A spatial downscaling method and device for regional nitrogen and phosphorus input, and electronic equipment

Abstract

The application discloses a spatial downscaling method and device for regional nitrogen and phosphorus input, an electronic device and a storage medium. The method analyzes nitrogen and phosphorus input, social and economic data and geographic information data at two different resolutions, performs weight evaluation, and calculates high-resolution nitrogen input by using low-resolution nitrogen input data, so as to realize the effect of spatial downscaling. Based on low-resolution nitrogen input data, social and economic data and geographic information data sets, and high-resolution social and economic data and geographic information data sets, the application can predict and evaluate high-resolution nitrogen input, thereby providing a high-precision evaluation method for regional nitrogen and phosphorus management. The application is suitable for regions with social and economic data and geographic information data required for predicting high-resolution nitrogen input data.

Description

Technical Field

[0001] The present application relates to the technical field of socioeconomic data reconstruction, and specifically to a spatial downscaling method and device, and electronic equipment for regional nitrogen and phosphorus inputs, which are suitable for nitrogen and phosphorus input prediction in data-deficient areas. Technical Background

[0002] With the increasing human activities today, various environmental problems frequently occur. How to accurately grasp the regional nutrient input is crucial for environmental pollution control and related decision-making. The nitrogen and phosphorus input in the region mainly comes from agricultural activities and human life, including farming, livestock and poultry farming, domestic pollution sources and atmospheric nitrogen deposition. Studies have shown that excessive nitrogen and phosphorus inputs can cause a series of environmental problems and have become important indicators for regional nitrogen and phosphorus management and prevention and control measures. At present, in order to improve the accuracy of regional nitrogen and phosphorus input assessments, a large amount of detailed socioeconomic data statistics are needed. However, in some data-scarce areas, such high-precision data is difficult to obtain. Therefore, how to use limited and low-precision data to improve the spatial resolution of regional nitrogen and phosphorus inputs has become a key technical issue that needs to be urgently addressed in regional environmental governance. Summary of the Invention

[0003] The purpose of the embodiments of the present application is to provide a spatial downscaling method and device, and electronic equipment for regional nitrogen and phosphorus inputs to solve the technical problem of low spatial resolution of traditional regional nitrogen and phosphorus input assessments.

[0004] According to a first aspect of an embodiment of the present application, a spatial downscaling method for regional nitrogen and phosphorus inputs is provided, comprising:

[0005] Collect socioeconomic data and geographic information data at two different spatial resolutions, and divide them into low-resolution and high-resolution socioeconomic data and geographic information data according to the accuracy of data resolution;

[0006] Calculate low-resolution and high-resolution nitrogen and phosphorus inputs based on the low-resolution and high-resolution socioeconomic data;

[0007] Calculating correlation coefficients between nitrogen and phosphorus inputs and the socioeconomic data and geographic information data based on the high-resolution nitrogen and phosphorus inputs, socioeconomic data, and geographic information data;

[0008] Calculating the weights of high-resolution socioeconomic data and geographic information data based on the correlation coefficient;

[0009] Based on the low-resolution nitrogen and phosphorus inputs, socioeconomic data, geographic information data and the weights, high-resolution nitrogen and phosphorus inputs in the low-resolution area are calculated.

[0010] Optionally, the two types of socioeconomic data and geographic information data of different resolutions are divided into low-resolution and high-resolution socioeconomic data and geographic information data according to the relative resolution, including: low-resolution and high-resolution year-end permanent population, year-end livestock and poultry output and year-end stock, annual total output of various livestock and poultry products, annual total sown area and annual total output of crops, pure amount of nitrogen fertilizer, phosphate fertilizer and compound fertilizer, atmospheric nitrate nitrogen deposition, low-resolution and high-resolution night light data; geographic information data include: low-resolution and high-resolution slope data, vegetation normalized index data, and land use data;

[0011] Optionally, calculating low-resolution and high-resolution nitrogen and phosphorus inputs based on the low-resolution and high-resolution socioeconomic data includes:

[0012] Calculate the low-resolution and high-resolution net food / feed nitrogen and phosphorus inputs based on the low-resolution and high-resolution year-end permanent population, the year-end output and inventory of livestock and poultry, the annual total output of various livestock and poultry products, and the annual total output of crops;

[0013] Calculating nitrogen fixation inputs of low-resolution and high-resolution crops based on the low-resolution and high-resolution crop planting areas;

[0014] Calculating the nitrogen and phosphorus inputs of the low-resolution and high-resolution chemical fertilizers according to the pure amounts of the low-resolution and high-resolution nitrogen fertilizers, phosphorus fertilizers, and compound fertilizers;

[0015] Calculating low-resolution and high-resolution atmospheric nitrogen deposition based on the atmospheric nitrate nitrogen deposition;

[0016] Calculate low-resolution and high-resolution nitrogen and phosphorus inputs based on the low-resolution and high-resolution net food / feed nitrogen and phosphorus inputs, crop nitrogen fixation inputs, fertilizer nitrogen and phosphorus inputs, and atmospheric nitrogen deposition;

[0017] Optionally, calculating the correlation coefficient between nitrogen and phosphorus inputs and the socioeconomic data and geographic information data based on the high-resolution nitrogen and phosphorus inputs, socioeconomic data, and geographic information data includes:

[0018] Calculating the significance level between the high-resolution nitrogen and phosphorus inputs, socioeconomic data, and geographic information data based on the high-resolution nitrogen and phosphorus inputs, socioeconomic data, and geographic information data, and screening out all high-resolution nitrogen and phosphorus inputs, socioeconomic data, and geographic information data with a significance level less than P;

[0019] The correlation coefficients between the high-resolution nitrogen and phosphorus inputs, socioeconomic data, and geographic information data with all significance levels less than P were calculated.

