CN115266612A - Southern hilly area cultivated land available phosphorus drawing method based on high-resolution environment variable - Google Patents
Southern hilly area cultivated land available phosphorus drawing method based on high-resolution environment variable Download PDFInfo
- Publication number
- CN115266612A CN115266612A CN202210889387.5A CN202210889387A CN115266612A CN 115266612 A CN115266612 A CN 115266612A CN 202210889387 A CN202210889387 A CN 202210889387A CN 115266612 A CN115266612 A CN 115266612A
- Authority
- CN
- China
- Prior art keywords
- soil
- variables
- data
- variable
- remote sensing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- OAICVXFJPJFONN-UHFFFAOYSA-N Phosphorus Chemical compound [P] OAICVXFJPJFONN-UHFFFAOYSA-N 0.000 title claims abstract description 86
- 229910052698 phosphorus Inorganic materials 0.000 title claims abstract description 86
- 239000011574 phosphorus Substances 0.000 title claims abstract description 86
- 238000000034 method Methods 0.000 title claims abstract description 54
- 239000002689 soil Substances 0.000 claims abstract description 151
- 238000005527 soil sampling Methods 0.000 claims abstract description 47
- 230000007613 environmental effect Effects 0.000 claims abstract description 35
- 238000012216 screening Methods 0.000 claims abstract description 23
- 238000011160 research Methods 0.000 claims abstract description 21
- 238000005070 sampling Methods 0.000 claims abstract description 17
- 238000011156 evaluation Methods 0.000 claims abstract description 16
- 238000007781 pre-processing Methods 0.000 claims abstract description 15
- 238000010276 construction Methods 0.000 claims abstract description 4
- 239000002344 surface layer Substances 0.000 claims description 32
- 238000013507 mapping Methods 0.000 claims description 28
- 238000007637 random forest analysis Methods 0.000 claims description 19
- 238000012937 correction Methods 0.000 claims description 17
- 238000001556 precipitation Methods 0.000 claims description 15
- 230000001419 dependent effect Effects 0.000 claims description 12
- 239000000126 substance Substances 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 9
- 238000012795 verification Methods 0.000 claims description 9
- 230000008030 elimination Effects 0.000 claims description 8
- 238000003379 elimination reaction Methods 0.000 claims description 8
- 238000010220 Pearson correlation analysis Methods 0.000 claims description 6
- UIIMBOGNXHQVGW-UHFFFAOYSA-M Sodium bicarbonate Chemical compound [Na+].OC([O-])=O UIIMBOGNXHQVGW-UHFFFAOYSA-M 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 230000003595 spectral effect Effects 0.000 claims description 4
- 238000012935 Averaging Methods 0.000 claims description 3
- 238000012952 Resampling Methods 0.000 claims description 3
- 238000001632 acidimetric titration Methods 0.000 claims description 3
- 229910052787 antimony Inorganic materials 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000002310 reflectometry Methods 0.000 claims description 3
- 229910000030 sodium bicarbonate Inorganic materials 0.000 claims description 3
- 235000017557 sodium bicarbonate Nutrition 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000004737 colorimetric analysis Methods 0.000 claims description 2
- 238000012360 testing method Methods 0.000 claims description 2
- 238000013316 zoning Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 description 7
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 4
- 230000006872 improvement Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 229910052757 nitrogen Inorganic materials 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 239000010410 layer Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000005416 organic matter Substances 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/27—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands using photo-electric detection ; circuits for computing concentration
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/75—Systems in which material is subjected to a chemical reaction, the progress or the result of the reaction being investigated
- G01N21/77—Systems in which material is subjected to a chemical reaction, the progress or the result of the reaction being investigated by observing the effect on a chemical indicator
- G01N21/78—Systems in which material is subjected to a chemical reaction, the progress or the result of the reaction being investigated by observing the effect on a chemical indicator producing a change of colour
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/24—Earth materials
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/02—Instruments for indicating weather conditions by measuring two or more variables, e.g. humidity, pressure, temperature, cloud cover or wind speed
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Health & Medical Sciences (AREA)
- Analytical Chemistry (AREA)
- Environmental & Geological Engineering (AREA)
- Pathology (AREA)
- Immunology (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Biochemistry (AREA)
- Environmental Sciences (AREA)
- Geology (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Ecology (AREA)
- Biodiversity & Conservation Biology (AREA)
- Atmospheric Sciences (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Plasma & Fusion (AREA)
- Remote Sensing (AREA)
- Food Science & Technology (AREA)
- Medicinal Chemistry (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of soil available phosphorus research, in particular to a method for drawing available phosphorus in farmland soil in southern hilly areas based on high-resolution environmental variables, which mainly comprises the following steps of S1: acquiring and establishing soil sampling point data, and S2: acquisition of environmental data, S3: data preprocessing, S4: extracting and screening environmental variables, and S5: model construction and precision evaluation and S6: predicting the effective phosphorus content (0-15 cm) of surface soil of all cultivated lands in the region to be detected in the same period as the sampling time of the soil sampling points: the method for drawing the available phosphorus in the farmland soil in the southern hilly areas based on the high-resolution environmental variables can obviously improve the drawing precision of the available phosphorus in the farmland soil in the southern hilly areas, quickly realize the prediction and spatial distribution of the available phosphorus content in the farmland soil in large areas in the southern hilly areas, and is suitable for further popularization and application.
Description
Technical Field
The invention relates to the technical field of research on soil available phosphorus, in particular to a method for drawing soil available phosphorus in cultivated land in southern hilly areas based on high-resolution environmental variables.
Background
The traditional soil available phosphorus content analysis and drawing needs to collect a large amount of soil sampling points, so that time and labor are wasted, the cost is high, the ecological environment risk is increased, and the drawing precision facing a large area is low. The satellite remote sensing image can reflect soil surface layer information and is used for digital soil mapping research, a plurality of satellite remote sensing images can be obtained freely, remote sensing data space coverage has continuity, the satellite remote sensing image can be used as a secondary data source to improve mapping efficiency of soil attributes of sparse soil sampling points, and the satellite remote sensing image and other environment variables are combined to become an important means for large-area digital soil mapping of the sparse soil sampling points. However, the spatial resolution of the free remote sensing satellite data which can be obtained and used before is low, so that the final digital soil mapping precision is not high enough. With the appearance of new remote sensing data such as a Sentinel-2 satellite which is higher in spatial resolution and easy to acquire, compared with common Landsat data, the Sentinel-2 data has higher spatial resolution and 3 red-edge bands, and the red-edge bands have obvious advantages in soil properties such as organic matter and total nitrogen prediction. Relevant researches prove that the prediction accuracy of the phosphorus in the soil can be effectively improved by adding remote sensing data. Therefore, the introduction of the Sentinel-2 data can improve the accuracy of soil phosphorus mapping.
