the threshold is reached, thereby achieving adaptive optimization and long-term steady-state maintenance of the community structure. It has good engineering operability and promotion value.

[0155] This invention first constructs an ecological characteristic database based on the ecological characteristics of various aquatic plants (including water depth adaptation range, root type, reproductive method, and nitrogen absorption capacity). Then, it quantifies the niche complementarity relationships among different aquatic plants by constructing a complementarity index model. Combining the NSGA-II multi-objective genetic algorithm, with complementarity and purification capacity as dual objectives, it screens aquatic plant combinations suitable for different water level zones and plants them in actual water bodies according to high, medium, and low water level zones. Furthermore, a dynamic feedback mechanism is introduced, which can intelligently adjust and optimize the aquatic plant combinations based on vegetation growth status, water quality monitoring results, or external disturbances, thereby improving water purification effects and community stability while ensuring biodiversity. This method has advantages such as clear logic, quantifiable parameters, strong regulation mechanism, and high adaptability. It is applicable to ecological restoration projects of various types of water bodies such as lakes, wetlands, and rivers, and has broad application value in ensuring stable project operation and improving ecological functions.

[0156] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. Since the above embodiments are substantially similar to the method embodiments, they are described relatively simply; relevant parts can be referred to the descriptions of the method embodiments.

[0157] The above embodiments provide a detailed description of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for screening and planting aquatic plant combinations in lakeside zones based on complementary ecological niches, characterized in that, The method includes the following steps: An ecological characteristic database of aquatic plants in the lakeside zone is constructed. The ecological characteristic database shall include at least the suitable water depth range, ecological niche width, root type, propagation method, nutrient absorption capacity and purification capacity parameters of each aquatic plant. Based on the ecological parameters of each aquatic plant in the aforementioned ecological feature database, the niche complementarity index between aquatic plant pairs is calculated. The calculation process includes: The attribute difference degree is determined based on the ecological parameters of each aquatic plant in the ecological characteristic database, including: Determining attribute variability based on differences in plant root system type If the two plants have the same root system type, then If they are completely different, then ; Determining attribute difference based on differences in plant propagation methods If both plants reproduce by seeds or by vegetative propagation, then If the reproduction methods are completely different, then ; The attribute difference was determined by normalizing the maximum value of the difference in total nitrogen uptake rate per unit time and per unit biomass between the two plants. The expression is: ； in, and Plants and nitrogen absorption rate; For all pairwise combinations within the candidate species set The maximum value; The niche complementarity index between aquatic plant pairs was calculated based on attribute differences. ,Right now: ； in, Aquatic plants with aquatic plants Niche complementarity index between them; The degree of overlap between the suitable water depth ranges of two species is defined as the ratio of the length of the overlapping range to the length of the union range. 、 and These represent the differences in root system type, propagation method, and nutrient absorption capacity between the two aquatic plants, with values ​​ranging from 0 to 1. 、 and The weight coefficients for the corresponding features satisfy... ; A complementarity matrix was established to quantify the niche complementarity among different plants, and a non-dominated sorting genetic algorithm was used to screen candidate aquatic plant combinations to obtain the optimal aquatic plant combination. According to the perennial water depth zoning of the lakeside zone, the optimal combination of aquatic plants was planted in the high water level zone, the medium water level zone and the low water level zone respectively; and natural fiber mats or biodegradable fixing materials were used as substrate materials to complete the construction of the lakeside plant community during the planting process. After the lakeside plant community is established, ecological monitoring will be conducted every 30 days for at least 6 months, recording aquatic plant coverage, diversity index, and water TN and TP concentrations. Adjustments will be made when any of the following conditions are met: Coverage has decreased by more than 20% for two consecutive years; the Shannon-Wiener diversity index is below 1.0; and the effluent TN or TP concentration is 20% higher than the initial value for two consecutive monitoring sessions.