[0020] Optionally, the weights of the socioeconomic data and geographic information data are calculated based on the correlation coefficients. The specific steps are as follows:

[0021] Based on the correlation coefficients, calculating the squares of all correlation coefficients;

[0022] Based on the square of the correlation coefficient, the weight of each socioeconomic data or geographic information data can be expressed as: the proportion of the square of the correlation coefficient between the high-resolution socioeconomic data or geographic information data corresponding to the high-resolution nitrogen and phosphorus input in the sum of the squares of the correlation coefficients of all socioeconomic data and geographic information data corresponding to the high-resolution nitrogen and phosphorus input, calculated using the following formula:

[0023]

[0024] Where w j is the weight of the socio-economic data and geographic information data, r j is the correlation coefficient between nitrogen input and socioeconomic data and geographic information data, and j is the type of socioeconomic data or geographic information data.

[0025] Optionally, high-resolution nitrogen and phosphorus inputs in a low-resolution area are calculated based on the low-resolution nitrogen and phosphorus inputs, socioeconomic data, geographic information data, and weights of the socioeconomic data and geographic information data, in the following specific steps:

[0026] Normalization processing is performed based on the high-resolution socioeconomic data and geographic information data. If the correlation coefficient is a positive number, normalization processing is performed according to the following formula:

[0027]

[0028] If the correlation coefficient is negative, normalization is performed according to the following formula:

[0029]

[0030] Where S i is the normalized value of the i-th sample in the high-resolution socioeconomic data and geographic information data, X, X min 、X max They are respectively the value of the i-th sample of the high-resolution socio-economic data and geographic information data, the minimum value and the maximum value in the total samples.

[0031] The predicted nitrogen and phosphorus inputs are calculated based on the low-resolution socioeconomic data, geographic information data, and weights of the socioeconomic data and geographic information data, using the following formula:

[0032]

[0033] Where EN is the high-resolution nitrogen and phosphorus input in the low-resolution area, TN is the low-resolution nitrogen and phosphorus input, and S ij For the high-resolution socio-economic data or geographic information data, the EN i It should include high-resolution net nitrogen and phosphorus inputs from food / feed, nitrogen fixation inputs from crops, nitrogen and phosphorus inputs from fertilizers, and atmospheric nitrogen deposition. TN should include low-resolution net nitrogen and phosphorus inputs from food / feed, nitrogen fixation inputs from crops, nitrogen and phosphorus inputs from fertilizers, and atmospheric nitrogen deposition. j and J are the type and number of the socioeconomic data and geographic information data, respectively. i and I are the i-th sample and the total number of samples, respectively. w is the weight of the socioeconomic data or geographic information data.

[0034] According to a second aspect of an embodiment of the present application, a spatial downscaling device for regional nitrogen and phosphorus inputs is provided, comprising:

[0035] The acquisition module is used to collect socioeconomic data and geographic information data of two different spatial resolutions, which are divided into low-resolution and high-resolution socioeconomic data and geographic information data according to the accuracy of data resolution;

[0036] a first calculation module for calculating low-resolution and high-resolution nitrogen and phosphorus inputs based on the low-resolution and high-resolution socioeconomic data;

[0037] A second calculation module is used to calculate the correlation coefficient between nitrogen and phosphorus input and the socioeconomic data and geographic information data based on the high-resolution nitrogen and phosphorus input, socioeconomic data and geographic information data;

[0038] A third calculation module is used to calculate the weights of high-resolution socioeconomic data and geographic information data based on the correlation coefficient;

[0039] The fourth calculation module is used to calculate the high-resolution nitrogen and phosphorus inputs in the low-resolution area based on the low-resolution nitrogen and phosphorus inputs, socioeconomic data and geographic information data and the weights.

[0040] According to a third aspect of the embodiments of the present application, there is provided an electronic device, including:

[0041] one or more processors;

[0042] a memory for storing one or more programs;

[0043] When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in the first aspect.

[0044] According to a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, on which computer instructions are stored. When the instructions are executed by a processor, the steps of the method described in the first aspect are implemented.