In addition, the conventional digital soil drawing method is mainly oriented to the cultivated land in northern flat areas, the area of a single field is large, the field is in a standard mode, the spectral characteristics and the terrain of the cultivated land are uniform, and therefore the effect of the conventional digital soil effective phosphorus drawing method is good. However, southern cultivated land is mostly in hilly areas, the relief of the land is large, the cultivated land is broken, the area of a single field is small, and the precision of the traditional digital drawing is extremely poor. Therefore, the current commonly used digital soil mapping method is not suitable for southern hilly areas. Many scholars have significantly improved the accuracy of digital soil carbon and nitrogen mapping by using remote sensing variables or terrain variables with higher accuracy, i.e. higher spatial resolution.
However, the improvement of the phosphorus drawing precision of the digital soil by using the remote sensing variable or the terrain variable with higher spatial resolution is still blank, so that the improvement of the effective phosphorus drawing precision of the digital soil is expected by using the remote sensing variable or the terrain variable with higher spatial resolution.
Disclosure of Invention
In order to solve the problems, the invention provides a method for drawing available phosphorus in the soil of the cultivated land in the southern hilly area based on a high-resolution environment variable, so as to improve the drawing accuracy of the available phosphorus in the soil on the surface (0-15 cm) of the cultivated land in the southern hilly area.
In order to solve the technical problems, the invention adopts the technical scheme that:
a soil available phosphorus mapping method for cultivated land in southern hilly areas based on high-resolution environmental variables comprises the following steps:
s1, acquiring soil sampling point data and building a library: acquiring physical and chemical attribute data of surface soil sampling points of cultivated land in a region to be detected, and establishing a soil sampling point database in ArcGIS10.2 software;
s2, acquiring environmental data: acquiring environmental data of the area in synchronization with the soil sampling point, wherein the environmental data comprises medium and high resolution remote sensing variables, medium and high resolution topographic data and meteorological data;
s3, data preprocessing: uniformly projecting the remote sensing data, the topographic data and the meteorological data to a coordinate system of Western-style land 1980, checking the consistency of the environmental data and the attribute space matching of soil sampling points, carrying out space attribute connection after error is avoided, and obtaining the surface soil of the cultivated land of the region to be detected by using a vector cutting function in ArcGIS10.2 software through an administrative region drawing of the region to be detected;
s4, extracting and screening environmental variables: the selected forecasting variables comprise remote sensing variables, terrain variables, meteorological variables and soil pH values; in order to simplify model input, an auxiliary prediction factor is optimized before modeling, a variable with obvious correlation with soil available phosphorus is reserved through Pearson correlation analysis, and an auxiliary variable which finally participates in modeling is reserved through a backward elimination method;
s5, model construction and precision evaluation: taking the soil available phosphorus content of the surface layer of the cultivated land of the preset number in the area to be tested as a dependent variable, taking the screened medium and high resolution remote sensing variable, the screened topographic variable, the meteorological variable and the soil pH value as independent variables, respectively executing random forest model operation on a Python platform based on the medium and high resolution environmental variable, automatically screening a random forest model with the best prediction precision and analyzing the importance of the random forest model, and respectively obtaining an expected model of the soil available phosphorus content of the surface layer of the cultivated land of the area to be tested based on the medium and high resolution environmental variable and the sampling time of the soil sample point;
s6, predicting the content of available phosphorus in surface soil (0-15 cm) of all cultivated lands in the region to be tested in the same period as the soil sampling time: according to an optimal model constructed by the environment variable with high resolution (10 meters) and the input environment variable with high precision (10 meters), executing a random forest algorithm in Python software, predicting the effective phosphorus content (0-15 cm) of all the farmland surface soil in the region to be detected at the same time, and obtaining the effective phosphorus spatial distribution map (0-15 cm) of all the farmland surface soil in the region to be detected at the time.
Further, S1 specifically includes the following steps:
s11: researching physical and chemical attribute data of soil sample points on the surface layer of a preset number of cultivated lands in the area to be tested, wherein the physical and chemical attribute data are derived from cultivated land quality detection and evaluation sample point data of the national agricultural rural area, and the data comprise soil available phosphorus, soil pH value, geographic coordinates and elevation;
s12: according to the geographic coordinates and physical and chemical property data of the soil sampling points, a soil sampling point database is established in ArcGIIS10.2 software, and a soil sampling point space distribution map under projection of a France 1980 coordinate system is obtained.
Further, the pH value of the soil in the S11 is measured by an acidimetry method, and the available phosphorus in the soil is measured by a sodium bicarbonate leaching-molybdenum-antimony anti-colorimetric method.
Further, the topographic data in S2 comprises a DEM with high resolution (12.5 meters) and a DEM with spatial resolution of 30 meters, which is commonly used for digital soil mapping;
the remote sensing data comprise a 4-scene Sentinel-2A remote sensing image with high resolution (10 meters) and a Landsat-8OLI image with spatial resolution of 30 meters, which is commonly used for digital soil mapping;
the meteorological data are raster data of 1km multiplied by 1km of historical monthly average precipitation and monthly average air temperature of the region to be measured.
Further, S3 specifically includes the following steps:
s31, remote sensing data preprocessing: the remote sensing data preprocessing comprises radiometric calibration, atmospheric correction, geometric fine correction and image mosaic; for 4-scene Sentinel-2A remote sensing images in a research area, respectively carrying out radiometric calibration on each scene image in SNAP software, then carrying out atmospheric correction, converting DN value of the remote sensing image into surface reflectivity, then carrying out image mosaic splicing, then carrying out geometric fine correction on the remote sensing image by utilizing a fine-corrected SPOT5 reference image in ENVI5.3 software, and cutting according to an administrative zoning map of an area to be detected; the Landsat-8OLI remote sensing image is a geometrically refined image, atmospheric correction is carried out in ENVI5.3 software, the geometrically refined error of all the remote sensing images is controlled within 1 pixel, and the geometrically refined error is uniformly projected to a 1980 coordinate system of Western Ann;
s32, preprocessing of terrain variables: firstly, checking DEM data with the spatial resolution of 12.5m and 30m without errors, uniformly converting the DEM data into a coordinate system of Western Ann 1980 through projection transformation in ArcGIS10.2 software, and producing the DEM with the spatial resolution of 12.5m into DEM data with the spatial resolution of 10m by using a nearest neighbor method;
s33, preprocessing meteorological variables: uniformly projecting meteorological data to a 1980 coordinate system of Western Ann by using arcgis10.2 software;
s34, checking the consistency of attribute space matching of the remote sensing data, the topographic data, the meteorological data and the soil sampling point data in ArcGIS10.2 software, and establishing geographical connection between the spatial data and the soil attribute after checking without errors.
