CN114781242A - Remote sensing monitoring method for total amount of algae in true light layer of eutrophic lake - Google Patents
Remote sensing monitoring method for total amount of algae in true light layer of eutrophic lake Download PDFInfo
- Publication number
- CN114781242A CN114781242A CN202210227376.0A CN202210227376A CN114781242A CN 114781242 A CN114781242 A CN 114781242A CN 202210227376 A CN202210227376 A CN 202210227376A CN 114781242 A CN114781242 A CN 114781242A
- Authority
- CN
- China
- Prior art keywords
- algae
- remote sensing
- data
- total amount
- lake
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/01—Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention provides a remote sensing monitoring method for the total amount of algae in a true light layer of an eutrophic lake, which comprises the following steps: based on the synchronous experiment of the field and the field, the water body R of the highly turbid eutrophic water body in different vertical distribution types of algae in the true light layer is contrastively analyzedrsSelecting MODIS B1-B7 wave band R according to spectral characteristicsrcData and index of floating algae, high turbidity index and ratio of near infrared to red light are constructed based on RrcThe random forest machine learning algorithm of the total amount of the algae in the true light layer of the data realizes the satellite remote sensing monitoring of the total amount of the algae in the true light layer of the eutrophic lake. Based on the method, the annual and inter-lunar change rules and the spatial distribution of the total amount of the algae in the true light layer of the eutrophic lake can be accurately obtained.
Description
Technical Field
The invention relates to the technical field of remote sensing, in particular to a satellite remote sensing monitoring method for the total amount of algae in a true light layer of a large eutrophic lake.
Background
The remote sensing technology provides the possibility of rapidly monitoring the blue algae in a large range. The water body spectral characteristics are changed due to the fact that the content of chlorophyll in the water body is obviously increased when the cyanobacteria blooms burst, and the spectral characteristics of the cyanobacteria coverage area are usually obviously different from those of the non-cyanobacteria lake surface. The lake water color remote sensing can utilize multifarities to detect and invert water color element parameters such as chlorophyll of inland water bodies and the like at a sensor. Therefore, the cyanobacterial bloom can be monitored by using satellite remote sensing data. Currently, remote sensing data such as MODIS, CBERS-1, TM, ETM, IRS-P6, LISS-3 and the like are widely used for monitoring cyanobacterial bloom (Langpowen, 2008).
Currently, related scholars have developed various methods for estimating the algae content in the surface water body of lakes (marronghua et al, 2010). In fact, the remote monitoring of the area of algal blooms can vary greatly in a short time. At the same time, changes in external hydrodynamic or environmental factors alter the vertical distribution of algae, causing seemingly short bursts or disappearing of algal blooms (Beaver et al, 2013; Blotti re et al, 2013; Ndong et al, 2014). Therefore, the variation of the vertical structure of algae makes the monitoring of algal blooms on the water surface alone unable to reflect the eutrophication condition of the whole water body, and also affects the precision of remote sensing inversion of the optical parameters of the water body (Stramska and Stramski,2005) and estimation of pigment biomass (sillwane et al, 2010).
The signals detected by remote sensing not only include the information of the surface layer of the water body, but also reflect the structure of the underwater optical field in a certain depth, and the remote sensing reflection ratio has response to the vertical non-uniform distribution of the optical components of the water body in the true optical layer (Xuekun, 2016). Compared with the algae vertical uniform distribution, the algae vertical non-uniform distribution influences the size and the spectral shape of the remote sensing reflectance (Kutser et al, 2008), so that the calculation of the algae total amount in the true light layer becomes the basis for estimating the algae total amount in the whole water column by remote sensing.
The current method for estimating the total algae amount mainly adopts an empirical algorithm, and the depth definition of a 'surface layer' needs to be further defined besides the accuracy of the method for estimating the total algae amount by using the chlorophyll concentration of the surface layer needs to be improved. The influence of different vertical distribution types of algae on the apparent optical quantity of the water body is complex, and the empirical algorithm cannot fully explain the mechanism and the interaction of the algae. In addition, a large amount of field monitoring data show that the vertical distribution type of algae is complex, and the original classification result cannot cover the actual vertical distribution type of algae, so that the original empirical algorithm has larger errors. With the continuous development of the machine learning algorithm, an important technical support is provided for the research of the machine learning algorithm of the total algae. For eutrophic lakes, a true light layer total algae amount estimation model based on remote sensing reflectance is constructed, and the total amount of blue-green algae in the lakes is inverted by a satellite remote sensing method, so that the change and the time-space distribution information of the total amount of the blue-green algae in the lakes can be mastered, the eutrophication condition of the whole lakes can be better reflected, and important technical support is provided for ecological disaster monitoring and early warning of blue-green algae blooms in the lakes.