[0045] The technical solutions provided by the embodiments of the present application may have the following beneficial effects:

[0046] The nitrogen and phosphorus input assessment method of the present application is a main method for regional nitrogen and phosphorus risk assessment, and plays an important role in the nitrogen and phosphorus risk assessment in many regions. The accuracy of regional nitrogen and phosphorus input assessment currently depends mainly on the spatial resolution of the collected socioeconomic data. The traditional method improves the accuracy of regional nitrogen and phosphorus input assessment by collecting high-resolution socioeconomic data. Although it is the most reliable assessment method, it also has the problems of low efficiency and large data demand. With the development of satellite remote sensing technology, high-resolution geographic information data and socioeconomic data can be obtained more conveniently, providing a feasible solution to this problem. The present invention proposes a spatial downscaling method for regional nitrogen and phosphorus input, which is achieved by collecting socioeconomic data and geographic information data of two different resolutions and realizing the conversion of lower resolution to higher resolution nitrogen and phosphorus input through their weight analysis. This method avoids complicated statistical work to a certain extent, and also improves the accuracy of regional nitrogen and phosphorus input, providing more accurate guidance for regional nitrogen and phosphorus governance.

[0047] The present invention mainly relies on socioeconomic data and geographic information data of two different resolutions to realize the conversion of nitrogen and phosphorus inputs from lower resolution to higher resolution, achieving the effect of spatial downscaling and providing a more convenient method for regional nitrogen and phosphorus input assessment.

[0048] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0050] Figure 1 The present invention is a flowchart showing spatial downscaling of regional nitrogen and phosphorus inputs according to an exemplary embodiment.

[0051] Figure 2 The figure shows the correlation coefficient and significance level of the nitrogen input with the socioeconomic data and geographic information data according to an exemplary embodiment.

[0052] Figure 3 FIG. 1 is a diagram showing a comparison of nitrogen input between predicted nitrogen input and actual nitrogen input according to an exemplary embodiment.

[0053] Figure 4 The present invention is a block diagram of a device for predicting spatial downscaling of regional nitrogen and phosphorus inputs according to an exemplary embodiment. DETAILED DESCRIPTION

[0054] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0055] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. As used in this application and the appended claims, the singular forms "a," "an," "the," and "the" are intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0056] Figure 1 is a flowchart showing a spatial downscaling of regional nitrogen and phosphorus inputs according to an exemplary embodiment. Figure 1 As shown, the method may include the following steps:

[0057] S1: Collect socioeconomic data and geographic information data at two different spatial resolutions, and divide them into low-resolution and high-resolution socioeconomic data and geographic information data according to the accuracy of the data resolution;

[0058] Specifically, the low-resolution and high-resolution socioeconomic data and geographic information data include: low-resolution and high-resolution year-end permanent population, year-end livestock and poultry output and year-end inventory, annual total output of various livestock and poultry products, annual total sown area and annual total output of crops, pure amount of nitrogen fertilizer, phosphate fertilizer and compound fertilizer, atmospheric nitrate nitrogen deposition, low-resolution and high-resolution night light data; geographic information data include: low-resolution and high-resolution slope data, vegetation normalization index data, cultivated land area data, and construction land area data;

[0059] The study area selected in this embodiment is Guangdong Province, China, where the low-resolution data is low-resolution (Guangdong Province) and the high-resolution data is high-resolution (all counties in Guangdong Province, a total of 123). The time scale of the socioeconomic data collected in this embodiment is 2015, and the data includes: low-resolution and high-resolution year-end permanent population; low-resolution and high-resolution pure amount of nitrogen fertilizer and compound fertilizer; low-resolution and high-resolution year-end inventory and year-end output of various livestock and poultry, mainly including: pigs, cattle, sheep, and poultry; low-resolution and high-resolution output of various livestock and poultry products, including: pork, beef, mutton, poultry, eggs, and milk; low-resolution and high-resolution output of various agricultural products, including: rice, peanuts, vegetables, potatoes, and fruits; low-resolution and high-resolution annual total sown area of ​​crops, including: rice and peanuts; night light data is 500m resolution night light data obtained by spatial statistical analysis of low-resolution and high-resolution night light data; geographic information data includes low-resolution and high-resolution slope data, vegetation normalization index data, cultivated land area data, and construction land area data;

[0060] Alternatively, since the evaluation of nitrogen input and phosphorus input is similar, only nitrogen input is selected as one of the examples in this case.

[0061] S2: Calculating low-resolution and high-resolution nitrogen and phosphorus inputs based on the low-resolution and high-resolution socioeconomic data; this step includes the following sub-steps:

[0062] S21: Calculate the low-resolution and high-resolution net food / feed nitrogen and phosphorus inputs based on the low-resolution and high-resolution year-end permanent population, the year-end livestock and poultry output and inventory, the annual total output of various livestock and poultry products, and the annual total output of crops:

[0063] Specifically, the low-resolution and high-resolution net nitrogen and phosphorus inputs of food / feed (N im ) is calculated as follows:

[0064] N im =HC+AC-(GP+AP)×(1-η)

[0065] Where HC, AC, GP, and AP are the high-resolution nitrogen intake of humans, nitrogen intake of livestock and poultry, total nitrogen production of crop products, and total nitrogen production of livestock and poultry products, respectively, all in kg; η represents the percentage of loss of crop products and livestock and poultry products during production, processing, and transportation, which is 10% in this case;