Further, the backward elimination method in S4 specifically includes:
and screening the variables according to the increase and decrease of the root mean square error after each variable is sequentially excluded from the model during modeling, keeping the variables when the RMSE is increased, and otherwise, eliminating the variables.
Further, S4 specifically includes the following steps:
s41, extracting remote sensing variables: respectively extracting remote sensing variables from Sentinel-2A and Landsat-8OLI in ArcGIS10.2 software to calculate 46 spectral indexes for digital soil mapping;
s42, extracting terrain variables: extracting 8 terrain variables commonly used for digital soil mapping from the resampled DEM with the resolution of 10 meters and the resampled DEM with the resolution of 30 meters respectively by ArcGISI 10.2 software;
s43, extracting meteorological variables: calculating annual average precipitation and monthly average air temperature data in ArcGIS10.2 software, calculating the average value of annual average precipitation and annual average temperature in a research area by using a grid calculator to obtain annual average temperature and annual average precipitation data of the research area for many years, and resampling the annual average precipitation and annual average temperature data to grid data with the spatial resolution of 10 meters and 30 meters by adopting a nearest neighbor method;
s44, acquiring a soil pH spatial distribution map: interpolating the pH value of the soil in the research area by using a kriging interpolation method in ArcGISI 10.2 software to obtain a spatial distribution map of the pH value;
s45, carrying out geographical association on the soil sampling points and the remote sensing variables, the terrain variables, the meteorological variables and the soil pH data in ArcGIS10.2 software to obtain all environment variables corresponding to the soil sampling points in the research area;
s46, screening of environment variables participating in modeling: performing Pearson correlation analysis on a Python platform, screening variables with significant correlation with soil available phosphorus, screening the variables according to increase and decrease of Root Mean Square Error (RMSE) after each variable is sequentially excluded from a model during modeling through a backward elimination method, reserving the variables when the RMSE is increased, and otherwise, reserving auxiliary variables which finally participate in modeling, and finally obtaining a sample set which participates in modeling.
Further, S5 specifically includes the following steps:
s51, constructing a training sample set and a verification sample set: randomly dividing the sample set into a training set and a verification set according to a proportion, wherein samples of the training set are used for modeling, and samples of the verification set are used for testing the prediction accuracy of the model;
s52, determining dependent variables and independent variables of the model: taking the soil available phosphorus content of the surface layer of the cultivated land of the soil sampling point of the region to be detected as a dependent variable, taking a screened high-resolution (10 m) remote sensing variable related to significance, a high-resolution (10 m) terrain variable, a meteorological variable and a soil pH value as independent variables, executing a random forest algorithm on a Python platform, and automatically screening an expected model of the soil available phosphorus content of the surface layer of the cultivated land of the region to be detected based on a high-precision environment variable and the same period as the sampling time of the soil sampling point;
s53, model prediction accuracy evaluation: by R2(coefficient of determination), MAE (mean absolute error) and RMSE (root mean square error) to screen the best predictive model, R2The closer to 1,MAE and RMSE are, the smaller the prediction model precision is; the precision evaluation formula (1-3) is as follows:
wherein n is the number of sampling points, Oi、PiThe measured value and the predicted value of the sampling point i,
is the average value of measured values;
s54, taking the soil available phosphorus content of the surface layer (0-15 cm) of the cultivated land of the soil sampling point as a dependent variable, taking the remote sensing variable with medium resolution (30 meters), the terrain variable with medium resolution (30 meters), the meteorological variable and the soil pH value which are related to the screening significance as independent variables, and repeatedly executing the steps of S51 to obtain a common optimal random forest prediction model of the soil (0-15 cm) available phosphorus content of the surface layer of the cultivated land based on the environment variable with medium resolution and the sampling time of the soil sampling point;
s55, obtaining a relative importance value of the variable in the process of predicting the content of the available phosphorus in the soil by averaging through a multi-iteration RF model;
the relative importance of the variables modeled based on the environment variables with the resolution of 10m shows that the relative importance scores of the meteorological variables, the terrain variables, the remote sensing variables and the soil pH value are 30.64%, 30.38%, 22.87% and 16.11% in sequence; according to the importance degree of a single variable, the importance degree of the single variable is sequentially annual average temperature, pH value, topographic humidity index, DEM value, annual average rainfall, enhanced vegetation index, first principal component and red edge wave band 6;
based on the variable relative importance display of the environment variable with the spatial resolution of 30 meters, the relative importance scores of the meteorological variable, the terrain variable, the remote sensing variable and the soil pH value are 25.86%, 32.44%, 21.31% and 20.39% in sequence; according to the importance degree, the vegetation index is the elevation (DEM), the pH value, the annual average rainfall, the enhanced vegetation index, B5 and the first main component in turn from large to small.
The invention has the beneficial effects that:
the method for drawing the available phosphorus in the farmland soil in the southern hilly areas based on the high-resolution environmental variables can obviously improve the drawing precision of the available phosphorus in the farmland soil in the southern hilly areas, quickly realize the prediction and spatial distribution of the available phosphorus content in the farmland soil in large areas in the southern hilly areas, and is suitable for further popularization and application.
Drawings
FIG. 1 is a distribution diagram of sample points of surface soil (0-15 cm) of cultivated land;
FIG. 2 is a spatial distribution diagram of the available phosphorus content of soil based on a research area;
FIG. 3 is a schematic flow chart of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the embodiments of the present invention, and it is obvious that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, which can be obtained by a person skilled in the art without inventive step based on the embodiments of the present invention, are within the scope of protection of the present invention.
