Mango garden planting area runoff estimation method based on SCS-CN model correction

Abstract

The invention discloses a mango garden planting area runoff estimation method based on SCS-CN model correction, and relates to the technical field of mango planting runoff estimation, and the method comprises the following steps: obtaining spatial data in a mango planting area, carrying out the preprocessing through ArcGIS, uniformly setting the spatial data as projection coordinates, and carrying out the calculation of the projection coordinates; collecting a basic element data set of a mango garden planting area; constructing an SCS-CN model, reading a basic element data set of a mango garden planting area, and calling a pre-configured algorithm to determine CN parameters of the SCS-CN model; according to the method, an exhaustion method is adopted, the rate period and the verification period of the runoff initial loss coefficient in the SCS-CN model are obtained respectively, the verified SCS-CN model is obtained and utilized, the runoff loss amount of the mango garden planting area in the corresponding period is estimated, and the surface runoff loss amount of the mango planting area is rapidly and effectively predicted.

Description

Technical Field

[0001] The invention relates to the technical field of mango planting runoff estimation, and in particular to a runoff estimation method for a mango orchard planting area based on SCS-CN model correction. Background Art

[0002] Mango is one of the famous tropical fruits. It has delicate flesh, unique flavor and rich nutrition. It is deeply loved by people and is known as the "King of Tropical Fruits". Mango has the advantages of large domestic and foreign demand and broad development space in market competition. At present, mango is one of the seven major characteristic fruit industries with high quality and key development in Guangxi's modern characteristic agriculture. Baise, Guangxi, as a famous mango hometown in my country, is a Chinese characteristic agricultural product advantage zone created by Guangxi. "Baise Mango" has been included in the sixth batch of national agricultural product geographical indication registered products. Baise is also a typical karst mountainous area, with mountainous areas accounting for about 95.4% of the total area (rocky mountains account for 30%, earthen mountains account for 65.4%), and hills and plains account for only 4.6%; due to natural conditions, mangoes are mostly cultivated in wastelands and sparse forest slopes in hilly and mountainous areas; the parent rocks of this type of slopes are mostly Quaternary red soil, limestone and sandstone, and the topsoil layer is shallow , the soil is acidic, thin, poorly structured, and has poor water and fertilizer retention capacity; at the same time, fruit farmers generally spray herbicides to suppress weeds and ignore the protection of vegetation; abundant rainfall, the planting method of steep slope reclamation and the exposed surface lead to a high frequency of surface runoff in mango orchards and serious soil erosion; soil and water loss is not only one of the important forms of agricultural non-point source pollution, but also a carrier of non-point source pollutants and an important way to cause water pollution; relevant studies have shown that the mango planting area produces an average of 21.5 surface runoffs per year, and the runoff and sediment loss of single-crop mangoes are as high as 10590.5m3 / hm2 and 4440.3kg / hm2; data show that the area of ​​soil and water loss in Baise is 6438.56km2, and the annual soil loss is 5.3008 million tons; it can be seen that accurate prediction of runoff is crucial for soil and water conservation and water ecological safety management in mango producing areas;

[0003] At present, the main methods for estimating surface runoff include in-situ observation of runoff plots, automatic online flow monitoring, and model simulation; the in-situ observation of runoff plots can effectively grasp the impact of a single factor on surface runoff, but it is labor-intensive; automatic online flow monitoring can monitor runoff in real time, but it is difficult to promote due to the limited instrument installation conditions and high maintenance costs; model simulation is usually used, and the SCS-CN model is a runoff curve number model summarized by the Ministry of Agriculture of other countries based on practical experience. It is one of the most widely used surface runoff models at home and abroad; the runoff initial loss coefficient is the ratio of the initial loss of the rainfall runoff process to the maximum potential water storage capacity in the SCS-CN model. Due to the heterogeneity of geographic space, parameters such as λ and CN values ​​obtained based on the geographical and climatic characteristics of other countries are not fully applicable in China;

[0004] At present, domestic research on λ optimization in the SCS-CN model is mainly concentrated in the Loess Plateau and loess hilly areas, and no research on the southern red soil hilly areas has been reported. The CN and λ values ​​of the traditional SCS-CN model cannot be corrected and optimized, and thus cannot adapt to actual needs. As a result, the applicability and accuracy of the traditional SCS-CN model in simulating surface runoff in mango planting areas needs to be improved. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] In view of the deficiencies in the prior art, the present invention provides a method for estimating runoff in a mango orchard planting area based on the SCS-CN model correction, which can quickly and effectively predict the surface runoff loss in the mango planting area and solve the problems raised in the background technology.

[0007] (II) Technical solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] A method for estimating runoff in a mango orchard planting area based on SCS-CN model correction, the method comprising the following steps:

[0010] Select a mango planting area, obtain the spatial data in the mango planting area, use ArcGIS for preprocessing, set the spatial data to projection coordinates, and collect the basic element data set of the mango planting area;

[0011] Build the SCS-CN model, read the basic element dataset of the mango orchard planting area, and call the pre-configured algorithm to determine the CN parameters of the SCS-CN model;

[0012] The exhaustive method is adopted, the annual rainfall-runoff data of the monitoring station in the Nth year is used as the rate period of the runoff initial loss coefficient of the SCS-CN model, and the annual rainfall-runoff data of the monitoring station in the N+1th year is used as the verification period of the runoff initial loss coefficient of the SCS-CN model to obtain the verified SCS-CN model;

[0013] The validated SCS-CN model was used to estimate the runoff loss in the mango orchard planting area during the corresponding period.

[0014] Furthermore, during the operation of the SCS-CN model based on ArcGIS, various spatial data were uniformly set as projection coordinates, and the spatial data projection coordinates were set to WGS_1984_UTM_Zone_48N in combination with the location of the mango planting area.

[0015] Furthermore, the basic element data set includes but is not limited to mango orchard distribution data, planting area soil data, DEM data, NDVI values, and rainfall data.

[0016] Furthermore, the process of obtaining the mango orchard distribution data is as follows:

[0017] Firstly, the river direction is drawn on the Google Earth platform to form KML data. The KML data is imported into the ArcGIS 10.2 platform based on the DEM data. The watershed range is generated using the hydrological analysis module. The hydrological analysis module is implemented through the steps of depression filling, flow direction analysis, flow statistics and watershed extraction.

[0018] Secondly, the Sentinel-2 remote sensing images of the basin were obtained in the GEE platform. Combined with the survey mechanism, M land types were determined. Sample data were constructed according to the surface characteristics presented by the M land types, and the land use distribution of the basin was obtained according to the random forest algorithm.

[0019] Furthermore, the process of obtaining soil data in the planting area is as follows:

[0020] The DEM data of the mango-growing area was obtained from the NASA Earth Science Data Website. The HWSD soil dataset and the soil type dataset were superimposed to obtain the soil property database of the mango-growing area for the soil hydrological grouping calculation required by the SCS-CN model.

[0021] Furthermore, DEM data are used to calculate terrain factors and distance factors. NDVI values: Based on the regional normalized vegetation index dataset of the data center, the inverse distance weighted method is used to interpolate the NDVI values ​​in ArcGIS to calculate the CN values ​​of the SCS-CN model and the vegetation factors of nitrogen and phosphorus inflow into the river. Rainfall data: The monthly precipitation dataset is obtained from the data center.

[0022] Furthermore, the SCS-CN model is an empirical model based on two assumptions and the water balance equation;

[0023] The first assumption is that the ratio of actual runoff infiltration to the potential maximum water storage is equal to the ratio of surface runoff depth to the potential maximum runoff depth, see formula (1); the second assumption is that the initial runoff loss is proportional to the potential maximum water storage, see formula (2); the water balance equation is: the total rainfall is equal to the sum of the initial runoff loss, the actual runoff infiltration and the surface runoff depth, see formula (3);

[0024]

[0025] I a =λS(2

[0026] P=I a +F+Q(3)

[0027] The empirical formula of rainfall and runoff can be derived from equations (1) to (3), as shown in equations (4) to (5). When the rainfall is greater than the initial runoff loss, there is surface runoff. Otherwise, the surface runoff is 0:

[0028]

[0029] Q=0,P a (5

[0030] Where: F is the actual runoff infiltration; S is the potential maximum water storage capacity, that is, the ability of the surface to intercept rainfall; P is the rainfall; I a is the initial loss of runoff; λ is the initial loss coefficient; Q is the surface runoff depth; S can be calculated as follows:

[0031]

[0032] In formula (6), CN is dimensionless and its value range is [0, 100].

[0033] Furthermore, the steps for obtaining the value of CN are as follows:

[0034] The first step is to determine the minimum soil infiltration rate. The saturated hydraulic conductivity of the upper and lower layers of each soil type is calculated in the SPAW software. The corresponding soil sand content is selected according to the minimum hydraulic conductivity. The minimum soil infiltration rate is calculated according to formula (7) and formula (8). The soil hydrological grouping of the mango planting area is determined according to the minimum infiltration rate.

[0035]

[0036] X = (20 × Y) 1.8 (8)

[0037] Where: SD is the percentage of soil sand content; Y is the average soil particle diameter; X is the minimum infiltration rate;

[0038] The second step is to classify the vegetation coverage type. In ArcGIS, the vegetation coverage of the mango planting area is reclassified into three types. When the NDVI value is less than 0.5, the vegetation type is low vegetation coverage, and the reclassification value is 1; when 0.5≤NDVI value≤0.75, the vegetation type is medium vegetation coverage, and the reclassification value is 2; when the NDVI value is 0>0.75, the vegetation type is high vegetation coverage, and the reclassification value is 3;

[0039] The results of the first and second steps are combined to trigger a review mechanism to determine the CN value of the mango growing area.

[0040] ​Furthermore, when the verified SCS-CN model was obtained, the simulated runoff depths under different rainfalls were calculated in sequence with a step size of 0.01, based on the regression equation: λ = 0.0007*Pm+0.071, to obtain the values ​​of λ under different rainfall amounts;

[0041] Where Pm represents monthly rainfall, 0.071 represents the conventional coefficient, and λ represents the runoff initial loss coefficient.

[0042] (III) Beneficial effects

[0043] The present invention provides a method for estimating runoff in a mango orchard planting area based on SCS-CN model correction, which has the following beneficial effects:

[0044] The present invention corrects and optimizes the CN and λ values ​​of the SCS-CN model, and uses the verified SCS-CN model to evaluate the surface runoff loss in the mango planting area, thereby effectively improving the runoff estimation accuracy in the mango orchard planting area and shortening the non-point source pollution assessment time; compared with the traditional SCS-CN model, the λ of the present invention is optimized, thereby improving the applicability and accuracy of the SCS-CN model in simulating the surface runoff in the mango planting area. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 The land use distribution map of the mango planting area in the present invention;

[0046] Figure 2 This is the soil distribution map of the mango planting area in the present invention;

[0047] Figure 3 This is a DEM schematic diagram of the mango planting area in the present invention;

[0048] Figure 4 It is the slope map of the mango planting area in the present invention;

[0049] Figure 5 This is one of the NDVI distribution maps of the mango planting area in the present invention;

[0050] Figure 6 This is the second NDVI distribution map of the mango planting area in the present invention;

[0051] Figure 7 This is the monthly rainfall distribution map of the mango planting area in 2021 in the present invention;

[0052] Figure 8 This is the monthly rainfall distribution map of the mango planting area in 2022 in the present invention;

[0053] Fig. 9 This is the monthly CN value distribution map of the mango planting area in 2021 in the present invention;

[0054] Fig.10This is the monthly CN value distribution map of the mango planting area in 2022 in the present invention;

[0055] Fig.11 Schematic diagram of calibration and verification of λ in the present invention;

[0056] Fig.12 This is one of the schematic diagrams of surface runoff characteristics in the mango planting area of ​​the present invention;

[0057] Fig.13 This is the second schematic diagram of surface runoff characteristics in the mango planting area of ​​the present invention;

[0058] Fig.14 This is a schematic diagram of the average monthly runoff depth in the mango planting area of ​​the present invention;

[0059] Fig.15 A schematic diagram of rainfall and surface runoff at a monitoring station in the present invention;

[0060] Fig.16 It is a line graph showing the monthly CN value changes in the mango planting area from 2021 to 2022 in the present invention;

[0061] Fig.17 It is a line chart for the calibration and verification of λ in the present invention. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] See also Figures 1 to 17 As shown: A runoff estimation method for mango orchard planting areas based on SCS-CN model correction is provided, and the estimation method includes the following specific steps:

[0064] S1. Select a mango planting area, obtain spatial data in the mango planting area, use ArcGIS for preprocessing, set the spatial data to projection coordinates, and collect basic element data sets of the mango orchard planting area;

[0065] Among them, a mango planting area was selected as the research area, which was selected based on actual needs;

[0066] During the operation of the SCS-CN model based on ArcGIS, various spatial data need to be projected in a unified coordinate system. The spatial data projection coordinate system is set to WGS_1984_UTM_Zone_48N in combination with the location of the study area. This step is crucial because it ensures data consistency and analysis accuracy.

[0067] WGS_1984_UTM_Zone_48N is a specific spatial reference system (or coordinate reference system, CRS) used in geographic information systems (GIS); this coordinate system is one of the zones of the Universal Transverse Mercator (UTM) projection based on the WGS 84 coordinate system; the following is an explanation of the various parts of this coordinate system:

[0068] WGS 84: WGS 84 is a global geographic coordinate system widely used for GPS positioning. It provides a reference frame that allows geographic location to be accurately determined worldwide;

[0069] UTM: Universal Transverse Mercator is a geographic projection method that divides the earth's surface into multiple regions (or "zones") and creates a separate plane coordinate system for each region. This method reduces the distortion caused by map projections over large geographic areas.

[0070] Zone 48N: This refers to the 48th northern hemisphere projection zone in the UTM projection system. The earth is divided into multiple zones by the UTM system, each covering 6 degrees of longitude; these zones are numbered from west to east, starting with 1, so Zone 48N roughly covers a certain range of longitude in the northern hemisphere;

[0071] Setting the projection coordinates of spatial data to WGS_1984_UTM_Zone_48N means that all data will be transformed into this specific coordinate reference system to ensure the consistency and accuracy of the data, which is especially important in GIS analysis and map making because it allows users to make accurate spatial comparisons and analyses between different layers and datasets.

[0072] The basic element dataset includes but is not limited to mango orchard distribution data, planting area soil data, DEM data, NDVI value, and rainfall data. This basic element dataset reflects the climate, topography, hydrology, soil and other elements of the mango orchard planting area;

[0073] Mango orchard distribution data:

[0074] Since the mango planting basin is small in area, it is difficult to identify the direction of the river in the basin on remote sensing images, so it is impossible to obtain a more accurate basin range. Therefore, the river direction is first drawn on the Google Earth platform to form KML data. The KML data is imported on the ArcGIS 10.2 platform based on the 12.5m precision DEM data. The basin range is generated using the hydrological analysis module (refer to Figure 1 );

[0075] Secondly, for Figure 1Actual land use distribution map of mango growing areas;

[0076] The Sentinel-2 remote sensing image (10 m) of the basin was obtained in the GEE platform. Combined with the survey mechanism, that is, the actual survey, seven land types were determined, including mango planting area, water area, construction land, forest land, rice-vegetable rotation area, and sugarcane planting area. Therefore, the value of M is 7. Sample data was constructed based on the surface characteristics presented by the seven land types, and the land use distribution of the basin was obtained according to the random forest algorithm (refer to Figure 1 ), the overall coefficient reached 0.96, and the Kappa coefficient reached 0.74, which was consistent with the current land distribution characteristics of the basin;

[0077] It should be noted that when using the hydrological analysis module of the ArcGIS 10.2 platform to generate the watershed range of the mango planting area based on the terrain characteristics, the following algorithm or mechanism was used:

[0078] Filling depressions: Since DEM data may contain depressions due to interpolation or real terrain, these areas may lead to unreasonable or wrong water flow directions when simulating surface water flow. Therefore, it is necessary to first fill the depressions in the DEM data to eliminate these data defects and obtain a DEM without depressions. This is an important step to ensure the accuracy of subsequent hydrological analysis.

[0079] Flow Direction Analysis: After the depression is filled, flow direction analysis is performed. ArcGIS usually uses the D8 algorithm to determine the flow direction. This algorithm assumes that there is infinite precipitation and rainwater continuously forms runoff on the surface. Through this algorithm, a flow direction can be calculated for each grid cell to generate a flow direction raster data.

[0080] Flow Accumulation: Flow statistics is the calculation of the cumulative flow of each grid. This cumulative amount represents how many grids upstream have water flows that eventually converge through this grid. The larger the cumulative flow value, the easier it is for the area to form surface runoff, which is crucial for determining the scope of the watershed.

[0081] Watershed extraction: After determining the direction of water flow and flow accumulation, this information can be used to extract the extent of the watershed. Typically, this is done by using the Watershed tool or other related tools that can delineate different watersheds based on flow direction and flow data;

[0082] Data conversion and visualization: Finally, in order to facilitate further analysis and visualization, it may be necessary to convert the raster data into vector data format, which can be achieved through tools such as raster river network vectorization or raster to polygon conversion.

[0083] In summary, the hydrological analysis module of ArcGIS can generate the watershed range of the mango planting area based on high-precision DEM data through the steps of depression filling, flow direction analysis, flow statistics and watershed extraction. These steps make full use of the terrain feature information and provide important decision support for the planning and management of mango planting areas.

[0084] The land use distribution of the basin is shown in Table 1:

[0085]

[0086] Soil data of planting area:

[0087] The soil raster data comes from the HWSD global high-resolution (1km) soil dataset (http: / / westdc.westgis.ac.cn / data / ). The Chinese region is based on the results of the second soil survey in the country by a local soil research institute, using the FAO-90 classification system, which includes the physical and chemical characteristics of various soil types at different depths. It is overlapped with the Chinese soil type dataset (resolution of 30m, data from the "Geographic Remote Sensing Ecological Network Scientific Data Registration and Publishing System (www.gisrs.cn)") to obtain the soil distribution in mango planting areas. The soil type distribution in mango planting areas is as follows: Figure 2 As shown, paddy soil accounts for 13%, neutral purple soil accounts for 23%, red soil accounts for 26%, and red soil accounts for 38%; soil data will be used for soil hydrological grouping calculation in the SCS-CN model, and soil hydrological grouping calculation is performed in SPAW software;

[0088] DEM data:

[0089] The digital elevation model (DEM) was obtained from the NASA Earth Science Data website (https: / / nasadaacs.eos.nasa.gov / ) with a resolution of 12.5 m and was used to calculate the terrain factor and distance factor. The DEM and slope of the mango planting area are as follows: Figure 3 and Figure 4 As shown, the slope distribution is 0 to 56.4°, with an average slope of 13.8°;

[0090] NDVI value:

[0091] Normalized difference vegetation index (NDVI) can be used to reflect changes in vegetation cover. The monthly NDVI data from 2021 to 2022 were obtained from the 250-meter normalized difference vegetation index dataset (2000-2022) in China from the National Tibetan Plateau Science Data Center (http: / / data.tpdc.ac.cn). The NDVI data of the mango-growing area were interpolated using the inverse distance weighted method in ArcGISs to obtain 10-meter resolution data, which was used to calculate the CN value of the SCS-CN model and the vegetation factor of nitrogen and phosphorus into the river. The monthly NDVI values ​​of the mango-growing area from 2021 to 2022 are distributed as follows: Figure 5 and Figure 6 As shown;

[0092] Rainfall data:

[0093] The rainfall data comes from the 1km resolution monthly precipitation dataset of China from 1901 to 2022 from the National Earth System Science Data Center (https: / / www.geodata.cn); after comparison with the actual rainfall station data, it basically conforms to the rainfall distribution characteristics during this study period; Kriging interpolation in ArcGIS obtains 10m resolution data for calculating rainfall erosion factors. In the example, the monthly rainfall distribution in the mango planting area from 2021 to 2022 is as follows Figure 7 and Figure 8 shown.

[0094] S2, construct the SCS-CN model, read the basic element data set of the mango orchard planting area, and call the preconfigured algorithm to determine the CN parameters of the SCS-CN model;

[0095] The SCS-CN model is an empirical model based on two assumptions and the water balance equation;

[0096] The first assumption is that the ratio of actual runoff infiltration to the potential maximum water storage is equal to the ratio of surface runoff depth to the potential maximum runoff depth, as shown in formula (1); the second assumption is that the initial runoff loss is proportional to the potential maximum water storage, as shown in formula (2); the water balance equation is: the total rainfall is equal to the sum of the initial runoff loss, the actual runoff infiltration and the surface runoff depth, as shown in formula (3);

[0097]

[0098] I a =λS(2

[0099] P=I a +F+Q(3)

[0100] The empirical formula of rainfall and runoff can be derived from equations (1) to (3), as shown in equations (4) to (5). When the rainfall is greater than the initial runoff loss, there is surface runoff. Otherwise, the surface runoff is 0:

[0101]

[0102] Q=0,P a (5

[0103] Where: F is the actual runoff infiltration, in mm; S is the potential maximum water storage capacity, that is, the ability of the surface to intercept rainfall, in mm; P is the rainfall, in mm; I a is the initial loss of runoff, in mm; λ is the initial loss coefficient, dimensionless; Q is the surface runoff depth, in mm; S can be calculated as:

[0104]

[0105] In formula (6), CN is dimensionless and its value range is [0, 100]. The CN value can be determined by referring to the "XX National Process Manual". The CN value is affected by multiple factors such as soil moisture, land use type, and vegetation cover. This study assumes that the soil moisture in the mango planting area is in a general state, so the steps for determining the CN value are as follows:

[0106] The first step is to determine the minimum soil infiltration rate. The saturated hydraulic conductivity of the upper and lower layers of each soil type is calculated in SPAW software. The corresponding soil sand content is selected according to the minimum hydraulic conductivity. The minimum soil infiltration rate is calculated according to formula (7) and formula (8). The soil hydrological grouping of the mango planting area is determined according to the minimum infiltration rate. The grouping basis is shown in Table 2.

[0107]

[0108] X = (20 × Y) 1.8 (8)

[0109] Where: SD is the percentage of soil sand content, unit is %; Y is the average soil particle diameter, unit is mm; X is the minimum infiltration rate, the greater the minimum infiltration rate, the lower the surface runoff capacity, unit is mm.h-1;

[0110] Table 2 shows the soil hydrology groups:

[0111]

[0112] ​The second step is to classify the vegetation coverage type. In ArcGIS, the vegetation coverage of the mango planting area is reclassified into three types. When the NDVI value is less than 0.5, the vegetation type is low vegetation coverage, and the reclassification value is 1; when 0.5≤NDVI value≤0.75, the vegetation type is medium vegetation coverage, and the reclassification value is 2; when the NDVI value is 0>0.75, the vegetation type is high vegetation coverage, and the reclassification value is 3;

[0113] Please refer to Table 3 below for details (division of vegetation types in mango planting areas):

[0114]

[0115] Combine the results of the first and second steps to trigger the retrieval mechanism to determine the CN value of the mango growing area;

[0116] Among them, the access mechanism is to consult the "XX National Process Manual";

[0117] Specifically, the surface runoff characteristics in mango planting areas are as follows:

[0118] Characteristics of rainfall and surface runoff: Fig.15 As shown in the figure, the cumulative rainfall observed by the monitoring station from 2021 to 2022 was 1660.73mm and 1421.30mm respectively; the cumulative surface runoff depth was 273.04mm and 253.11mm respectively. As can be seen from the figure, the annual precipitation is mainly concentrated in May to August, and the total cumulative precipitation in this period accounts for 55.6% and 60.8% of 2021 and 2022 respectively. Among them, the maximum cumulative rainfall from 2021 to 2022 occurred in May, which were 344.57mm and 336.10mm respectively; the minimum values ​​occurred in December, which were 14.61mm and 6.40mm respectively. The surface runoff depth and rainfall showed the same trend, with the maximum values ​​occurring in May, which were 106.05mm and 98.821mm respectively; the minimum values ​​were 0.187mm in January in 2021 and 0.204mm in December in 2022. In October 2021, affected by the "La Niña" phenomenon, rainfall and surface runoff depth showed historical abnormal values, 233.28 mm and 34.194 mm respectively;

[0119] Distribution of CN values ​​in mango-growing areas: According to the "XX National Process Manual", the CN value is determined by the comprehensive influence of land use type, vegetation type and soil texture. The calculation results of the CN value in mango-growing areas are shown in Table 5. The soil texture of neutral purple soil is mainly sandy soil, loamy sandy soil, and sandy loam. The minimum infiltration rate is the largest, which is 7.55; the CN value is the smallest, with values ​​of 32, 43 and 57. The soil texture of paddy soil and red soil is mainly loam and silt loam, with moderate minimum infiltration rates of 4.98 and 6.63, respectively; the CN values ​​are 58, 65 and 73. The soil texture of red soil is mainly sandy clay loam, with the smallest minimum infiltration rate of 2.65; the CN values ​​are 72, 76 and 82;

[0120] Table 5 shows (CN value of mango planting area):

[0121]

[0122] The CN value range is generally [0, 100]. The larger the CN value, the smaller the soil moisture infiltration capacity and the more surface runoff. The monthly CN value distribution in the mango planting area from 2021 to 2022 is as follows: Fig.16 As shown. The CN values ​​in mango-growing areas are unevenly distributed in space, and generally show a spatial distribution characteristic of high in the north and south and low in the middle. The CN average value shows an obvious monthly variation characteristic, and the CN average value shows a pattern of first decreasing, then increasing, and then decreasing. The monthly CN average value in 2021 ranged from 59.7 to 64.5, with the minimum value occurring in May and the maximum value occurring in August; the monthly CN average value in 2022 ranged from 59.3 to 63.7, with the minimum value occurring in June and the maximum value occurring in August;

[0123] Calibration and verification of optimized λ: The applicability of the SCS-CN model in mango growing areas was verified using rainfall-runoff data from monitoring stations, such as Fig.16 As shown, the CN value of the monitoring station is 43 and remains unchanged. The value range of the initial loss coefficient λ is generally [0, 0.4]. After many experimental observations, Country A obtained the classic value of λ: 0.2. However, this value has been proven to be applicable to Region A. Due to differences in environmental factors in other regions, it may not be applicable to other parts of the world. Therefore, this study adopts the exhaustive method, taking 2021 as the rate period and 2022 as the verification period, and calculating the simulated runoff depth of different rainfalls in turn with a step size of 0.01, and summarizing the λ values ​​suitable for mango growing areas, see Table 4;

[0124] Depend on Fig.17It can be seen that the Ens coefficients of the rate period and the verification period are both 0.972, and the NRMSE coefficients are 0.233 and 0.226 respectively, and the accuracy reaches a high level; if the value of λ is 0.2, the Ens coefficients of the rate period and the verification period are 0.923 and 0.875 respectively, and the NRMSE coefficients are 0.388 and 0.487 respectively; after λ is optimized, the Ens coefficients of the SCS-CN model in the rate period and the verification period are increased by 5.34% and 11.10% respectively compared with the classic value, and the NRMSE coefficients are increased by 40.05% and 53.62% respectively; it can be seen that the optimized λ value is suitable for mango planting areas, and the optimized SCS-CN model is suitable for surface runoff calculation in mango planting areas.

[0125] S3. Use the exhaustive method to use the annual rainfall-runoff data of the monitoring station in the Nth year as the rate period of the runoff initial loss coefficient of the SCS-CN model, and use the annual rainfall-runoff data of the monitoring station in the N+1th year as the verification period of the runoff initial loss coefficient of the SCS-CN model to obtain the verified SCS-CN model; the value of N is usually 1;

[0126] Among them, the simulated runoff depths of different rainfalls were calculated in turn with a step size of 0.01, based on the regression equation: λ = 0.0007*Pm + 0.071, to obtain the values ​​of λ under different rainfall amounts;

[0127] The above regression equation is a fitting equation, which is the relationship obtained by fitting the monthly rainfall and initial loss coefficient;

[0128] In the formula, Pm represents monthly rainfall, 0.071 represents the conventional coefficient, and λ represents the runoff initial loss coefficient;

[0129] It is concluded that when the rainfall is less than 50 mm, the λ value suitable for mango growing areas is 0.07, when it is 50-12 mm, the λ value is 0.13, when it is 120-200 mm, the λ value is 0.19, when it is 200-300 mm, the λ value is 0.25, when it is >300 mm, the λ value is 0.3. After the correction and optimization of λ, the Ens coefficient of the SCS-CN model in the rate period and the verification period is increased by 5.34% and 11.10% respectively compared with the classic value, and the NRMSE coefficient is increased by 40.05% and 53.62% respectively;

[0130] Compared with the classic SCS-CN model, the optimized λ of the present invention improves the applicability and accuracy of the SCS-CN model in simulating surface runoff in mango planting areas;

[0131] The λ value suitable for mango growing areas is obtained by referring to Table 4 below:

[0132]

[0133] In the above table, R2 is the determination coefficient. A value exceeding 0.9 indicates that the model prediction is very accurate. The P value is the coefficient of variation. P<0.001 indicates that there is a very significant correlation between monthly rainfall and initial loss coefficient.

[0134] S4. Use the validated SCS-CN model to estimate the runoff loss in the mango orchard planting area during the corresponding period;

[0135] The corresponding period includes the whole year or different growing seasons.

[0136] Example:

[0137] A method for estimating runoff in a mango orchard planting area based on SCS-CN model correction includes the following specific steps:

[0138] 1. Determine the unified projection coordinates of spatial data

[0139] During the operation of the SCS-CN model based on ArcGIS, various spatial data need to be projected in a unified coordinate system. The spatial data projection coordinates are set to WGS_1984_UTM_Zone_48N in combination with the location of the study area.

[0140] (II) Selection of mango planting areas

[0141] This case study selected the mango growing area in Guangxi ( Figure 1 -A).

[0142] 3. Land use distribution in mango growing areas

[0143] First, the river course was drawn on the Google Earth platform to form KML data. The KML data was imported based on the 12.5m precision DEM data on the ArcGIS 10.2 platform. The watershed range ( Figure 1 -B), then, the Sentinel-2 remote sensing image (10m) of the basin was obtained in the GEE platform, and combined with the actual survey, seven land types were determined, including mango planting area, water area, construction land, forest land, rice-vegetable rotation area, and sugarcane planting area. Sample data were constructed based on the surface characteristics of the seven land types, and the land use distribution of the basin was obtained according to the random forest algorithm ( Figure 1 -C), the overall coefficient reached 0.96, and the Kappa coefficient reached 0.74, which was consistent with the current land distribution characteristics of the basin.

[0144] (IV) Soil data of mango growing areas

[0145] The soil raster data of mango-growing areas were obtained from the HWSD global high-resolution (1km) soil dataset, and overlapped with the Chinese soil type dataset (resolution of 30m) to obtain the soil distribution of mango-growing areas ( Figure 2 ). The soil data will be used for soil hydrology group calculation in the SCS-CN model, and the soil hydrology group calculation is performed in the SPAW software.

[0146] (V) DEM data of mango growing areas

[0147] The DEM and slope data of the mango growing area with a resolution of 12.5 m were obtained from the NASA Earth Science Data website ( Figure 3 and Figure 4 ).

[0148] (VI) NDVI value in mango growing areas

[0149] The NDVI values ​​were obtained from the 250-meter normalized difference vegetation index dataset for China in the National Tibetan Plateau Data Center, and the values ​​were interpolated using the inverse distance weighted method in ArcGIS to obtain 10-meter resolution NDVI data ( Figure 5 and Figure 6 ).

[0150] (VII) Rainfall data in mango growing areas

[0151] The monthly precipitation data set of China with a resolution of 1 km from 1901 to 2022 from the National Earth System Science Data Center was compared with the actual rainfall station data, and Kriging interpolation was performed in ArcGIS to obtain the precipitation data with a resolution of 10 m ( Figure 7 and Figure 8 ).

[0152] (VIII) CN Values ​​of the SCS-CN Model

[0153] The saturated hydraulic conductivity of the upper and lower layers of each soil type was calculated using SPAW software. The corresponding soil sand content was selected according to the minimum hydraulic conductivity, the minimum soil infiltration rate was calculated, and the soil hydrological grouping of the mango planting area was determined (Table 2). The vegetation cover of the mango planting area was reclassified into three types using ArcGIS and NDVI values ​​(Table 3). Based on the soil hydrological grouping and vegetation cover type of the mango planting area, the CN value of the mango planting area was determined by consulting the XX National Process Manual ( Fig. 9 and Fig.10 ).

[0154] (IX) Determination and verification of λ

[0155] The exhaustive method was used to take 2021 as the rate period and 2022 as the validation period. The simulated runoff depths of different rainfalls were calculated in turn with a step size of 0.01, and the λ values ​​suitable for mango planting areas were summarized (Table 4). After λ was optimized, the Ens coefficients of the SCS-CN model in the rate period and validation period were increased by 5.34% and 11.10% respectively, and the NRMSE coefficients were increased by 40.05% and 53.62% respectively compared with the classic values ​​( Fig.11 ).

[0156] (X) Characteristics of surface runoff in mango planting areas

[0157] The surface runoff characteristics of the mango planting area increase with the cumulative rainfall in time, and show a characteristic of more in the north and less in the south in space ( Fig.12 and Fig.13 ); In 2021, the surface runoff depth in the mango planting area varied from 0 to 189.55 mm, and the total average surface runoff depth was 690.99 mm; the monthly average surface runoff depth was ( Fig.14 ), the average surface runoff depth in January and December was close to 0, which were 0.38mm and 0.18mm respectively, and the surface runoff was 16,400 tons and 7,800 tons respectively; the average surface runoff depth in February, March, September and November was less than 30mm, which were 19.79mm, 18.65mm, 27.84mm and 21.56mm respectively, and the cumulative surface runoff was 852,700 tons, 803,600 tons and 1,199,500 tons respectively. t and 928,900 t; the average surface runoff depths in April, May, June, July, August and October were 42.54 mm, 181.76 mm, 134.81 mm, 58.79 mm, 75.14 mm and 109.55 mm respectively, and the cumulative surface runoff were 1.8329 million t, 7.8313 million t, 5.8084 million t, 2.5330 million t, 3.2375 million t and 4.7201 million t respectively.

[0158] In 2022, the surface runoff depth in the mango planting area varied from 0 to 258.24 mm, with a total average surface runoff depth of 539.15 mm. Among the monthly average surface runoff depths, the average surface runoff depth in December was close to 0, and the surface runoff was 3,900 tons; the average surface runoff depths in January, February, March, April, October and November were less than 30 mm, which were 20.45 mm, 22.32 mm, 17.54 mm, 25.25 mm, 10.50 mm and 9.55 mm respectively, and the cumulative surface runoff was 881,100 tons, 961,700 tons, 755,700 tons, 452,400 tons and 41,000 tons respectively. .15000 tons, of which January and October were 53.81 times and 0.10 times that of the same period in 2021; the average surface runoff depths in May, June, July, August and September were 183.69mm, 127.81mm, 39.14mm, 49.30mm and 33.51mm respectively, and the cumulative surface runoff was 7.9145 million tons, 5.5068 million tons, 1.6864 million tons, 2.1241 million tons, 1.4438 million tons and 4.7201 million tons respectively. The total surface runoff from May to August accounted for 65.20% and 74.18% of that in 2021 and 2022 respectively. It can be seen that May to August is the main period of concentrated surface runoff in mango planting areas;

[0159] Table 6 shows that the surface runoff of different soil types is different under the same rainfall conditions. The unit surface runoff depths of neutral purple soil, paddy soil, red soil, and red soil were 283.84 mm, 769.21 mm, 713.12 mm, and 959.97 mm in 2021, and 214.57 mm, 605.68 mm, 545.41 mm, and 761.41 mm in 2022, respectively. Under different monthly cumulative rainfall conditions from 2021 to 2022, the unit surface runoff depth of red soil is 2.47-18.2 times that of neutral purple soil, 1.15-11.5 times that of paddy soil, and 1.23-23 times that of red soil; the unit surface runoff depth of paddy soil is 2.15-13.56 times that of neutral purple soil, 1-2 times that of red soil; the unit surface runoff depth of red soil is 1.92-11.39 times. It can be seen that under the same rainfall conditions, the surface runoff capacity of red soil> paddy soil ≥ red soil> neutral purple soil.

[0160] Table 6 shows (the monthly unit surface runoff depth of different soil types from 2021 to 2022):

[0161]

[0162]

[0163] In the application, the several formulas involved are all calculated by taking their numerical values ​​after removing the dimensions, and the formulas are established by collecting a large amount of data and performing software simulation to obtain a formula for the most recent actual situation. Some coefficients or weights in the formulas are set by technical personnel in this field according to actual conditions, so they will not be elaborated here.

[0164] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0165] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0166] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A method for estimating runoff in a mango orchard based on SCS-CN model correction, characterized in that: The estimation method includes the following steps: Select a mango planting area, obtain the spatial data in the mango planting area, use ArcGIS for preprocessing, set the spatial data to projection coordinates, and collect the basic element data set of the mango planting area; Build the SCS-CN model, read the basic element dataset of the mango orchard planting area, and call the preconfigured algorithm to determine the CN parameters of the SCS-CN model; The exhaustive method is adopted, the annual rainfall-runoff data of the monitoring station in the Nth year is used as the rate period of the runoff initial loss coefficient of the SCS-CN model, and the annual rainfall-runoff data of the monitoring station in the N+1th year is used as the verification period of the runoff initial loss coefficient of the SCS-CN model to obtain the verified SCS-CN model; The validated SCS-CN model was used to estimate the runoff loss in the mango orchard planting area during the corresponding period.

2. A method for estimating runoff in a mango orchard planting area based on SCS-CN model correction according to claim 1, characterized in that: During the operation of the SCS-CN model based on ArcGIS, various spatial data were uniformly set as projection coordinates, and the spatial data projection coordinates were set to WGS_1984_UTM_Zone_48N in combination with the location of the mango planting area.

3. A method for estimating runoff in a mango orchard planting area based on SCS-CN model correction according to claim 1, characterized in that: The basic element data sets include but are not limited to mango orchard distribution data, planting area soil data, DEM data, NDVI values, and rainfall data.

4. A method for estimating runoff in a mango orchard planting area based on SCS-CN model correction according to claim 3, characterized in that: The process of obtaining mango orchard distribution data is as follows: Firstly, the river direction is drawn on the Google Earth platform to form KML data. The KML data is imported into the ArcGIS 10.2 platform based on the DEM data. The watershed range is generated using the hydrological analysis module. The hydrological analysis module is implemented through the steps of depression filling, flow direction analysis, flow statistics and watershed extraction. Secondly, the Sentinel-2 remote sensing images of the basin were obtained in the GEE platform. Combined with the survey mechanism, M land types were determined. Sample data were constructed according to the surface characteristics presented by the M land types, and the land use distribution of the basin was obtained according to the random forest algorithm.

5. A method for estimating runoff in a mango orchard planting area based on SCS-CN model correction according to claim 3, characterized in that: The process of obtaining soil data in the planting area is as follows: The DEM data of the mango-growing area was obtained from the NASA Earth Science Data Website. The HWSD soil dataset and the soil type dataset were superimposed to obtain the soil property database of the mango-growing area for the soil hydrological grouping calculation required by the SCS-CN model.

6. A method for estimating runoff in a mango orchard planting area based on SCS-CN model correction according to claim 3, characterized in that: DEM data is used to calculate terrain factors and distance factors; NDVI value: Based on the regional normalized vegetation index dataset of the data center, the inverse distance weighted method is used to interpolate the NDVI value in ArcGIS, which is used to calculate the CN value of the SCS-CN model and the vegetation factor of nitrogen and phosphorus into the river; rainfall data: The monthly precipitation dataset is obtained from the data center.

7. A method for estimating runoff in a mango orchard planting area based on SCS-CN model correction according to claim 6, characterized in that: The SCS-CN model is an empirical model based on two assumptions and the water balance equation; The first assumption is that the ratio of actual runoff infiltration to the potential maximum water storage is equal to the ratio of surface runoff depth to the potential maximum runoff depth, see formula (1); the second assumption is that the initial runoff loss is proportional to the potential maximum water storage, see formula (2); the water balance equation is: the total rainfall is equal to the sum of the initial runoff loss, the actual runoff infiltration and the surface runoff depth, see formula (3); I a =λS(2 P=I a +F+Q(3) The empirical formula of rainfall and runoff can be derived from equations (1) to (3), as shown in equations (4) to (5). When the rainfall is greater than the initial runoff loss, there is surface runoff. Otherwise, the surface runoff is 0: Q=0,P<I a (5) Where: F is the actual runoff infiltration; S is the potential maximum water storage capacity, that is, the ability of the surface to intercept rainfall; P is the rainfall; I a is the initial loss of runoff; λ is the initial loss coefficient; Q is the surface runoff depth; S can be calculated as follows: In formula (6), CN is dimensionless and its value range is [0, 100].

8. The method for estimating runoff in a mango orchard planting area based on SCS-CN model correction according to claim 7, characterized in that: The steps for obtaining the value of CN are as follows: The first step is to determine the minimum soil infiltration rate. The saturated hydraulic conductivity of the upper and lower layers of each soil type is calculated in the SPAW software. The corresponding soil sand content is selected according to the minimum hydraulic conductivity. The minimum soil infiltration rate is calculated according to formula (7) and formula (8). The soil hydrological grouping of the mango planting area is determined according to the minimum infiltration rate. X=(20×Y) 1.8 (8) Where: SD is the percentage of soil sand content; Y is the average soil particle diameter; X is the minimum infiltration rate; The second step is to classify the vegetation coverage type. In ArcGIS, the vegetation coverage of the mango planting area is reclassified into three types. When the NDVI value is less than 0.5, the vegetation type is low vegetation coverage, and the reclassification value is 1; when 0.5≤NDVI value≤0.75, the vegetation type is medium vegetation coverage, and the reclassification value is 2; when the NDVI value is 0>0.75, the vegetation type is high vegetation coverage, and the reclassification value is 3; The results of the first and second steps are combined to trigger a review mechanism to determine the CN value of the mango growing area.

9. The method for estimating runoff in a mango orchard planting area based on SCS-CN model correction according to claim 8, characterized in that: When the verified SCS-CN model was obtained, the simulated runoff depths of different rainfalls were calculated in sequence with a step size of 0.01, based on the regression equation: λ = 0.0007*Pm+0.071, to obtain the values ​​of λ under different rainfall amounts; Where Pm represents monthly rainfall, 0.071 represents the conventional coefficient, and λ represents the runoff initial loss coefficient.

Images