Disclosure of Invention
The invention aims to provide a remote sensing estimation method for the total amount of algae in a true light layer of an eutrophic lake water body based on a machine learning algorithm.
In order to achieve the technical purpose, the invention adopts the following technical scheme:
a remote sensing monitoring method for the total amount of algae in a true light layer of an eutrophic lake is characterized by comprising the following steps:
1) based on the field satellite-ground synchronous monitoring data, the vertical distribution cluster analysis of algae in the true light layer of the highly turbid eutrophic water body is carried out, the main vertical distribution types of algae are screened out, and the water body R of different vertical distribution types of algae is analyzedrsSpectral features;
2) water body R vertically distributed from different algaersSpectral feature and two angle screening model input parameters of the easily-mixed ground object are as follows:
on the basis of fully considering the spatial resolution and the wave band setting, Rrc data of B1-B7 wave bands of MODIS are selected as alternative input parameters to reflect the visible light-near infrared reflection spectrum characteristics of algae under different vertical distributions;
considering that the spectral characteristics of the high-turbidity water body in red light and near infrared wave bands are similar to the water body with higher algae content and the difference of the remote sensing reflection spectral characteristics of the water body under the vertical distribution type of the algae, selecting floating algae index FAI and high-turbidity water bodyTurbidity index TWI and near-infrared and red light ratio index as alternative input parameters, wherein the near-infrared and red light ratio index refers to near-infrared band RrcData and red band RrcThe ratio of the data;
3) arranging and combining the alternative input parameters to obtain various model input parameters, selecting a random forest machine learning algorithm, and constructing various remote sensing estimation models of the total algae amount in the true light layer by combining a model construction data set screened from field satellite and ground synchronous monitoring data; and detecting the precision of the various models by using verification data, and selecting an optimal model as a final remote sensing estimation model of the total amount of the euglena according to the calculation precision and speed so as to obtain the total amount of the euglena in the euglena layer of the lake whole water area and the spatial distribution of the total amount of the euglena.
As a further improvement of the invention, in the step 1), based on the field satellite-to-ground synchronous experimental data, the vertical distribution type of the algae is determined by performing cluster analysis on the vertical distribution of the chlorophyll a.
As a further improvement of the present invention, in the step 1), the algae vertical distribution is divided into three categories in the step 1): respectively of uniform type, surface layer accumulation type and subsurface highest type;
wherein: dividing the exponential type, power exponential type and Gaussian distribution of the highest concentration of the surface layer into surface layer accumulation types;
the highest subsurface type is defined as a Gaussian having a concentration of 25 to 50cm below the water surface.
As a further improvement of the invention, in the step 2), RrcThe data is data after Rayleigh scattering correction, and belongs to data corrected by inaccurate atmosphere.
In the step 2), the calculation modes of the FAI, the TWI and the near infrared to red light ratio index based on the Rrc data of the MODIS are as follows:
FAI=Rrc(B2)-Rrc(B1)-[Rrc(B5)-Rrc(B5)]×(λB2-λB1)/(λB5-λB1)
TWI=Rrc(B1)-Rrc(B5)
BR=Rrc(B2)/Rrc(B1) (2)
b1, B2 and B5 in the formula respectively refer to central wavelength 645nm, 859nm and 1240nm of MODIS; lambda is the central wavelength of each waveband of the MODIS; BR refers to a near infrared to red ratio index, and B1 and B2 wave bands are selected as two wave bands of BR, so that the spatial resolution is optimal;
as a further improvement of the invention, in the step 3), the alternative model input parameters (4 wave bands R) are inputrcData and 3 remote sensing indexes) to construct various random forest machine learning algorithms in a permutation and combination mode.
As a further improvement of the invention, in the step 3), based on the verification data in the field and satellite synchronous monitoring data, a random forest machine learning algorithm with the highest prediction precision is screened out together as a final inversion model through evaluating a judgment coefficient R2, a root mean square error RMSE and a relative analysis error RPD.