2. The method for screening and planting aquatic plant combinations in lakeside areas according to claim 1, characterized in that, The process of using a non-dominated sorting genetic algorithm to screen candidate aquatic plant combinations to obtain the optimal aquatic plant combination includes: Set the basic parameters for the optimization process of the non-dominated sorting genetic algorithm, including: When initializing the population, the population size of the candidate aquatic plant combinations is set to be no less than 50; the number of generations is set to be no less than 100; a tournament selection strategy and a crowding distance sorting mechanism are adopted to maintain the diversity and distribution balance of the solutions. The objective function for the optimization process of the non-dominated sorting genetic algorithm is established, including: Maximize the average niche complementarity index among aquatic plants in the combination The value, expressed as: ； The expression for maximizing the total ecological function value of the lakeside plant community is: ； ； in, Indicating aquatic plant assemblage The average niche complementarity index of all plant pairs in the study; This represents the total number of aquatic plant pairs in the combination; Number of compositions; Indicating aquatic plant assemblage The total ecological function value; For species Biomass percentage in aquatic plant assemblages; For species Purification capacity per unit biomass; and For species The normalized value of the biomass removal capacity per unit; For adjustable weights, satisfying ; During the optimization process, crowding distance is introduced. To maintain the diversity of solution sets, the calculation formula is as follows: ； in, For the first One objective function value; Represents an individual index; and For this individual in the objective function The target values ​​of two adjacent individuals; and These are the maximum and minimum values ​​of the target value in the current population, respectively. The number of objective functions; Individuals located at both ends of each non-dominated front, crowding distance Take infinity or the maximum value to preserve the boundary solution; The non-dominated sorting in the non-dominated sorting genetic algorithm uses the following criterion: If the solution satisfy and If at least one of them is strictly greater than 0, then the solution is... Dominant Solution .

3. The method for screening and planting aquatic plant combinations in lakeside areas according to claim 2, characterized in that, The total ecological function value is the normalized weighted average of the total nitrogen and total phosphorus removal capacity per unit biomass of each aquatic plant in the combination, i.e. .

4. The method for screening and planting aquatic plant combinations in lakeside areas according to claim 2, characterized in that, The individual encoding method for the aquatic plant combination in the non-dominated sorting genetic algorithm is as follows: the length is... ,in , Indicates the first in the combination The aquatic plant numbers selected for each water depth zone.

5. The method for screening and planting aquatic plant combinations in lakeside areas according to claim 1, characterized in that, The water depth ranges from 0 to 20 cm in the high water level zone, from 20 to 50 cm in the medium water level zone, and from 50 to 100 cm in the low water level zone.

6. The method for screening and planting aquatic plant combinations in lakeside areas according to claim 1, characterized in that, The process of adjusting the combination includes updating the species and distribution of aquatic plants based on ecological monitoring results, and the specific rules include: If the aquatic plant cover in any water depth zone declines by more than 20% continuously, the worst-growing aquatic plant species in that zone should be replaced, with priority given to introducing species whose niche complementarity index is higher than that of existing species. The largest candidate species; If the Shannon-Wiener diversity index is below 1.0, at least one new species with different ecological functions or significant differences in niche breadth should be introduced to increase community heterogeneity. If the TN or TP concentration in the effluent exceeds the initial value by 20% for two consecutive monitoring tests, the plant with the lowest purification capacity index should be replaced first, and the weighting coefficient of the purification capacity factor should be increased. and Participate in portfolio updates.

7. The method for screening and planting aquatic plant combinations in lakeside areas according to claim 1, characterized in that, The selection and configuration of the substrate materials used for planting satisfy the following synergistic optimization conditions: The surface porosity of the natural fiber mats or biodegradable matrix materials used is controlled between 40% and 80% to ensure root penetration and water exchange efficiency. The initial tensile strength of the matrix material is not less than 15MPa to ensure structural stability under wind and wave disturbance conditions; The quality retention rate after continuous immersion in water for 30 days is no less than 80% to meet the support period for early plant growth. The residual mass of the substrate material after degradation 6-12 months after planting is no more than 10%, the absolute value of the pH change of the water body during the degradation process does not exceed 0.5, and the leaching concentration of heavy metal ions meets the relevant national standard limits. The materials are matched with the type of plant root system. Plants with fibrous roots use a high-porosity lightweight fiber matrix, while plants with taproots use a medium-density substrate.

8. The method for screening and planting aquatic plant combinations in lakeside areas according to claim 1, characterized in that, During the planting process, at least two types of aquatic plants shall be planted in the high, medium, or low water level zones, and the overall combination shall include at least one submerged plant, one emergent plant, and one floating-leaved plant.

9. The method for screening and planting aquatic plant combinations in lakeside areas according to claim 1, characterized in that, The weighting coefficient , and The model is determined by a regression model trained based on historical monitoring data. Specific methods include: A training set was constructed using a historical aquatic plant assemblage dataset, which included aquatic plant assemblage structure, water purification effect, aquatic plant growth performance, and community stability indicators. Using community purification efficiency and stability as dependent variables and ecological parameter variability as independent variables, a multiple linear regression model or ridge regression model is established. The regression model is as follows: ； in, Indicators representing the overall effectiveness of a community; , and For attribute difference; This is the error term; After normalizing the regression coefficients, they were used as the values ​​in the niche complementarity index model. 、 and Parameter values ​​are adjusted to improve the scientific rigor and adaptability of parameter settings. The multiple linear regression model or ridge regression model is updated periodically. When the accumulation of new monitoring data exceeds a set threshold, the weight coefficients are retrained and updated.

Images