CN115308386B - Soil salinity inversion method and system based on CYGNSS satellite data - Google Patents
Soil salinity inversion method and system based on CYGNSS satellite data Download PDFInfo
- Publication number
- CN115308386B CN115308386B CN202210843401.8A CN202210843401A CN115308386B CN 115308386 B CN115308386 B CN 115308386B CN 202210843401 A CN202210843401 A CN 202210843401A CN 115308386 B CN115308386 B CN 115308386B
- Authority
- CN
- China
- Prior art keywords
- soil
- target area
- data
- satellite data
- cygnss
- 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.)
- Active
Links
Images
Classifications
-
- 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
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- 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
Abstract
The application relates to the technical field of systems for changing incident light according to tested material properties, and provides a soil salinity inversion method and system based on CYGNSS satellite data. The method comprises the following steps: calculating the surface roughness of different land types in a target area according to ICESat-2 satellite data, and correcting NBRCS data in the CYGNSS satellite data according to the surface roughness parameters and the pre-acquired vegetation optical thickness to obtain first Fresnel reflectivity of different incidence angles of the target area; performing incident angle normalization processing on the first Fresnel reflectivity to obtain a second Fresnel reflectivity of the target area; performing gridding processing on the second Fresnel reflectivity to obtain a third Fresnel reflectivity of each grid in the target area; constructing a soil salinity inversion model according to the third Fresnel reflectivity; and inverting the soil salinity of the target area based on the soil salinity inversion model.
Description
Technical Field
The application relates to the technical field of systems for changing incident light according to tested material properties, in particular to a soil salinity inversion method and system based on CYGNSS satellite data.
Background
In recent years, global soil salinization has become a major environmental problem leading to global land degradation, and land areas affected by salinization are rapidly increasing year by year. The traditional soil salinization monitoring method comprises three methods: the first method is a sampling point-based method, which obtains soil salinization information of sampling points based on field sampling combined with indoor assay analysis, but the method is time-consuming and labor-consuming, can only obtain soil salinization information of a certain time period in a small space range, and cannot monitor large-area real-time dynamic soil salinization distribution and change rules. The second soil salinization monitoring method is optical remote sensing, and the optical remote sensing means monitors soil salinization based on establishing an empirical model between different spectral reflectances and soil salinity, however, the method is affected by weather conditions, has poor universality, and cannot be popularized to different research areas or even the whole world; meanwhile, optical remote sensing cannot monitor the underlying surface due to the influence of the vegetation canopy, soil information is difficult to directly reflect, and monitoring precision is not high. The third soil salinization monitoring method is microwave remote sensing, and although research shows that the backscattering coefficient of partial microwave remote sensing has certain sensitivity with soil salinity, the spatial resolution of microwave remote sensing data is low, so that the requirements of practical application are difficult to meet.
Therefore, there is a need to provide an improved solution to the above-mentioned deficiencies of the prior art.
Disclosure of Invention
The application aims to provide a soil salinity inversion method and system based on CYGNSS satellite data, so as to solve or alleviate the problems in the prior art.
In order to achieve the above purpose, the present application provides the following technical solutions:
the application provides a soil salinity inversion method based on CYGNSS satellite data, which comprises the following steps:
calculating the earth surface roughness MSS under different earth types in a target area based on the pre-acquired land classification information in the SMAP satellite data according to the ICESat-2 satellite data;
based on a geometric optical model, correcting the normalized double-base radar cross section data in the CYGNSS satellite data according to the surface roughness parameter MSS and the pre-acquired vegetation optical thickness VOD to obtain first Fresnel reflectivity of different incidence angles of the target area;
carrying out incident angle normalization processing on the first Fresnel reflectivity to obtain a second Fresnel reflectivity of the target area;
performing gridding processing on the second Fresnel reflectivity to obtain a third Fresnel reflectivity of each grid in the target area;
constructing a soil salinity inversion model based on a gradient lifting stochastic tree algorithm according to the third Fresnel reflectivity, the pre-acquired soil moisture data, the pre-acquired earth surface temperature data and the pre-acquired soil property parameters; inverting the soil salinity of the target area by using a soil salinity-based inversion model;
wherein the soil property parameters are other influence factor sets of the soil complex permittivity of the target area besides the soil moisture data and the surface temperature data in the influence factor sets of the soil complex permittivity of the target area.
Preferably, the calculating, according to the ICESat-2 satellite data, the land surface roughness MSS under different land types in the target area based on the land classification information in the pre-acquired SMAP satellite data specifically is:
according to the formula:
calculating the surface roughness MSS of different land types in the target area;
in the formula: s is the root mean square height of the earth surface calculated according to the information of the earth surface height of each photon in the ICESat-2 satellite data and used for representing the vertical roughness of the earth surface under different land types in the target area; l is the land surface correlation length obtained by calculating a Gaussian correlation function according to the land surface height information of each photon in the ICESat-2 satellite data; for characterizing surface level roughness under different land types in the target region.
Preferably, the geometric optical model-based method corrects the normalized bistatic radar cross section data in the CYGNSS satellite data according to the surface roughness parameter MSS and the pre-obtained vegetation optical thickness VOD to obtain the first fresnel reflectivity at different incident angles in the target area, and specifically includes:
fusing the CYGNSS satellite data with the pre-acquired SMAP satellite data based on the point position in the CYGNSS satellite data to obtain fused data;
based on the fusion data, according to the normalized bistatic radar cross section data in the CYGNSS satellite data, according to a formula:
calculating to obtain first Fresnel reflectivity gamma of different incidence angles of the target area rl ；
In the formula: sigma 0 The normalized bistatic radar cross section data in the CYGNSS satellite data is obtained; theta is an incident angle of the CYGNSS satellite data; τ is the pre-acquired vegetation optical thickness VOD; MSS represents the surface roughness parameter.