Referring to fig. 3, in the present embodiment, effective phosphorus mapping of cultivated land soil is performed with respect to cultivated land of Ou City built in Fujian province. The method comprises the following specific steps:
s1, acquiring soil sampling point data and building a database. Acquiring physicochemical attribute data of soil sample points on the surface layer (0-15 cm) of cultivated land of Ou City built in Fujian province in 2017, and establishing a soil sample point database in ArcGIS10.2 software. The method specifically comprises the following steps:
s11, physicochemical attribute data of surface soil (0-20 cm) sampling points of 97 cultivated lands in the research area are derived from the data of the quality detection and evaluation sampling points of the cultivated lands at the end of 2017 years in the Ministry of agricultural and rural areas, and comprise soil available phosphorus, soil pH value, geographic coordinates, elevation and the like. Wherein, the pH value of the soil is measured by an acidimetry method, and the available phosphorus of the soil is measured by a sodium bicarbonate leaching-molybdenum-antimony colorimetric-resistance method. And eliminating 1 sampling point affected by cloud layers on the satellite remote sensing image in the same period, and totally adopting 96 soil sampling points.
S12, establishing a soil sampling point database in ArcGIS10.2 software according to the geographic coordinates, the effective phosphorus content, the pH value and other physical and chemical properties of 96 soil sampling points to obtain a soil sampling point space distribution map under the projection of a 1980 coordinate system of Xian, which is shown in figure 1.
And S2, acquiring environmental data. And acquiring environmental data of the area in synchronization with the soil sampling point, wherein the environmental data comprises medium and high resolution remote sensing variables, medium and high resolution topographic data and meteorological data. The topographic data comprise a high-resolution (12.5 meters) DEM in Ou City of Fujian province in 2017 and a DEM with a spatial resolution of 30m, which is commonly used for digital soil mapping. The remote sensing data are a high-resolution (10 m) 4-scene Sentinel-2A remote sensing image synchronous with the sampling time of the soil sampling point and a Landsat-8OLI image with the spatial resolution of 30m, which is commonly used for digital soil mapping. The meteorological data is 1km multiplied by 1km raster data of the average precipitation (1-12 months) and the average temperature (1-12 months) in the month of 1970-2000.
And S3, preprocessing data. Uniformly projecting the remote sensing data, the terrain data and the meteorological data to a coordinate system of Xian 1980, checking the consistency of spatial matching of all environment data and soil sample point attributes, carrying out spatial attribute connection after error is avoided, and obtaining the soil on the surface layer (0-15 cm) of the cultivated land of the Ou city by utilizing a vector cutting function in ArcGIIS10.2 software through an administrative region drawing of the Ou city. The method specifically comprises the following steps:
and S31, preprocessing the remote sensing data. Mainly comprises radiometric calibration, atmospheric correction, geometric fine correction and image mosaic. For 4-scene Sentinel-2A remote sensing images in a research area, radiation calibration is carried out on each scene of image respectively in SNAP software, then atmospheric correction is carried out, DN values of the remote sensing images are converted into earth surface reflectivity, then image mosaic splicing is carried out, then geometric fine correction is carried out on the remote sensing images by utilizing the fine corrected SPOT5 reference images in ENVI5.3 software, and cutting is carried out according to an administrative region drawing of Ou of Fujian province. The Landsat-8OLI remote sensing image is a geometrically refined image, and atmospheric correction is carried out in ENVI5.3 software. All the errors of the geometric fine correction of the remote sensing images are controlled within 1 pixel, and the geometric fine correction errors are uniformly projected to a 1980 coordinate system of Western An.
And S32, preprocessing the terrain variable. First, after checking the DEM data with the spatial resolution of 12.5 meters and 30 meters without errors, in ArcGIS10.2 software, the DEM data with the spatial resolution of 10 meters is produced by uniformly converting the DEM data into a France 1980 coordinate system through projection transformation and utilizing a nearest neighbor method, wherein the DEM data with the spatial resolution of 12.5 meters is generated.
S33, preprocessing meteorological variables: meteorological data was projected uniformly to the 1980 coordinates system for west ann using arcgis10.2 software.
And S34, checking the attribute space matching consistency of the remote sensing data, the topographic data, the meteorological data and the soil sampling point data in ArcGIS10.2 software, and establishing geographical connection between the spatial data and the soil attribute after checking without errors.
And S4, extracting and screening the environmental variables. The predicted variables selected in the study include remote sensing variables, terrain variables, meteorological variables and soil pH values. To further simplify the model input, the auxiliary predictor needs to be optimized before modeling. Variables with obvious correlation with soil effective phosphorus are reserved through Pearson correlation analysis, further through a backward elimination method, the variables are screened according to increase and decrease of Root Mean Square Error (RMSE) after each variable is sequentially eliminated from a model during modeling, the variables are reserved when the RMSE is increased, and auxiliary variables which finally participate in modeling are reserved after RMSE is eliminated. The method specifically comprises the following steps:
and S41, extracting the remote sensing variable. Remote sensing variables were extracted from Sentinel-2A and Landsat-8OLI, respectively, in ArcGIS10.2 software to calculate 46 spectral indices for digital soil mapping.
And S42, extracting the terrain variable. The 8 terrain variables commonly used for digital soil mapping were extracted from the resampled 10 meter resolution DEM and 30 meter DEM respectively in the arcgis10.2 software.
And S43, extracting meteorological variables. And calculating the average precipitation and average temperature data of the multiple years and months in ArcGISI 10.2 software, and calculating the average precipitation and average temperature of the multiple years and months in the research area by using a grid calculator to obtain the annual average temperature and annual average precipitation data of the multiple years in the research area. And finally, resampling the annual average precipitation and annual average temperature data to grid data with the spatial resolution of 10 meters and 30 meters by adopting a nearest neighbor method.
And S44, acquiring a soil pH spatial distribution map. And (3) interpolating the soil pH value of the research area by using a kriging interpolation method in ArcGISI 10.2 software to obtain a spatial distribution map of the soil pH value.
And S45, carrying out geographical association on the soil sampling points and the spatial data such as the remote sensing variables, the terrain variables, the meteorological variables, the soil pH and the like in ArcGIS10.2 software to obtain all environment variables corresponding to the soil sampling points in the research area.
And S46, participating in screening of the modeling environment variables. Performing Pearson correlation analysis on a Python platform, screening variables with significant correlation with soil available phosphorus, further screening the variables by a backward elimination method according to increase and decrease of Root Mean Square Error (RMSE) after each variable is sequentially eliminated from a model during modeling, reserving the variables when the RMSE is increased, and otherwise, reserving auxiliary variables finally participating in modeling, and finally obtaining a sample set participating in modeling, wherein the auxiliary variables participating in modeling are shown in Table 1.
TABLE 1 soil available phosphorus modeling auxiliary variables
And S5, model construction and precision evaluation. Taking the effective phosphorus content of the soil on the surface layers (0-15 cm) of 96 cultivated lands in the Ou-built market as dependent variables, taking the selected medium-high resolution remote sensing variables, terrain variables, meteorological variables and soil pH values as independent variables, respectively executing random forest model operation on a Python platform based on the medium-high resolution environment variables, automatically screening a random forest model with optimal prediction precision and analyzing the importance of the random forest model to obtain expected effective phosphorus content models of the Ou-built cultivated land surface layers (0-15 cm) in 2017 based on the medium-high resolution environment variables; and the prediction accuracy of the model is evaluated. The method specifically comprises the following steps:
and S51, constructing a training sample set and a verification sample set. And (3) performing the following steps on the sample set according to the ratio of 9:1, randomly dividing a sample set into a training set and a verification set, wherein the samples of the training set are used for modeling, and the samples of the verification set are used for verifying the prediction accuracy of the model.
And S52, determining dependent variables and independent variables of the model. The method comprises the steps of taking the effective phosphorus content of soil on the surface layers (0-15 cm) of 96 cultivated lands in the Ou-built city as dependent variables, taking the previously screened high-resolution (10 m) remote sensing variables, high-resolution (10 m) terrain variables, meteorological variables and soil pH values related to significance as independent variables, executing a random forest algorithm on a Python platform, and automatically screening an optimal random forest prediction model of the effective phosphorus content of the soil on the surface layers (0-15 cm) of the Ou-built city cultivated lands in 2017 based on high-precision environment variables.
And S53, evaluating the model prediction accuracy. By R2(coefficient of determination), MAE (mean absolute error) and RMSE (root mean square error) to screen the best predictive model, R2Closer to 1, less MAE and RMSE indicate higher accuracy of the prediction model. The precision evaluation formula (1-3) is as follows:
wherein n is the number of sampling points, Oi、PiThe measured value and the predicted value of the sampling point i,
are the average values of the measured values.
Prediction precision index R for predicting effective phosphorus content of soil on surface layer (0-15 cm) of cultivated land in Ou city based on high-precision (10 m) environment variable20.59, MAE 19.04 mg.kg-1RMSE error of 25.26 mg/kg-1。
And S54, taking the effective phosphorus content of the soil on the surface layers (0-15 cm) of 96 cultivated lands in the Ou-built city as a dependent variable, taking the screened remote sensing variable with medium resolution (30 meters), the terrain variable with medium resolution (30 meters), the meteorological variable and the soil pH value which are obviously related as independent variables, and repeatedly executing the same steps of S51 to obtain the common optimal random forest prediction model of the effective phosphorus content of the soil on the surface layers (0-15 cm) of the Ou-built city cultivated lands in 2017 based on the environment variable with medium resolution.
Prediction precision index R of effective phosphorus content of soil on surface layer (0-15 cm) of Ou city cultivated land based on common environment variable with medium resolution (30 meters)20.42, MAE error of 21.84mg kg-1RMSE error of 30.18 mg/kg-1。
S55, obtaining the relative importance value of the variable in the process of predicting the content of the available phosphorus in the soil by averaging through a multi-iteration RF model. The relative importance of the variables modeled based on the environment variables with a resolution of 10m showed that the relative importance scores of the meteorological, topographical, remote sensed and soil pH values were 30.64%, 30.38%, 22.87% and 16.11% in that order. According to the importance degree of a single variable, the annual average temperature, the pH value, the topographic humidity index, the DEM value, the annual average rainfall, the enhanced vegetation index, the first main component and the red edge wave band 6 are arranged in sequence from large to small.
The relative importance of variables based on environmental variables with spatial resolution of 30 meters showed that the relative importance scores of meteorological variables, terrain variables, remote sensing variables, and soil pH were 25.86%, 32.44%, 21.31%, and 20.39% in order. According to the importance degree, the vegetation index is the elevation (DEM), the pH value, the annual average rainfall, the enhanced vegetation index, B5 and the first main component in turn from large to small.
And S6, evaluating the effect of the method. Based on the environmental variable with the spatial resolution of 30 meters, the modeling of the effective phosphorus of the soil on the surface layer (0-15 cm) of the cultivated land of Ou city built in 2017 is predicted as a reference standard, and the effective phosphorus is respectively measured from R2The three indexes of MAE and RMSE are used for carrying out effect evaluation on the accuracy of predicting the effective phosphorus content of soil on the surface layer (0-15 cm) of cultivated land of Ou city built in 2017 based on the high-resolution (10 m) environment variable. The result shows that the precision of the combination of the 10-meter remote sensing data and the 12.5-meter terrain data with other environment variables is remarkably improved compared with the precision of the combination of the common 30-meter remote sensing data and the 30-meter terrain data with other variables, the modeling precision R2 is improved by 40.5% (from 0.42 to 0.59), and the MAE error is reduced by 12.8% (from 21.84 mg.kg)-1Reduced to 19.04 mg/kg-1) The RMSE error is reduced by 16.3% (from 30.18 mg. Kg)-1Reduced to 25.26 mg/kg-1). The method comprises the following specific steps:
and S61, constructing a method improvement effect index. Based on the prediction of the modeling of the effective phosphorus in the soil on the surface layer (0-15 cm) of the cultivated land of Ou city built in 2017 by using the environmental variable with the spatial resolution of 30 meters as reference, the effective phosphorus in the soil is respectively predicted from R2The three indexes of MAE and RMSE are used for carrying out effect evaluation on the accuracy of predicting the effective phosphorus content of soil on the surface layer (0-15 cm) of cultivated land of Ou city built in 2017 based on the high-resolution (10 m) environment variable. The formula of the method improvement effect P is as follows:
P＝(X10m-X30m)/X30m*100％
Wherein P represents R2Improved ratio or reduced error ratio of MAE and RMSE, X10 meters represents the precision of predicting the available phosphorus content of soil on the surface layer (0-15 cm) of cultivated land in southern hilly area based on high-resolution (10 m) environmental variables, and X30 meters represents 30 meters based on conventional useThe method can predict the precision of the effective phosphorus content of the soil on the surface layer (0-15 cm) of the cultivated land in southern hilly areas by the environmental variables.
S62, calculating R2The effect is improved. According to the formula, the modeling precision R is calculated in the excel table240.5% (from 0.42 to 0.59).
And S63, calculating the error reduction ratio. The reduction ratios of the MAE error and RMSE error were calculated in excel tables according to the above formula. MAE error is reduced by 12.8% (from 21.84 mg.kg)-1Reduced to 19.04 mg/kg-1) The RMSE error is reduced by 16.3 percent (from 30.18 mg. Kg)-1Reduced to 25.26 mg/kg-1)。
S64, the result shows that in the aspect of three evaluation indexes, the prediction accuracy of the combination of the remote sensing data based on 10 meters and the terrain data based on 12.5 meters and other environment variables is remarkably improved compared with the common prediction accuracy of the combination of the remote sensing data based on 30 meters and the terrain data based on 30 meters and other variables.
TABLE 2 improved evaluation results of the method
And S7, predicting the effective phosphorus content of surface soil (0-15 cm) of all cultivated lands in Ou city built in 2017. According to an optimal model constructed by the environment variable with high resolution (10 meters) and the input environment variable with high precision (10 meters), a random forest algorithm is executed in Python software, the effective phosphorus content of all farmland surface soil (0-15 cm) in the city built in 2017 is predicted, and the effective phosphorus spatial distribution map of all farmland surface soil (0-15 cm) in the city built in 2017 is obtained and is shown in figure 2.
The above examples are only intended to illustrate the technical solution of the present invention, and not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (8)
1. A method for drawing available phosphorus in cultivated land soil in southern hilly areas based on high-resolution environmental variables is characterized by comprising the following steps:
s1, acquiring soil sampling point data and building a library: acquiring physical and chemical attribute data of soil sampling points on the surface layer of the cultivated land in the area to be detected, and establishing a soil sampling point database in ArcGIS10.2 software;
s2, acquiring environmental data: acquiring environmental data of the area in synchronization with the soil sampling point, wherein the environmental data comprises medium and high resolution remote sensing variables, medium and high resolution topographic data and meteorological data;
s3, data preprocessing: uniformly projecting the remote sensing data, the topographic data and the meteorological data to a coordinate system of Western Ann 1980, checking the consistency of the environmental data and the attribute space matching of soil sampling points, performing space attribute connection after no error, and obtaining the surface soil of the cultivated land of the region to be tested by using a vector cutting function in ArcGIS10.2 software through an administrative zoning map of the region to be tested;
s4, extracting and screening environmental variables: the selected prediction variables comprise remote sensing variables, terrain variables, meteorological variables and soil pH values; in order to simplify model input, an auxiliary prediction factor is optimized before modeling, a variable with obvious correlation with soil available phosphorus is reserved through Pearson correlation analysis, and an auxiliary variable which finally participates in modeling is reserved through a backward elimination method;
s5, model construction and precision evaluation: taking the soil available phosphorus content of the surface layer of the cultivated land of the preset number in the area to be tested as a dependent variable, taking the screened medium and high resolution remote sensing variable, the screened topographic variable, the meteorological variable and the soil pH value as independent variables, respectively executing random forest model operation on a Python platform based on the medium and high resolution environmental variable, automatically screening a random forest model with the best prediction precision and analyzing the importance of the random forest model, and respectively obtaining an expected model of the soil available phosphorus content of the surface layer of the cultivated land of the area to be tested based on the medium and high resolution environmental variable and the sampling time of the soil sample point;
s6, predicting the content of available phosphorus in surface soil (0-15 cm) of all cultivated lands in the region to be tested in the same period as the soil sampling time: according to an optimal model constructed by the environment variable with high resolution (10 meters) and the input environment variable with high precision (10 meters), executing a random forest algorithm in Python software, predicting the effective phosphorus content (0-15 cm) of all the farmland surface soil in the region to be detected at the same time, and obtaining the effective phosphorus spatial distribution map (0-15 cm) of all the farmland surface soil in the region to be detected at the time.
2. The method for mapping available phosphorus in soil of cultivated land in southern hilly area based on high resolution environment variable as claimed in claim 1, wherein S1 comprises the following steps:
s11: researching physical and chemical attribute data of soil sample points on the surface layer of a preset number of cultivated lands in the area to be tested, wherein the physical and chemical attribute data are derived from cultivated land quality detection and evaluation sample point data of the national agricultural rural area, and the data comprise soil available phosphorus, soil pH value, geographic coordinates and elevation;
s12: according to the geographic coordinates and physical and chemical property data of the soil sampling points, a soil sampling point database is established in ArcGIIS10.2 software, and a soil sampling point space distribution map under projection of a France 1980 coordinate system is obtained.
3. The method for mapping available phosphorus in soil of farmland in southern hilly areas based on high resolution environmental variables of claim 2, wherein the pH of the soil in S11 is measured by acidimetry and the available phosphorus in the soil is measured by sodium bicarbonate leaching-molybdenum-antimony colorimetry.
4. The method for mapping available phosphorus in soil of farmland in southern hilly areas based on high resolution environment variables of claim 1, wherein the terrain data in S2 comprises high resolution (12.5 m) DEM and DEM with spatial resolution of 30m commonly used for digital soil mapping;
the remote sensing data comprise a 4-scene Sentinel-2A remote sensing image with high resolution (10 meters) and a Landsat-8OLI image with spatial resolution of 30 meters, which is commonly used for digital soil mapping;
the meteorological data are raster data of 1km multiplied by 1km of historical monthly average precipitation and monthly average air temperature of the region to be measured.
