A method and device for identifying a water source conservation and ecological restoration area

Abstract

The application discloses a water source conservation and ecological restoration area identification method and device. The method comprises the following steps: first, obtaining the water source conservation amount and water source conservation influencing factors of a target area; screening and detecting the water source conservation amount and the water source conservation influencing factors to obtain optimal water source conservation values, dominant factors and background factors; clustering and classifying the background factors and the dominant factors after superimposing the background factors to perform partition processing on the background factors and the dominant factors, thereby obtaining a plurality of effective ecological partition units; combining the optimal water source conservation values to quantify the water source conservation amount of each effective ecological partition unit, thereby obtaining water source conservation potential values and potential surplus amounts; and finally, dividing restoration priority levels according to the water source conservation potential values and the potential surplus amounts to identify restoration areas. The application makes up for the problems of low identification accuracy and weak priority targeting of traditional methods, improves the scientificity of partition and evaluation, and effectively improves the water source conservation and ecological restoration efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ecological environment restoration, and in particular to a method and device for identifying water conservation and ecological restoration areas. BACKGROUND

[0002] With the problems of water conservation function degradation and ecological system service capacity decline becoming increasingly prominent, accurately identifying key water conservation areas and ecological restoration priority areas has become a core requirement for ensuring regional ecological security and improving ecological system stability. However, in the current identification of water conservation and ecological restoration areas, although existing methods attempt to analyze water conservation influencing factors, there are still obvious limitations. On the one hand, multiple influencing factors such as terrain, soil, and climate are not systematically selected and detected, resulting in low utilization efficiency of factors and inaccurate positioning of key influencing factors. On the other hand, the clustering and classification superposition logic in the partition process is extensive and cannot scientifically quantify the restoration potential and surplus gap of each partition, ultimately resulting in weak targeting of ecological restoration priority classification, misallocation of restoration resources, and suboptimal restoration results.

[0003] Therefore, how to achieve the coordination of effective ecological zoning and scientific potential assessment, and then accurately identify water conservation and ecological restoration areas, is a key problem that needs to be solved. SUMMARY

[0004] Therefore, the method and device for identifying water conservation and ecological restoration areas provided by the embodiments of the present application can improve the scientificity of zoning and assessment and effectively improve the efficiency of water conservation and ecological restoration. The method and device for identifying water conservation and ecological restoration areas provided by the embodiments of the present application are implemented as follows:

[0005] The method for identifying water conservation and ecological restoration areas provided by the embodiments of the present application comprises the following steps:

[0006] Obtaining the water conservation amount of a target area and water conservation influencing factors, wherein the water conservation influencing factors include at least one of terrain, soil, climate, vegetation, and human activities;

[0007] Screening and detecting the water conservation amount and the water conservation influencing factors to obtain optimal water conservation values, dominant factors, and background factors;

[0008] Clustering and classifying the background factors to obtain processed background factors, and partitioning the processed background factors and the dominant factors to obtain multiple effective ecological partition units;

[0009] The optimal water conservation value and the water conservation amount corresponding to each effective ecological zoning unit are statistically quantified to obtain a water conservation potential value corresponding to each effective ecological zoning unit and a potential surplus amount;

[0010] An ecological restoration priority level is divided according to the water conservation potential value and the potential surplus amount, and a water conservation and ecological restoration area is identified based on the ecological restoration priority level to obtain an identification result.

[0011] In some embodiments, the water conservation amount and the water conservation influencing elements are screened and detected to obtain an optimal water conservation value, a dominant factor and a background factor, including:

[0012] The water conservation amount of the target area is subjected to time series screening to obtain an optimal water conservation value, which is used to represent the optimal state of the ecological environment in the observation period of the target area;

[0013] The water conservation influencing elements are screened to obtain terrain basic elements, soil hydrological elements, climate condition elements, vegetation ecological elements and human intervention elements;

[0014] The terrain basic elements, the soil hydrological elements, the climate condition elements, the vegetation ecological elements and the human intervention elements are removed and integrated to obtain an initial factor pool;

[0015] The optimal water conservation value and each factor in the initial factor pool are detected to obtain a dominant factor and a background factor.

[0016] In some embodiments, the background factor is clustered and classified and superimposed to obtain a processed background factor, and the processed background factor and the dominant factor are subjected to zoning processing to obtain a plurality of effective ecological zoning units, including:

[0017] An optimal discretization scheme is obtained, and the background factor and the optimal discretization scheme are preprocessed and clustered to obtain an ecological background unit;

[0018] The dominant factor and the ecological background unit are subjected to zoning processing to obtain a theoretical joint zoning;

[0019] The theoretical joint zoning and the dominant factor are subjected to invalid factor combination removal processing to obtain a plurality of effective ecological zoning units.

[0020] In some embodiments, the optimal discretization scheme is obtained, and the background factor and the optimal discretization scheme are preprocessed and clustered to obtain an ecological background unit, including:

[0021] obtaining an optimal discretization scheme generated by an optimal parameter geographical detector;

[0022] performing matching preprocessing on the background factor and the optimal discretization scheme to obtain a preprocessed background factor;

[0023] performing clustering processing on the preprocessed background factor by a clustering algorithm to obtain an ecological background unit, the clustering algorithm taking minimizing intra-class variance and maximizing inter-class difference as a target.

[0024] In some embodiments, the detecting processing on the optimal water conservation value and each factor in the initial factor pool to obtain the dominant factor and the background factor comprises:

[0025] performing factor detecting processing on the optimal water conservation value and each factor in the initial factor pool to obtain the dominant factor, the dominant factor being a factor whose explanatory power on the optimal water conservation value exceeds a preset explanatory power threshold;

[0026] performing interactive detecting processing on each factor in the initial factor pool to obtain the background factor, the background factor being a factor whose nonlinear enhancement action frequency is higher than a preset frequency threshold.

[0027] In some embodiments, the partition processing on the dominant factor and the ecological background unit to obtain a theoretical joint partition comprises:

[0028] performing matching preprocessing on the dominant factor and the optimal discretization scheme to obtain a preprocessed dominant factor;

[0029] performing partition processing on the preprocessed dominant factor and the ecological background unit to obtain a theoretical joint partition.

[0030] In some embodiments, the effective ecological partition unit comprises a priority restoration area, a key improvement area, a moderate optimization area, and a present situation maintenance area.