Preferably, the incident angle normalization processing is performed on the first fresnel reflectivity to obtain a second fresnel reflectivity of the target area, specifically:
based on a preset correction function, according to a formula:
Γ rl (ε,0)＝Γ rl (ε,θ)/f(θ)
calculating to obtain a second Fresnel reflectivity of the target area;
in the formula: r rl (epsilon, 0) is the second fresnel reflectivity, namely the fresnel reflectivity obtained after the incident angle normalization processing of the CYGNSS satellite data in the target area is 0 degree; r rl (epsilon, theta) represents a first fresnel reflectivity of said target region at an angle of incidence theta of said CYGNSS satellite data; f (theta) is the preset correction function; epsilon represents the complex permittivity of the soil of the target area; θ is an angle of incidence of the CYGNSS satellite data in the target region.
Preferably, the grid processing is performed on the second fresnel reflectivity to obtain a third fresnel reflectivity of each grid in the target region, and the specific steps are as follows:
dividing the target area into a plurality of grids based on the preset grid size;
and performing grid processing on the second Fresnel reflectivity to obtain a third Fresnel reflectivity of each grid in the target area.
Preferably, a soil salinity inversion model is constructed based on a gradient lifting random tree algorithm according to the third fresnel reflectivity, the pre-acquired soil moisture data, the pre-acquired earth surface temperature data and the pre-acquired soil property parameters; and inverting the soil salinity of the target area by using a soil salinity-based inversion model, specifically:
constructing the soil salinity inversion model by taking the third Fresnel reflectivity, the pre-acquired soil moisture data, the pre-acquired surface temperature data and the pre-acquired soil property parameters as independent variables of a gradient lifting random tree algorithm and the soil conductivity of the target area as dependent variables;
training the soil salinity inversion model by taking the grids in the target area as a sample set to obtain the trained soil salinity inversion model;
and inverting the soil salinity of the target area based on the trained soil salinity inversion model.
Preferably, the training the soil salinity inversion model with the multiple grids in the target area as a sample set to obtain the trained soil salinity inversion model includes:
taking the grids in the target area as a sample set, and dividing the sample set according to a preset proportion to obtain a training set and a verification set of the soil salinity inversion model;
training the soil salinity inversion model based on the training set and the verification set of the soil salinity inversion model to obtain the trained soil salinity inversion model.
Preferably, the inverting the soil salinity of the target area based on the trained soil salinity inversion model includes:
determining the soil conductivity of each grid in the target area based on the trained soil salinity inversion model;
and carrying out interpolation processing on the soil conductivity of each grid in the target area to obtain the soil salinity distribution information of the target area.
Preferably, the soil property parameters comprise one or more of a soil sand content ratio, a soil viscosity content ratio and a soil volume weight.
The embodiment of the application provides a soil salinity inversion system based on CYGNSS satellite data, includes:
the computing unit is configured to compute the surface roughness MSS under different land types in the target area according to the ICESat-2 satellite data and based on land classification information in the SMAP satellite data acquired in advance;
the correcting unit is configured to correct the normalized double-base radar cross section data in the CYGNSS satellite data according to the surface roughness parameter MSS and the pre-acquired vegetation optical thickness VOD based on a geometric optical model to obtain first Fresnel reflectivity of different incidence angles of the target area;
the normalization unit is configured to perform incident angle normalization processing on the first Fresnel reflectivity to obtain a second Fresnel reflectivity of the target area;
the gridding unit is configured to perform gridding processing on the second Fresnel reflectivity to obtain a third Fresnel reflectivity of each grid in the target area;
the inversion unit is configured to construct a soil salinity inversion model based on a gradient lifting random tree algorithm according to the third Fresnel reflectivity, the pre-acquired soil moisture data, the pre-acquired earth surface temperature data and the pre-acquired soil property parameters; inverting the soil salinity of the target area by using a soil salinity-based inversion model;
the soil property parameters are other influence factor sets of the soil complex permittivity of the target area besides the soil moisture data and the surface temperature data in the influence factor sets of the soil complex permittivity of the target area.
Has the advantages that:
in the method, firstly, according to ICESat-2 satellite data, land surface roughness MSS under different land types in a target area is calculated based on land classification information in pre-acquired SMAP satellite data, and therefore corresponding relations between the different land types and the land surface roughness MSS in the target area are established; then based on a geometric optical model, correcting the normalized double-base radar cross section data in the CYGNSS satellite data according to the surface roughness parameter MSS and the pre-acquired vegetation optical thickness VOD to obtain first Fresnel reflectivity of different incidence angles of the target area; the incident angle normalization processing is carried out on the first Fresnel reflectivity to obtain a second Fresnel reflectivity of the target area, so that the data precision is improved through the incident angle normalization processing; performing gridding processing on the second Fresnel reflectivity to obtain a third Fresnel reflectivity of each grid in the target area; constructing a soil salinity inversion model based on a gradient lifting stochastic tree algorithm according to the third Fresnel reflectivity, the pre-acquired soil moisture data, the pre-acquired earth surface temperature data and the pre-acquired soil property parameters; and inverting the soil salinity of the target area based on the soil salinity inversion model.