5. The method for mapping available phosphorus in cultivated land in southern hilly areas according to claim 1, wherein S3 comprises the following steps:
s31, remote sensing data preprocessing: the remote sensing data preprocessing comprises radiometric calibration, atmospheric correction, geometric fine correction and image mosaic; for 4-scene Sentinel-2A remote sensing images in a research area, respectively carrying out radiometric calibration on each scene image in SNAP software, then carrying out atmospheric correction, converting DN value of the remote sensing image into surface reflectivity, then carrying out image mosaic splicing, then carrying out geometric fine correction on the remote sensing image by using a fine-corrected SPOT5 reference image in ENVI5.3 software, and cutting according to an administrative region map of a region to be detected; the Landsat-8OLI remote sensing image is a geometrically refined image, atmospheric correction is carried out in ENVI5.3 software, the geometrically refined error of all the remote sensing images is controlled within 1 pixel, and the geometrically refined error is uniformly projected to a 1980 coordinate system of Western Ann;
s32, preprocessing of terrain variables: firstly, after checking DEM data with the spatial resolution of 12.5m and 30m without errors, uniformly converting the DEM data into a France 1980 coordinate system through projection transformation in ArcGIS10.2 software, and producing the DEM data with the spatial resolution of 10m from the DEM data with the spatial resolution of 12.5m by using a nearest neighbor method;
s33, preprocessing meteorological variables: uniformly projecting meteorological data to a Cian 1980 coordinate system by using arcgis10.2 software;
s34, checking the consistency of attribute space matching of the remote sensing data, the topographic data, the meteorological data and the soil sampling point data in ArcGIS10.2 software, and establishing geographical connection between the spatial data and the soil attribute after checking without errors.
6. The method for mapping available phosphorus in farmland soil in southern hilly areas based on high resolution environmental variables according to claim 1, wherein the backward elimination method in S4 is specifically:
and screening the variables according to the increase and decrease of the root mean square error after each variable is sequentially excluded from the model during modeling, keeping the variables when the RMSE is increased, and otherwise, eliminating the variables.
7. The method for mapping available phosphorus in cultivated land in southern hilly areas according to claim 6, wherein S4 comprises the following steps:
s41, extracting remote sensing variables: respectively extracting remote sensing variables from Sentinel-2A and Landsat-8OLI in ArcGISI 10.2 software to calculate 46 spectral indexes for digital soil mapping;
s42, extracting terrain variables: extracting 8 terrain variables commonly used for digital soil mapping from the resampled DEM with the resolution of 10 meters and the resampled DEM with the resolution of 30 meters respectively by ArcGISI 10.2 software;
s43, extracting meteorological variables: calculating annual average precipitation and monthly average air temperature data in ArcGIS10.2 software, calculating the average value of annual average precipitation and annual average temperature in a research area by using a grid calculator to obtain annual average temperature and annual average precipitation data of the research area for many years, and resampling the annual average precipitation and annual average temperature data to grid data with the spatial resolution of 10 meters and 30 meters by adopting a nearest neighbor method;
s44, acquiring a soil pH spatial distribution map: interpolating the soil pH value of the research area by using a kriging interpolation method in ArcGISI 10.2 software to obtain a spatial distribution map of the soil pH value;
s45, carrying out geographical association on the soil sampling points and the remote sensing variables, the terrain variables, the meteorological variables and the soil pH data in ArcGIS10.2 software to obtain all environment variables corresponding to the soil sampling points in the research area;
s46, screening of environment variables participating in modeling: performing Pearson correlation analysis on a Python platform, screening variables with obvious correlation with soil available phosphorus, screening the variables by a backward elimination method according to increase and decrease of Root Mean Square Error (RMSE) after each variable is sequentially eliminated from a model during modeling, reserving the variables when the RMSE is increased, otherwise, eliminating, reserving auxiliary variables which finally participate in modeling, and finally obtaining a sample set which participates in modeling.
8. The method for mapping available phosphorus in cultivated land soil in southern hilly areas according to claim 7, wherein S5 comprises the following steps:
s51, constructing a training sample set and a verification sample set: randomly dividing the sample set into a training set and a verification set according to a proportion, wherein samples of the training set are used for modeling, and samples of the verification set are used for testing the prediction accuracy of the model;
s52, determining dependent variables and independent variables of the model: taking the soil available phosphorus content of the surface layer of the cultivated land of the soil sampling point of the region to be detected as a dependent variable, taking a screened high-resolution (10 m) remote sensing variable related to significance, a high-resolution (10 m) terrain variable, a meteorological variable and a soil pH value as independent variables, executing a random forest algorithm on a Python platform, and automatically screening an expected model of the soil available phosphorus content of the surface layer of the cultivated land of the region to be detected based on a high-precision environment variable and the same period as the sampling time of the soil sampling point;
s53, model prediction accuracy evaluation: by R2(coefficient of determination), MAE (mean absolute error) and RMSE (root mean square error) to screen the best predictive model, R2The closer to 1,MAE and RMSE are, the smaller the prediction model precision is, the higher the prediction model precision is; the precision evaluation formula (1-3) is as follows:
wherein n is the number of sampling points, Oi、PiIs the measured value and the predicted value of the sampling point i,
is the average value of measured values;
s54, taking the soil available phosphorus content of the surface layer (0-15 cm) of the cultivated land of the soil sampling point as a dependent variable, taking a remote sensing variable with medium resolution (30 meters), a terrain variable with medium resolution (30 meters), a meteorological variable and a soil pH value which are related to the screening significance as independent variables, and repeatedly executing the steps of S51 the same as the steps to obtain a common optimal random forest prediction model of the available phosphorus content of the cultivated land surface layer soil (0-15 cm) based on the environmental variable with medium resolution and the sampling time of the soil sampling point;
s55, obtaining a relative importance score of the variable in the process of predicting the effective phosphorus content of the soil by averaging through a multi-iteration RF model;
