A method for rapid classification of landform types on the Loess Plateau

Abstract

The present invention discloses a method for rapid classification of landform types on the Loess Plateau, comprising the following steps: S1, determining a minimum processing unit data set: clipping the basic raster data of the target basin with a single sub-basin as the boundary, obtaining the basic raster data of the sub-basin range as the minimum processing unit data set for landform classification; S2, sub-basin landform classification: for each minimum processing unit data set, tracing the source of the simulated river network one by one, and preliminarily dividing the sub-basin raster into gully, gully slope, and gully-tableland types; merging and / or correcting the isolated island type, source gully-valley type, and abnormal gully-tableland in the preliminary division results; and starting from the source of the river network raster, correcting the upstream gully-tableland river network to achieve sub-basin landform classification. The advantages are: the extracted gully-tableland-gully-slope-gully-valley types can ensure the upstream and downstream hydraulic connection between the three, and can be better applied to the unit division of distributed water and sediment models.

Description

Technical Field

[0001] The present invention relates to the technical field of landform type identification, and in particular to a method for quickly classifying landform types on the Loess Plateau. Background Art

[0002] The topography of the Loess Plateau is a crucial underlying surface for studying gully erosion in its watersheds and a key factor influencing the accuracy of distributed water-sediment models. Different landforms, such as gully plateaus, gully slopes, and gully valleys, have distinct influences on water-sediment processes and therefore need to be distinguished in water-sediment models.

[0003] Existing research primarily uses ultra-high-precision DEMs and other data to identify landform types. This has proven difficult to apply to large watersheds on the Loess Plateau, primarily due to the difficulty in obtaining ultra-high-precision DEM data covering the entire region and the computational dimensionality penalty associated with such large data volumes. Furthermore, existing algorithms often classify landforms based on the morphological characteristics of individual grid cells, failing to reflect upstream and downstream connections between landform types (for example, potentially demarcating gully slopes flowing into gully plateaus), making them inadequate for hydrological modeling. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for quickly classifying landform types on the Loess Plateau, thereby solving the above-mentioned problems existing in the prior art.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] A method for rapidly classifying landform types on the Loess Plateau comprises the following steps:

[0007] S1. Determination of the minimum processing unit data set:

[0008] The basic raster data of the target basin is clipped with a single sub-basin as the boundary, and the basic raster data of the sub-basin range is obtained as the minimum processing unit dataset for landform classification;

[0009] S2. Sub-basin landform classification:

[0010] For each minimum processing unit dataset, the river network is simulated and traced back to its source, and the sub-basin grid is preliminarily divided into gully, gully slope, and gully-plateau types. The island type, source gully-valley type, and abnormal gully-plateau in the preliminary division results are merged and / or corrected respectively. Starting from the source of the river network grid, the upstream gully-plateau river network is corrected to achieve sub-basin landform classification.

[0011] Preferably, step S2 specifically includes the following contents:

[0012] S21. Trace back each simulated river network grid and divide the sub-basin grid into three types: gully valley, gully slope, and gully plateau;

[0013] S22. Use the GIS software's mesh filter to merge the island types smaller than the area threshold in the preliminary classification results into the surrounding types, and simulate the river network tracing and process the corrected types one by one to ensure that the slope grid connection satisfies the upstream and downstream relationship of gully plateau-gully slope-gully valley.

[0014] S23. Change the sub-basin watershed grid of gully type to gully-plateau type. Then, recursively follow the grid flow direction down to the simulated river network grid and change the type of all gully-plateau type grids in between to gully-plateau type. Start from the simulated river network and trace back to the source grid one by one. If a gully-slope type grid is encountered, change the gully-plateau type to a gully-slope type grid one by one along the grid flow direction.

[0015] S24, identifying the watershed grid of the gully-plateau type, and modifying the watershed grid whose surrounding grids are all gully-valley or gully-slope types to the gully-valley type;

[0016] S25. Starting from the source of the river network grid, change the river network grid without gully-slope type upstream to gully-plateau type, and end at the river network grid with gully-slope type upstream.

[0017] Preferably, step S21 specifically includes marking the river network grid as a gully type, tracing the source according to the flow direction, marking the slope grid with a slope less than the slope-valley slope threshold as a gully type, continuing to trace upward, marking the slope grid with a slope greater than the plateau slope threshold as a gully-slope type, and marking the slope grid with a slope less than the plateau slope threshold as a gully-plateau type;

[0018] If there is a gentle slope area in the middle of the gully slope type, it needs to be marked as the gully slope type, that is, ensure that the division result is distributed from top to bottom as gully plateau-gully slope-gully valley type.

[0019] Preferably, after step S2, the method further includes: after all the landforms in the sub-basins are classified, integrating the sub-basin classification results according to the relative positions of the sub-basins in the target basin to obtain a complete landform classification result of the target basin.

[0020] Preferably, GIS software is used to calculate the basic grid data of the target watershed, such as the grid flow direction, runoff accumulation, simulated river network, slope and sub-basin unit of the target watershed based on the DEM data of the target watershed.

[0021] Preferably, the grid flow direction is extracted using the D8 algorithm.

[0022] Preferably, the simulated river network is extracted from the cumulative number of confluence according to a given river network threshold, and the values greater than the threshold are set to 1, indicating a river network grid, and the rest are set to 0, indicating a slope grid; the river network threshold is determined by trial calculation according to the basin conditions.

[0023] Preferably, the sub-basin unit divides the area belonging to the same river section into the same sub-basin according to the simulated river network confluence control range.

[0024] The beneficial effects of the present invention are as follows: 1. By splitting the raster data of a large watershed into numerous relatively independent sub-watersheds for landform type classification, the present invention can use a parallel algorithm to perform calculations on different sub-watersheds simultaneously, effectively improving the classification rate. 2. Compared with existing algorithms, the present invention can use a relatively low-precision DEM for landform type classification and can ensure a certain degree of accuracy in the classification results. 3. The gully-plateau-gully-slope-gully-valley types extracted by the present invention can ensure the upstream and downstream hydraulic connection between the three, and can be better applied to the unit division of distributed water and sediment models. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a principle flow chart of the method in the embodiment of the present invention;

[0026] Figure 2 Schematic diagram of dividing cases in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0028] like Figure 1 As shown, in this embodiment, a method for rapid classification of landform types of the Loess Plateau is provided, which includes the following three parts:

[0029] 1. Basic raster data acquisition

[0030] GIS software is used to calculate the basic raster data of the target basin, such as grid flow direction, confluence accumulation number, simulated river network, slope and sub-basin unit, based on the DEM data of the target basin.

[0031] Grid flow direction was extracted using the D8 algorithm. Simulated river networks were extracted from the cumulative runoff according to a given river network threshold, with values above the threshold set to 1 (for river network grids) and all others set to 0 (for slope grids). The river network threshold was determined through trial calculations based on the watershed conditions. Sub-basin units were divided into sub-basins based on the simulated river network runoff control range, with areas belonging to the same river reach divided into the same sub-basin.

[0032] 2. Determination of the minimum processing unit data set:

[0033] The basic raster data of the target basin is clipped with a single sub-basin as the boundary, and the basic raster data of the sub-basin range is obtained as the minimum processing unit dataset for landform classification.

[0034] 3. Sub-basin landform classification:

[0035] For each minimum processing unit dataset, the river network is simulated and traced back to its source, initially dividing the sub-basin grid into gully, gully slope, and gully-plateau types. The isolated island types, source gully-valley types, and abnormal gully-plateau types in the preliminary division results are merged and / or corrected. Starting from the source of the river network grid, the upstream gully-plateau river network is corrected to achieve sub-basin landform classification. Specifically, the following five steps are required for each minimum processing unit dataset:

[0036] 2.1. Tracing back through the simulated river network grids, the sub-basin grids were divided into three types: gully-valley, gully-slope, and gully-plateau. The specific division process is as follows: Mark the river network grid as gully-valley type, trace back along the flow direction, and mark the slope grids with a slope gradient less than the slope-valley threshold as gully-valley type. Continuing upstream, mark the slope grids with a slope gradient greater than the plateau slope threshold as gully-slope type, and mark the slope grids with upstream slope gradients less than the plateau slope threshold as gully-plateau type. Note that if there are areas with gentle slopes within the gully-slope type, they should be marked as gully-slope type. This ensures that the division results in a gully-plateau-gully-slope-gully type distribution from top to bottom.

[0037] 2.2. The GIS software's mesh filter was used to merge the island types smaller than the area threshold in the preliminary classification results into the surrounding types. The river network was then simulated and traced back one by one. The corrected types were processed so that the slope grid connection satisfied the upstream and downstream relationship of gully plateau-gully slope-gully valley.

[0038] 2.3. Modify the sub-basin watershed grid of gully type to gully-plateau type, and recursively modify all gully-type grids in the middle to gully-plateau type along the grid flow direction until the simulated river network grid is reached. Trace the source grid by grid from the simulated river network, and if a gully-slope type is encountered, modify the gully-plateau type to gully-slope type along the grid flow direction.

[0039] 2.4. Identify the watershed grids of gully-plateau type and modify the watershed grids whose surrounding grids are all gully or gully-slope type to gully type.

[0040] 2.5. Starting from the source of the river network grid, change the river network grid without gully-slope type upstream to gully-plateau type, and end at the river network grid with gully-slope type upstream.

[0041] In this embodiment, after all the landforms in the sub-basins are classified, the sub-basin classification results are integrated according to their relative positions in the target basin to obtain the complete landform classification results of the target basin. Figure 2 .

[0042] In this embodiment, a parallel algorithm may be used to perform calculations on different sub-watersheds simultaneously, thereby increasing the classification rate.

[0043] By adopting the above technical solution disclosed in the present invention, the following beneficial effects are obtained:

[0044] The present invention provides a method for rapidly classifying landform types on the Loess Plateau. This method classifies landform types by splitting raster data from a large watershed into numerous relatively independent sub-watersheds. Parallel algorithms can be used to simultaneously perform calculations on different sub-watersheds, effectively increasing the classification rate. Compared to existing algorithms, the present invention can use relatively low-precision DEMs for landform classification and ensure a certain degree of accuracy in the classification results. The gully-plateau-gully-slope-gully-valley types extracted by the present invention ensure upstream and downstream hydraulic connections between the three, enabling better application in the unit division of distributed water and sediment models.

[0045] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for rapid classification of landform types on the Loess Plateau, characterized by: The following steps are included: S1. Determination of the minimum processing unit data set: The basic raster data of the target basin is clipped with a single sub-basin as the boundary, and the basic raster data of the sub-basin range is obtained as the minimum processing unit dataset for landform classification; S2. Sub-basin landform classification: For each minimum processing unit data set, the river network is traced back one by one, and the sub-basin grid is preliminarily divided into gully, gully slope, and gully plateau types; the island type, source gully valley type, and abnormal gully plateau in the preliminary division results are merged and / or corrected respectively; and starting from the source of the river network grid, the upstream gully plateau river network is corrected to achieve sub-basin landform classification; Step S2 specifically includes the following contents, S21. Trace back each simulated river network grid and divide the sub-basin grid into three types: gully valley, gully slope, and gully plateau; S22. Use the GIS software's mesh filter to merge the island types smaller than the area threshold in the preliminary classification results into the surrounding types, and simulate the river network tracing and process the corrected types one by one to ensure that the slope grid connection satisfies the upstream and downstream relationship of gully plateau-gully slope-gully valley. S23, modify the sub-basin watershed grid of gully type to gully-plateau type, and recursively modify all gully type grids in the middle to gully-plateau type along the grid flow direction until the simulated river network grid is reached; S24, starting from the simulated river network, tracing back to the source grid by grid, if a gully-slope type is encountered, recursively change all gully-plateau types to gully-slope types along the grid flow direction; S25. Starting from the source of the river network grid, change the river network grid without gully-slope type upstream to gully-plateau type, and end at the river network grid with gully-slope type upstream.

2. The method for rapid classification of landform types of the Loess Plateau according to claim 1 is characterized by: Step S21 specifically involves marking the river network grid as a gully-valley type, tracing the source along the flow direction, marking the slope grids with a slope less than the slope-valley threshold as a gully-valley type, and continuing upwards to mark the slope grids with a slope greater than the plateau slope threshold as a gully-slope type, and marking the slope grids with a slope less than the plateau slope threshold as a gully-plateau type. If there is a gentle slope area in the middle of the gully slope type, it needs to be marked as the gully slope type, that is, ensure that the division result is distributed from top to bottom as gully plateau-gully slope-gully valley type.

3. The method for rapid classification of landform types of the Loess Plateau according to claim 1 is characterized by: The step S2 also includes, after all the landforms in the sub-basins are classified, integrating the sub-basin classification results according to the relative positions of the sub-basins in the target basin to obtain a complete landform classification result of the target basin.

4. The method for rapid classification of landform types on the Loess Plateau according to any one of claims 1 to 3, characterized in that: GIS software is used to calculate the basic raster data of the target basin, such as grid flow direction, confluence accumulation number, simulated river network, slope and sub-basin unit, based on the DEM data of the target basin.

5. The method for rapid classification of landform types of the Loess Plateau according to claim 4 is characterized by: The grid flow direction is extracted using the D8 algorithm.

6. The method for rapid classification of landform types of the Loess Plateau according to claim 4, characterized in that: The simulated river network is extracted from the cumulative number of confluence according to a given river network threshold, and the values greater than the threshold are set to 1, representing the river network grid, and the rest are set to 0, representing the slope grid; the river network threshold is determined by trial calculation based on the basin conditions.

7. The method for rapid classification of landform types of the Loess Plateau according to claim 4, characterized in that: The sub-basin unit divides the area belonging to the same river section into the same sub-basin according to the simulated river network confluence control range.

Images