[0031] The embodiment of the present application provides a water conservation and ecological restoration area identification device, which comprises:

[0032] An acquisition module is configured to acquire a water conservation amount and water conservation influence elements of a target area, the water conservation influence elements comprising at least one of terrain, soil, climate, vegetation, and human activities;

[0033] A processing module is configured to perform screening and detecting processing on the water conservation amount and the water conservation influence elements to obtain an optimal water conservation value, a dominant factor, and a background factor;

[0034] The processing module is further configured to cluster and classify the background factors to obtain processed background factors, and to perform partition processing on the processed background factors and the dominant factors to obtain a plurality of effective ecological partition units.

[0035] The processing module is further configured to perform statistical quantification processing on the optimal water conservation value and the water conservation amount of each effective ecological partition unit to obtain a water conservation potential value and a potential surplus amount corresponding to each effective ecological partition unit.

[0036] The identification module is further configured to divide an ecological restoration priority level according to the water conservation potential value and the potential surplus amount, and to identify a water conservation and ecological restoration area based on the ecological restoration priority level to obtain an identification result.

[0037] The computer device provided in the embodiments of the present application includes a memory and a processor, the memory stores a computer program capable of running on the processor, and the processor implements the method provided in the embodiments of the present application when executing the program.

[0038] The computer readable storage medium provided in the embodiments of the present application stores a computer program, and the computer program is executed by a processor to implement the method provided in the embodiments of the present application.

[0039] The method and device for identifying a water conservation and ecological restoration area provided in the embodiments of the present application can obtain a water conservation amount and a water conservation influencing element of a target area, filter and detect the water conservation amount and the water conservation influencing element to obtain an optimal water conservation value, dominant factors and background factors, perform partition processing on the background factors and the dominant factors after clustering and classifying the background factors to obtain a plurality of effective ecological partition units, statistically quantify the water conservation amount of each effective ecological partition unit in combination with the optimal water conservation value to obtain a water conservation potential value and a potential surplus amount, and finally divide a restoration priority level according to the water conservation potential value and the potential surplus amount to identify a restoration area. In this way, the method can make up for the low identification accuracy and weak priority targeting of the traditional method, improve the scientific nature of partition and evaluation, effectively improve the water conservation and ecological restoration efficiency, and solve the technical problems in the background technology. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments of the present application or the prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without any creative labor.

[0041] Figure 1An implementation flowchart of a water source conservation and ecological restoration area identification method provided by an embodiment of the present application is shown in FIG. 1.

[0042] Figure 2 An implementation flowchart of obtaining optimal water source conservation values, dominant factors and background factors provided by an embodiment of the present application is shown in FIG. 4.

[0043] Figure 3 A structural diagram of a water source conservation and ecological restoration area identification device provided by an embodiment of the present application is shown in FIG. 5. DETAILED DESCRIPTION

[0044] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of, but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0045] The following describes some technologies related to the embodiments of the present application to help understanding, which should be considered only as exemplary. Therefore, a person of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. Also, for the sake of clarity and conciseness, the description of some well-known functions and structures is omitted in the following description.

[0046] Figure 1 An implementation flowchart of a water source conservation and ecological restoration area identification method provided by an embodiment of the present application is shown in FIG. 1, which includes steps 101 to 105. In the steps, Figure 1 The execution sequence shown in the embodiments of the present application is only one execution sequence, and does not represent the only execution sequence of a water source conservation and ecological restoration area identification method. As long as the final result can be achieved, Figure 1 The steps shown can be executed in parallel or in reverse.

[0047] Step 101: Obtain the water source conservation amount and water source conservation influence elements of a target area.

[0048] In the embodiments of the present application, the water source conservation amount of the target area is obtained by using the water balance method, and the formula is (1). In formula (1), is the water source conservation amount, P is the rainfall amount, ET is the actual evapotranspiration amount, and Q is the surface runoff amount.

[0049] Five elements were selected: topography, soil, climate, vegetation, and human activities. Topography included elevation and slope; soil included porosity and organic carbon content; climate included annual rainfall and average annual temperature; vegetation included annual maximum NDVI (Normalized Difference Vegetation Index) and vegetation cover; and human activities included land use type and population density. All elements were standardized to a 1km×1km resolution and a time scale consistent with the conservation capacity.

[0050] Step 102 involves screening and detecting the water conservation capacity and its influencing factors to obtain the optimal water conservation value, dominant factors, and background factors.

[0051] In this embodiment of the application, the annual average water conservation capacity is calculated by statistically analyzing the regional average value grid by grid over time. The value with the largest value that has been verified by ecological and climatic conditions is selected as the optimal water conservation value.

[0052] Factors influencing water conservation were first categorized into five types based on their ecological function correlation: topographical foundation, soil and hydrology, etc. Factors with a data missing rate >20% and an absolute value of the correlation coefficient for water conservation <0.1 were then removed, forming an initial factor pool. An optimal parameter geographic detector was used, with the optimal water conservation value as the dependent variable and the factors in the factor pool as independent variables. Factor detection was used to screen factors with an explanatory power (q-value) ≥30% and a significance test (p) <0.01 as dominant factors. Interaction detection was used to screen factors with the highest frequency of nonlinear enhancing interactions, which were then used as background factors.

[0053] Step 103: Cluster and classify the background factors to obtain the processed background factors. Then, partition the processed background factors and the dominant factors to obtain multiple effective ecological partition units.

[0054] In this embodiment, the background factors are first processed by obtaining the optimal discretization scheme that maximizes the q value based on the optimal parameter geographic detector, and then the k-means clustering algorithm is used to form ecological background units.

[0055] After preprocessing the dominant factors using the same optimal discretization scheme, they are layered with ecological background units according to the explanatory power of the dominant factors from high to low, forming a theoretical joint partition. Two types of invalid combinations are eliminated: combinations without actual spatial distribution and combinations that violate ecological laws (such as high-altitude areas corresponding to low rainfall). Finally, effective ecological partition units with spatial continuity and clear ecological significance are obtained.

[0056] Step 104, the optimal water conservation value and the water conservation amount corresponding to each effective ecological partition unit are statistically quantified to obtain the water conservation potential value corresponding to each effective ecological partition unit and the potential surplus amount.

[0057] In the embodiment of the application, the average value, 75% quantile, 90% quantile and other statistical indicators of the water conservation amount of each effective ecological partition unit are calculated. In combination with the optimal water conservation value, the 90% quantile close to the optimal level and resistant to abnormal fluctuations is selected as the water conservation potential value. The potential surplus amount is obtained through the potential value-unit current conservation amount average calculation.

[0058] Step 105, according to the water conservation potential value and the potential surplus amount, the ecological restoration priority level is divided, and the water conservation and ecological restoration area is identified based on the ecological restoration priority level to obtain an identification result.

[0059] In the embodiment of the application, the ecological restoration priority level is divided according to the percentage of the potential surplus amount. The priority restoration area (potential surplus amount≥80% quantile), the key improvement area (50% quantile≤potential surplus amount<80% quantile), the moderate optimization area (20% quantile≤potential surplus amount<50% quantile), and the current maintenance area (potential surplus amount<20% quantile). The level is superimposed with the regional administrative boundary, and the spatial distribution map and attribute table of the water conservation and ecological restoration area are output.

[0060] The embodiment of the application solves the problems of process fragmentation and loose connection of each link in the traditional identification method by constructing a full-process standardized technical framework covering data acquisition-screening detection-clustering partition-potential evaluation-restoration identification, and ensures the reproducibility and stability of the technology landing. The optimal water conservation value is introduced as the core reference to unify the factor detection target and the potential evaluation scale, avoiding the problems of inaccurate dominant factor positioning and fuzzy restoration potential caused by the lack of unified reference in traditional methods. Through the clustering superposition of background factors and dominant factors and the classification of potential surplus amount, the partition of uniform ecological properties and the precise restoration priority linkage are realized, and the problems of unclear ecological significance of traditional partition and mismatch of restoration resources are solved.

[0061] On the basis of the above Figure 1 The application also provides an implementation process diagram for obtaining the optimal water conservation value, the dominant factor and the background factor, as shown in Figure 2 , comprising steps 201 to 204:

[0062] Step 201, the time series screening processing is performed on the water conservation amount of the target area to obtain the optimal water conservation value.

[0063] In the embodiment of the present application, based on the time series data of water source conservation amount, the annual value of the water source conservation amount grid data is calculated by using the ArcGIS grid calculator, and the annual average water source conservation amount of the target region is calculated by using the partition statistical tool to generate the year-annual conservation amount corresponding table.

[0064] According to the descending order of the annual conservation amount, the maximum value is selected as the candidate optimal value.

[0065] The vegetation coverage (≥65%, the highest in the observation period) and the human disturbance index (the construction land ratio ≤8.3%, the lowest in the observation period) of the candidate year are extracted to ensure that the ecosystem is in the optimal state.

[0066] According to the spatial distribution mean of precipitation / surface runoff / actual evapotranspiration for many years, the annual rainfall (≥1350mm, 14.3% higher than the mean value of the observation period) and the annual actual evapotranspiration (≤980mm, 3.4% lower than the mean value of the observation period) of the candidate year are beneficial to water conservation.

[0067] After the above verification, the candidate value can represent the optimal state of the ecological environment in the observation period of the target region, and is determined as the optimal water conservation value.

[0068] In step 202, the water source conservation influencing factors are screened to obtain terrain basic factors, soil hydrological factors, climate condition factors, vegetation ecological factors and human intervention factors.

[0069] In the embodiment of the present application, the factors that have no direct correlation or weak correlation with the water source conservation function are removed, and the terrain-soil-climate-vegetation-human full-dimensional core factors are retained.

[0070] The terrain basic factors include elevation and slope, which reflect the influence of terrain on runoff path and infiltration.

[0071] The soil hydrological factors include soil porosity and soil organic carbon content, which directly determine the soil water retention and infiltration capacity.

[0072] The climate condition factors include annual rainfall and annual average temperature, which provide the hydrological input and energy background of water conservation.

[0073] The vegetation ecological factors include annual maximum NDVI and vegetation coverage, which regulate the water interception and transpiration process. Among them, the vegetation coverage is calculated according to the formula (2), wherein is the NDVI value of the non-vegetation coverage area, is the NDVI value of the pure vegetation coverage area, , and the NDVI value in the confidence interval of 5%-95% is set.

[0074] Human intervention factors include land use type and population density, reflecting the disturbance intensity of human activities on the ecosystem.

[0075] In step 203, the terrain base factor, soil hydrological factor, climate condition factor, vegetation ecological factor and human intervention factor are removed and integrated to obtain an initial factor pool.

[0076] In the embodiment of the present application, the terrain base factor, soil hydrological factor, climate condition factor, vegetation ecological factor and human intervention factor are subjected to data quality verification-format integration to construct an initial factor pool.

[0077] The missing rate of each factor grid data is counted, and the factors with a missing rate greater than 20% are removed, and the factors with a missing rate less than or equal to 20% are filled by the neighborhood mean method.

[0078] The Pearson correlation coefficient of each factor and water conservation capacity is calculated, and the factors with an absolute value of the correlation coefficient less than 0.1 are removed, and the factors with an absolute value of the correlation coefficient greater than or equal to 0.1 are retained.

[0079] The retained factors are uniformly converted into ASCII grid format, the spatial resolution is 1km x 1km, the time scale is unified to the ecological environment optimal state year data, an initial factor pool is formed, and each factor is numbered (such as F1-elevation, F2-slope, F3-soil porosity, etc.).

[0080] In step 204, the optimal water conservation value and each factor in the initial factor pool are subjected to detection processing to obtain dominant factors and background factors.

[0081] In the embodiment of the present application, the optimal parameter geographic detector is used to detect the initial factor pool with the optimal water conservation value as the core reference.

[0082] The factors of the initial factor pool and the optimal water conservation value grid data are generated by the ArcGIS random point tool to generate 1000 evenly distributed sample points, and the multi-value to point tool is used to assign the factors and the optimal water conservation value attributes to the sample points to generate a CSV format sample data set.

[0083] The factor detection module is used to calculate the explanatory power of each factor to the optimal water conservation value according to formula (3) (3), wherein q is the explanatory power, L is the number of classification and classification of the variable factor, N is the number of samples in different classification regions and the whole region, and the variance of different classification regions and the whole region, h is the hth classification region of the influencing factor, h=1, 2, …, L.

[0084] An interactive detection module is used to calculate the interaction q value of each factor with the remaining 9 factors, to determine the interaction type (nonlinear enhancement, synergistic enhancement, etc.), and to focus on the frequency of nonlinear enhancement (interaction q value > maximum q value of single factor, which is nonlinear enhancement).

[0085] Referring to the factor detection map heat map result, select the factors with the top 30% of nonlinear enhancement frequency as background factors (for example, elevation and annual rainfall, vegetation coverage, all produce nonlinear enhancement, frequency = 3; soil porosity, elevation, and land use produce nonlinear enhancement, frequency = 3, all meet the standard).

[0086] Identify the dominant factors (such as annual rainfall, annual maximum NDVI) and background factors (such as elevation, soil porosity, land use type).

[0087] The embodiments of the application optimize the optimal water conservation value screening logic to avoid misjudging the abnormally high conservation capacity caused by accidental hydrological events as the optimal value, and to ensure that the optimal value can truly represent the optimal state of regional ecology, providing reliable reference for subsequent links. By eliminating invalid elements and integrating formats to build an initial factor pool, redundant and low-quality data are eliminated to avoid interference from invalid data in subsequent detection, improving factor detection efficiency and accuracy; at the same time, the data preparation process before detection is clarified to reduce the technical obstacles of non-uniform formats of multi-source data.

[0088] In some embodiments, the background factors are clustered and classified and superimposed to obtain processed background factors, and the processed background factors and the dominant factors are partitioned to obtain a plurality of effective ecological partition units, including: obtaining an optimal discretization scheme, preprocessing and clustering the background factors and the optimal discretization scheme to obtain an ecological background unit.

[0089] Specifically, an optimal parameter geographic detector is used to generate a scheme with the core principle of maximizing the explanatory power (q value) of the background factors on the optimal water conservation value, and the background factors and the optimal water conservation value are input.

[0090] For numerical background factors, 3-6 natural breakpoint methods, quantile methods, equal interval methods, geometric interval methods, and standard deviation methods are tried respectively; for categorical background factors (such as land use type), the original categories are combined according to ecological relevance (such as combining 8 categories into 5 categories of farmland, forest land, grassland, construction land, and water area).

[0091] The q value corresponding to each discretization scheme is calculated, and the scheme with the maximum q value is selected as the optimal discretization scheme. For example:

[0092] Elevation: 3-class division (<500m, 500-1000m, >1000m), q value = 0.28 (higher than 0.25 of 4-class division).

[0093] Soil porosity: 3-class division (<40%, 40%-50%, >50%), q value=0.21 (higher than 0.18 of 5-class division).

[0094] Land use type: 5-class combination (farmland=1, forest land=2, grassland=3, construction land=4, water area=5), q value=0.25 (higher than 0.20 of original 8-class).

[0095] According to the optimal discretization scheme, the continuous value grid of altitude and soil porosity is converted into a classified value grid, such as altitude <500m assigned value 1, 500-1000m assigned value 2, >1000m assigned value 3.

[0096] Convert the text category of land use type (such as forest land) into corresponding digital code (such as forest land=2).

[0097] Through the ArcGIS grid algebra tool, ensure that the resolution, coordinate system and spatial range of all background factor grids are completely consistent to avoid pixel misplacement.

[0098] Using k-means clustering algorithm, the objective is to minimize intra-class variance and maximize inter-class difference:

[0099] Calculate the intra-class sum of squares when k=3-7 by elbow rule, maximum iteration number=100, convert the preprocessed 3 background factor grids into pixel-attribute matrix, input k-means algorithm, and reach convergence condition after 28 iterations.

[0100] For the output 5-class ecological background unit, calculate the standard deviation of the background factors within the unit by ArcGIS partition statistics tool. For example, in the high altitude forest land type unit (number 3), the standard deviation of altitude is 180m, and the standard deviation of soil porosity is 3.2%, both <5%, meeting the requirement of uniform ecological properties.

[0101] Combine the factor combination characteristics to name, such as low altitude farmland type (number 1), high altitude grassland type (number 2), high altitude forest land type (number 3), low altitude construction land type (number 4), and water area surrounding type (number 5), to form the final ecological background unit.

[0102] Further, the dominant factor and ecological background unit are partitioned to obtain the theoretical joint partition.

[0103] Specifically, the optimal parameter geoprocesser consistent with the background factor is used to complete the discretization and format adaptation.

[0104] For the determined dominant factor (such as annual rainfall, annual maximum NDVI), the maximum q value screening scheme is selected:

[0105] Annual precipitation: 3-class division (<800mm, 800-1200mm, >1200mm), q value = 0.42 (higher than 0.39 of 4-class division).

[0106] Annual maximum NDVI: 3-class division (<0.6, 0.6-0.8, >0.8), q value = 0.38 (higher than 0.35 of 3-class natural breakpoint division).

[0107] Classification value conversion: Convert the continuous value grids of annual precipitation and NDVI into classification values (e.g., annual precipitation <800mm = 1, 800-1200mm = 2, >1200mm = 3), ensuring consistency with the classification value format of the background cells.

[0108] Dominant factor combination: According to the priority of annual precipitation-annual maximum NDVI (higher q value of annual precipitation, stronger dominant effect), generate 9 kinds of dominant factor combinations (e.g., 1-1: <800mm + <0.6; 2-3: 800-1200mm + >0.8, etc.), and assign a unique combination number (1-9).

[0109] Use the raster calculator tool in ArcGIS to generate theoretical division numbers according to the formula of ecological background cell number x 10 + dominant factor combination number, example:

[0110] High altitude forest type (3) + annual precipitation >1200mm (3) + NDVI >0.8 (3) → division number = 3x10 + 9 = 39 (9 is the number of 3-3 combination).

[0111] Calculate the average water conservation capacity of each theoretical division by ArcGIS division statistics tool, and merge combinations with average conservation capacity difference <5% and consistent land use type. For example, division 39 (average conservation capacity 50.2mm) and division 38 (average conservation capacity 48.5mm), difference = 3.4% <5% and both are forest land, merged into 1 type.

[0112] Finally, 15 theoretical joint divisions are obtained, each of which is labeled with background cell-dominant factor combination characteristics (e.g., 3-3-3: high altitude forest + high precipitation + high NDVI).

[0113] Further, the invalid factor combination is removed from the theoretical joint division to obtain multiple effective ecological division units.

[0114] Specifically, remove combinations that have no actual spatial distribution in the target area. For example, high altitude forest type (3) + annual precipitation <800mm (1).

[0115] Remove combinations that violate ecological laws. For example, low-altitude construction land type (4) + NDVI > 0.8 (3), low vegetation coverage in construction land concentrated area, NDVI generally <0.6, this combination does not conform to the actual ecological characteristics.

[0116] Based on the above criteria, 3 invalid combinations (such as 3-1, 4-3, 5-2) are removed from 15 theoretical joint partitions, and 12 candidate partitions are retained.

[0117] Check the spatial continuity of the candidate partitions by the ArcGIS field analysis tool. Remove the broken partitions with a maximum connected area <100km². In this embodiment, the 12 candidate partitions all meet the requirements.

[0118] Determine 12 valid ecological partition units, and form a correspondence table of partition number-space range-ecological characteristics-average capacity.

[0119] The embodiment of the present application solves the problem of traditional superposition directly mixing multiple factors and confusing ecological significance of partition by splitting the ecological background unit, constructing theoretical partition, and removing invalid combinations. The new invalid factor combination removal link removes combinations that have no actual geographical distribution and violate ecological laws (such as high altitude + low rainfall), avoids the problem that traditional partition contains invalid units and cannot be applied, and clearly defines the superposition logic of the dominant factor and the background factor. The partition reflects the innate ecological endowment (background factor) and the core driving force (dominant factor), solves the problem that traditional partition only focuses on a single factor and cannot reflect the synergistic effect of multiple factors, and improves the ecological representativeness of the partition.

[0120] In some embodiments, the optimal discretization scheme is obtained, the background factor and the optimal discretization scheme are preprocessed and clustered to obtain the ecological background unit, including: obtaining the optimal discretization scheme generated by the optimal parameter geographic detector.

[0121] Specifically, based on the determined background factor and the optimal water source capacity value, the preprocessed 1km×1km grid data is extracted to ensure that the data area has been filled with the neighborhood mean value.

[0122] The merged 5-type coded (farmland = 1, forest land = 2, grassland = 3, construction land = 4, water area = 5) grid data and the optimal water source capacity value (such as 52.3mm) corresponding to the grid data are used to calculate the factor explanatory power.

[0123] According to the core principle of maximizing the explanatory power (q value) of the optimal water source capacity value of the factor, the optimal parameter geographic detector is executed for discretization.

[0124] Altitude: Try 3 classes (<500m, 500-1000m, >1000m), 4 classes (<500m, 500-800m, 800-1200m, >1200m), 5 classes (<400m, 400-700m, 700-1000m, 1000-1300m, >1300m) equidistant division, calculate the q value of each class scheme. 3 classes q=0.28, 4 classes q=0.25, 5 classes q=0.23, so choose 3 classes as the optimal scheme.

[0125] Soil porosity: Try 3 classes (<40%, 40%-50%, >50%), 4 classes (<35%, 35%-45%, 45%-55%, >55%) natural breakpoint division, 3 classes q=0.21 (higher than 4 classes 0.18), determined as the optimal scheme.

[0126] Land use type has been merged into 5 categories according to ecological relevance, and the q value of this classification scheme is 0.25 (higher than the original 8 categories 0.20), which is determined as the optimal scheme.

[0127] Organize the optimal discretization rules of each background factor to form a factor-discretization class-encoding correspondence table.

[0128] Further, the background factors and the optimal discretization scheme are matched and preprocessed to obtain the preprocessed background factors. Based on the discretization rules, the continuous values are converted to classification values through the ArcGIS raster calculator tool:

[0129] Altitude: Input formula Con("altitude grid" <500, 1, Con("altitude grid" <1000, 2, 3)), generate altitude classification value grid.

[0130] Soil porosity: Input formula Con("porosity grid" <40, 1, Con("porosity grid" <50, 2, 3)), generate porosity classification value grid.

[0131] After conversion, it needs to be verified by attribute table statistics: such as the proportion of the number of grids with altitude classification value 1 (<500m) is about 35%, which is consistent with the proportion of low altitude area, to ensure the accuracy of conversion.

[0132] Check if the encoding in the attribute table contains only 1-5 (corresponding to cultivated land, forest land, grassland, construction land, water area respectively), and remove abnormal encoding (such as 0 or 6) and fill it with neighborhood mode.

[0133] The grid with water area code 5 needs to coincide with the spatial range of rivers and lakes in the figure, and the grid with construction land code 4 needs to match the urban built-up area range, to ensure that the classification conforms to the actual geographical features.

[0134] The row / column resolution coordinate system space range of the three background factor grids of elevation, soil porosity, and land use type is unified by using the ArcGIS raster algebra tool to ensure that each pixel corresponds to one another (without pixel misplacement).

[0135] Each of the three factor grids is converted into an ASCII format, and each ASCII file contains the basic information of the grid (ncols, nrows, xllcorner, yllcorner, cellsize, NODATA_value) and the classification value matrix.

[0136] The final output is the preprocessed background factor dataset, which includes three ASCII format classification value grids.

[0137] Further, the preprocessed background factors are clustered by a clustering algorithm to obtain ecological background units, and the clustering algorithm aims to minimize the within-class variance and maximize the between-class difference.

[0138] Specifically, the preprocessed three background factor grids are converted into a pixel-attribute matrix (each row represents one pixel, and the three columns are the classification values of elevation, soil porosity, and land use type), and the within-class sum of squares (WCSS) is calculated when k=3-7: WCSS=8500 when k=3, WCSS=6200 when k=4, WCSS=4800 when k=5, and WCSS=4600 when k=6 (the decrease amplitude is <5%), so k=5 is selected (both the within-class homogeneity and the avoidance of excessive subdivision are ensured).

[0139] The maximum number of iterations is 100, and the convergence threshold (within-class variance change) is 0.001 (when the within-class variance change of two consecutive iterations is <0.001, the iteration is stopped).

[0140] The k-means++ algorithm is used to repeat clustering 5 times, and the result with the smallest WCSS is selected as the final clustering model.

[0141] The pixel-attribute matrix and the above parameters are input to perform clustering. After 28 iterations, the convergence condition is reached, and the clustering label (1-5) of each pixel is output.

[0142] The clustering label matrix is converted into a grid data, and the clustering rationality is preliminarily judged: for example, the label 3 grid is concentrated in the high elevation (>1000m), high porosity (>50%), and forest land (code 2) area, which is consistent with the ecological characteristics of high elevation forest land.

[0143] The standard deviations of the background factors of the five clustering results are calculated:

[0144] Tag 1 (low-altitude cultivated land type): standard deviation of altitude = 120m, standard deviation of soil porosity = 2.8%, mode proportion of land use type = 85% (cultivated land code 1).

[0145] Tag 3 (high-altitude forest type): standard deviation of altitude = 180m, standard deviation of soil porosity = 3.2%, mode proportion of land use type = 82% (forest code 2).

[0146] In all categories, the standard deviation of altitude <200m, the standard deviation of soil porosity <4%, and the mode proportion of land use >80%.

[0147] Combined with the factor combination characteristics of the clustering results, the five categories of units are named in the format of altitude-land use-core attribute:

[0148] Tag 1: low-altitude cultivated land type (altitude <500m, cultivated land proportion 85%, soil porosity <40%).

[0149] Tag 2: mid-altitude grassland type (altitude 500-1000m, grassland proportion 80%, soil porosity 40%-50%).

[0150] Tag 3: high-altitude forest type (altitude >1000m, forest proportion 82%, soil porosity >50%).

[0151] Tag 4: low-altitude construction land type (altitude <500m, construction land proportion 75%, soil porosity <40%).

[0152] Tag 5: water area surrounding type (altitude <500m, water area and surrounding cultivated land proportion 90%, soil porosity 40%-50%).

[0153] The final output is the ecological background unit raster data and the unit number-factor combination-space range correspondence table.

[0154] The embodiments of the present application generate a discretization scheme by using an optimal parameter geographic detector, aiming to maximize the explanatory power of factors on optimal conservation values, avoid the problems of subjective classification and weak association with conservation functions in traditional discretization, and ensure the scientificity of background factor discretization. The matching preprocessing of new background factors and discretization scheme is added to unify the factor format and spatial scale, solve the problem of distorted clustering results caused by non-uniform input data format and pixel misplacement, and improve the spatial consistency of ecological background units. The elbow rule is used to determine the value of k to minimize the intra-class variance and maximize the inter-class difference, avoid the problems of subjective setting of the number of classes and large differences in ecological properties within units in traditional clustering, and ensure the uniformity of ecological background units.

[0155] In some embodiments, the optimal water conservation value and each factor in the initial factor pool are subjected to detection processing to obtain a dominant factor and a background factor, including: subjecting the optimal water conservation value and each factor in the initial factor pool to factor detection processing to obtain a dominant factor, the dominant factor being a factor that has an explanation power exceeding a preset explanation power threshold for the optimal water conservation value.

[0156] Specifically, the optimal water conservation value is stored in a grid format according to the screening. The constructed initial factor pool is all in a pretreated ASCII format classification / numerical grid. Through the random point tool of ArcGIS, 1000 sample points are generated in the target area; then the multi-value-to-point tool is used to assign the optimal water conservation value and the attribute values of the 10 factors to the sample points to generate a CSV format sample data set, and the invalid samples containing NoData are removed.

[0157] When the q value is greater than the preset explanation power threshold or the p value is less than 0.01 (significance test by 99% confidence level), the annual rainfall (q=0.42), the annual maximum NDVI (q=0.38), and the vegetation coverage (q=0.31) meet the standard and are determined as the dominant factors.

[0158] Further, each factor in the initial factor pool is subjected to interactive detection processing to obtain a background factor, the background factor being a factor that has a non-linear enhancement action frequency higher than a preset frequency threshold.

[0159] Specifically, the interactive detection judges the action type by calculating the interactive explanation power of the combination of two factors, wherein the action type includes non-linear enhancement, synergistic enhancement, independent action, and antagonistic action.

[0160] The background factor is a factor that has a non-linear enhancement action frequency higher than a preset frequency threshold, and the preset frequency threshold is 2 times.

[0161] The elevation (frequency 3), the soil porosity (frequency 2), and the land use type (frequency 2) meet the non-linear enhancement frequency > 2 times standard and are determined as the background factors.

[0162] The embodiments of the application screen the dominant factors by using factor detection plus q value threshold plus significance test, replace subjective judgment with quantification and statistics test, solve the problem of experience selection and lack of quantitative basis in traditional dominant factor screening, and ensure that the dominant factors can truly reflect the core driving action. The background factors are screened by using interactive detection plus non-linear enhancement frequency threshold, focus on the elements that have strong synergistic effect with other factors, solve the problem of ignoring the synergistic effect between factors and only looking at the influence of a single factor in traditional background factor identification, clarify the supporting action of the background factors on the ecological system, and perfect the dominant-background factor system.

[0163] In some embodiments, the dominant factors and the ecological background unit are partitioned to obtain a theoretical joint partition, comprising: matching preprocessing of the dominant factors and the optimal discretization scheme to obtain preprocessed dominant factors.

[0164] Specifically, based on the determined dominant factors (annual rainfall, annual maximum NDVI, vegetation coverage), the same optimal parameter geographic detector as the background factor is used to generate a scheme with the principle of maximizing the explanatory power (q value) of the factor to the optimal water conservation value.

[0165] The data input is the original raster data of the dominant factors (annual rainfall: mm, annual maximum NDVI: unitless, vegetation coverage: %) and the determined optimal water conservation value raster.

[0166] For each dominant factor, 3-4 types of equidistant / natural breakpoint division are tried, and the q value of each type of scheme is calculated.

[0167] The scheme with the maximum q value is selected as the optimal discretization scheme.

[0168] The three dominant factor classification value rasters are all resampled to 1km×1km.

[0169] The dominant factor rasters are cropped with the target region boundary to remove invalid data outside the region and ensure that the spatial range of the dominant factors completely coincides with that of the ecological background unit.

[0170] Finally, three preprocessed dominant factor rasters are output to provide standardized input for classification superposition.

[0171] Further, the preprocessed dominant factors and the ecological background unit are partitioned to obtain a theoretical joint partition.

[0172] According to the q value of the dominant factor (annual rainfall 0.42> annual maximum NDVI 0.38> vegetation coverage 0.31), the superposition priority is determined: annual rainfall (first priority)→ annual maximum NDVI (second priority)→ vegetation coverage (third priority). The priority setting is based on the fact that the higher the q value, the stronger the dominant role of the factor in water conservation, and superpositioning first can ensure that the influence of the partition is given priority, avoiding interference from secondary factors (such as first dividing high rainfall and low rainfall areas, and then dividing high NDVI and low NDVI areas in the same rainfall area).

[0173] For example, the partition of high altitude forest type (3) + high rainfall (3) + high NDVI (3) (number 333.3) needs to be concentrated in the high rainfall area and the forest concentration area of the hydrological mean value map, with a spatial coincidence degree of ≥85%, to ensure that it meets the actual ecological characteristics.

[0174] The final output theoretical joint partition result includes:

[0175] Theoretical joint zoning grid: 1km x 1km resolution, WGS84 coordinate system, and the pixel value is the unique zoning number (1 decimal place is reserved).

[0176] Zoning attribute table: contains fields such as zoning number-ecological background unit-leading factor combination-naming-pixel quantity, which clearly defines the core characteristics of each zoning.

[0177] Spatial distribution map: render the zoning grid through ArcGIS, different numbers correspond to different colors, and the spatial distribution of the zoning is intuitively displayed.

[0178] The embodiment of the present application matches the optimal discretization scheme through the preprocessing of the leading factor, ensures the consistency of the discretization logic of the leading factor and the background factor, solves the problem of inconsistent discretization standards of traditional superimposed leading factors and poor adaptability to background units, and improves the logicality of the superimposed result. The superimposition priority is determined according to the high and low of the leading factor q value, the influence of the core driving element is ensured to be given priority to, the problem of random sorting and interference of secondary factors to the core difference in traditional superimposition is avoided, the theoretical zoning can clearly reflect the hierarchical influence of the core driving-background endowment.

[0179] Although the present application provides method operation steps as described in the embodiments or flowcharts, more or fewer operation steps can be included based on conventional or non-inventive labor. The order of steps listed in the present embodiment is only one of the many step execution orders, and does not represent the only execution order. When the device or client product is executed in practice, it can be executed in sequence or in parallel (for example, in a parallel processor or multi-threaded processing environment) according to the method order shown in the present embodiment or the accompanying drawings.

[0180] As shown in Figure 3 The present application also provides a water conservation and ecological restoration area identification device 300. The device comprises:

[0181] The acquisition module 301 is configured to acquire the water conservation amount of the target area and the water conservation influencing factors, wherein the water conservation influencing factors include at least one of terrain, soil, climate, vegetation, and human activities.

[0182] The processing module 302 is configured to filter and detect the water conservation amount and the water conservation influencing factors to obtain the optimal water conservation value, the leading factor, and the background factor.

[0183] The processing module 302 is further configured to perform clustering and classification superimposition processing on the background factor to obtain the processed background factor, and perform zoning processing on the processed background factor and the leading factor to obtain a plurality of effective ecological zoning units.

[0184] The processing module 302 is further configured to perform statistical and quantitative processing on the optimal water conservation value and the water conservation amount corresponding to each valid ecological zoning unit, to obtain a water conservation potential value corresponding to each valid ecological zoning unit and a potential surplus amount.

[0185] The identification module 303 is further configured to divide an ecological restoration priority level according to the water conservation potential value and the potential surplus amount, identify the water conservation and ecological restoration area based on the ecological restoration priority level, and obtain an identification result.

[0186] In some embodiments, the processing module 302 is further configured to perform time series screening processing on the water conservation amount of the target area, to obtain the optimal water conservation value, and the optimal water conservation value is used to represent an optimal state of an ecological environment in an observation period of the target area.

[0187] The processing module 302 is further configured to perform screening processing on the water conservation influencing factors, to obtain terrain basic factors, soil hydrological factors, climate condition factors, vegetation ecological factors, and human intervention factors.

[0188] The processing module 302 is further configured to perform elimination and integration processing on the terrain basic factors, the soil hydrological factors, the climate condition factors, the vegetation ecological factors, and the human intervention factors, to obtain an initial factor pool.

[0189] The processing module 302 is further configured to perform detection processing on the optimal water conservation value and each factor in the initial factor pool, to obtain a dominant factor and a background factor.

[0190] In some embodiments, the acquisition module 301 is further configured to acquire an optimal discretization scheme, and perform preprocessing and clustering processing on the background factor and the optimal discretization scheme, to obtain an ecological background unit.

[0191] The processing module 302 is further configured to perform zoning processing on the dominant factor and the ecological background unit, to obtain a theoretical joint zoning.

[0192] The processing module 302 is further configured to perform invalid factor combination elimination processing on the theoretical joint zoning, to obtain a plurality of valid ecological zoning units.

[0193] In some embodiments, the acquisition module 301 is further configured to acquire an optimal discretization scheme generated by an optimal parameter geographic detector.

[0194] The processing module 302 is further configured to perform matching preprocessing on the background factor and the optimal discretization scheme, to obtain a preprocessed background factor.

[0195] The processing module 302 is further configured to perform clustering processing on the preprocessed background factor by using a clustering algorithm, to obtain an ecological background unit, and the clustering algorithm aims to minimize intra-class variance and maximize inter-class difference.

[0196] In some embodiments, the processing module 302 is further configured to perform factor detection on each factor in the initial factor pool and the optimal water source conservation value to obtain a dominant factor, the dominant factor being a factor that has an explanation power exceeding a preset explanation power threshold for the optimal water source conservation value.

[0197] The processing module 302 is further configured to perform interaction detection on each factor in the initial factor pool to obtain a background factor, the background factor being a factor that has a frequency of nonlinear enhancement higher than a preset frequency threshold.

[0198] In some embodiments, the processing module 302 is further configured to perform matching preprocessing on the dominant factor and the optimal discretization scheme to obtain a preprocessed dominant factor.

[0199] The processing module 302 is further configured to perform partitioning on the preprocessed dominant factor and the ecological background unit to obtain a theoretical joint partitioning.

[0200] Some of the modules in the apparatus described in the present application can be described in the general context of computer-executable instructions, such as program modules, which are executed by computers. Generally, program modules include routines, programs, objects, components, data structures, classes, and the like, which perform particular tasks or implement particular abstract data types. The present application can also be practiced in distributed computing environments where tasks are performed by remote processing devices that are connected through a communication network. In a distributed computing environment, program modules can be located in both local and remote computer storage media including the storage devices.

[0201] The apparatus or modules described in the above embodiments can be implemented by computer chips or entities, or by products having certain functions. For the convenience of description, the above apparatus is described in various modules with functions. In the implementation of the embodiments of the present application, the functions of each module can be implemented in the same or multiple software and / or hardware. Of course, the modules implementing certain functions can also be implemented by multiple sub-modules or sub-units.

[0202] The methods, apparatuses or modules described in the present application can be implemented in a computer readable program code in any appropriate manner, for example, the controller can take the form of, for example, a microprocessor or processor and a computer readable medium storing computer readable program code (for example, software or firmware) executable by the (micro)processor, logic gates, switches, application specific integrated circuits (ASIC), programmable logic controllers and embedded microcontrollers, examples of the controller include but are not limited to the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicone Labs C8051F320, the memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that in addition to implementing the controller in pure computer readable program code, the same function can be achieved by logically programming the method steps in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers and embedded microcontrollers. Therefore, such a controller can be considered as a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both a software module for implementing the method and a structure within the hardware component.

[0203] The embodiments of the present application also provide a device, which comprises: a processor; a memory for storing processor executable instructions; and the processor implements the method as described in the embodiments of the present application when executing the executable instructions.

[0204] The embodiments of the present application also provide a non-volatile computer readable storage medium, which stores a computer program or instructions, and when the computer program or instructions are executed, the method as described in the embodiments of the present application is implemented.

[0205] In addition, the functional modules in each of the embodiments of the present application can be integrated in one processing module, or each module can exist independently, or two or more modules can be integrated in one module.

[0206] The storage medium described above includes but is not limited to random access memory (RAM), read-only memory (ROM), cache, hard disk drive (HDD) or memory card. The memory can be used to store computer program instructions.

