A method for restoring and constructing a lake submerged plant community based on a coupling model

Abstract

This invention proposes a method for restoring and constructing submerged plant communities in lakes based on a coupled model, belonging to the field of ecological restoration engineering technology. The method includes: obtaining a pixel-by-pixel spatiotemporal depth dataset of the restoration area based on underwater elevation data; collecting growth characteristic data of each target species and estimating the plant height of each target species at any time point; calculating the basic light compensation depth of each target species based on underwater light intensity attenuation patterns and light compensation points of each target species, and performing canopy correction based on plant height to obtain the pixel-by-pixel spatiotemporal dynamic maximum suitable water depth for each target species; comparing the spatiotemporal depth dataset with the spatiotemporal dynamic maximum suitable water depth pixel-by-pixel to delineate suitable light environment zones; and combining the light adaptability characteristics of each target species with the temporal variation patterns of the light environment to formulate a spatial layout and planting sequence scheme for the submerged plant community. This invention can improve the success rate of large-scale submerged plant restoration and reduce the cost of aquatic ecosystem restoration.

Description

Technical Field

[0001] This invention relates to the field of ecological restoration engineering technology, and in particular to a method for restoring and constructing submerged plant communities in lakes based on a coupling model. Background Technology

[0002] In recent decades, driven by human activities and climate change, many lakes in my country have experienced intensified eutrophication and the decline of large submerged plants. This has led to a transformation of lake ecosystems from grass-based clear water to algae-based turbid water, causing ecological and environmental problems such as biodiversity loss, algal blooms, and water quality deterioration. Domestic and international scholars, through long-term theoretical and practical research, believe that the reconstruction and restoration of submerged plants is a necessary step in restoring lakes to their original clear water state. Submerged plant restoration is not simply "planting aquatic plants." Not all areas are suitable and / or unsuitable for submerged plant restoration. Only by restoring submerged plants in suitable areas can the restoration and reconstruction of aquatic plant communities be promoted. Therefore, strengthening research on the restoration of submerged plant communities in lakes and providing technical support for the scientific and orderly implementation of submerged plant restoration work is essential.

[0003] The restoration of submerged plants in lakes is influenced by environmental factors such as light, water quality, and water depth. Light is the primary limiting factor for submerged plant growth; underwater light intensity must exceed the plant's light compensation point for them to establish and survive. Water quality (transparency, turbidity, etc.) and water depth are important factors affecting the light compensation point of submerged plants. Chinese patent application CN103778319A discloses a method for determining the submerged plant restoration area in a target water body. This method uses the ratio of the submerged plant's light compensation depth to the deepest water depth to screen suitable target water areas for submerged plant restoration. Chinese patent application CN116282556A discloses a method for restoring submerged plant communities in lakes based on natural restoration suitability diagnosis. This method constructs a curve showing the relationship between the average water depth and cumulative flooding duration of the target lake restoration area to obtain the suitable water depth range for submerged plant growth. However, these methods do not fully consider the complex heterogeneity of water level, water depth, and water quality in time and space, as well as the differences in the physiological needs of different submerged plants at different growth stages. This makes the selection of the above-mentioned submerged plant community restoration areas lack precise spatial positioning and sufficient scientific basis.

[0004] Therefore, in order to meet the high-quality development requirements of current ecological restoration work, a method for constructing lake submerged plant communities based on a coupling model is proposed. Summary of the Invention

[0005] In view of this, the present invention provides a method for constructing a lake submerged plant community restoration based on a coupled model, so as to realize the pixel-by-pixel and time-by-time dynamic evaluation of the light environment suitability of the restoration area, and provide systematic technical support for the accurate delineation of submerged plant restoration areas, species spatial configuration and planting sequence planning, thereby improving the success rate of large-scale submerged plant restoration and reducing the cost of aquatic ecosystem restoration.

[0006] The technical solution of this invention is implemented as follows: This invention provides a method for constructing a lake submerged plant community based on a coupled model, comprising: S1. Based on the underwater elevation data of the restoration area, an underwater elevation raster model is generated by spatial interpolation. A water level time series model is constructed by combining the time series water level data of the target lake. The water level time series model and the underwater elevation raster model are superimposed and calculated to obtain the spatiotemporal water depth dataset of each pixel in the restoration area. S2. Collect growth characteristic data of each target species, construct biological curve models of each target species through curve fitting, and use biological curve models to estimate the plant height of each target species at any time point. S3. Based on the underwater light intensity attenuation law and the light compensation point of each target species, calculate the basic light compensation depth of each target species, and use the plant height to correct the basic light compensation depth by the canopy layer to obtain the spatiotemporal dynamic maximum suitable water depth of each target species per pixel. S4. Compare the spatiotemporal water depth dataset with the spatiotemporal dynamic maximum suitable water depth pixel by pixel, perform time-series statistics on the light environment suitability of each pixel in the restoration area, and delineate the light environment suitability zone. S5. Based on the suitable light environment zone, and combined with the light adaptability characteristics of each target species and the temporal change pattern of the light environment, formulate a spatial layout and planting sequence plan for submerged plant communities.

[0007] Preferably, step S1 includes: Based on the measured underwater topographic data of the restoration area, an underwater elevation raster model is generated using spatial interpolation. Each pixel in the raster is identified by its spatial coordinates (x, y), and the corresponding underwater elevation value is stored. ; Collect time-series water level data of the target lake, with t representing the time node, and construct a water level time-series model. The water level time series model uses time and water level values ​​as its core fields; by and The actual water depth of the repair area is calculated pixel-by-pixel and time-by-time node by performing differential calculations. We obtained a spatiotemporal water depth dataset.

[0008] Preferably, before step S1, the method further includes a step of determining the target recovery range of the repair area: The current distribution range of submerged vegetation in the target lake is obtained through current status survey. A remote sensing extraction model of submerged vegetation is constructed based on remote sensing images. The coverage range of submerged vegetation in each quarter of the historical baseline period is inverted. Spatial overlay analysis of the coverage range of each historical quarter is performed to eliminate accidental distribution areas and determine the areas where submerged vegetation has appeared multiple times in history as the target restoration range.

[0009] Preferably, in step S2, growth height data of each target species during the germination, rapid growth, growth stagnation, and withering stages are obtained through field surveys. S-curves are used to fit the continuous growth height data of each target species throughout the growing season, and the optimal biocurve model for each species is selected through accuracy verification. Where the subscript i is the target species number, , where n is the total number of target species; using Calculate the plant height of species i at any time point t. .

[0010] Preferably, the S-shaped curve family includes the Logistic model, the Gompertz model, the Richards model, and a model with a wilting period correction that superimposes a linear decreasing term on the above models during the wilting period. The above models are fitted to the field-measured altitude data of each target species, and the optimal biocurve model for each species is selected based on the goodness of fit. .

[0011] Preferably, step S3 includes: Based on the principle that underwater light intensity decreases exponentially with water depth, the baseline light compensation depth for each target species is calculated using the photosynthetic-respiratory critical light intensity and water light attenuation parameters as inputs. , where i is the species number; the basic light compensation depth characterizes the water depth at which the plant leaf is exactly located at the light compensation point without considering the plant height; Based on plant height The canopy was modified to adjust the base light compensation depth, converting the light requirements of the plant canopy location to the water surface to obtain the maximum suitable water depth for growth in a spatiotemporal dynamic manner. ,in t represents the pixel spatial coordinates, and t represents the time node.

[0012] Preferably, the basic optical compensation depth The calculation formula is: ; In the formula: Water transparency, in meters (m). Effective photosynthetic radiation incident on the water surface; denoted as the critical light intensity for the light compensation point of species i; a is the transparency attenuation coefficient, ranging from 0.8 to 1.0, with 0.8 for eutrophic lakes and 1.0 for mesotrophic lakes.

[0013] Preferably, the maximum suitable water depth in spatiotemporal dynamics. The calculation formula is: ; In the formula: The effective photosynthetically active radiation incident on the water surface at time t; The critical light intensity for the light compensation point of species i; For the pixel at time t The light attenuation coefficient of the water body at that location; Let be the plant height of species i at time t; Let t be the canopy height of species i at time t.

[0014] Preferably, step S4 includes: For each cell At each time point t, when the actual water depth The appropriate light environment is determined at a given time point. The frequency percentage of time points with appropriate light environment for each pixel throughout the year or the entire growing season is statistically analyzed. Based on the frequency percentage, three types of suitable light environment zones are defined: pixels with a frequency percentage of not less than 80% are defined as long-term suitable zones, pixels with a frequency percentage of 50% to 80% are defined as phased suitable zones, and pixels with a frequency percentage of less than 50% are defined as unsuitable zones. in, The maximum suitable water depth for life in time and space dynamics.

[0015] Preferably, in step S5, spatial matching is performed according to the light environment suitability zone category of each pixel and the light tolerance characteristics of the species, placing light-loving species in long-term suitable zones and low-light-tolerant species in temporary suitable zones; based on the biocurve model of each target species... By estimating the germination time and rapid growth period, and combining the temporal changes in the light environment, the optimal planting time window for each species is determined, and the germination time sequence of multiple species is planned to form a planting sequence scheme covering the entire growing season.

[0016] The present invention has the following advantages over the prior art: (1) This invention introduces plant height and canopy height correction terms into the basic light compensation depth calculation model, incorporating the actual growth status of submerged plants into the dynamic calculation process of suitable water depth. Compared with the method of using a fixed light compensation depth threshold, this correction model can more objectively reflect the actual light conditions of the plant canopy at different growth stages, thus making the delineation results of suitable restoration areas closer to the actual planting needs of plants, and helping to reduce planting failures caused by bias in light condition assessment.

[0017] (2) This invention deeply couples a high-precision underwater topographic model (DEM), long-term water level data, and a species biographies model to achieve stepwise calculation of the light environment suitability for each raster pixel and each time node in the restoration area. Based on the frequency of excellent light environment throughout the growing season, the restoration area is divided into three categories: long-term suitable area, periodically suitable area, and unsuitable area. Compared with static evaluation methods based on average water depth or water depth in a single time period, this system fully reflects the dynamic impact of water level fluctuations and seasonal changes in water quality on underwater illumination conditions, giving the zoning results higher resolution and reliability in both space and time.

[0018] (3) This invention correlates the light environment suitability zoning results with the light requirements of each target species, prioritizing the configuration of light-loving species in long-term suitable areas and configuring low-light-tolerant species in phased suitable areas, thereby achieving a quantitative match between species and local light environment. This configuration method is based on model calculation results, avoiding the problem of species selection not matching the actual light environment in empirical planting schemes, and helps to improve the survival rate of planting in the initial stage of community establishment and the stability of community structure in the later stage.

[0019] (4) This invention utilizes the seasonal dynamics of light environment suitability to rationally match the germination periods of different species. Light-loving species are planted during periods of relatively abundant light, while light-tolerant species are planted during periods of relatively limited light. By coordinating the planting rhythms of early-germinating and late-germinating species, full-season vegetation coverage is achieved. Compared to a uniform planting scheme that does not consider temporal changes in the light environment, this temporal optimization method can, to a certain extent, avoid high-risk planting periods and reduce seedling losses caused by insufficient light in certain periods.

[0020] (5) This invention extracts the FAI index and performs spatial overlay analysis on multiple historical remote sensing images to identify areas where submerged vegetation has been stably distributed in history, serving as an important basis for the target restoration range and eliminating interference from sporadic distribution areas. This approach incorporates historical vegetation baseline information into the early spatial planning of the restoration plan, ensuring that the delineation of the target restoration range is based on the long-term ecological baseline of the lake, thereby improving the scientific nature of the restoration plan in terms of ecological suitability. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1This is a flowchart of the method of the present invention; Figure 2 This refers to the target restoration area of ​​submerged vegetation in Lake A in this embodiment of the invention. Figure 3 This is an underwater DEM topographic map of lake A in an embodiment of the present invention; Figure 4 This is a water level change diagram of Lake A in an embodiment of the present invention; Figure 5 This is a growth curve of different submerged plants in lake A in an embodiment of the present invention. Detailed Implementation

[0023] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0024] This invention provides a method for constructing a lake submerged plant community based on a coupled model, comprising: S1. Based on the underwater elevation data of the restoration area, an underwater elevation raster model is generated by spatial interpolation. A water level time series model is constructed by combining the time series water level data of the target lake. The water level time series model and the underwater elevation raster model are superimposed and calculated to obtain the spatiotemporal water depth dataset of each pixel in the restoration area. S2. Collect growth characteristic data of each target species, construct biological curve models of each target species through curve fitting, and use biological curve models to estimate the plant height of each target species at any time point. S3. Based on the underwater light intensity attenuation law and the light compensation point of each target species, calculate the basic light compensation depth of each target species, and use the plant height to correct the basic light compensation depth by the canopy layer to obtain the spatiotemporal dynamic maximum suitable water depth of each target species per pixel. S4. Compare the spatiotemporal water depth dataset with the spatiotemporal dynamic maximum suitable water depth pixel by pixel, perform time-series statistics on the light environment suitability of each pixel in the restoration area, and delineate the light environment suitability zone. S5. Based on the suitable light environment zone, and combined with the light adaptability characteristics of each target species and the temporal change pattern of the light environment, formulate a spatial layout and planting sequence plan for submerged plant communities.

[0025] Specific embodiments of the present invention are as follows: Figure 1As shown, the overall process includes six core steps: determining the target restoration range, constructing a spatiotemporal water depth model, constructing a bio-curve model, constructing a spatiotemporal dynamic light compensation model, delineating suitable light environment zones, and formulating community restoration plans. These steps are sequentially linked, with data transferred step-by-step. Before each calculation begins, all input data must be preprocessed: the underwater elevation raster model, water level time-series data, plant height raster sequence, and light environment parameter datasets are converted to the same raster spatial resolution and geographic coordinate system. This ensures that each data layer is strictly aligned spatially during subsequent pixel-by-pixel overlay calculations, eliminating systematic errors introduced by coordinate deviations.