In the step 3), the method for calculating the total amount of algae in the real light layer of the whole lake comprises the following specific steps:
3.1) acquiring a remote sensing image and carrying out image preprocessing;
3.2) obtaining R based on remote sensing imagercData, calculating the ratio index of FAI, TWI, near infrared and red light by pixel;
3.3) operating a remote sensing estimation model of the total amount of the eupatorium algae pixel by pixel;
and according to the flow, obtaining the spatial distribution condition of the total amount of algae in the true light layer of the whole lake.
As a further improvement of the present invention, in step 3.1), geometric correction and radiometric calibration calculation are performed on the acquired image; the geometric correction adopts Geogaphic Lat/Lon projection and combines longitude and latitude information in the 1B data to correct; a lake vector boundary is utilized in the ERDAS, a lake water area is extracted through a masking technology, and the influence of islands is removed; and determining the distribution vector boundary of the aquatic plants by combining the spatial distribution characteristics of the aquatic plants in the lake in 2000-2020, and removing the influence of the aquatic plants by using a mask technology.
As a further improvement of the invention, the step 3) further comprises the step of obtaining the annual and lunar change rule and the spatial distribution of the total amount of algae in the true light layer of the eutrophic lake after processing a plurality of time-series satellite images.
Based on the field and field synchronous experiment, the invention contrasts and analyzes the water body R of the highly turbid eutrophic water body in different vertical distribution types of algae in the true light layerrsSelecting MODIS B1-B7 waveband R according to spectral characteristicsrcData and index of floating algae, high turbidity index and ratio of near infrared to red light are constructed based on RrcThe random forest machine learning algorithm of the total amount of algae in the true photo layer of the data realizes the satellite remote sensing monitoring of the total amount of algae in the true photo layer of the eutrophic lake. Based on the method, the annual and inter-lunar change rules and the spatial distribution of the total amount of the algae in the true light layer of the eutrophic lake can be accurately obtained.
It should be understood that all combinations of the foregoing concepts and additional concepts described in greater detail below can be considered as part of the inventive subject matter of this disclosure unless such concepts are mutually inconsistent. In addition, all combinations of claimed subject matter are considered a part of the presently disclosed subject matter.
The foregoing and other aspects, embodiments and features of the present teachings will be more fully understood from the following description taken in conjunction with the accompanying drawings. Additional aspects of the present invention, such as features and/or advantages of exemplary embodiments, will be apparent from the description which follows, or may be learned by practice of specific embodiments in accordance with the teachings of the present invention.
Drawings
The figures are not intended to be drawn to scale. In the drawings, in which each identical or nearly identical component that is illustrated in various figures may be represented by a like numeral, and in which not every component is labeled for clarity, embodiments of various aspects of the present invention will now be described, by way of example, with reference to the accompanying drawings, in which:
FIG. 1 is a vertical distribution pattern of algae based on field monitoring data.
FIG. 2 shows a water body R of different vertical types of algaersSpectral curve.
Fig. 3 is a comparison of the predicted results and the measured results of the first twelve models with the highest accuracy.
FIG. 4 is a structure of a random forest regression model.
FIG. 5 is a schematic diagram of remote sensing monitoring of total algae in the true light layer based on MODIS satellite data.
In the above-mentioned diagrams 1-4, the coordinates, marks or other representations expressed in english are all known in the art and are not described in detail in this embodiment.
Detailed Description
In order to better understand the technical content of the present invention, specific embodiments are described below with reference to the accompanying drawings.
In this disclosure, aspects of the present invention are described with reference to the accompanying drawings, in which a number of illustrative embodiments are shown. Embodiments of the present disclosure are not necessarily intended to encompass all aspects of the invention. It should be appreciated that the various concepts and embodiments described above, as well as those described in greater detail below, may be implemented in any of numerous ways, and that the concepts and embodiments disclosed herein are not limited to any one implementation. Additionally, some aspects of the present disclosure may be used alone, or in any suitable combination with other aspects of the present disclosure.
This example further describes the method of the present invention, taking a nested lake as an example.
The invention provides a satellite remote sensing monitoring method for the total amount of algae in a true light layer of an eutrophic lake, which is realized by the following steps: constructing a machine learning algorithm of the total amount of algae in the eutrophic lake water body based on satellite-ground synchronous monitoring data;
analyzing water body R of the highly turbid eutrophic water body in different vertical distribution types of algae in true light layer based on field measured datarsSpectral characteristics;
screening model input parameters, and constructing different input parameter combinations;
and selecting a random forest machine learning algorithm, and constructing various remote sensing estimation models of the total algae amount in the true light layer according to different input parameter combinations.
And selecting a random forest algorithm with highest precision and highest calculation speed as a final pholioma alga total remote sensing estimation method through the field and field monitoring data.
The implementation of the foregoing method is specifically described below, as an exemplary description, with reference to the figures.
water R in different vertical distribution types of algae in true light layer of high-turbidity eutrophic water bodyrsSpectral characteristics, based on field satellite-ground synchronous data, performing cluster analysis on vertical distribution of algae, and screening 3 main algae vertical distribution types (figure 1), namely a homogeneous type, a surface accumulation type and a subsurface highest type, wherein the surface accumulation type mainly refers to the highest content of surface algae, the content of the surface accumulation type is gradually reduced along with the increase of water depth, specifically comprises a Gaussian type with the highest concentration on the surface layer in an original classification result, an exponential type, a power exponential type and the like, the subsurface highest type mainly refers to a Gaussian type with the highest concentration 25-50cm below a water meter, and a specific expression form is shown in table 1. Analyzing water body R of different alga vertical distribution typesrsSpectral characteristics, which provide a theoretical basis for determining model input parameters;
TABLE 1 algal vertical distribution types
In Table 1, z is the depth of water, z0Depth at maximum concentration, C chlorophyll concentration, h parameter related to peak intensity, σ standard deviation, C0、m1、m2、n1、n2Are structural parameters.
screening model input parameters from two angles of reflection spectrum characteristics and miscible ground objects: selecting R of B1-B7 wave bands of MODIS on the basis of fully considering spatial resolution and wave band settingrcData, reflecting algaeThe visible light-near infrared reflection spectrum characteristics under different vertical distributions are similar; considering that the spectral characteristics of the high-turbidity water body in red light and near infrared bands are similar to those of the water body with higher algae content, selecting a floating algae index FAI, a high-turbidity index TWI and a near infrared to red light ratio index BR as one of input parameters;
the calculation modes of FAI, TWI and BR based on the Rrc data of MODIS are as follows:
FAI=Rrc(B2)-Rrc(B1)-[Rrc(B5)-Rrc(B5)]×(λB2-λB1)/(λB5-λB1)
TWI=Rrc(B1)-Rrc(B5)
BR=Rrc(B2)/Rrc(B1) (1)
b1, B2 and B5 in the formula respectively refer to wavebands with central wavelengths of 645nm, 859nm and 1240nm of MODIS; λ is the central wavelength of each band of MODIS.
and (3) inputting parameters (7 wave band Rrc data and 3 remote sensing indexes) into the alternative model, and constructing various random forest machine learning algorithms in a permutation and combination mode.
In this embodiment, only rayleigh scattering correction is performed on the MODIS image, that is, the optical information of the top of the atmospheric layer is free from the influence of rayleigh scattering, and still contains aerosol information and ground information. For inland water bodies, an atmosphere accurate correction algorithm is not completely mature, and a universal atmosphere accurate correction algorithm of classical inland water bodies does not exist. If a precise atmosphere correction algorithm is selected, large errors and uncertainties are likely to be introduced, so that the accuracy of a final analysis result is influenced. The rayleigh scatter correction process is as follows (Hu et al, 2004):
in the formula (I), the compound is shown in the specification,is the sensor radiance after correcting the absorption effect of ozone and other gases, F0Is the solar irradiance outside the atmosphere, theta, at which data is acquired0Is the zenith angle of the sun, RrIs the rayleigh reflectance predicted using 6S (Vermote et al, 1997).
Based on radiative transfer theory and assuming an uncoupled marine-atmospheric system, RrcCan be expressed as:
Rrc=Ra+t0tRtarget (3)
in the formula, RaIs the aerosol reflectance (including interactions from aerosol molecules), RtargetIs the surface reflectivity, t, of the field measured target0Is the atmospheric transmission from the sun to the target, and t is the atmospheric transmission from the target to the satellite sensor.
Step 4, determining a remote sensing estimation model of the total amount of the alga in the true light layer based on a machine learning algorithm;
firstly, geometric correction and radiometric calibration calculation are carried out on the obtained MODIS image.
The geometric correction adopts the GeogrAN _ SNhic Lat/Lon projection, the longitude and latitude information in the 1B data is combined for correction, and the position precision after correction reaches 0.5 pixel. The lake vector boundary is utilized in the ERDAS, the lake water area is extracted through the mask technology, and the influence of island vegetation is removed; and determining a uniform aquatic plant distribution vector boundary by combining the spatial distribution characteristics of the Taihu lake aquatic plants in 2000-2020, and removing the influence of the aquatic plants by using a mask technology.
② the MODIS image is subjected to pixel-by-pixel calculation on R of B1(645nm), B2(859nm), B3(469nm), B4(555nm), B5(1240nm), B6(1640nm) and B7(2140nm)rcA value;
calculating FAI, TWI and BR values one by one according to the formula (1);
fourthly, a random forest machine learning algorithm with highest pixel-by-pixel operation precision is operated;
the remote sensing estimation model is constructed by adopting the method, the input parameters of each model are shown in the table 2, and the model structure is shown in fig. 4.
TABLE 2 model input parameters
Model (model) | Input the method | N | Tree num | R2 | RMSE(mg) | RPD | MAE | MAPE(%) |
RF1 | 7bands | 90 | 80 | 0.6813 | 9.8932 | 1.7735 | 7.2415 | 33.35 |
RF2 | 7bands+FAI | 90 | 60 | 0.8273 | 7.2593 | 2.4169 | 5.4101 | 23.77 |
RF3 | 7bands+TWI | 90 | 60 | 0.6570 | 10.239 | 1.7136 | 7.0304 | 32.56 |
RF4 | 7bands+FAI+TWI | 90 | 90 | 0.8214 | 7.3793 | 2.3770 | 5.4751 | 24.02 |
RF5 | 5bands+FAI+TWI | 90 | 70 | 0.8371 | 7.0454 | 2.4903 | 5.3441 | 24.66 |
RF6 | 4bands+FAI+TWI | 90 | 60 | 0.8347 | 7.1095 | 2.4679 | 5.3745 | 25.64 |
RF7 | 4bands+FAI+BR+TWI | 90 | 60 | 0.8447 | 6.8795 | 2.5504 | 5.0584 | 22.68 |
RF8 | 5bands+FAI+BR+TWI | 90 | 60 | 0.8363 | 7.0601 | 2.4852 | 5.1587 | 22.85 |
RF9 | 4bands+FAI+BR | 90 | 90 | 0.8355 | 7.0825 | 2.4773 | 5.1164 | 23.27 |
RF10 | 4bands+BR+TWI | 90 | 70 | 0.7643 | 8.4790 | 2.0693 | 6.0875 | 27.85 |
RF11 | 4bands+TWI | 90 | 60 | 0.7228 | 9.2357 | 1.8997 | 6.6528 | 32.12 |
RF12 | 4bands+BR | 90. | 80 | 0.7812 | 8.1711 | 2.1473 | 5.8702 | 26.37 |
Wherein, a remote sensing monitoring chart of the total algae in the true light layer based on MODIS satellite data is shown in figure 5.
Based on the steps and the method, the method is applied to MODIS satellite image data subjected to Rayleigh scattering correction, and based on the method, the annual and lunar change rules and the spatial distribution of the total amount of algae in the eukaryosphere of the eutrophic lake are obtained after a plurality of time-series satellite images are processed.
Although the present invention has been described with reference to the preferred embodiments, it is not intended to be limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention. Therefore, the protection scope of the present invention should be defined by the appended claims.
Claims (10)
1. A remote sensing monitoring method for the total amount of algae in a true photo layer of an eutrophic lake is characterized by comprising the following steps:
1) based on the field satellite-ground synchronous monitoring data, the vertical distribution cluster analysis of algae in the true light layer of the highly turbid eutrophic water body is carried out, the main vertical distribution types of algae are screened out, and the water body R of different vertical distribution types of algae is analyzedrsSpectral features;
2) water body R vertically distributed from different algaersSpectral feature and two angle screening model input parameters of the easily-mixed ground object are as follows: selecting R of B1-B7 wave bands of MODISrcTaking data, a floating algae index FAI, a high turbidity index TWI and a near infrared to red light ratio index as alternative input parameters; the near infrared and red light ratio index indicates a near infrared band RrcData and red band RrcThe ratio of the data;
3) arranging and combining the alternative input parameters to obtain various model input parameters, selecting a random forest machine learning algorithm, and constructing a remote sensing estimation model of the total algae amount in various real light layers by combining a model construction data set screened from the field satellite-ground synchronous monitoring data;
and detecting the precision of the various models by using the verification data, and selecting an optimal model as a real-light layer algae total quantity remote sensing estimation model according to the calculation precision and speed so as to obtain the algae total quantity in the real-light layer of the lake whole water area and the spatial distribution of the algae total quantity.
2. The remote sensing monitoring method for the total amount of algae in the eudiophora of an eutrophic lake according to claim 1, wherein in the step 1), the vertical distribution type of the algae is determined by performing cluster analysis on the vertical distribution of chlorophyll a based on the field satellite-ground synchronous experimental data.
3. The remote sensing monitoring method for the total amount of algae in the euhedral layer of the eutrophic lake according to claim 1, wherein in the step 1), the vertical distribution of the algae is divided into three categories: respectively of uniform type, surface layer accumulation type and subsurface highest type;
wherein: dividing the exponential type, the power exponential type and the Gaussian distribution with the highest concentration on the surface layer into a surface layer accumulation type;
the highest subsurface type is defined as a Gaussian having a concentration of 25 to 50cm below the water surface.
4. The remote sensing estimation method for the total algae inventory in the eubiotic layer of an eutrophic lake according to claim 1, wherein in the step 2), R isrcThe data is data after Rayleigh scattering correction, and belongs to data corrected by inaccurate atmosphere.
5. The method as claimed in claim 1, wherein in step 2), MODIS R is used as a basis for remote sensing of the total amount of algae in the euhedral layer of the eutrophic lakercThe FAI, TWI, and near infrared to red ratio indices of the data were calculated as follows:
FAI=Rrc(B2)-Rrc(B1)-[Rrc(B5)- Rrc(B5)]×(λB2-λB1)/( λB5-λB1)
TWI=Rrc(B1)-Rrc(B5)
BR=Rrc(B2)/Rrc(B1) (1)
b1, B2 and B5 in the formula respectively refer to central wavelength 645nm, 859nm and 1240nm of MODIS; λ is the central wavelength of each band of MODIS, and BR refers to the near infrared to red ratio index.
6. The method as claimed in claim 1, wherein in step 3), the parameters of the alternative model, i.e. the 7 bands R, are input into the remote sensing monitoring method for the total amount of algae in the eupatorium of the eutrophic lakercAnd constructing various random forest machine learning algorithms by the data and the 3 remote sensing indexes in a permutation and combination mode.
7. The method as claimed in claim 1, wherein in step 3), the coefficient R is determined by evaluation based on the validation data in the field satellite-ground synchronous monitoring data2And the root mean square error RMSE and the relative analysis error RPD are used for screening out a random forest machine learning algorithm with the highest prediction precision as a final inversion model.
8. The remote sensing monitoring method for the total amount of algae in the phoma layer of the eutrophic lake according to claim 1, wherein the method for calculating the total amount of algae in the phoma layer of the whole lake in the step 3) is as follows:
3.1) obtaining a remote sensing image and carrying out image preprocessing;
3.2) obtaining R based on remote sensing imagercData, calculating FAI, TWI, near infrared and red light ratio index pixel by pixel;
3.3) operating a real photosphere algae total amount remote sensing estimation model pixel by pixel;
and according to the flow, obtaining the spatial distribution condition of the total amount of algae in the true light layer of the whole lake.
9. The remote sensing monitoring method for the total amount of algae in the eutrophication lake stratum according to claim 8, wherein in the step 3.1), geometric correction and radiometric calibration calculation are performed on the obtained image; the geometric correction adopts the Geogranic Lat/Lon projection and combines the longitude and latitude information in the 1B data to carry out correction; and the lake vector boundary is utilized in the ERDAS, the lake water area is extracted through the mask technology, and the influence of islands is removed; and determining the distribution vector boundary of the aquatic plants by combining the space distribution characteristics of the aquatic plants in the lake in 2000-2020, and removing the influence of the aquatic plants by using a mask technology.
10. The method as claimed in claim 1, wherein the step 3) further comprises obtaining the annual and lunar change rules and the spatial distribution of the total amount of algae in the eupatorium of the eutrophic lake after processing a plurality of time-series satellite images.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210227376.0A CN114781242A (en) | 2022-03-08 | 2022-03-08 | Remote sensing monitoring method for total amount of algae in true light layer of eutrophic lake |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210227376.0A CN114781242A (en) | 2022-03-08 | 2022-03-08 | Remote sensing monitoring method for total amount of algae in true light layer of eutrophic lake |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114781242A true CN114781242A (en) | 2022-07-22 |
Family
ID=82423199
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210227376.0A Pending CN114781242A (en) | 2022-03-08 | 2022-03-08 | Remote sensing monitoring method for total amount of algae in true light layer of eutrophic lake |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114781242A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117093806A (en) * | 2023-10-16 | 2023-11-21 | 国家卫星气象中心(国家空间天气监测预警中心) | Remote sensing-based full-space coverage offshore atmosphere CO 2 Column concentration calculation method |
-
2022
- 2022-03-08 CN CN202210227376.0A patent/CN114781242A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117093806A (en) * | 2023-10-16 | 2023-11-21 | 国家卫星气象中心(国家空间天气监测预警中心) | Remote sensing-based full-space coverage offshore atmosphere CO 2 Column concentration calculation method |
CN117093806B (en) * | 2023-10-16 | 2024-02-09 | 国家卫星气象中心(国家空间天气监测预警中心) | Remote sensing-based full-space coverage offshore atmosphere CO 2 Column concentration calculation method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108152289B (en) | Remote sensing indirect monitoring method for total amount of extra-phoma algae in eutrophic lake | |
Loisel et al. | Challenges and new advances in ocean color remote sensing of coastal waters | |
Tilstone et al. | Variability in specific-absorption properties and their use in a semi-analytical ocean colour algorithm for MERIS in North Sea and Western English Channel Coastal Waters | |
CN111024618A (en) | Water quality health monitoring method and device based on remote sensing image and storage medium | |
CN105203466B (en) | Algae total inventory remote sensing estimation method under the conditions of a kind of non-algal tufa of eutrophic lake | |
CN112666328B (en) | Chlorophyll a remote sensing inversion method and device based on genetic neural network model | |
CN103712955A (en) | Class-II water atmospheric correction method based on neural network quadratic optimization | |
CN113298086A (en) | Red tide multispectral detection method based on U-Net network | |
CN117152636B (en) | Shallow sea substrate reflectivity remote sensing monitoring method based on dual-band relation | |
Harkort et al. | Estimation of dissolved organic carbon from inland waters at a large scale using satellite data and machine learning methods | |
Mélin | Uncertainties in ocean colour remote sensing | |
CN110196239A (en) | Turbid water body phytoplankton absorption coefficients spectral remote sensing inversion method | |
Cao et al. | Ship fuel sulfur content prediction based on convolutional neural network and ultraviolet camera images | |
CN114781242A (en) | Remote sensing monitoring method for total amount of algae in true light layer of eutrophic lake | |
CN108088805B (en) | Satellite remote sensing monitoring method for total amount of algae in true optical layer of eutrophic lake | |
Zhang et al. | A multiband model with successive projections algorithm for bathymetry estimation based on remotely sensed hyperspectral data in Qinghai Lake | |
CN108596819A (en) | A kind of inland optics Complex water body bloom spectrum reconstruction method based on sparse expression | |
CN116482317A (en) | Lake water nutrition state real-time monitoring method, system, equipment and medium | |
CN111426643A (en) | Atmospheric nitrogen dioxide column concentration inversion stripe noise self-adaptive window correction method | |
CN117115669B (en) | Object-level ground object sample self-adaptive generation method and system with double-condition quality constraint | |
CN105445233B (en) | The South China coastal Disaster And Prevention Measures of Red Tides emergency monitoring and method for early warning | |
CN116879192B (en) | Water bloom prediction method, device, equipment and medium based on satellite remote sensing data | |
CN117035066A (en) | Ground surface temperature downscaling method coupling geographic weighting and random forest | |
Jones et al. | Moving out of the linear rut: the possibilities of generalized additive models | |
Miao et al. | Modeling strategies and influencing factors in retrieving canopy equivalent water thickness of mangrove forest with Sentinel-2 image |
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 |