Therefore, the soil salinity inversion method based on the CYGNSS satellite data under the incoherent hypothesis is provided, the soil salinity inversion method is based on the contribution of soil salinity to the reflectivity of the CYGNSS satellite signals, the characteristics of high space-time resolution, wide coverage range, multi-angle monitoring and fast updating of the CYGNSS satellite data are fully utilized, the soil conductivity with high resolution in the global range is obtained through fast inversion, the whole soil salinity distribution condition of a research area is dynamically obtained in real time, and the universality of the method is improved.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, are included to provide a further understanding of the application, and the description of the exemplary embodiments and illustrations of the application are intended to explain the application and are not intended to limit the application. Wherein:
fig. 1 is a schematic flow diagram of a soil salinity inversion method based on CYGNSS satellite data provided in accordance with some embodiments of the present application;
FIG. 2 is a schematic diagram of a numerical solution form of an f (θ) function provided in accordance with some embodiments of the present application;
FIG. 3 is a schematic flow chart of soil salinity inversion model training provided in accordance with some embodiments of the present application;
fig. 4 is a schematic structural diagram of a soil salinity inversion system based on CYGNSS satellite data according to some embodiments of the present application.
Detailed Description
The present application will be described in detail below with reference to the embodiments with reference to the attached drawings. The various examples are provided by way of explanation of the application and are not limiting of the application. In fact, it will be apparent to those skilled in the art that modifications and variations can be made in the present application without departing from the scope or spirit thereof. For instance, features illustrated or described as part of one embodiment, can be used with another embodiment to yield a still further embodiment. It is therefore intended that the present application cover the modifications and variations of this invention provided they come within the scope of the appended claims and their equivalents.
In the following description, references to the terms "first/second/third" merely distinguish between similar items and do not denote a particular order, but rather the terms "first/second/third" may, where permissible, be interchanged with a particular order or sequence, such that embodiments of the application described herein may be practiced in other than the order shown or described herein.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. The terminology used herein is for the purpose of describing embodiments of the disclosure only and is not intended to be limiting of the disclosure.
Before further detailed description of the embodiments of the present disclosure, terms and expressions mentioned in the embodiments of the present disclosure are explained, and the terms and expressions mentioned in the embodiments of the present disclosure are explained as follows.
1) ICESat-2 satellite: used for detecting the heights of the ice layer, the cloud layer and the surface layer of the earth. An ICESat-2 satellite carries an advanced laser device ATLAS (advanced terrain laser altimeter system), can provide data related to the exact position and speed of ice melting, measures the change of the ice on the earth, and monitors the influence caused by climate change.
2) Cyclone Global Navigation Satellite System (CYGNSS): as a Satellite-borne GNSS-R (Global Navigation Satellite System reflection) subtask, the signal can be used for surface parameter monitoring. Compared with other remote sensing means, the CYGNSS has the advantages of high space-time resolution, wide coverage range, multi-angle monitoring and the like.
3) Gradient Boosting Decision Tree (Gradient Boosting Tree) algorithm: the method belongs to a boosting algorithm in an integrated algorithm, and the loss function optimization is completed by connecting a plurality of CART regression trees in series.
Exemplary method
The embodiment of the application provides a soil salinity inversion method based on CYGNSS satellite data, and as shown in FIG. 1, the method comprises the following steps:
and S101, calculating the surface roughness MSS of different land types in the target area according to the ICESat-2 satellite data and on the basis of land classification information in the SMAP satellite data acquired in advance.
It should be noted that, the ICESat-2 satellite is equipped with a laser device, and the acquired data is strip-shaped laser data and does not cover the ground surface of the target area completely, so when calculating the ground surface roughness MSS, the ground surface roughness MSS of different ground types in the target area needs to be calculated by combining the ground classification information in the SMAP satellite data, so as to obtain a lookup table of the ground types and the corresponding ground surface roughness MSS. Based on the lookup table and land classification information in the SMAP satellite data which fully covers the target area, the surface roughness MSS of any position of the target area can be obtained, and parameters are provided for later CYGNSS surface roughness correction.
Wherein, the surface roughness MSS values under different land types are shown in Table 1, and Table 1 is as follows:
TABLE 1 mss values for different land types
In some embodiments, the calculating the surface roughness MSS under different land types in the target area from the ICESat-2 satellite data is specifically:
firstly, acquiring the position information and the surface height z of each photon i in ICESat-2 satellite data i . Wherein the position information of each photon i in ICESat-2 satellite data is used for subsequent data fusion based on the position information, and the earth surface height z i And the parameters are used for calculating the surface roughness MSS.
Then, according to the root-mean-square height of the ground surface and the correlation length of the ground surface of the target area, according to the formula:
calculating the surface roughness MSS of different land types in the target area;
in the formula: s is the root mean square height of the earth surface calculated according to the earth surface height information of each photon in the ICESat-2 satellite data and used for representing the vertical roughness of the earth surface under different land types in the target area; l is the land surface correlation length calculated by a Gaussian correlation function according to the height information of each photon land surface in the ICESat-2 satellite data; for characterizing the surface level roughness of different types of earth in a target area.
The method comprises the following steps of calculating the land surface correlation length l:
under the one-dimensional discrete condition, the gaussian correlation function has the following form:
wherein N represents the total number of photons in the ICESat-2 satellite data for the target region, z i 、z j+i-1 Representing the height of the earth's surface.
Wherein:
ξ＝(j-1)△x
in the formula, Δ x represents a horizontal interval between two adjacent photons.
When gaussian correlation function ρ (ξ) = e -1 And the value of xi is the related length l of the earth surface of the target area.
The formula for calculating the root mean square height s of the earth surface is as follows:
in the formula (I), the compound is shown in the specification,
represents the mean surface height of the ICESat-2 satellite data in the target area.
Step S102, based on a geometric optical model, according to the surface roughness parameter MSS and the pre-acquired vegetation optical thickness VOD, correcting normalized biradical radar cross section data (NBRCS) in the CYGNSS satellite data to obtain first Fresnel reflectivity of different incidence angles of the target area.