[0026] In one embodiment of the present invention, before step S1, a step of determining the target recovery range of the repair area is included.

[0027] A combined transect and quadrat method was used to conduct a survey of the current status of submerged vegetation in the entire lake. The distribution range, species composition, and relationships with habitat factors such as water depth, substrate, and transparency of the submerged plants were systematically recorded. The survey results served as ground samples for subsequent remote sensing model construction. Based on the survey data and combined with the spectral characteristics of multi-temporal remote sensing images of the target lake, a remote sensing extraction model for submerged vegetation adapted to the lake was constructed. This remote sensing extraction model uses the FAI (Floating Algae Index) automatic thresholding method as its core. The accuracy of the extraction results was verified and the threshold was corrected using ground-based measurement points to ensure it was adapted to the water optical characteristics and vegetation distribution features of the target lake.

[0028] In selecting historical baseline periods, based on historical documents, long-term ecological monitoring data, and lake environmental evolution data, historical years with high submerged plant coverage, relatively stable community structure, and suitable habitat conditions were selected as baseline periods. The number of baseline periods was no less than 10 to ensure the representativeness of historical information. Remote sensing images of each baseline period underwent standardized preprocessing by quarter (spring, summer, autumn, winter, or high-water season, normal-water season, low-water season, determined according to the growth season characteristics of submerged plants in the target lake), including radiometric calibration, atmospheric correction, and geometric fine correction, to ensure the comparability of images from different years and quarters. Subsequently, the FAI index automatic thresholding method was used to extract vector maps of submerged vegetation coverage for each year and quarter scene by scene. The coverage vector maps of all historical baseline periods for each quarter were spatially overlaid and analyzed in a unified geographic coordinate system. Distribution areas that only appeared occasionally in individual years or quarters were eliminated, and the core areas where submerged vegetation repeatedly appeared in history were determined as the target restoration range for lake submerged vegetation. Taking Lake A as an example, if... Figure 2 As shown in the above spatial overlay analysis, the target restoration range of submerged vegetation in Lake A is mainly distributed in the shallow water area in the central and southeastern part of the lake area. The northern part of the lake area and the deep water area are not included in the restoration range due to the low frequency of historical submerged vegetation occurrence.

[0029] In one embodiment of the present invention, step S1 includes: Based on measured underwater topographic data of the restoration area, an underwater elevation raster model was generated using spatial interpolation. Ordinary Kriging was preferentially used for spatial interpolation, as it balances spatial autocorrelation structure and interpolation smoothness, making it suitable for high-precision continuous reconstruction of lakebed topography. Measured underwater topographic data were acquired using shipborne multibeam bathymetry or single-beam bathymetry. The density of measurement points was determined based on the area and topographic complexity of the restoration area, prioritizing interpolation accuracy. In the generated underwater elevation raster model, each pixel is represented by spatial coordinates. Identify its planar location and store the corresponding underwater elevation value. The unit is meters (absolute elevation). Taking lake A as an example... Figure 3 As shown, the underwater DEM elevation values ​​of Lake A range from 2166.48 m to 2178.64 m, showing an overall distribution pattern of lower elevation in the lake center and higher elevation in the shoreline. Shallow water areas are mainly concentrated in the southeast of the lake, which is the main potential area for the restoration of submerged plants.

[0030] Time-series water level data of the target lake were collected, with data sourced from actual measurements at hydrological stations in the lake area or from long-term deployed water level gauges. The time resolution was daily or ten-day periods, with t representing the time node, to construct a water level time-series model. This model uses time and water level as core fields to construct a one-dimensional time series dataset. Taking lake A as an example... Figure 4 As shown, the water level of Lake A exhibits a single-peak variation trend throughout the year, with the water level being lower during the dry season (late spring to early summer) and reaching its peak during the wet season (late summer to early autumn). The annual variation in water level is approximately 0.8 m. This water level variation pattern has a significant seasonal regulatory effect on the water depth suitability of submerged plants.

[0031] Water level time series model With underwater elevation grid model The actual water depth of the repaired area is calculated pixel-by-pixel and time-by-time node according to the following formula: ; In the formula: For the pixel at time t The actual water depth at the location, in meters; The water level at time t, in meters (m). For pixels The underwater elevation values ​​at the location are given in meters. The above calculation results constitute a spatiotemporal depth dataset for each pixel in the restoration area, stored in a three-dimensional "space × time" format, providing basic depth data for subsequent light environment suitability assessment.

[0032] In one embodiment of the present invention, step S2 includes: Growth characteristic data of various target species in the target lake for restoration were obtained through a systematic field survey. The survey employed a fixed-point sampling method, covering four typical growth stages of the target species: germination, rapid growth, stagnant growth, and withering. Plant height data for each species were recorded at ten-day or monthly time steps, ultimately forming a continuous growth height data sequence for each species throughout the entire growing season. Habitat parameters such as water depth, transparency, and substrate type were simultaneously recorded during data collection to aid in subsequent model applicability verification.

[0033] Based on continuous growth height data throughout the growing season, a family of S-shaped curves was used to fit the growth height of each target species. The S-shaped curve family includes the Logistic model, the Gompertz model, the Richards model, and a modified model with a lagging period, which adds a linear decreasing term to the above models. The basic forms of each model are as follows: Logistic model: Its core characteristics are: symmetrical S-shaped, with the inflection point at t=c, and stable in the later stage; suitable for species that do not show obvious wilting during the growth cessation period; Gompertz model: Its core characteristics are: asymmetrical S-shaped growth, slow growth in the early stage and stable growth in the later stage; suitable for species with slow germination period and significant rapid growth period. Richards model: Its core features are: flexible S-shape, adjustable curve symmetry; suitable for species with growth characteristics between Logistic and Gompertz. Model with Wilt Correction: Based on the values ​​calculated by any of the above S-shaped models, at the start of the wilt period... Introducing segmented processing, the model including the wilt period correction (taking Logistic as an example, the same applies to other models) takes the following form: ; In the above formulas: a is the theoretical maximum plant height of the species, in cm; b is the growth rate parameter; c is the inflection point of growth (corresponding to the midpoint of the rapid growth period, in ten-day periods); k is the shape parameter of the Richards model, the value of which controls the symmetry of the S-curve, approaching the Logistic model when k→1, and approaching the Gompertz model when k→+∞; d is the linear decline slope during the wilt period, in cm / ten-day periods. The core characteristics of the Logistic modified model (including the wilt period) are: an S-shape in the first half and a linear decline in the second half; it is suitable for species that show significant wilt after the growth cessation period.

[0034] The initial parameter values ​​for the above models are set according to the following principles: the initial value of a is the maximum plant height measured in the field throughout the entire growing season; the initial value of b is estimated based on the daily growth rate of plant height during the rapid growth period; and the initial value of c is the midpoint between the start and end times of the rapid growth period. After determining the range of initial parameter values, the nonlinear least squares method is used to estimate the parameters of each model. The solution process uses the Levenberg-Marquardt (LM) algorithm as the core iterative method. This algorithm combines the advantages of gradient descent and Gauss-Newton methods, and can still converge robustly even when the initial parameter values ​​have large deviations. It is suitable for nonlinear parameter estimation of S-shaped growth curves.

[0035] The field-measured plant height data for each target species were fitted using the four models described above, with the coefficient of determination used as the basis for the fitting. Accuracy was validated using the root mean square error (RMSE) as a criterion to select the optimal biomolecular curve model for each species. Where the subscript i is the target species number, , where n is the total number of target species. Using the optimal model... Calculate the plant height of species i at any time point t. Taking lake A as an example, such as Figure 5 As shown, the plant height of the eight target species, including Vallisneria natans, Myriophyllum spicatum, Chara, Potamogeton pectinatum, Potamogeton bambusoides, Potamogeton schreberi, Ceratophyllum demersum, and Hydrilla verticillata, all exhibited a typical "growth-stagnation-decline" single-peak curve pattern throughout the entire growing season. The timing and magnitude of the peak plant height varied significantly among the species, reflecting the significant differentiation in growth rhythm and morphological specifications among different species. This difference has a direct impact on the subsequent canopy correction calculation.

[0036] In one embodiment of the present invention, step S3 includes: Underwater light intensity decreases exponentially with depth, following the Beer-Lambert law. When the photosynthetically active radiation at a certain depth equals the critical photosynthetic-respiratory light intensity (i.e., the light compensation point) of a submerged plant, that depth is the light compensation depth for that species; that is, the maximum water depth at which the plant leaves just maintain photosynthetic-respiratory balance without considering plant height. The baseline light compensation depths for each target species are as follows: The calculation formula is: ; In the formula: Water transparency, in meters (m), obtained through on-site measurement using a Secchi disk. For photosynthetically active radiation incident on the water surface, unit It can be calculated from meteorological station radiation data or measured synchronously on-site; The critical light intensity for the light compensation point of species i, in units of The values ​​were obtained through indoor photosynthesis-respiration response curves; α is the transparency attenuation coefficient, ranging from 0.8 to 1.0, with 0.8 for eutrophic lakes and 1.0 for meso-oligotrophic lakes.

[0037] Transparency in the above formula It is a comprehensive indicator characterizing the overall optical state of a water body. In practical applications, the processing path should be selected according to the data conditions: when the remediation area is small and the spatial homogeneity of the water quality is strong, the spatial average of the measured values ​​from multiple Seymourne disks can be used as a unified input; when there is significant spatial heterogeneity in the water quality of the remediation area, it is necessary to... Spatialization, and then calculation of the per-pixel light attenuation coefficient. This is to ensure that subsequent canopy correction calculations possess complete spatiotemporal dynamic characteristics.

[0038] There are two methods for spatial extrapolation: one is to use an empirical formula method, based on the synchronously monitored chlorophyll a concentration. (unit ), suspended particulate matter concentration (unit ) and the absorption coefficient of colored dissolved organic matter (unit Let ) be the independent variable, and establish a system of the form The empirical formula for multiple linear regression, with regression coefficients derived from synchronously measured values ​​of the target lake. —First, water quality parameter pairing data calibration; second, using the water quality model method, the time-by-time and space-by-space water quality concentration fields output by the calibrated water quality model (such as EFDC, MIKE21, etc.) are substituted into the above empirical formulas or the model's embedded optical module to directly output the data. The spatiotemporal distribution field. Both methods can provide pixel-by-pixel, time-by-time node data for subsequent raster operations. The specific data selection is determined based on the density of the water quality monitoring network and the current status of the model construction for the target lake.

[0039] The above These are static baseline values ​​and do not consider the influence of plant height on light interception. Photosynthesis in submerged plants occurs in the leaf canopy, not the roots, and the water depth within the canopy is less than the water depth at the plant's roots. Therefore, the actual suitable planting depth is greater than [missing value]. Therefore, the plant height calculated in step S2 is used as a reference. The canopy was modified to adjust the base light compensation depth, converting the light requirements of the plant canopy to the water surface to obtain the maximum suitable water depth for growth in a spatiotemporal dynamic manner. The calculation formula is: ; In the formula: Let be the incident photosynthetically active radiation at time t, in units of ; The critical light intensity for the light compensation point of species i, in units of ; For the pixel at time t Light attenuation coefficient of water body at a given location, in units of The calculation method is described above; For species i at time t, in pixel Plant height at this location, in meters, is derived from the biocurve model. Calculation; Let be the canopy height of species i at time t, in meters, defined as the height of the apical functional leaf from the base of the rhizome. The first term on the right side of the formula is the light compensation depth, precisely calculated based on the Beer-Lambert light attenuation law, i.e., the light intensity attenuates from the water surface to the species' light compensation point. The corresponding water depth at that time; the second item The stem length from the base of the canopy to the bottom mud is used to convert the light compensation position (canopy top) into the depth of the bottom mud where the plant roots are located, thereby obtaining the maximum suitable water depth based on the water surface.

[0040] Canopy height The canopy height was directly measured through field surveys: In the plant height survey in step S2, the height of the top functional leaves of each species was recorded simultaneously at each survey time point, covering the entire process from germination to growth cessation; the species-specific canopy ratio coefficient was calculated based on the ratio of canopy height to the plant height at the same time. The average value is taken in segments according to the growth stages (germination period, rapid growth period, and growth stagnation period); in actual calculations, the average value is taken according to the growth stages (germination period, rapid growth period, and growth stagnation period). The canopy height is determined at each time point.

[0041] The above calculations were performed using ArcGIS Model Builder to perform batch raster operations. Underwater elevation raster data, water level time-series data, plant height raster sequences, and light environment parameters were input one by one to complete the processing of each pixel and time point in the restoration area. Batch calculations generate spatiotemporal dynamic maximum suitable water depth raster sequences.

[0042] In one embodiment of the present invention, step S4 includes: For each cell in the repair area At each time point t, the actual water depth obtained in step S1 is... The spatiotemporal dynamic maximum suitable water depth obtained in step S3 Perform pixel-by-pixel comparison: when When, it is determined that the light environment of that pixel at that time point is suitable for species i; when If the light environment is unsuitable at that time point, it is determined that the light environment at that time point is unsuitable.

[0043] Pixel-by-pixel analysis was conducted to determine the proportion of times with suitable light conditions throughout the year or the entire growing season. Based on this proportion, three categories of suitable light environments were defined: pixels with a proportion of at least 80% were designated as long-term suitable areas, where light conditions meet the photosynthetic needs of the target species for most of the growing season; pixels with a proportion of 50% to 80% were designated as periodically suitable areas, where light conditions meet the requirements for specific growing seasons; and pixels with a proportion of less than 50% were designated as unsuitable areas, where water depth is too high or water transparency is insufficient, and the overall light environment does not meet the establishment requirements of the target species. The 80% and 50% thresholds mentioned above are determined based on the following: Studies on the effective photosynthetic accumulation of submerged plants show that when the light satisfaction rate is not less than 80% throughout the growing season, the plants can maintain positive net primary productivity and achieve population expansion; when the satisfaction rate is between 50% and 80%, the plants can complete the accumulation of effective substances during periods of suitable light, but due to water level fluctuations, they need to have a certain ability to adapt to weak light during population establishment; when the satisfaction rate is less than 50%, the cumulative negative effect of insufficient light on the plant's carbon balance is sufficient to inhibit population establishment, making it unsuitable as a planting area. In practical applications, if the target lake has historical vegetation distribution data, the above thresholds can also be calibrated to ensure that the delineation results are consistent with the historical patterns of local species establishment. The above raster overlay statistical calculations were also completed in batches on the ArcGIS platform using the model builder.

[0044] In one embodiment of the present invention, step S5 includes: Based on the suitable light environment zones defined in step S4, spatial configuration is carried out according to the light tolerance characteristics of the species. Light-loving species (such as *Myriophyllum spicatum* and *Potamogeton malaianus*) are placed in the long-term suitable zone, where the light conditions are consistently sufficient, which is conducive to the germination, establishment, and vigorous growth of light-loving species. Low-light-tolerant species (such as *Vallisneria natans* and *Ceratophyllum demersum*) are placed in the periodic suitable zone, where the light conditions meet the photosynthetic needs of low-light-tolerant species during specific growth periods, enabling effective establishment.

[0045] In terms of planting sequence planning, it is based on the biomass curve model of each target species. The germination time and rapid growth period of each species are estimated. Combined with the spatiotemporal dynamic light environment suitability raster sequence generated in step S4, the optimal planting time window for each species in each suitable area is determined.

[0046] The specific logic for determining the planting time window is as follows: For each target pixel in the temporal suitability raster sequence, the search starts from the end time of germination calculated by the biological curve model, and the first consecutive N time nodes in the subsequent time series that meet the suitable light environment conditions (i.e., The starting time of the pixel is the recommended planting time for that pixel; the value of N corresponds to the minimum number of days of continuous light required for the root system of the target species to complete the initial establishment, which is determined by indoor planting experiments or literature data for each species, and is generally 7 to 14 days.

[0047] The aforementioned decision-making logic ensures a sufficiently long and continuous window of suitable light after the selected planting time, providing stable photosynthetic support to transplanted plants during their initial, less resilient stage, thereby improving survival rates. By planning the combined planting of early-germinating species (such as Vallisneria natans) and late-germinating species (such as Myriophyllum spicatum), the rapid growth periods of different species are complementary in time, forming a planting sequence plan that covers the entire growing season and reducing seasonal gaps in vegetation cover within the restoration area. Finally, by combining the spatial distribution of suitable light environments with the planting sequence plan, and overlaying auxiliary layers such as water level, water quality, and historical baseline data, the core restoration area of ​​the submerged plant community is comprehensively delineated, and a zoned and time-sequential planting implementation plan is formed.

[0048] The aforementioned method deeply couples the underwater elevation model, the water level time series model, and the plant biocurve model, transforming the assessment of light environment suitability from traditional static water depth discrimination to dynamic calculation on a pixel-by-pixel and time-node-by-time basis. This overcomes the limitations of previous methods that used fixed water depth thresholds or static light compensation depths to delineate recovery zones. The introduction of a canopy correction term makes the estimation of the maximum suitable water depth more consistent with the light interception mechanism of submerged plants under actual growth conditions, helping to reduce the underestimation of suitable area due to neglecting plant height. The suitability classification and grading method based on light environment frequency statistics quantifies the impact of annual dynamic changes in water level on plant light availability into an operational spatial layer, providing a practically supported basis for differentiated species allocation and time series planning.

[0049] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for constructing a lake submersed plant community restoration based on a coupling model, characterized in that, include: S1. Based on the underwater elevation data of the restoration area, an underwater elevation raster model is generated by spatial interpolation. A water level time series model is constructed by combining the time series water level data of the target lake. The water level time series model and the underwater elevation raster model are superimposed and calculated to obtain the spatiotemporal water depth dataset of each pixel in the restoration area. S2. Collect growth characteristic data of each target species, construct biological curve models of each target species through curve fitting, and use biological curve models to estimate the plant height of each target species at any time point. S3. Based on the underwater light intensity attenuation law and the light compensation point of each target species, calculate the basic light compensation depth of each target species, and use the plant height to correct the basic light compensation depth by the canopy layer to obtain the spatiotemporal dynamic maximum suitable water depth of each target species per pixel. S4. Compare the spatiotemporal water depth dataset with the spatiotemporal dynamic maximum suitable water depth pixel by pixel, perform time-series statistics on the light environment suitability of each pixel in the restoration area, and delineate the light environment suitability zone. S5. Based on the suitable light environment zone, and combined with the light adaptability characteristics of each target species and the temporal change pattern of the light environment, formulate a spatial layout and planting sequence plan for submerged plant communities.

2. The method of claim 1, wherein, Step S1 includes: Based on the measured underwater topographic data of the restoration area, an underwater elevation raster model is generated using spatial interpolation. Each pixel in the raster is identified by its spatial coordinates (x, y), and the corresponding underwater elevation value is stored. ; Collect time-series water level data of the target lake, using t to represent time points, and construct a water level time-series model. The water level time series model uses time and water level values ​​as its core fields; by and The actual water depth of the repair area is calculated pixel-by-pixel and time-by-time node by performing differential calculations. We obtained a spatiotemporal water depth dataset.

3. The method according to claim 1 or 2, characterized in that, Before step S1, the method further includes determining the target recovery range of the repair area: The current distribution range of submerged vegetation in the target lake is obtained through current status survey. A remote sensing extraction model of submerged vegetation is constructed based on remote sensing images. The coverage range of submerged vegetation in each quarter of the historical baseline period is inverted. Spatial overlay analysis of the coverage range of each historical quarter is performed to eliminate accidental distribution areas and determine the areas where submerged vegetation has appeared multiple times in history as the target restoration range.

4. The method according to claim 1, characterized in that, In step S2, growth height data for each target species during the germination, rapid growth, growth stagnation, and withering stages are obtained through field surveys. S-curves are used to fit the continuous growth height data of each target species throughout the growing season, and the optimal biocurve model for each species is selected through accuracy verification. Where the subscript i is the target species number, , where n is the total number of target species; utilizing Calculate the plant height of species i at any time point t. .

5. The method according to claim 4, characterized in that, The S-shaped biocurve family includes the Logistic model, the Gompertz model, the Richards model, and a modified model with a lagging period that adds a linear decreasing term to the above models. These models were fitted to field-measured altitude data for each target species, and the optimal biocurve model for each species was selected based on goodness of fit. .

6. The method according to claim 1, characterized in that, Step S3 includes: Based on the exponential decay of underwater light intensity with water depth, the baseline light compensation depth for each target species is calculated using the photosynthetic-respiratory critical light intensity and water light attenuation parameters as inputs. , where i is the species number; the basic light compensation depth characterizes the water depth at which the plant leaf is exactly located at the light compensation point without considering the plant height; Based on plant height The canopy was modified to adjust the base light compensation depth, converting the light requirements of the plant canopy location to the water surface to obtain the maximum suitable water depth for growth in a spatiotemporal dynamic manner. ,in t represents the pixel space coordinates, and t represents the time node.

7. The method according to claim 6, characterized in that, Basic optical compensation depth The calculation formula is: ； In the formula: Water transparency, in meters (m). Effective photosynthetic radiation incident on the water surface; denoted as the critical light intensity for the light compensation point of species i; a is the transparency attenuation coefficient, ranging from 0.8 to 1.0, with 0.8 for eutrophic lakes and 1.0 for mesotrophic lakes.

8. The method according to claim 6, characterized in that, Spatiotemporal dynamic maximum suitable water depth The calculation formula is: ； In the formula: The effective photosynthetic radiation incident on the water surface at time t; The critical light intensity for the light compensation point of species i; For the pixel at time t The light attenuation coefficient of the water body at that location; Let be the plant height of species i at time t; Let t be the canopy height of species i at time t.

9. The method according to claim 2, characterized in that, Step S4 includes: For each cell At each time point t, when the actual water depth The appropriate light environment is determined at a given time point. The frequency percentage of time points with appropriate light environment for each pixel throughout the year or the entire growing season is statistically analyzed. Based on the frequency percentage, three types of suitable light environment zones are defined: pixels with a frequency percentage of not less than 80% are defined as long-term suitable zones, pixels with a frequency percentage of 50% to 80% are defined as phased suitable zones, and pixels with a frequency percentage of less than 50% are defined as unsuitable zones. in, The maximum suitable water depth for life in time and space dynamics.

10. The method according to claim 1, characterized in that, In step S5, based on the suitable light environment zone category of each pixel, spatial matching is performed according to the species' light tolerance characteristics. Light-loving species are placed in long-term suitable zones, and low-light-tolerant species are placed in temporary suitable zones. This is done based on the biomolecular curve model of each target species. By estimating the germination time and rapid growth period, and combining the temporal changes in the light environment, the optimal planting time window for each species is determined, and the germination time sequence of multiple species is planned to form a planting sequence scheme covering the entire growing season.

Images