[0207] From the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software and the necessary hardware. Based on such an understanding, the technical solutions of the present application can be embodied in the form of a software product or in the form of data migration. The computer software product can be stored in a storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a plurality of instructions for causing a computer device (which can be a personal computer, a mobile terminal, a server, or a network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0208] The various embodiments in the specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. The whole or part of the present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld devices or portable devices, tablet devices, mobile communication terminals, multi-processor systems, microprocessor-based systems, programmable electronic devices, network PCs, small computers, large computers, distributed computing environments including any of the above systems or devices, etc.

[0209] The above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the present application.

Claims

1. A method for identifying water conservation and ecological restoration zones, characterized in that, include: The water conservation capacity and water conservation influencing factors of the target area are obtained. The water conservation influencing factors include at least one of topography, soil, climate, vegetation and human activities. The water conservation capacity and the factors influencing water conservation are screened and detected to obtain the optimal water conservation value, dominant factor and background factor. The background factors are clustered and classified overlayed to obtain processed background factors. The processed background factors and the dominant factors are then partitioned to obtain multiple effective ecological partition units. The optimal water conservation value and the water conservation amount corresponding to each effective ecological zone unit are statistically and quantitatively processed to obtain the water conservation potential value and potential surplus of each effective ecological zone unit. Based on the water conservation potential value and the potential surplus, the priority level of ecological restoration is divided, and the water conservation and ecological restoration areas are identified based on the ecological restoration priority level to obtain the identification results; The process of screening and detecting the water conservation capacity and the factors influencing water conservation to obtain the optimal water conservation value, dominant factors, and background factors includes: The water conservation capacity of the target area is subjected to time series screening to obtain the optimal water conservation value, which is used to characterize the optimal ecological environment state of the target area during the observation period. The factors affecting water conservation were screened and processed to obtain basic topographical factors, soil hydrological factors, climate conditions factors, vegetation ecological factors, and human intervention factors. The topographical elements, soil and hydrological elements, climate conditions, vegetation ecology elements, and human intervention elements are removed and integrated to obtain an initial factor pool. The optimal water conservation value and each factor in the initial factor pool are detected and processed to obtain the dominant factor and background factor. The background factors are clustered and classified overlaid to obtain processed background factors. These processed background factors are then partitioned with the dominant factors to obtain multiple effective ecological partition units, including: The optimal discretization scheme is obtained, and the background factors and the optimal discretization scheme are preprocessed and clustered to obtain ecological background units. The dominant factors and the ecological background units are partitioned to obtain a theoretical joint partition; The theoretical joint partition and the dominant factor are subjected to invalid factor combination elimination processing to obtain multiple valid ecological partition units.

2. The method according to claim 1, characterized in that, The process of obtaining the optimal discretization scheme involves preprocessing the background factors and the optimal discretization scheme, as well as performing clustering to obtain ecological background units, including: Obtain the optimal discretization scheme generated by the optimal parameter geodetector; The background factor and the optimal discretization scheme are matched and preprocessed to obtain the preprocessed background factor. The preprocessed background factors are clustered using a clustering algorithm to obtain ecological background units. The clustering algorithm aims to minimize intra-class variance and maximize inter-class differences.

3. The method according to claim 1, characterized in that, The process of detecting and processing the optimal water conservation value and each factor in the initial factor pool to obtain the dominant factor and background factor includes: Factor detection processing is performed on the optimal water conservation value and each factor in the initial factor pool to obtain the dominant factor. The dominant factor is the factor whose explanatory power for the optimal water conservation value exceeds a preset explanatory power threshold. Interactive detection processing is performed on each factor in the initial factor pool to obtain the background factor, which is a factor whose nonlinear enhancement frequency is higher than a preset frequency threshold.

4. The method according to claim 1, characterized in that, The partitioning of the dominant factor and the ecological background unit to obtain a theoretical joint partition includes: The dominant factor and the optimal discretization scheme are matched and preprocessed to obtain the preprocessed dominant factor. The preprocessed dominant factors and the ecological background units are partitioned to obtain a theoretical joint partition.

5. The method according to claim 1, characterized in that, The effective ecological zoning units include priority restoration zones, key improvement zones, moderate optimization zones, and status quo maintenance zones.

6. A device for identifying water conservation and ecological restoration zones for implementing the method described in any one of claims 1-5, characterized in that, include: The acquisition module is used to acquire the water conservation capacity of the target area and the factors affecting water conservation, wherein the factors affecting water conservation include at least one of topography, soil, climate, vegetation and human activities; The processing module is used to screen and detect the water conservation capacity and the factors affecting water conservation, so as to obtain the optimal water conservation value, dominant factor and background factor. The processing module is also used to perform clustering and classification superposition processing on the background factors to obtain processed background factors, and to perform partitioning processing on the processed background factors and the dominant factors to obtain multiple effective ecological partition units. The processing module is also used to perform statistical quantification on the optimal water conservation value and the water conservation amount corresponding to each effective ecological zone unit to obtain the water conservation potential value and potential surplus of each effective ecological zone unit. The identification module is also used to classify the priority level of ecological restoration according to the water source conservation potential value and the potential surplus, and to identify the water source conservation and ecological restoration areas based on the ecological restoration priority level to obtain the identification result; The processing module is further used to screen and detect the water conservation capacity and the factors influencing water conservation, to obtain the optimal water conservation value, dominant factor, and background factor, wherein: The water conservation capacity of the target area is subjected to time series screening to obtain the optimal water conservation value, which is used to characterize the optimal ecological environment state of the target area during the observation period. The factors affecting water conservation were screened and processed to obtain basic topographical factors, soil hydrological factors, climate conditions factors, vegetation ecological factors, and human intervention factors. The topographical elements, soil and hydrological elements, climate conditions, vegetation ecology elements, and human intervention elements are removed and integrated to obtain an initial factor pool. The optimal water conservation value and each factor in the initial factor pool are detected and processed to obtain the dominant factor and background factor. The processing module is further configured to perform clustering and classification overlay processing on the background factors to obtain processed background factors, and to perform partitioning processing on the processed background factors and the dominant factors to obtain multiple effective ecological partition units, wherein: The optimal discretization scheme is obtained, and the background factors and the optimal discretization scheme are preprocessed and clustered to obtain ecological background units. The dominant factors and the ecological background units are partitioned to obtain a theoretical joint partition; The theoretical joint partition and the dominant factor are subjected to invalid factor combination elimination processing to obtain multiple valid ecological partition units.

7. A computer device comprising a memory and a processor, the memory storing a computer program executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 5.

Images