In some embodiments, based on a geometric optical model, according to the surface roughness parameter MSS and a pre-obtained vegetation optical thickness VOD, a normalized biradical radar cross-section data (NBRCS) in the CYGNSS satellite data is corrected to obtain first fresnel reflectivities at different incident angles of a target area, specifically: fusing the CYGNSS satellite data and the pre-acquired SMAP satellite data based on the point position in the CYGNSS satellite data to obtain fused data; based on the fusion data, according to the normalized bistatic radar cross section data in the CYGNSS satellite data, according to the formula:
calculating to obtain first Fresnel reflectivity gamma of different incidence angles of the target area rl ；
In the formula: sigma 0 The method comprises the steps of obtaining normalized bistatic radar cross section data in CYGNSS satellite data; theta is an incident angle of CYGNSS satellite data; tau is the pre-acquired vegetation optical thickness VOD and is extracted from SMAP satellite data; MSS denotes the surface roughness parameter.
The above equation may also be referred to as a geometric-optical model of the GNSS signal transmission process.
In specific implementation, first, NBRCS and incidence angle parameters in the cymgnss L1 data set and vegetation optical thickness (VOD) in the SMAP data set are extracted.
The SMAP data set is observation data of a pre-acquired SMAP satellite, the satellite acquires data by using a radar acquisition unit and a radiometer at the same time, and soil moisture content data of 5cm on the surface layer of the soil can be provided under the time resolution of 1-3 d and the space resolution of 3, 10 and 40 km. In addition to vegetation optical thickness (VOD), the SMAP dataset may also include land classification information, soil moisture data.
And then fusing the CYGNSS satellite data and the pre-acquired SMAP satellite data based on the point position in the CYGNSS satellite data to obtain fused data.
It should be noted that the CYGNSS satellite data is discrete point data, which may be understood as dotted vector data, each point has spatial position information, and the CYGNSS satellite data does not implement global coverage. The SMAP satellite data has better coverage on the earth surface in the global range, and better earth surface coverage data can be obtained under 40 degrees of north and south latitude.
In specific implementation, according to the position information of discrete points in the CYGNSS satellite data, SMAP satellite data closest to the discrete points are searched, and the vegetation optical thickness, the land classification information and the soil moisture data in the SMAP satellite data are associated with the discrete points, so that fusion data comprising the vegetation optical thickness, the land classification information and the soil moisture data are obtained.
It is understood that the same procedure can be used to fuse the surface roughness MSS calculated based on the ICESat-2 satellite data with the CYGNSS satellite data, so as to establish the association between the points in the CYGNSS satellite data and the surface roughness MSS, and obtain fused data containing the surface roughness MSS.
Finally, based on the fusion data, according to the normalized bistatic radar cross section data in the CYGNSS satellite data and according to a geometric optical model of the GNSS signal transmission process, the surface roughness and the vegetation attenuation are corrected, the Fresnel reflectivity is calculated, and the first Fresnel reflectivity gamma of different incidence angles of the target area can be obtained rl 。
And step S103, carrying out incident angle normalization processing on the first Fresnel reflectivity to obtain a second Fresnel reflectivity of the target area. Wherein the second fresnel reflectivity is also referred to as the normalized fresnel reflectivity.
It should be noted that, when the incident angle is large, the quality of the satellite signal may be significantly affected, so that before the first fresnel reflectivity is calculated in step S102, the CYGNSS satellite data with the incident angle greater than 65 ° is removed, and only the fresnel reflectivity of the signal with the incident angle below 65 ° is retained.
Wherein the fresnel reflectivity is a function of the signal complex permittivity epsilon and the angle of incidence theta, expressed as f rl (ε, θ). In the embodiment of the application, the incidence angles of all fresnel reflectivities of signals below 65 ° are normalized to 0 ° by introducing the correction function f (θ), and the second fresnel reflectivity of the target area is obtained.
As an embodiment of the present application, the incident angle normalization processing is performed on the first fresnel reflectivity to obtain a second fresnel reflectivity of the target area, and the specific processing is as follows:
based on a preset correction function, according to a formula:
Γ rl (ε,0)＝Γ rl (ε,θ)/f(θ)
calculating to obtain a second Fresnel reflectivity of the target area;
in the formula: r rl (epsilon, 0) is a second Fresnel reflectivity, namely the Fresnel reflectivity is obtained after the incident angle normalization processing of the CYGNSS satellite data in the target area is 0 degree; r' s rl (epsilon, theta) represents a first Fresnel reflectivity of the target area when the incident angle of the CYGNSS satellite data is theta; f (theta) is a preset correction function; epsilon represents the complex permittivity of the soil in the target area; theta is the incidence angle of the CYGNSS satellite data in the target region.
As can be seen from the fitting of the correction function f (θ) by setting different complex permittivities, the correction function f (θ) is hardly affected by the complex permittivity and can be expressed in the form of a numerical solution, as shown in fig. 2.
Through normalization processing of the first Fresnel reflectivity, the influence of the satellite signal incidence angle on the Fresnel reflectivity can be eliminated, only the influence of the complex dielectric constant on the Fresnel reflectivity is reserved, and a foundation is laid for subsequent model training based on the independent variable of the complex dielectric constant determination model.
And step S104, performing grid processing on the second Fresnel reflectivity to obtain a third Fresnel reflectivity of each grid in the target area.
In some embodiments, the gridding processing is performed on the second fresnel reflectivity to obtain a third fresnel reflectivity of each grid in the target region, specifically: dividing a target area into a plurality of grids based on the preset grid size; and performing grid processing on the second Fresnel reflectivity to obtain a third Fresnel reflectivity of each grid in the target area.
In specific implementation, the size of the preset grid is determined according to the size of the target region and the research purpose. Illustratively, in the embodiment of the present application, the target area is divided into basic cells of 0.1 ° × 0.1 °, i.e., the grid size is 0.1 ° × 0.1 °.
In gridding the second fresnel reflectivity, each grid may include a plurality of discrete points of the CYGNSS satellite data, where the gridding includes: eliminating abnormal data in discrete points, and then for each gridAveraging the second Fresnel reflectivity with discrete points to obtain the third Fresnel reflectivity of the grid
The abnormal data determination rule is as follows: if the second fresnel reflectivity corresponding to the discrete point is greater than a preset upper threshold or less than a preset lower threshold, the point is considered to be an outlier, or called abnormal data, and needs to be removed.
Through grid processing, the normalized Fresnel reflectivity of each grid is obtained, so that a sample set of the soil salinity inversion model is formed, and a foundation is provided for subsequent soil salinity inversion model training.
S105, constructing a soil salinity inversion model based on a gradient lifting random tree algorithm according to the third Fresnel reflectivity, the pre-acquired soil moisture data, the pre-acquired earth surface temperature data and the pre-acquired soil property parameters; inverting the soil salinity of the target area by using the soil salinity-based inversion model; the soil property parameters are other influence factor sets of the soil complex dielectric constant of the target area besides the soil moisture data and the earth surface temperature data in the influence factor sets of the soil complex dielectric constant of the target area.
In an alternative embodiment, the soil property parameters include one or more of a soil Sand content Sand ratio Sand, a soil Sand content Clay, a soil Bulk weight BD (Bulk Density).
It should be noted that, through the foregoing normalization process, the normalized fresnel reflectivity of each grid (i.e., the third fresnel reflectivity)
) Determined only by the complex permittivity of the soil.
The applicant researches and discovers that besides being influenced by Soil salinity, soil Moisture SM (Soil Moisture), soil Sand content Sand, soil Sand content Clay, soil Bulk Density BD (Bulk Density) and Surface Temperature LST (Land Surface Temperature LST) jointly determine the value of the complex dielectric constant.
In some embodiments, a soil salinity inversion model is constructed based on a gradient-boosting random tree (GBRT) algorithm according to the third fresnel reflectivity, the pre-acquired soil moisture data, the pre-acquired surface temperature data, and the pre-acquired soil property parameters; soil salinity in order to invert the soil salinity in target area based on soil salinity inversion model specifically does: constructing a soil salinity inversion model by taking the soil Conductivity (EC) of the target area as a dependent variable and taking the soil moisture data, the pre-obtained surface temperature data and the pre-obtained soil property parameters as independent variables of the gradient lifting random tree algorithm; training a soil salinity inversion model by taking a plurality of grids in a target area as a sample set to obtain the trained soil salinity inversion model; and inverting the soil salinity of the target area based on the trained soil salinity inversion model.
In the embodiment of the application, the
Inputting SM, sand%, clay, BD and LST as characteristic parameters of independent variables and soil conductivity EC as dependent variables into a GBRT algorithm to construct a soil salinity inversion model, and training the soil salinity inversion model by taking a plurality of grids in a target area as a sample set to obtain the trained soil salinity inversion model.
Wherein, the soil moisture data is extracted from the SMAP data set; extracting the soil sand content ratio and the soil volume weight of the soil sand content ratio from a HWSD database file provided by a world soil database; the surface temperature is obtained from MODIS satellite data.
A schematic flow chart of soil salinity inversion model training is shown in fig. 3. As can be seen from FIG. 3, based on the GBRT algorithm, the soil salinity inversion model forms a vector x by independent variables i Input into the model by residual error r 1i ，r 2i 823060 \ 8230and adaptive value c 1i ，c 2i 823060, multiple-base learning device f 0 (x)，f 1 (x)、f 2 (x) Correction is carried out to finally obtain the trained soil salinity inversion model F (x).
In another embodiment, training a soil salinity inversion model by using a plurality of grids in a target region as a sample set to obtain a trained soil salinity inversion model includes: taking a plurality of grids in a target area as a sample set, and dividing the sample set according to a preset proportion to obtain a training set and a verification set of a soil salinity inversion model; and training the soil salinity inversion model based on the training set and the verification set of the soil salinity inversion model to obtain the trained soil salinity inversion model.
In specific implementation, 70% of samples in the sample set are used as a training set, the rest 30% of samples are used as a verification set, and the soil salinity inversion model is trained based on the divided training set and the divided verification set to obtain the trained soil salinity inversion model.
In another embodiment, the inversion of the soil salinity of the target area based on the trained soil salinity inversion model includes: determining the soil conductivity of each grid in the target area based on the trained soil salinity inversion model; and carrying out interpolation processing on the soil conductivity of each grid in the target area to obtain the soil salinity distribution information of the target area.