the relative importance of the variables modeled based on the environment variables with the resolution of 10m shows that the relative importance scores of the meteorological variables, the terrain variables, the remote sensing variables and the soil pH value are 30.64%, 30.38%, 22.87% and 16.11% in sequence; according to the importance degree of a single variable, the importance degree of the single variable is from large to small, namely annual average temperature, pH value, topographic humidity index, DEM value, annual average rainfall, enhanced vegetation index, first principal component and red edge wave band 6;
the relative importance of the variables based on the environment variables with the spatial resolution of 30 meters shows that the relative importance scores of the meteorological variables, the terrain variables, the remote sensing variables and the soil pH value are 25.86%, 32.44%, 21.31% and 20.39% in sequence; according to the importance degree, the vegetation index is the elevation (DEM), the pH value, the annual average rainfall, the enhanced vegetation index, B5 and the first main component in turn from large to small.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210889387.5A CN115266612A (en) | 2022-07-27 | 2022-07-27 | Southern hilly area cultivated land available phosphorus drawing method based on high-resolution environment variable |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210889387.5A CN115266612A (en) | 2022-07-27 | 2022-07-27 | Southern hilly area cultivated land available phosphorus drawing method based on high-resolution environment variable |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115266612A true CN115266612A (en) | 2022-11-01 |
Family
ID=83769631
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210889387.5A Pending CN115266612A (en) | 2022-07-27 | 2022-07-27 | Southern hilly area cultivated land available phosphorus drawing method based on high-resolution environment variable |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115266612A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118429564A (en) * | 2024-07-03 | 2024-08-02 | 江西省地质局地理信息工程大队 | Machine learning-based three-dimensional modeling method for soil of south hills |
| CN118537441A (en) * | 2024-05-27 | 2024-08-23 | 浙江大学 | Digital soil mapping method and device based on soil space neighborhood information |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3340130A1 (en) * | 2016-12-23 | 2018-06-27 | Hexagon Technology Center GmbH | Method for prediction of soil and/or plant condition |
| CN109342697A (en) * | 2018-12-07 | 2019-02-15 | 福建农林大学 | Based on random forest-normal stabilizing pile soil organic carbon prediction technique |
| CN110413666A (en) * | 2019-05-31 | 2019-11-05 | 河南省科学院地理研究所 | A kind of multi-source heterogeneous data integration method of farmland quality |
| CN111879915A (en) * | 2020-08-04 | 2020-11-03 | 北京师范大学 | High-resolution monthly soil salinity monitoring method and system for coastal wetland |
| CN113671130A (en) * | 2021-08-24 | 2021-11-19 | 上海交通大学 | Method for researching prevention and control mechanism of interface barrier on weeds in paddy field |
| CN113902580A (en) * | 2021-10-18 | 2022-01-07 | 四川农业大学 | Historical farmland distribution reconstruction method based on random forest model |
| CN114324823A (en) * | 2021-12-30 | 2022-04-12 | 河南农业大学 | Time-space prediction mapping method for soil organic carbon content |
-
2022
- 2022-07-27 CN CN202210889387.5A patent/CN115266612A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3340130A1 (en) * | 2016-12-23 | 2018-06-27 | Hexagon Technology Center GmbH | Method for prediction of soil and/or plant condition |
| CN109342697A (en) * | 2018-12-07 | 2019-02-15 | 福建农林大学 | Based on random forest-normal stabilizing pile soil organic carbon prediction technique |
| CN110413666A (en) * | 2019-05-31 | 2019-11-05 | 河南省科学院地理研究所 | A kind of multi-source heterogeneous data integration method of farmland quality |
| CN111879915A (en) * | 2020-08-04 | 2020-11-03 | 北京师范大学 | High-resolution monthly soil salinity monitoring method and system for coastal wetland |
| CN113671130A (en) * | 2021-08-24 | 2021-11-19 | 上海交通大学 | Method for researching prevention and control mechanism of interface barrier on weeds in paddy field |
| CN113902580A (en) * | 2021-10-18 | 2022-01-07 | 四川农业大学 | Historical farmland distribution reconstruction method based on random forest model |
| CN114324823A (en) * | 2021-12-30 | 2022-04-12 | 河南农业大学 | Time-space prediction mapping method for soil organic carbon content |
Non-Patent Citations (2)
| Title |
|---|
| 张钟亿: "《余庆耕地》", vol. 1, 31 December 2020, 贵州大学出版社, pages: 21 - 24 * |
| 曹佳萍等: "基于稀疏样点的南方丘陵地区耕地土壤有效磷制图", 《中国生态农业学报（中英文）》, vol. 30, no. 2, 28 February 2022 (2022-02-28), pages 290 - 301 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118537441A (en) * | 2024-05-27 | 2024-08-23 | 浙江大学 | Digital soil mapping method and device based on soil space neighborhood information |
| CN118429564A (en) * | 2024-07-03 | 2024-08-02 | 江西省地质局地理信息工程大队 | Machine learning-based three-dimensional modeling method for soil of south hills |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109884664B (en) | Optical microwave collaborative inversion method and system for urban overground biomass | |
| CN108764255B (en) | Method for extracting winter wheat planting information | |
| CN115266612A (en) | Southern hilly area cultivated land available phosphorus drawing method based on high-resolution environment variable | |
| Conti et al. | Evaluation and comparison of satellite precipitation estimates with reference to a local area in the Mediterranean Sea | |
| US7305983B1 (en) | Assessment of solar energy potential on existing buildings in a region | |
| CN105677890A (en) | Urban greening quantity digital map manufacturing and displaying method | |
| CN114019579B (en) | High space-time resolution near-surface air temperature reconstruction method, system and equipment | |
| CN106501186B (en) | A kind of soil moisture content product NO emissions reduction method | |
| Han et al. | A new daily gridded precipitation dataset for the Chinese mainland based on gauge observations | |
| CN110927120B (en) | Early warning method for vegetation coverage | |
| CN106991529A (en) | City night lights economic index evaluation method based on cross-domain multidimensional big data | |
| He et al. | Forest stand biomass estimation using ALOS PALSAR data based on LiDAR-derived prior knowledge in the Qilian Mountain, western China | |
| CN108154193B (en) | Long-time sequence precipitation data downscaling method | |
| Wu et al. | Improved CASA model based on satellite remote sensing data: simulating net primary productivity of Qinghai Lake basin alpine grassland | |
| Hufkens et al. | Impacts and uncertainties of upscaling of remote-sensing data validation for a semi-arid woodland | |
| CN114461983B (en) | Satellite precipitation product space downscaling method based on water balance principle | |
| CN116522090A (en) | Soil moisture downscaling method considering vegetation memory and fine-scale vegetation index | |
| CN114511170A (en) | Soil nutrient estimation method and device based on multispectral image | |
| Han et al. | A new daily gridded precipitation dataset based on gauge observations across mainland China | |
| CN117035066A (en) | Ground surface temperature downscaling method coupling geographic weighting and random forest | |
| CN111489063A (en) | Method and system for evaluating influence of wind driven generator on surrounding environment | |
| Yu et al. | Merging precipitation scheme design for improving the accuracy of regional precipitation products by machine learning and geographical deviation correction | |
| Bi et al. | A deep learning-based model for tropospheric wet delay prediction based on multi-layer 1D convolution neural network | |
| CN115830476A (en) | Terrain factor space downscaling method | |
| CN109885808A (en) | A kind of near surface meteorological element calculation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |