A spatial discretization method based on sub-hill slopes

Abstract

The present invention discloses a spatial discretization method based on sub-hill slopes, which relates to the fields of resource and environmental management and geographic national conditions census. The spatial discretization method based on sub-hill slopes includes the following steps: extracting river sections based on digital elevation models, and making the height difference between the beginning and the end of each river section less than a preset threshold; extracting sub-hill slope units that directly flow into the river section; calculating the nearest neighbor river channel relative height value of each pixel, and dividing the slope units according to the value; splitting some discontinuous slope units into spatially continuous slope units based on the neighborhood search method; traversing all slope units along the flow direction from the sub-hill slope source unit, and merging the units that do not meet the area threshold into the downstream unit until the merged unit is greater than the preset area threshold. The present invention breaks through the problems of excessive height difference and complex confluence relationship of slope units divided by traditional methods, and has broad application prospects in the fields of disaster emergency response and water resources management.

Description

Technical Field

[0001] The invention relates to the field of resource and environmental management and geographic national conditions survey, and in particular to a sub-hill slope-based spatial discretization method. Background Art

[0002] Spatial discretization is one of the important links in distributed hydrological modeling. Currently, the commonly used spatial discretization methods include: grid unit, hydrological response unit and slope unit. The grid hydrological model divides the watershed into a series of grid units of equal size. Each grid unit has a set of parameters and variables related to hydrological processes, such as rainfall, evapotranspiration, runoff, etc. These parameters and variables can be estimated or determined based on the characteristics of the watershed and actual observation data. The hydrological response of the entire watershed is calculated based on the water exchange between adjacent grids. Due to its discretization characteristics, a higher spatial resolution is required to accurately simulate the hydrological processes in the watershed. In addition, the grid hydrological model has limited simulation effects on some complex hydrological processes, such as soil moisture migration and groundwater interconnection.

[0003] Hydrological response units (HRUs) can reflect meteorological and underlying surface factors such as precipitation and topography, simplify the structure and calculation process of simulation units, and provide a data basis for hydrological simulation at the scale of large watersheds. They are usually obtained by directly superimposing or clustering environmental data. HRUs can reflect the vegetation and land types within the unit, but the confluence relationship between HRUs is complex and the spatial position is inaccurate, which affects the simulation accuracy. The slope unit division is to divide the sub-basin into left and right slopes and source slopes according to the river network, clarify the upstream area that flows into the river channel, and then divide different slope units in the order from the river valley to the ridge, which can reflect the differences in hydrological conditions in the vertical direction and the slope confluence process. The division method based on slope units can characterize the differences in land hydrological conditions. For example, on the vertical gradient, the bottom of the valley is wetter than the ridge; on the horizontal gradient, the sunny slope is warmer and drier than the shady slope. In areas with greater water stress, this difference in hydrological conditions also causes systematic differences in soil and vegetation between valleys and ridges, and between sunny and shady slopes. However, when dividing slope units, existing methods directly merge the ridges on both sides of the valley downstream without distinguishing between the left and right slopes, which may lead to huge differences in vegetation on the north and south slopes in some well-developed valleys. In addition, the slope units directly divided at the slope scale have large horizontal and vertical spans, which will cause huge height differences, making the environmental conditions such as temperature in the same unit vary greatly, that is, there is still strong heterogeneity within the unit, which affects the rationality of basin modeling and simulation.

[0004] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0005] In view of the problems in the related art, the present invention proposes a sub-hillslope-based spatial discretization method to overcome the above-mentioned technical problems existing in the existing related art.

[0006] To this end, the specific technical solution adopted by the present invention is as follows:

[0007] A sub-hillslope-based spatial discretization method, the sub-hillslope-based spatial discretization method comprising the following steps:

[0008] S1. Based on the flow direction, initial river network and sub-watershed extracted from the digital elevation model, river sections are extracted, and the height difference between the beginning and the end of each river section is made less than a preset height difference threshold.

[0009] S2. Based on the river section and flow direction data in the sub-basin, the upstream catchment area of ​​each river pixel is taken as the sub-hillslope unit. Based on the sub-hillslope unit area, the sub-hillslope units whose areas are smaller than the preset sub-hillslope unit area threshold and are spatially adjacent are merged to obtain the sub-hillslope unit of the river section.

[0010] S3. Calculate the relative height value of the nearest neighboring river channel for each pixel, and divide the slope position units according to the relative height value of the nearest neighboring river channel.

[0011] S4. Based on the neighborhood search method, some discontinuous slope units are split into spatially continuous slope units.

[0012] S5. Traverse all slope units along the flow direction according to the sub-hill slope source unit, and merge the slope units into downstream units based on the slope unit area until the area of ​​the merged slope unit is greater than the preset slope unit area threshold, so as to obtain a discretized unit with clear flow direction and spatial continuity.

[0013] Furthermore, extracting river sections and making the height difference between the beginning and the end of each river section smaller than a preset height difference threshold includes the following steps:

[0014] S11. Trace back the node pixels of the river network in the sub-basin, which include the source of the river network and the river pixels into which two or more river pixels converge;

[0015] S12, taking each node as the source of the river section, tracing back the downstream river network pixels according to the flow direction data, until the downstream pixels are node pixels and a river section is formed, and assigning a code to the river section;

[0016] S13, calculating the height difference between the beginning and the end of each river section, and if the height difference between the beginning and the end exceeds a preset height difference threshold, calculating the number of breakpoints;

[0017] S14, re-using the first river pixel whose elevation is lower than the preset value in the river section as a node and tracing downstream until the elevation of the downstream river pixel meets the condition, and re-coding the river section;

[0018] S15. If the height difference between the beginning and the end is less than the preset height difference threshold, the process is skipped and the next river section is judged until all river sections are judged.

[0019] Furthermore, the formula for calculating the number of breakpoints is:

[0020] N = Int (dH / Threshold);

[0021] In the formula, N represents the number of breakpoints;

[0022] dH represents the height difference between the beginning and the end of the river section;

[0023] Threshold indicates the maximum height difference threshold.

[0024] Furthermore, the calculation formula for elevation is:

[0025] A=Head_H+Threshold*k (0<=k<=N);

[0026] In the formula, A represents elevation;

[0027] Head_H represents the elevation of the first pixel of the river section;

[0028] Threshold indicates the maximum height difference threshold;

[0029] k is an integer value between 0 and N, where N represents the number of breakpoints.

[0030] Further, until the elevation of the downstream river pixel meets the conditions, the conditions include that the elevation is greater than Head_H+Threshold*(k+1), the downstream river pixel belongs to other rivers, and is an invalid value;

[0031] Among them, Head_H represents the elevation of the first pixel of the river section, Threshold represents the maximum height difference threshold, k represents an integer value between 0 and N, and N represents the number of breakpoints.

[0032] Furthermore, based on the river network and flow direction data in the sub-basin, the upstream catchment area of ​​each river pixel is taken as a sub-hillslope unit, and based on the sub-hillslope unit area, the sub-hillslope units whose areas are smaller than a preset sub-hillslope unit area threshold and spatially adjacent are merged to obtain the sub-hillslope units of the river section, including the following steps:

[0033] S21, based on the river network data extracted from the digital elevation model, tracing back to the source of the river network in the sub-basin and the entire river from the source to the downstream, and obtaining a list of river pixels arranged in the order of upstream and downstream;

[0034] S22. Based on the flow direction data, trace back the upstream catchment area of ​​each river pixel, take the upstream catchment area as the initial sub-hill slope, assign different codes to each sub-hill slope, and calculate the area of ​​the sub-hill slope;

[0035] S23, the coded sub-hillsides whose overall combined area is smaller than the sub-hillside unit area threshold are obtained as a set consisting of some sub-hillsides, and are coded as special values;

[0036] S24. Split the subset in the set into a number of spatially continuous sub-hillslope units through a neighborhood search method.

[0037] Furthermore, splitting the sub-set in the set into a plurality of spatially continuous sub-hillslope units by using a neighborhood search method includes the following steps:

[0038] S241, traverse all pixels of the set to be split, and use the unsearched pixel as the starting unit, and use the breadth-first search algorithm to find the pixel with the same code as the starting unit;

[0039] S242, assign a new code to the starting unit and mark it as searched, and stop after all pixels have been searched.

[0040] Furthermore, splitting the slope unit into spatially continuous units based on the neighborhood search method includes the following steps:

[0041] S41, traverse all pixels of the sub-hill slope, if the pixel has not been searched, create a new queue and list, and add the pixel to the queue and list;

[0042] S42, sequentially remove the pixels in the queue as the central pixel, search for pixels in the neighborhood of the central pixel with the same code as the central pixel, and if the pixel has not been searched, enter the queue and list, and mark it as a searched pixel;

[0043] S43, if the queue is empty, encode the pixels in the list as new slope units until all the grids of the sub-slope are marked as searched pixels, and then stop.

[0044] Furthermore, all slope units are traversed along the flow direction according to the sub-hill source unit, and the slope units are merged into downstream units based on the area of ​​the slope units until the area of ​​the merged slope units is greater than the preset slope unit area threshold, and the discretized units with clear flow direction and spatial continuity are obtained, which includes the following steps:

[0045] S51, searching for the sub-hill slope source unit, taking the slope unit without upstream water flow input as the source unit and as the starting unit to be determined;

[0046] S52, processing the unit to be determined based on the slope unit area, and making the unit to be determined meet the slope unit area threshold;

[0047] S53: If the area of ​​the unit to be determined meets the slope unit area threshold, continue to determine whether the downstream slope units meet the slope unit area threshold until all slope units meet the slope unit area threshold, and then stop.

[0048] Further, processing the unit to be determined based on the slope unit area and making the unit to be determined meet the slope unit area threshold includes the following steps:

[0049] S521: If the area of ​​the unit to be determined is smaller than the area threshold of the slope unit, search for the downstream slope unit according to the flow direction, and calculate the area of ​​the unit to be determined and the downstream slope unit after merging;

[0050] S522: If the area after the merger is larger than the area threshold of the slope unit, the merged unit is used as a new slope unit; otherwise, the search downstream is continued until the area after the merger is larger than the area threshold of the slope unit.

[0051] The beneficial effects of the present invention are:

[0052] (1) The present invention overcomes the problems of large height difference and complex confluence relationship of slope units divided by traditional methods, and has broad application prospects in the fields of disaster emergency response and water resources management.

[0053] (2) The present invention uses sub-hillside units as the initial units for division, which reduces the longitudinal scale of the HRU and avoids excessive height differences within the HRU. The HRU is divided within the sub-hillside unit and its flow direction is controlled by computer graphics methods, ensuring that each HRU has only one downstream and accurate spatial position, which has broad application prospects in water resource scheduling management, watershed simulation, etc. It solves the problems that the HRU divided by the existing spatial discretization method is often inaccurate in spatial position, the confluence relationship between HRUs is unclear, and the terrain differences within the HRU are large. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0055] Figure 1 is a flow chart of a sub-hillslope-based spatial discretization method according to an embodiment of the present invention;

[0056] Figure 2 This is one of the working principle flow charts of a sub-hillslope-based spatial discretization method according to an embodiment of the present invention;

[0057] Figure 3 This is the second working principle flow chart of a sub-hillslope-based spatial discretization method according to an embodiment of the present invention;

[0058] Figure 4 This is a third flowchart of the working principle of a spatial discretization method based on sub-hill slopes according to an embodiment of the present invention;

[0059] Figure 5 This is a schematic diagram of the spatial discretization results based on sub-hill slopes in a certain place. DETAILED DESCRIPTION

[0060] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in the field should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0061] According to an embodiment of the present invention, a sub-hillslope-based spatial discretization method is provided.

[0062] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to the sub-hillside-based spatial discretization method of an embodiment of the present invention, the sub-hillside-based spatial discretization method includes the following steps:

[0063] S1. Based on the flow direction, initial river network and sub-basin extracted from the Digital Elevation Model (DEM), river sections are extracted, and the height difference between the beginning and the end of each river section is made less than the preset height difference threshold.

[0064] In one embodiment, extracting river sections and making the height difference between the beginning and the end of each river section smaller than a preset height difference threshold comprises the following steps:

[0065] S11. Based on the river network in the sub-basin, trace back the node pixels of the river network in the sub-basin (the source of the river network, and the river pixels into which two or more river pixels converge).

[0066] S12. Take each node as the source of the river section, trace back the downstream river network pixels according to the flow direction data, until the downstream pixels are node pixels and a river section is formed, and assign a code to the river section (the code starts from 0).

[0067] S13, calculating the height difference between the beginning and the end of each river section, if the height difference between the beginning and the end exceeds a preset height difference threshold (the default is 100m), then calculating the number of breakpoints.

[0068] S14. The first river pixel in this river section whose elevation is lower than the preset value (the default value is 100m) is re-set as a node and traced downstream until the elevation of the downstream river pixel meets the conditions, and the river section is recoded.

[0069] S15. If the height difference between the beginning and the end is less than the preset height difference threshold, the process is skipped and the next river section is judged until all river sections are judged.

[0070] In one embodiment, the formula for calculating the number of breakpoints is:

[0071] N = Int (dH / Threshold);

[0072] Where N is the number of breakpoints; dH is the height difference between the beginning and the end of the river section; Threshold is the maximum height difference threshold.

[0073] In one embodiment, the calculation formula of elevation is:

[0074] A=Head_H+Threshold*k (0<=k<=N);

[0075] In the formula, A represents elevation; Head_H represents the elevation of the first pixel of the river section, which is a variable; Threshold represents the maximum height difference threshold (the default value is 100m); k represents an integer value between 0 and N, and N represents the number of breakpoints.

[0076] In one embodiment, until the elevation of the downstream river pixel meets the conditions, the conditions include that the elevation is greater than Head_H+Threshold*(k+1), the downstream river pixel belongs to other rivers and is an invalid value; wherein Head_H represents the elevation of the first pixel of the river section, Threshold represents the maximum height difference threshold, k represents an integer value between 0-N, and N represents the number of breakpoints.

[0077] S2. Based on the river section and flow direction data in the sub-basin, the upstream catchment area of ​​each river pixel is taken as the sub-hillslope unit. Based on the sub-hillslope unit area, the sub-hillslope units with an area smaller than the preset sub-hillslope unit area threshold and spatially adjacent are merged to obtain the sub-hillslope unit of the river section. 2 ) and spatially adjacent sub-hillslope units are merged to avoid excessive fragmentation of sub-hillslope units.

[0078] In one embodiment, based on the river network and flow direction data in the sub-watershed, the upstream catchment area of ​​each river pixel is used as a sub-hillslope unit, and based on the sub-hillslope unit area, the sub-hillslope units whose areas are smaller than a preset sub-hillslope unit area threshold and spatially adjacent are merged to obtain the sub-hillslope units of the river section, including the following steps:

[0079] S21. Based on the river network data extracted from the digital elevation model, trace the source of the river network in the sub-basin and the entire river from the source to the downstream to obtain a list of river pixels arranged in upstream and downstream order.

[0080] S22. Based on the flow direction data, trace back the upstream watershed of each river pixel, take the upstream watershed as the initial sub-hill slope, assign different codes to each sub-hill slope, and calculate the area of ​​the sub-hill slope.

[0081] S23. The coded sub-hillslopes whose overall merged area is smaller than the sub-hillslope unit area threshold are coded as special values ​​(such as "-1"), and a set consisting of some sub-hillslopes is obtained (the set may be composed of multiple patches that are not spatially adjacent).

[0082] S24. Split the subset in the set into a number of spatially continuous sub-hillslope units through a neighborhood search method.

[0083] In one embodiment, splitting a subset of a set into a plurality of spatially continuous sub-hillslope units by using a neighborhood search method comprises the following steps:

[0084] S241, traverse all pixels of the set to be split, and use the unsearched pixels as the starting unit, and use the breadth-first search algorithm to find the pixel with the same code as the starting unit.

[0085] S242, assign a new code to the starting unit and mark it as searched, and stop after all pixels have been searched.

[0086] S3. Calculate the height above nearest drainage (HAND) of each pixel, and divide the slope units according to the height above nearest drainage (the natural breakpoint method can be used to calculate the threshold or the user can specify the threshold for segmentation).

[0087] S4. Split (some discontinuous) slope units into spatially continuous units based on the neighborhood search method.

[0088] In one embodiment, splitting the slope unit into spatially continuous units based on the neighborhood search method includes the following steps:

[0089] S41. Traverse all pixels of the sub-hill slope. If the pixel has not been searched, create a new queue and list, and add the pixel to the queue and list.

[0090] S42, remove the pixels in the queue in sequence as the central pixel, search for pixels with the same code as the central pixel in the neighborhood (eight-dimensional area) of the central pixel, if the pixel has not been searched, enter the queue and list, and mark it as a searched pixel.

[0091] S43, if the queue is empty, the pixels in the list are encoded as new slope units (composed of spatially continuous pixels), until all the grids of the sub-slope are marked as searched pixels, and then stop.

[0092] S5. Traverse all slope units along the flow direction according to the sub-hill slope source unit, and merge the slope units into downstream units based on the area of ​​the slope units until the area of ​​the merged slope units is greater than the preset slope unit area threshold, and obtain a discretized unit with a clear flow direction and spatial continuity. That is, merge the slope units with an area less than the threshold into their downstream units until the area of ​​the merged unit is greater than the specified threshold (the default is the average area of ​​all slope units), and obtain a discretized unit with a clear flow direction and spatial continuity.

[0093] In one embodiment, all slope units are traversed along the flow direction according to the sub-hill source unit, and the slope units are merged into downstream units based on the area of ​​the slope units until the area of ​​the merged slope units is greater than a preset slope unit area threshold, and the discrete units with clear flow direction and spatial continuity are obtained, which includes the following steps:

[0094] S51, searching for the sub-hill slope source unit, taking the slope unit without upstream water flow input as the source unit and as the starting unit to be determined.

[0095] S52: Process the unit to be determined based on the slope unit area, and make the unit to be determined meet the slope unit area threshold.

[0096] S53: If the area of ​​the unit to be determined meets the slope unit area threshold, continue to determine whether the downstream slope units meet the slope unit area threshold until all slope units meet the slope unit area threshold, and then stop.

[0097] In one embodiment, processing the unit to be determined based on the slope unit area and making the unit to be determined meet the slope unit area threshold includes the following steps:

[0098] S521: If the area of ​​the unit to be determined is smaller than the area threshold of the slope unit (the average area of ​​all slope units by default), search for the downstream slope unit according to the flow direction, and calculate the area of ​​the unit to be determined and the downstream slope unit after merging.

[0099] S522: If the area after the merger is larger than the area threshold of the slope unit, the merged unit is used as a new slope unit; otherwise, the search downstream is continued until the area after the merger is larger than the area threshold of the slope unit.

[0100] Taking the spatial discretization of a certain river basin as an example, the specific implementation method of the present invention is described:

[0101] (1) The basin of a certain place is named after its gourd-shaped plane shape. The terrain is wider in the south and narrower in the north. The basin terrain is generally high in the south and low in the north, high in the west and low in the east, with a gradually decreasing slope, large terrain undulations, and large height differences in some river sections, which can better reflect the advantages and practicality of this method.

[0102] (2) Using Merit DEM (a set of data sets, in which a digital model with good performance was developed by removing multiple error components from existing satellite data systems. It represents topographic highlands with a resolution of 3SEC, covering the land area between 90N-60S, and referring to the EG96 geoid) data as input data, the D8 algorithm (the center pixel flows to the neighboring pixel with the largest slope, which is a recognized technology and will not be repeated here) is used to calculate the flow direction and the cumulative amount of confluence, and the original river network is proposed. By finding the node pixels in the original river network, the initial river network link data is constructed. At this time, there are still river sections with too large height differences. Subsequently, the initial river network link data is circulated to determine whether the height difference of each river section is greater than the threshold (the default is 100m). If it is greater than the threshold, a new node pixel is found for the river section. First, the number of breakpoints N is calculated according to the formula N=Int(dH / Threshold), where dH is the height difference between the beginning and the end of the river section, and Thresold is the maximum height difference threshold. Then, the first river pixel with an elevation less than or equal to Head_H+Thresold*k (0<=k<=N) in this river section is used as a node to trace downstream again until the downstream reaches a river pixel with an elevation greater than Head_H+Thresold*(k+1), other coded river pixels, or invalid value pixels. Then, the downstream tracing is stopped and the river section is recoded. The HAND of the study area is calculated using the river network link data and DEM.

[0103] (3) First, traverse the river network data within the sub-basin, search for the source of the river network, and then trace the entire river from the source to the downstream to obtain the river location arranged in upstream and downstream. Use the river code as the key and the river pixel row and column number list as the value to build a dictionary. Then, combined with the flow direction data, trace the upstream of each river raster and assign different codes, and calculate the upstream area corresponding to each code. 2 ) is recoded as "-1" to obtain the merged upstream, and then all pixels of the upstream data are traversed to search for pixels with the same code that are connected in the 4-neighborhood, merge them into a set and encode them. At this time, the sub-slope unit that distinguishes the left and right slopes and the source slope is obtained.

[0104] (4) Traverse each sub-hillslope unit, calculate the HAND value corresponding to the pixel inside the sub-hillslope, use the natural breakpoint method to calculate the breakpoint value under the division threshold (user-specified, the default is 3), and then use the pixel corresponding to the HAND value of each interval as the slope unit. At this time, these breakpoints divide the sub-hillslope unit into several groups with similar properties.

[0105] (5) Search for the source unit of the sub-hill slope, and regard the slope unit without upstream water flow input as the source unit as the starting unit to be judged. If the area of ​​the unit to be judged is less than the specified threshold, search for the downstream slope unit according to the flow direction, calculate the area of ​​the two slope units after merging, and if it is greater than the threshold, merge them as a new slope unit. Otherwise, continue to search downstream until the merged unit meets the threshold. If the area of ​​the unit to be judged meets the threshold, continue to judge the downstream slope units until the areas of all slope units meet the threshold, and then stop.

[0106] The principle of the present invention is that the longitudinal span of the sub-hill slope unit is small and there is no confluence relationship between the sub-hill slope units, that is, the river channel with a determined flow direction on each sub-hill slope is determined by the height difference of the river channel, which avoids the excessive height difference of the HRU caused by the area with a large height difference of the river channel, and there is no surface water flow exchange between adjacent sub-hill slopes, so only the flow direction inside the sub-hill slope needs to be considered, and each HRU has a unique downstream. The working principle flow chart of the present invention is as follows Figure 2-Figure 4 As shown, Figure 2 The arrow ① in Figure 3 The arrow ① in the middle connects, Figure 3 The arrow ② in Figure 4 Connect with arrow ② in the figure. Figure 5 This is a schematic diagram of the spatial discretization results based on sub-hill slopes in a certain area. Figure 5 The left half is the spatial discretization result without considering the sub-hillslope unit, and the right half is the spatial discretization result based on the sub-hillslope.

[0107] In summary, the present invention breaks through the problems of excessive height difference and complex confluence relationship of slope units divided by traditional methods, and has broad application prospects in the fields of disaster emergency response and water resources management. The present invention uses sub-hillside units as the initial units for division, reduces the longitudinal scale of the HRU, and avoids excessive height difference inside the HRU; divides the HRU inside the sub-hillside unit, and uses computer graphics methods to control its flow direction, ensuring that each HRU has only one downstream and accurate spatial position, and has broad application prospects in water resources scheduling management, watershed simulation and other aspects. It solves the problems that the HRU divided by the existing spatial discretization method often has inaccurate spatial position, unclear confluence relationship between HRUs, and large terrain differences inside the HRU.

[0108] 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 principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A spatial discretization method based on sub-hill slopes, characterized in that: The sub-hillslope-based spatial discretization method includes the following steps: S1, based on the flow direction, initial river network and sub-basin extracted from the digital elevation model, extract the river section, and make the height difference between the beginning and the end of each river section less than the preset height difference threshold; S2. Based on the river section and flow direction data in the sub-basin, the upstream catchment area of ​​each river pixel is taken as a sub-hillslope unit. Based on the sub-hillslope unit area, the sub-hillslope units whose areas are smaller than a preset sub-hillslope unit area threshold and spatially adjacent are merged to obtain the sub-hillslope unit of the river section. S3, calculating the relative height value of the nearest neighboring river channel of each pixel, and dividing the slope position unit according to the relative height value of the nearest neighboring river channel; S4, splitting the slope unit into spatially continuous units based on the neighborhood search method; S5, traversing all slope units along the flow direction according to the sub-hill slope source unit, and merging the slope units into downstream units based on the slope unit area, until the area of ​​the merged slope unit is greater than a preset slope unit area threshold, thereby obtaining a discretized unit with a clear flow direction and spatial continuity; The method of extracting river sections and making the height difference between the beginning and the end of each river section smaller than a preset height difference threshold comprises the following steps: S11. Trace back the node pixels of the river network in the sub-basin, which include the source of the river network and the river pixels into which two or more river pixels converge; S12, taking each node as the source of the river section, tracing back the downstream river network pixels according to the flow direction data, until the downstream pixels are node pixels and a river section is formed, and assigning a code to the river section; S13, calculating the height difference between the beginning and the end of each river section, and if the height difference between the beginning and the end exceeds a preset height difference threshold, calculating the number of breakpoints; S14, re-using the first river pixel whose elevation is lower than the preset value in the river section as a node and tracing downstream until the elevation of the downstream river pixel meets the condition, and re-coding the river section; S15, if the height difference between the beginning and the end is less than the preset height difference threshold, then skip and continue to judge the next river section until all river sections are judged; The method comprises the following steps: based on the river network and flow direction data in the sub-watershed, taking the upstream catchment area of ​​each river pixel as a sub-hillslope unit, and based on the sub-hillslope unit area, merging the sub-hillslope units whose areas are smaller than a preset sub-hillslope unit area threshold and are spatially adjacent to obtain the sub-hillslope units of the river section: S21, based on the river network data extracted from the digital elevation model, tracing back to the source of the river network in the sub-basin and the entire river from the source to the downstream, and obtaining a list of river pixels arranged in the order of upstream and downstream; S22. Based on the flow direction data, trace back the upstream catchment area of ​​each river pixel, take the upstream catchment area as the initial sub-hill slope, assign different codes to each sub-hill slope, and calculate the area of ​​the sub-hill slope; S23, the coded sub-hillsides whose overall combined area is smaller than the sub-hillside unit area threshold are obtained as a set consisting of some sub-hillsides, and are coded as special values; S24, splitting the sub-set in the set into a plurality of spatially continuous sub-hillslope units by using a neighborhood search method; The method of splitting the slope position unit into spatially continuous units based on the neighborhood search method comprises the following steps: S41, traverse all pixels of the sub-hill slope, if the pixel has not been searched, create a new queue and list, and add the pixel to the queue and list; S42, sequentially remove the pixels in the queue as the central pixel, search for pixels in the neighborhood of the central pixel with the same code as the central pixel, and if the pixel has not been searched, enter the queue and list, and mark it as a searched pixel; S43, if the queue is empty, the pixels in the list are encoded as new slope units until all the grids of the sub-slope are marked as searched pixels, and then stop.

2. The sub-hillslope-based spatial discretization method according to claim 1 is characterized in that: The formula for calculating the number of breakpoints is: N = Int (dH / Threshold); In the formula, N represents the number of breakpoints; dH represents the height difference between the beginning and the end of the river section; Threshold indicates the maximum height difference threshold.

3. The sub-hillslope-based spatial discretization method according to claim 1 is characterized in that: The calculation formula for the breakpoint elevation is: A=Head_H+Threshold*k, 0<=k<=N; In the formula, A represents elevation; Head_H represents the elevation of the first pixel of the river section; Threshold indicates the maximum height difference threshold; k is an integer value between 0 and N, where N represents the number of breakpoints.

4. The sub-hillslope-based spatial discretization method according to claim 1, characterized in that: The condition that the elevation of the downstream river pixel meets the conditions includes that the elevation is greater than Head_H+Threshold*(k+1), the downstream river pixel belongs to other rivers, and is an invalid value; Among them, Head_H represents the elevation of the first pixel of the river section, Threshold represents the maximum height difference threshold, k represents an integer value between 0 and N, and N represents the number of breakpoints.

5. The sub-hillslope-based spatial discretization method according to claim 1, characterized in that: The method of splitting the sub-sets in the set into a plurality of spatially continuous sub-hillslope units by using the neighborhood search method comprises the following steps: S241, traverse all pixels of the set to be split, and use the unsearched pixel as the starting unit, and use the breadth-first search algorithm to find the pixel with the same code as the starting unit; S242, assign a new code to the starting unit and mark it as searched, and stop after all pixels have been searched.

6. The sub-hillslope-based spatial discretization method according to claim 1, characterized in that: The method of traversing all slope units along the flow direction according to the sub-hillside source unit and merging the slope units into downstream units based on the slope unit area until the area of ​​the merged slope unit is greater than a preset slope unit area threshold to obtain a discretized unit with a clear flow direction and spatial continuity includes the following steps: S51, searching for the sub-hill slope source unit, taking the slope unit without upstream water flow input as the source unit and as the starting unit to be determined; S52, processing the unit to be determined based on the slope unit area, and making the unit to be determined meet the slope unit area threshold; S53: If the area of ​​the unit to be determined meets the slope unit area threshold, continue to determine whether the downstream slope units meet the slope unit area threshold until all slope units meet the slope unit area threshold, and then stop.

7. The sub-hillslope-based spatial discretization method according to claim 6, characterized in that: The method of processing the unit to be determined based on the slope unit area and making the unit to be determined meet the slope unit area threshold comprises the following steps: S521: If the area of ​​the unit to be determined is smaller than the area threshold of the slope unit, search for the downstream slope unit according to the flow direction, and calculate the area of ​​the unit to be determined and the downstream slope unit after merging; S522: If the area after the merger is larger than the area threshold of the slope unit, the merged unit is used as a new slope unit; otherwise, the search downstream is continued until the area after the merger is larger than the area threshold of the slope unit.

Images