To sum up, in the present application, firstly, according to the ICESat-2 satellite data, based on the land classification information in the SMAP satellite data acquired in advance, the land surface roughness MSS under different land types in the target area is calculated, so as to establish the corresponding relationship between the different land types and the land surface roughness MSS in the target area; then based on a geometric optical model, correcting the normalized double-base radar cross section data in the CYGNSS satellite data according to the surface roughness parameter MSS and the pre-acquired vegetation optical thickness VOD to obtain first Fresnel reflectivity of different incidence angles of the target area; the incident angle normalization processing is carried out on the first Fresnel reflectivity to obtain a second Fresnel reflectivity of the target area, so that the data precision is improved through the incident angle normalization processing; performing gridding processing on the second Fresnel reflectivity to obtain a third Fresnel reflectivity of each grid in the target area; constructing a soil salinity inversion model based on a gradient lifting stochastic tree algorithm according to the third Fresnel reflectivity, the pre-acquired soil moisture data, the pre-acquired earth surface temperature data and the pre-acquired soil property parameters; and inverting the soil salinity of the target area based on the soil salinity inversion model. Therefore, the soil salinity inversion method based on the CYGNSS satellite data under the incoherent hypothesis is provided, the soil salinity inversion method is based on the contribution of soil salinity to the reflectivity of the CYGNSS satellite signals, the characteristics of high space-time resolution, wide coverage range, multi-angle monitoring and fast updating of the CYGNSS satellite data are fully utilized, the soil conductivity with high resolution in the global range is obtained through fast inversion, the whole soil salinity distribution condition of a research area is dynamically obtained in real time, and the universality of the method is improved.
Exemplary System
Embodiments of the present application provide a soil salinity inversion system based on CYGNSS satellite data, and fig. 4 is a schematic structural diagram of a soil salinity inversion system based on CYGNSS satellite data according to some embodiments of the present application. As shown in fig. 4, the system includes: a calculation unit 401, a correction unit 402, a normalization unit 403, a gridding unit 404, and an inversion unit 405. Wherein:
a calculating unit 401 configured to calculate the surface roughness MSS of different land types in the target area based on the land classification information in the SMAP satellite data acquired in advance according to the ICESat-2 satellite data.
A correcting unit 402, configured to correct the normalized bi-base radar cross section data in the CYGNSS satellite data according to the surface roughness parameter MSS and the pre-acquired vegetation optical thickness VOD based on a geometric optical model, to obtain first fresnel reflectivities at different incident angles of the target area.
A normalizing unit 403, configured to perform an incident angle normalization process on the first fresnel reflectivity, so as to obtain a second fresnel reflectivity of the target area.
A gridding unit 404 configured to perform gridding processing on the second fresnel reflectivity to obtain a third fresnel reflectivity of each grid in the target region.
The inversion unit 405 is configured to construct a soil salinity inversion model based on a gradient-boosting random tree algorithm according to the third fresnel reflectivity, the pre-acquired soil moisture data, the pre-acquired earth surface temperature data and the pre-acquired soil property parameters; and inverting the soil salinity of the target area based on the soil salinity inversion model.
Wherein the soil property parameters are other influence factor sets of the soil complex permittivity of the target area besides the soil moisture data and the surface temperature data in the influence factor sets of the soil complex permittivity of the target area.
The soil salinity inversion system based on the CYGNSS satellite data can realize the steps and the process of any soil salinity inversion method based on the CYGNSS satellite data, achieves the same technical effect, and is not described in detail herein.
The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (9)
1. A soil salinity inversion method based on CYGNSS satellite data is characterized by comprising the following steps:
calculating the surface roughness MSS of different land types in the target area based on land classification information in the pre-acquired SMAP satellite data according to ICESat-2 satellite data;
based on a geometric optical model, correcting the normalized double-base radar cross section data in the CYGNSS satellite data according to the surface roughness parameter MSS and the pre-acquired vegetation optical thickness VOD to obtain first Fresnel reflectivity of different incidence angles of the target area;
carrying out incident angle normalization processing on the first Fresnel reflectivity to obtain a second Fresnel reflectivity of the target area;
performing gridding processing on the second Fresnel reflectivity to obtain a third Fresnel reflectivity of each grid in the target area;
constructing a soil salinity inversion model based on a gradient lifting stochastic tree algorithm according to the third Fresnel reflectivity, the pre-acquired soil moisture data, the pre-acquired ground surface temperature data and the pre-acquired soil property parameters; inverting the soil salinity of the target area by using a soil salinity-based inversion model;
wherein the set of influence factors of the soil complex permittivity of the target region comprises: the soil moisture data, the surface temperature data, and the soil property parameters; the soil trait parameters include: one or more of the sand content ratio of the soil, the viscosity content ratio of the soil and the volume weight of the soil.
2. The soil salinity inversion method based on CYGNSS satellite data as claimed in claim 1, wherein the calculating the surface roughness MSS of different land types in the target area based on the land classification information in the pre-obtained SMAP satellite data according to the ICESat-2 satellite data specifically comprises:
according to the formula:
calculating the surface roughness MSS of different land types in the target area;
in the formula:scalculating the root mean square height of the earth surface according to the earth surface height information of each photon in the ICESat-2 satellite data, and representing the vertical roughness of the earth surface under different land types in the target area;lcalculating the land surface correlation length obtained by a Gaussian correlation function according to the land surface height information of each photon in the ICESat-2 satellite data;for characterizing surface level roughness under different types of land in the target area.
3. The soil salinity inversion method based on CYGNSS satellite data according to claim 1, wherein the geometric optical model is used for correcting the normalized double-base radar cross section data in the CYGNSS satellite data according to the surface roughness parameter MSS and the pre-obtained vegetation optical thickness VOD to obtain first Fresnel reflectivities of different incidence angles of the target area, and specifically comprises:
fusing the CYGNSS satellite data with the pre-acquired SMAP satellite data based on the point position in the CYGNSS satellite data to obtain fused data;
based on the fusion data, according to the normalized bistatic radar cross section data in the CYGNSS satellite data, according to a formula:
calculating to obtain first Fresnel reflectivity of different incidence angles of the target areaГ rl ；
In the formula:σ 0 the normalized bistatic radar cross section data in the CYGNSS satellite data is obtained;θan angle of incidence for said CYGNSS satellite data;τthe optical thickness VOD of the pre-acquired vegetation; MSS represents the surface roughness parameter.