[0066] The low-resolution net food / feed nitrogen input (N im_T ) is the low-resolution net food / feed nitrogen input (N im) and, the calculation formula is as follows:

[0067] N im_T =∑(N im )

[0068] The high-resolution human nitrogen intake (HC) is the product of the high-resolution year-end permanent population and the per capita nitrogen intake, and is calculated as follows:

[0069] HC=HQ×HN

[0070] Where HQ is the high-resolution year-end permanent population, in units of persons; HN is the per capita nitrogen intake, in units of kg person -1 , here the value is 3.91kg·person -1 ;

[0071] The low-resolution human nitrogen intake (HC T ) is the sum of the high-resolution human nitrogen intake (HC), and is calculated as follows:

[0072] HC T =∑(HC)

[0073] The high-resolution livestock and poultry nitrogen intake (AC) is the product of the high-resolution livestock and poultry feeding number and the livestock and poultry nitrogen intake, and the calculation formula is as follows:

[0074]

[0075] Where, Inventory and Sales are the high-resolution year-end inventory and year-end sales of livestock and poultry, respectively, in units of individual units; Cycles is the livestock and poultry breeding cycle, which can be expressed as 365 divided by the number of livestock and poultry breeding days, dimensionless; AL is the number of livestock and poultry raised, in units of individual units; AN is the nitrogen intake of livestock and poultry, in units of kg / animal unit. -1 ; k is the type of livestock and poultry; K is the number of livestock and poultry species. In this case, a total of four livestock and poultry species are selected, namely pigs, cattle, sheep, and poultry. The values ​​of Cycles and AN in this embodiment are shown in Table 1:

[0076] Table 1 - Feeding cycle and nitrogen intake of various livestock and poultry

[0077] livestock and poultry species Breeding cycle <![CDATA[Nitrogen intake (kg·bird -1 )]]> pig 199 8.70 ox 365 51.30 sheep 365 5.97 poultry 158 0.255

[0078] The low-resolution nitrogen intake of livestock and poultry (AC T ) is the sum of the high-resolution livestock and poultry nitrogen intake (AC), and the calculation formula is as follows:

[0079] AC T =∑(AC)

[0080] The total nitrogen yield (GP) of the high-resolution crop product is the product of the annual total yield of the high-resolution crop and its nitrogen content, and is calculated as follows:

[0081]

[0082] Wherein, GW is the annual total yield of the high-resolution crops, in kg; GN is the nitrogen content of the crop product; i is the type of crop; I is the number of crop species. This case considers five agricultural products, namely rice, peanuts, vegetables, potatoes, and fruits. The values ​​of GN in this embodiment are shown in Table 2:

[0083] Table 2 - Nitrogen content of various agricultural products

[0084] paddy soybeans peanut vegetable potatoes fruit 1.24％ 5.60％ 1.92％ 0.51％ 0.33％ 0.09％

[0085] The total nitrogen yield (GNP) of the low-resolution crop products T ) is the sum of the total nitrogen production (GP) of the high-resolution crop products, and is calculated as follows:

[0086] GP T =∑(GP)

[0087] The total nitrogen production (AP) of the high-resolution livestock and poultry products is the product of the total annual production of the high-resolution livestock and poultry products and their nitrogen content, and the calculation formula is as follows:

[0088]

[0089] Wherein, AW is the total annual output of various livestock and poultry products with high resolution, in kg; APN is the nitrogen content of livestock and poultry products; m is the type of livestock and poultry products; and M is the number of livestock and poultry products. This case considers pork, beef, mutton, poultry, eggs, and milk, a total of 6 products. The APN values ​​in this embodiment are shown in Table 3:

[0090] Table 3 - Nitrogen content of various livestock and poultry products

[0091] pork beef mutton poultry milk eggs 2.42％ 3.20％ 2.96％ 2.81％ 0.53％ 2.31％

[0092] The low-resolution total nitrogen production (AP T ) is the sum of the total nitrogen production (AP) of the high-resolution livestock and poultry products, and is calculated as follows:

[0093] AP T =∑(AP)

[0094] S22: Calculating nitrogen fixation inputs of low-resolution and high-resolution crops based on the annual total sown areas of the low-resolution and high-resolution crops, this step includes:

[0095] Specifically, the high-resolution nitrogen fixation input (N fix ) is the product of the high-resolution crop sowing area and the crop nitrogen fixation rate, and is calculated as follows:

[0096]

[0097] Where S is the sown area of ​​crops, unit: km 2 ; N is the nitrogen fixation rate of crops, unit: kg km -2 ; j represents the type of crop, J represents the number of crop species. Here we consider two crops: rice and peanuts, so the value of N is 3000kg / km 2 , 8000kg / km 2 .

[0098] The low-resolution nitrogen fixation input of crops (N fix_T ) is the high-resolution nitrogen fixation input of crops (N fix ), the calculation formula is as follows:

[0099] N fix_T =∑(N fix )

[0100] S23: Calculating the nitrogen and phosphorus inputs of the low-resolution and high-resolution chemical fertilizers according to the low-resolution and high-resolution pure amounts of the nitrogen fertilizer, phosphorus fertilizer, and compound fertilizer;

[0101] Specifically, the high-resolution nitrogen input of fertilizer (N chem ) can be expressed as the sum of nitrogen input from nitrogen fertilizer and compound fertilizer, and the calculation formula is as follows:

[0102] N chem =NF+Cf×r

[0103] Wherein, NF and CF are the pure amounts of nitrogen fertilizer and compound fertilizer, both in kg; r represents the nitrogen content in compound fertilizer, which accounts for 32.64%.

[0104] The low-resolution nitrogen input of fertilizer (N chem_T ) is the nitrogen input of the high-resolution fertilizer (N chem ), the calculation formula is as follows:

[0105] N chem_T =∑(N chem )

[0106] Since atmospheric nitrogen deposition data can be obtained from raster data with higher accuracy, in order to eliminate its influence, the spatial downscaling is not analyzed in this case;

[0107] S24: Calculating low-resolution and high-resolution nitrogen and phosphorus inputs based on the low-resolution and high-resolution net food / feed nitrogen and phosphorus inputs, crop nitrogen fixation inputs, and fertilizer nitrogen and phosphorus inputs;

[0108] Specifically, the high-resolution nitrogen input (NANI) calculation formula is as follows:

[0109] NANI=N im +N fix +N chem

[0110] The low-resolution nitrogen input (NANI _T ) is the sum of the high-resolution nitrogen input (NANI), and is calculated as follows:

[0111] NANI T =∑(NANI)

[0112] The calculation results in this section are the nitrogen inputs for Guangdong Province (low resolution) and Guangdong Province (high resolution) in China in 2015. The results are shown in Table 4:

[0113] Table 4 - Nitrogen input to Guangdong Province, China (low resolution) and counties in Guangdong Province (high resolution) in 2015 (unit: tons)

[0114]

[0115] S3: Calculating correlation coefficients between nitrogen and phosphorus inputs, socioeconomic data, and geographic information data based on the high-resolution nitrogen and phosphorus inputs, socioeconomic data, and geographic information data. The specific steps are as follows:

[0116] S31: Divide the high-resolution nitrogen input, socioeconomic data, and geographic information data into a training set and a test set, as follows:

[0117] Specifically, based on high-resolution nitrogen input data, socioeconomic data, and geographic information data, we merged the columns into a matrix, with each row representing a sample. We randomly sampled 70% of the total rows without replacement to form the nitrogen input training set, designated traind, and the remaining rows to form the nitrogen input test set, designated testd. The results for traind and testd are shown in Tables 5 and 6, respectively.

[0118] Table 5 - Training set of nitrogen input - traind

[0119]

[0120]

[0121] To better characterize the impact of nighttime light data (light) and the Normalized Difference Vegetation Index (NDVI), we multiply the nighttime light data by the low-resolution or high-resolution area to represent the light intensity, and multiply the NDVI data by the low-resolution or high-resolution area to represent the vegetation cover.

[0122] Table 6 - Test set of nitrogen input - testd

[0123]

[0124]

[0125] S32: Calculating the significance level between the high-resolution nitrogen and phosphorus inputs and the socio-economic data and geographic information data based on the high-resolution nitrogen and phosphorus inputs and the socio-economic data and geographic information data, and screening out all high-resolution nitrogen and phosphorus inputs and the socio-economic data and geographic information data with a significance level less than P; and calculating the correlation coefficient between the high-resolution nitrogen and phosphorus inputs and the socio-economic data and geographic information data based on all the high-resolution nitrogen and phosphorus inputs and the socio-economic data and geographic information data with a significance level less than P, the specific steps are as follows:

[0126] Specifically, each column of traind is considered as a variable, and the correlation coefficient matrix between these variables is calculated. The specific formula is as follows:

[0127]

[0128] Cor=(ρ α,β )

[0129] Where, ρ α,β is the variable Y α and Y β Pearson correlation coefficient, Cov(Y α ,Y β ) is the variable Y α and Y β The covariance of and is the variable Y α and Y β The standard deviation of y i,α and y i,β are the variables Y in the i-th sample α and Y β The value of Cor is ρ α,β The correlation coefficient matrix composed of , I represents the total number of samples in traind, and α and β represent different variables.

[0130] Specifically, for each pair of variables (Y α , Y β) to perform a significance test on the correlation coefficient, assuming a significance level of α = 0.05 (corresponding to a confidence level of 1-α = 0.95), and calculate the corresponding p-value matrix res1 and confidence interval matrix. The specific steps are as follows:

[0131]

[0132] p α,β =2(1-T(|t α,β |,I-2))

[0133] Where t is the statistic of the correlation coefficient, T(|t α,β |,I-2)) is the cumulative distribution function value of the t distribution with I-2 degrees of freedom, p α,β is the variable Y α and Y β The significance level between .

[0134] In order to facilitate the display, the calculation results of this part are presented in a visual way, such as Figure 2 As shown in the figure, ns means not significant, *** means the significance level is <0.005, ** means the significance level is <0.01, * means the significance level is less than 0.05, and the numbers in the figure represent the correlation coefficient between the horizontal axis and the vertical axis.

[0135] S4: Calculating weights of the socioeconomic data and geographic information data based on the correlation coefficient;

[0136] Specifically, according to the result of S3, based on the square of the correlation coefficient, the weight of each type of socioeconomic data or geographic information data can be expressed as: the proportion of the square of the correlation coefficient between the high-resolution socioeconomic data or geographic information data corresponding to the high-resolution nitrogen and phosphorus input in the sum of the squares of the correlation coefficients of all socioeconomic data and geographic information data corresponding to the high-resolution nitrogen and phosphorus input, calculated using the following formula:

[0137]

[0138] Where w j is the weight of the socio-economic data and geographic information data, r j is the correlation coefficient between a certain nitrogen input and socioeconomic data or geographic information data in the nitrogen input training set, j is the type of socioeconomic data or geographic information data. The calculation results of this part are shown in Table 7.

[0139] Table 7 - Weights (w) of socioeconomic data and geographic information data in the training set for nitrogen input

[0140] HC AP AC Nchem GP Nfix Crop - 0.430 0.470 0.436 0.389 0.279 light 0.500 - - - - 0.050 Developed 0.500 - - - - 0.029 Slope - 0.320 0.267 0.301 0.309 0.329 NDVI - 0.250 0.263 0.263 0.302 0.313

[0141] S5: Calculate high-resolution nitrogen and phosphorus inputs in the low-resolution area based on the low-resolution nitrogen and phosphorus inputs, socioeconomic data, geographic information data, and the weights.

[0142] Specifically, the predicted nitrogen input is calculated based on the socioeconomic data, geographic information data, and nitrogen and phosphorus input of the administrative unit of Guangdong Province, the weight (w) of the socioeconomic data and geographic information number in the training set of nitrogen and phosphorus input, and the socioeconomic data and geographic information data in the test set of nitrogen input. The calculation steps are as follows:

[0143] Normalization processing is performed based on the high-resolution socioeconomic data and geographic information data. If the correlation coefficient is a positive number, normalization processing is performed according to the following formula:

[0144]

[0145] If the correlation coefficient is negative, normalization is performed according to the following formula:

[0146]

[0147] Where S i is the normalized value of the ith high-resolution value in the socioeconomic data and geographic information data, X, X min 、X max They are respectively the i-th high-resolution value, the minimum value, and the maximum value among all high-resolution values ​​of the socio-economic data and geographic information data.

[0148] The predicted nitrogen and phosphorus inputs are calculated based on the low-resolution socioeconomic data and geographic information data and the weights of the socioeconomic data and geographic information in the training set of nitrogen and phosphorus inputs, using the following formula:

[0149]

[0150] Where EN is the predicted nitrogen and phosphorus input, TN is the low-resolution nitrogen and phosphorus input, and S ij Enter the socioeconomic data or geographic information data in the test set for the nitrogen and phosphorus, where EN i It should include high-resolution net nitrogen and phosphorus inputs from food / feed, nitrogen fixation inputs from crops, nitrogen and phosphorus inputs from fertilizers, and atmospheric nitrogen deposition. TN should include low-resolution net nitrogen and phosphorus inputs from food / feed, nitrogen fixation inputs from crops, nitrogen and phosphorus inputs from fertilizers, and atmospheric nitrogen deposition. S is the low-resolution socioeconomic data or geographic information data, j and J are the type and number of the socioeconomic data and geographic information data, respectively. i and I are the i-th high-resolution and total high-resolution data, respectively. w is the weight of the socioeconomic data or geographic information data in the test set of nitrogen and phosphorus inputs.

[0151] The predicted nitrogen input is compared with the nitrogen input in the test set to verify the effect. The specific steps are as follows:

[0152]

[0153] Where R 2 is the coefficient of determination, ON i refers to the i-th high-resolution observation in the test set, is the average nitrogen input in the test set, EN i is the predicted nitrogen input, is the average of the predicted nitrogen inputs.

[0154] Draw a fitting curve based on the predicted nitrogen input and the nitrogen input in the test set to check the effect. The results of this section are as follows Figure 3 shown.

[0155] This method can use basic economic data and geographic information data to predict actual nitrogen input. test With N testd Fitted R 2 The value is 0.68, which makes it possible to evaluate high-precision nitrogen input data using low-precision nitrogen input.

[0156] Corresponding to the aforementioned embodiments of the method for spatial downscaling of regional nitrogen and phosphorus inputs, the present application also provides embodiments of a device for spatial downscaling of regional nitrogen and phosphorus inputs.

[0157] Figure 4 1 is a block diagram of a spatial downscaling device for regional nitrogen and phosphorus inputs according to an exemplary embodiment. Figure 4 , the device comprises:

[0158] Acquisition module 1 is used to collect socioeconomic data and geographic information data of two different spatial resolutions, and divide them into low-resolution and high-resolution socioeconomic data and geographic information data according to the accuracy of data resolution;

[0159] A first calculation module 2 is used to calculate low-resolution and high-resolution nitrogen and phosphorus inputs based on the low-resolution and high-resolution socioeconomic data;

[0160] A second calculation module 3 is used to calculate the correlation coefficient between nitrogen and phosphorus input and the socioeconomic data and geographic information data based on the high-resolution nitrogen and phosphorus input, socioeconomic data and geographic information data;

[0161] A third calculation module 4 is used to calculate the weights of high-resolution socioeconomic data and geographic information data based on the correlation coefficient;

[0162] The fourth calculation module 5 is used to calculate the high-resolution nitrogen and phosphorus inputs in the low-resolution area based on the low-resolution nitrogen and phosphorus inputs, socioeconomic data, geographic information data and the weights.

[0163] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0164] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present application scheme. A person of ordinary skill in the art can understand and implement it without paying any creative work.

[0165] Accordingly, the present application also provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the spatial downscaling method of regional nitrogen and phosphorus inputs as described above.

[0166] Accordingly, the present application also provides a computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the spatial downscaling method for regional nitrogen and phosphorus inputs as described above.

[0167] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the contents disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered merely as exemplary, and the true scope and spirit of the present application are indicated by the claims.

[0168] It should be understood that the present application is not limited to the exact structures described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for spatial downscaling of regional nitrogen and phosphorus inputs, characterized in that: include: Collect socioeconomic data and geographic information data at two different spatial resolutions, and divide them into low-resolution and high-resolution socioeconomic data and geographic information data according to the accuracy of data resolution; The socioeconomic data include: low-resolution and high-resolution year-end permanent population, year-end livestock and poultry output and inventory, annual total output of various livestock and poultry products, annual total sown area and annual total output of crops, pure nitrogen fertilizer, phosphate fertilizer and compound fertilizer usage, atmospheric nitrate nitrogen deposition, and nighttime light data; the geographic information data include: low-resolution and high-resolution slope data, vegetation normalization index data, cultivated land area data, and construction land area data; Calculate low-resolution and high-resolution nitrogen and phosphorus inputs based on the low-resolution and high-resolution socioeconomic data; Calculating correlation coefficients between nitrogen and phosphorus inputs and the socioeconomic data and geographic information data based on the high-resolution nitrogen and phosphorus inputs, socioeconomic data, and geographic information data; Calculating the weights of high-resolution socioeconomic data and geographic information data respectively according to the correlation coefficients; Calculating high-resolution nitrogen and phosphorus inputs in the low-resolution area based on the low-resolution nitrogen and phosphorus inputs, socioeconomic data, geographic information data, and the weights; Calculating the correlation coefficients between nitrogen and phosphorus inputs and the socioeconomic data and geographic information data based on the high-resolution nitrogen and phosphorus inputs, socioeconomic data, and geographic information data includes: Calculating the significance level between the high-resolution nitrogen and phosphorus inputs, socioeconomic data, and geographic information data based on the high-resolution nitrogen and phosphorus inputs, socioeconomic data, and geographic information data, and screening out all high-resolution nitrogen and phosphorus inputs, socioeconomic data, and geographic information data with a significance level less than P; Calculate the correlation coefficient between the high-resolution nitrogen and phosphorus inputs, socioeconomic data, and geographic information data with all the significance levels less than P; The high-resolution nitrogen and phosphorus inputs in the low-resolution area are calculated based on the low-resolution nitrogen and phosphorus inputs, socioeconomic data, geographic information data, and the weights, including: Normalization processing is performed based on the high-resolution socioeconomic data and geographic information data. If the correlation coefficient is a positive number, normalization processing is performed according to the following formula: ； If the correlation coefficient is negative, normalization is performed according to the following formula: ； Where S i is the normalized value of the i-th sample in the socioeconomic data and geographic information data, X, X min 、X max are the value of the i-th sample of the socio-economic data and geographic information data, the minimum value and the maximum value in the total samples respectively; Based on the low-resolution socioeconomic data, geographic information data, and the weights, high-resolution nitrogen and phosphorus inputs in the low-resolution area are calculated using the following formula: ； Where, is the high-resolution nitrogen and phosphorus input in the low-resolution area, TN is the low-resolution nitrogen and phosphorus input, S ij The high-resolution socioeconomic data or geographic information data, is the high-resolution net nitrogen and phosphorus input of food / feed, nitrogen fixation input of crops, nitrogen and phosphorus input of fertilizers, and atmospheric nitrogen deposition. TN includes the low-resolution net nitrogen and phosphorus input of food / feed, nitrogen fixation input of crops, nitrogen and phosphorus input of fertilizers, and atmospheric nitrogen deposition. j and J are the type and number of types of the socioeconomic data or geographic information data, respectively. i and I are the i-th sample and the total number of samples, respectively. w j is the weight of the socioeconomic data or geographic information data.

2. The method according to claim 1, characterized in that Calculate low-resolution and high-resolution nitrogen and phosphorus inputs based on the low-resolution and high-resolution socioeconomic data, including: Calculate the low-resolution and high-resolution net food / feed nitrogen and phosphorus inputs based on the low-resolution and high-resolution year-end permanent population, the year-end output and inventory of livestock and poultry, the annual total output of various livestock and poultry products, and the annual total output of crops; Calculate nitrogen fixation inputs to low-resolution and high-resolution crops based on the total annual sown areas of low-resolution and high-resolution crops; Calculate the nitrogen and phosphorus inputs of low-resolution and high-resolution chemical fertilizers based on the pure amounts of low-resolution and high-resolution nitrogen fertilizers, phosphorus fertilizers and compound fertilizers; Calculate low-resolution and high-resolution atmospheric nitrogen deposition based on low-resolution and high-resolution atmospheric nitrate nitrogen deposition; Low-resolution and high-resolution nitrogen and phosphorus inputs are calculated based on low-resolution and high-resolution net food / feed nitrogen and phosphorus inputs, crop nitrogen fixation inputs, fertilizer nitrogen and phosphorus inputs, and atmospheric nitrogen deposition.

3. The method according to claim 1, characterized in that The weights of the socio-economic data and geographic information data are calculated based on the correlation coefficients, including: Based on the correlation coefficients, calculating the squares of all correlation coefficients; According to the square of the correlation coefficient, the weight of each type of socioeconomic data and geographic information data is calculated as follows: ； Where w j is the weight of the socio-economic data or geographic information data, r j is the correlation coefficient between nitrogen input and socioeconomic data or geographic information data, and j is the type of socioeconomic data or geographic information data.

4. A spatial downscaling device for regional nitrogen and phosphorus inputs, characterized in that: include: The acquisition module is used to collect socioeconomic data and geographic information data of two different spatial resolutions, which are divided into low-resolution and high-resolution socioeconomic data and geographic information data according to the accuracy of data resolution; The socioeconomic data include: low-resolution and high-resolution year-end permanent population, year-end livestock and poultry output and inventory, annual total output of various livestock and poultry products, annual total sown area and annual total output of crops, pure nitrogen fertilizer, phosphate fertilizer and compound fertilizer usage, atmospheric nitrate nitrogen deposition, and nighttime light data; the geographic information data include: low-resolution and high-resolution slope data, vegetation normalization index data, cultivated land area data, and construction land area data; a first calculation module for calculating low-resolution and high-resolution nitrogen and phosphorus inputs based on the low-resolution and high-resolution socioeconomic data; A second calculation module is used to calculate the correlation coefficient between nitrogen and phosphorus input and the socioeconomic data and geographic information data based on the high-resolution nitrogen and phosphorus input, socioeconomic data and geographic information data; A third calculation module is used to calculate the weights of high-resolution socioeconomic data and geographic information data based on the correlation coefficient; a fourth calculation module, configured to calculate high-resolution nitrogen and phosphorus inputs in the low-resolution area based on the low-resolution nitrogen and phosphorus inputs, the socioeconomic data, the geographic information data, and the weights; Calculating the correlation coefficients between nitrogen and phosphorus inputs and the socioeconomic data and geographic information data based on the high-resolution nitrogen and phosphorus inputs, socioeconomic data, and geographic information data includes: Calculating the significance level between the high-resolution nitrogen and phosphorus inputs, socioeconomic data, and geographic information data based on the high-resolution nitrogen and phosphorus inputs, socioeconomic data, and geographic information data, and screening out all high-resolution nitrogen and phosphorus inputs, socioeconomic data, and geographic information data with a significance level less than P; Calculate the correlation coefficient between the high-resolution nitrogen and phosphorus inputs, socioeconomic data, and geographic information data with all the significance levels less than P; The high-resolution nitrogen and phosphorus inputs in the low-resolution area are calculated based on the low-resolution nitrogen and phosphorus inputs, socioeconomic data, geographic information data, and the weights, including: Normalization processing is performed based on the high-resolution socioeconomic data and geographic information data. If the correlation coefficient is a positive number, normalization processing is performed according to the following formula: ； If the correlation coefficient is negative, normalization is performed according to the following formula: ； Where S i is the normalized value of the i-th sample in the socioeconomic data and geographic information data, X, X min 、X max are the value of the i-th sample of the socio-economic data and geographic information data, the minimum value and the maximum value in the total samples respectively; Based on the low-resolution socioeconomic data, geographic information data, and the weights, high-resolution nitrogen and phosphorus inputs in the low-resolution area are calculated using the following formula: ； Where, is the high-resolution nitrogen and phosphorus input in the low-resolution area, TN is the low-resolution nitrogen and phosphorus input, S ij The high-resolution socioeconomic data or geographic information data, is the high-resolution net nitrogen and phosphorus input of food / feed, nitrogen fixation input of crops, nitrogen and phosphorus input of fertilizers, and atmospheric nitrogen deposition. TN includes the low-resolution net nitrogen and phosphorus input of food / feed, nitrogen fixation input of crops, nitrogen and phosphorus input of fertilizers, and atmospheric nitrogen deposition. j and J are the type and number of types of the socioeconomic data or geographic information data, respectively. i and I are the i-th sample and the total number of samples, respectively. w j is the weight of the socioeconomic data or geographic information data.

5. An electronic device, characterized in that: include: one or more processors; a memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 3.

6. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the instruction is executed by a processor, the steps of the method according to any one of claims 1 to 3 are implemented.

Images