4. The soil salinity inversion method based on CYGNSS satellite data according to claim 1, wherein the incident angle normalization processing is performed on the first Fresnel reflectivity to obtain a second Fresnel reflectivity of the target area, and specifically:
based on a preset correction function, according to a formula:
calculating to obtain a second Fresnel reflectivity of the target area;
in the formula:Г rl （ε，0）normalizing the incidence angle of the CYGNSS satellite data in the target area to 0 DEG to obtain a second Fresnel reflectivity;Г rl （ε，θ）representing an angle of incidence of said CYGNSS satellite data asθA first fresnel reflectivity of the target region;f（θ）is the preset correction function;εrepresenting the complex permittivity of the soil of the target area;θis an angle of incidence of the CYGNSS satellite data in the target region.
5. The soil salinity inversion method based on CYGNSS satellite data according to claim 1, wherein the grid processing is performed on the second Fresnel reflectivity to obtain a third Fresnel reflectivity of each grid in the target region, specifically:
dividing the target area into a plurality of grids based on the preset grid size;
and performing grid processing on the second Fresnel reflectivity to obtain a third Fresnel reflectivity of each grid in the target area.
6. The soil salinity inversion method based on CYGNSS satellite data according to claim 5, wherein the soil salinity inversion model is constructed based on a gradient-boosted stochastic tree algorithm according to the third Fresnel reflectivity, the pre-obtained soil moisture data, the pre-obtained surface temperature data and the pre-obtained soil property parameters; and inverting the soil salinity of the target area by using a soil salinity-based inversion model, specifically:
constructing the soil salinity inversion model by taking the soil conductivity of the target area as a dependent variable and the soil moisture data obtained in advance, the surface temperature data obtained in advance and the soil property parameters obtained in advance as independent variables of a gradient lifting random tree algorithm;
training the soil salinity inversion model by taking the grids in the target area as a sample set to obtain the trained soil salinity inversion model;
and inverting the soil salinity of the target area based on the trained soil salinity inversion model.
7. The method of claim 6, wherein the training of the soil salinity inversion model using the plurality of grids in the target region as a sample set to obtain the trained soil salinity inversion model comprises:
taking the grids in the target area as a sample set, and dividing the sample set according to a preset proportion to obtain a training set and a verification set of the soil salinity inversion model;
training the soil salinity inversion model based on the training set and the verification set of the soil salinity inversion model to obtain the trained soil salinity inversion model.
8. The method of claim 6, wherein the inversion of the soil salinity of the target region based on the trained soil salinity inversion model comprises:
determining the soil conductivity of each grid in the target area based on the trained soil salinity inversion model;
and carrying out interpolation processing on the soil conductivity of each grid in the target area to obtain the soil salinity distribution information of the target area.
9. A soil salinity inversion system based on CYGNSS satellite data is characterized by comprising:
the computing unit is configured to compute the surface roughness MSS under different land types in the target area according to the ICESat-2 satellite data and based on land classification information in the SMAP satellite data acquired in advance;
the correcting unit is configured to correct the normalized double-base radar cross section data in the CYGNSS satellite data according to the surface roughness parameter MSS and the pre-acquired vegetation optical thickness VOD based on a geometric optical model to obtain first Fresnel reflectivity of different incidence angles of the target area;
the normalization unit is configured to perform incident angle normalization processing on the first Fresnel reflectivity to obtain a second Fresnel reflectivity of the target area;
the gridding unit is configured to perform gridding processing on the second Fresnel reflectivity to obtain a third Fresnel reflectivity of each grid in the target area;
the inversion unit is configured to construct a soil salinity inversion model based on a gradient lifting random tree algorithm according to the third Fresnel reflectivity, the pre-acquired soil moisture data, the pre-acquired earth surface temperature data and the pre-acquired soil property parameters; inverting the soil salinity of the target area by using a soil salinity-based inversion model;
wherein the set of influence factors of the soil complex permittivity of the target region comprises: the soil moisture data, the surface temperature data, and the soil property parameters; the soil trait parameters include: one or more of the sand content ratio of the soil, the viscosity content ratio of the soil and the volume weight of the soil.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210843401.8A CN115308386B (en) | 2022-07-18 | 2022-07-18 | Soil salinity inversion method and system based on CYGNSS satellite data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210843401.8A CN115308386B (en) | 2022-07-18 | 2022-07-18 | Soil salinity inversion method and system based on CYGNSS satellite data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115308386A CN115308386A (en) | 2022-11-08 |
| CN115308386B true CN115308386B (en) | 2023-04-07 |
Family
ID=83856935
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210843401.8A Active CN115308386B (en) | 2022-07-18 | 2022-07-18 | Soil salinity inversion method and system based on CYGNSS satellite data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115308386B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116224379B (en) * | 2023-05-06 | 2023-09-12 | 中国科学院国家空间科学中心 | NBRCS correction method and device, electronic equipment and storage medium |
| CN116643029B (en) * | 2023-07-27 | 2023-09-26 | 中国科学院地理科学与资源研究所 | Method and system for monitoring soil salinity by using foundation GNSS-IR data |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110826173A (en) * | 2019-09-24 | 2020-02-21 | 广州地理研究所 | Soil moisture data acquisition method and system, storage medium and equipment |
| CN111678866A (en) * | 2020-05-28 | 2020-09-18 | 电子科技大学 | Soil water content inversion method for multi-model ensemble learning |
| CN112488008A (en) * | 2020-12-04 | 2021-03-12 | 中国科学院地理科学与资源研究所 | Soil moisture measurement method and device based on multi-source remote sensing data fusion |
| WO2021239972A1 (en) * | 2020-05-29 | 2021-12-02 | Basf Agro Trademarks Gmbh | A method for an "on-the-fly" treatment of an agricultural field using a soil sensor |
| WO2022011236A2 (en) * | 2020-07-10 | 2022-01-13 | The Board Of Trustees Of The University Of Illinois | Systems and methods for quantifying agroecosystem variables through multi-tier scaling from ground data, to mobile platforms, and to satellite observations |
| CN114139566A (en) * | 2021-08-27 | 2022-03-04 | 中国空间技术研究院 | Method for improving accuracy of measurement based on machine learning weighted average fusion feature extraction |
| CN114371182A (en) * | 2022-03-22 | 2022-04-19 | 中国科学院地理科学与资源研究所 | Satellite-borne GNSS-R high-precision soil moisture estimation method based on CYGNSS data |
| CN114563353A (en) * | 2022-01-21 | 2022-05-31 | 河海大学 | Soil heat flux prediction method based on multi-source satellite remote sensing data |
| CN114740180A (en) * | 2022-04-07 | 2022-07-12 | 中山大学 | Soil organic carbon estimation method and device based on multi-source remote sensing data |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10107770B2 (en) * | 2014-06-18 | 2018-10-23 | Texas Tech University System | Portable apparatus for soil chemical characterization |
-
2022
- 2022-07-18 CN CN202210843401.8A patent/CN115308386B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110826173A (en) * | 2019-09-24 | 2020-02-21 | 广州地理研究所 | Soil moisture data acquisition method and system, storage medium and equipment |
| CN111678866A (en) * | 2020-05-28 | 2020-09-18 | 电子科技大学 | Soil water content inversion method for multi-model ensemble learning |
| WO2021239972A1 (en) * | 2020-05-29 | 2021-12-02 | Basf Agro Trademarks Gmbh | A method for an "on-the-fly" treatment of an agricultural field using a soil sensor |
| WO2022011236A2 (en) * | 2020-07-10 | 2022-01-13 | The Board Of Trustees Of The University Of Illinois | Systems and methods for quantifying agroecosystem variables through multi-tier scaling from ground data, to mobile platforms, and to satellite observations |
| CN112488008A (en) * | 2020-12-04 | 2021-03-12 | 中国科学院地理科学与资源研究所 | Soil moisture measurement method and device based on multi-source remote sensing data fusion |
| CN114139566A (en) * | 2021-08-27 | 2022-03-04 | 中国空间技术研究院 | Method for improving accuracy of measurement based on machine learning weighted average fusion feature extraction |
| CN114563353A (en) * | 2022-01-21 | 2022-05-31 | 河海大学 | Soil heat flux prediction method based on multi-source satellite remote sensing data |
| CN114371182A (en) * | 2022-03-22 | 2022-04-19 | 中国科学院地理科学与资源研究所 | Satellite-borne GNSS-R high-precision soil moisture estimation method based on CYGNSS data |
| CN114740180A (en) * | 2022-04-07 | 2022-07-12 | 中山大学 | Soil organic carbon estimation method and device based on multi-source remote sensing data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115308386A (en) | 2022-11-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115308386B (en) | Soil salinity inversion method and system based on CYGNSS satellite data | |
| CN102539336B (en) | Method and system for estimating inhalable particles based on HJ-1 satellite | |
| CN111583214A (en) | Method for inverting sea surface wind speed by using marine radar image based on RBF neural network | |
| CN108303044B (en) | Leaf area index obtaining method and system | |
| CN108874734B (en) | Global land rainfall inversion method | |
| CN111060992B (en) | Equal-weight conjugate precipitation detection method and system for satellite-borne microwave dioxygen detection channel | |
| Zheng et al. | The retrieved urban LST in Beijing based on TM, HJ-1B and MODIS | |
| CN108827878B (en) | Passive microwave remote sensing detection method for surface air pressure | |
| CN114581791A (en) | Inversion method and system for atmospheric water vapor content based on MODIS data | |
| CN116912690A (en) | Forest leaf area index inversion acquisition method and system based on data fusion | |
| CN111191380B (en) | Atmospheric aerosol optical thickness estimation method and device based on measurement data of foundation spectrometer | |
| CN113340836A (en) | Atmospheric temperature and humidity profile inversion method for high-latitude complex underlying surface | |
| Balenzano et al. | A ground network for SAR-derived soil moisture product calibration, validation and exploitation in Southern Italy | |
| CN112285130B (en) | Inversion method, device and equipment of atmospheric thermal structure and storage medium | |
| CN108594335B (en) | Global precipitation detection method | |
| CN113887024A (en) | Surface soil moisture inversion method based on normalized temperature construction and drought index | |
| CN116519913B (en) | GNSS-R data soil moisture monitoring method based on fusion of satellite-borne and foundation platform | |
| CN114065931B (en) | Land water vapor inversion method and system based on physical model and neural network fusion | |
| CN113836731B (en) | Construction method and device of land surface stable target atmosphere top reflectivity model | |
| CN108897074B (en) | Global ocean rainfall inversion method | |
| CN111044489B (en) | Method for obtaining atmosphere refractive index height distribution profile based on multi-wavelength measurement | |
| CN109406911B (en) | Low-frequency channel radio frequency interference detection and correction method for satellite-borne microwave sensor | |
| Masabi et al. | Evaluation the efficiency of a parametric model based on MODIS data for solar radiation estimation in comparison with some empirical models | |
| Sun et al. | Construction of a spatially gridded heat flux map based on airborne flux Measurements using remote sensing and machine learning methods | |
| CN111751286A (en) | Soil moisture extraction method based on change detection algorithm |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |