CN113204543A - Machine learning-based carbon dioxide column concentration space-time sequence adjustment method - Google Patents
Machine learning-based carbon dioxide column concentration space-time sequence adjustment method Download PDFInfo
- Publication number
- CN113204543A CN113204543A CN202110451384.9A CN202110451384A CN113204543A CN 113204543 A CN113204543 A CN 113204543A CN 202110451384 A CN202110451384 A CN 202110451384A CN 113204543 A CN113204543 A CN 113204543A
- Authority
- CN
- China
- Prior art keywords
- fitting
- space
- parameters
- data
- carbon dioxide
- 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
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 title claims abstract description 42
- 238000000034 method Methods 0.000 title claims abstract description 29
- 229910002092 carbon dioxide Inorganic materials 0.000 title claims abstract description 22
- 239000001569 carbon dioxide Substances 0.000 title claims abstract description 21
- 238000010801 machine learning Methods 0.000 title claims abstract description 12
- 238000012826 global research Methods 0.000 claims abstract description 5
- 238000002372 labelling Methods 0.000 claims abstract description 3
- 238000005070 sampling Methods 0.000 claims description 6
- 230000005012 migration Effects 0.000 claims description 5
- 238000013508 migration Methods 0.000 claims description 5
- 238000011160 research Methods 0.000 claims description 5
- 230000001932 seasonal effect Effects 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 abstract description 4
- 230000001105 regulatory effect Effects 0.000 abstract description 2
- 239000000047 product Substances 0.000 description 8
- 239000005431 greenhouse gas Substances 0.000 description 6
- 238000010521 absorption reaction Methods 0.000 description 3
- 238000004177 carbon cycle Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 239000000443 aerosol Substances 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 239000002803 fossil fuel Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000013526 transfer learning Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2474—Sequence data queries, e.g. querying versioned data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Computing Systems (AREA)
- Strategic Management (AREA)
- Primary Health Care (AREA)
- Human Resources & Organizations (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Fuzzy Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Computational Linguistics (AREA)
- Development Economics (AREA)
- Remote Sensing (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a carbon dioxide column concentration space-time sequence adjusting method based on machine learning. Firstly, the conversion from discrete data points on the space property to a surface is completed through an empirical Bayesian Krigin interpolation (EBK) theory, and a space fitting value result is obtained. Secondly, a time parameter base is constructed, and satellite annual XCO is counted reversely2The method comprises the steps of regulating, fitting satellite data with a single pixel effective value in 10 to 12 months by adopting a specific formula, putting obtained parameters into a parameter library, and labeling corresponding point location information in the time parameter library; followed by migrationAnd matching the space fitting value result with the fitting value of the point corresponding to the step S2 by using the moving learning TCA technology, and distributing the obtained parameters in the time parameter library to each point of the global research area. And finally, taking each point location as a basic unit, bringing the distributed parameters into a specific formula, fitting again, wherein the fitting result is the data product after space-time adjustment.
Description
Technical Field
The invention relates to the field of atmospheric remote sensing, in particular to a carbon dioxide column concentration space-time sequence adjusting method based on machine learning, and further fills a data blank area in greenhouse gas satellite observation.
Background
Due to the rapid increase in the global emissions of major greenhouse gases, the greenhouse effect is increasingly affecting the health and economic prosperity of the earth's ecosystem. Since the industrial revolution of the last century, the global carbon dioxide concentration has risen from 278ppm before the industrial age to 410ppm in 2020. The increase in atmospheric carbon dioxide concentration has caused global climate change and has created a range of social impacts. Therefore, the scientific community has invested great efforts to understand the carbon cycle mechanism through practical measurements and advanced modeling tools. Burning fossil fuels and changing land use are major anthropogenic sources of greenhouse gases, while forests and oceans are major sinks. Thus, the scientific community is seeking accurate knowledge of the carbon dioxide spatiotemporal distribution to promote an in-depth understanding of the carbon cycle. Both active and passive measurement techniques are used to quantify global carbon dioxide. Active CO2The principle of concentration detection is integral path differential absorption, which takes two laser beams as light source to measure CO in laser transmission path2Absorption of (2), calculating CO by differential absorption2And (4) concentration.
At present, a set of mature passive remote sensing instruments are arranged on the satellite and used for measuring the spectrum of sunlight reflected by the earth surface. These spectra were used to retrieve the average carbon dioxide column concentration (XCO)2) (e.g., GOSAT-2, OCO-3, carbon satellites, etc.). However, cloud layers and aerosols can cause spectral interference on carbon dioxide signals, and inter-orbit numbers are also caused by satellite observation mechanismBlank. Thus, XCO2The number of products for inversion is small, and the data availability rate is low, so that the application of the valuable satellite data is limited. Therefore, we propose the present invention to further utilize satellite data to obtain XCO of high spatio-temporal resolution2A map.
Disclosure of Invention
The invention aims to solve the technical problem of filling up the data vacancy of the passive detection greenhouse gas satellite and providing high-quality data support for the research of the carbon dioxide concentration of each region.
The technical scheme adopted by the invention for solving the technical problems is as follows:
the invention provides a carbon dioxide column concentration space-time sequence adjusting method based on machine learning, which comprises the following steps:
step S1, carrying out interpolation on discrete data points on the spatial property to obtain a spatial fitting value result;
step S2, constructing a time parameter base, and counting the satellite annual XCO in a reverse direction2The satellite data of 10 to 12 months of effective values of a single pixel are fitted, the obtained parameters are put into a parameter library, and corresponding point location information in the time parameter library is marked;
step S3, matching the space fitting value result of the step S1 with the fitting value of the point corresponding to the step S2, and distributing the parameters in the time parameter library obtained in the step S2 to all points of the global research area;
and step S4, fitting the distributed parameters again by taking each point position as a basic unit, wherein the fitting result is the data product after space-time adjustment.
Further, in step S1, discrete data points on the spatial property are interpolated by an empirical bayesian kriging interpolation method.
Further, the specific method for constructing the time parameter library in step S2 is as follows;
firstly, counting grid point positions of original satellite observation data with effective month number exceeding 10 to 12, then fitting by adopting the following specific formula, storing fitted parameters b and c into a parameter base, and marking point position information of a selected parameter base, wherein the fitted specific formula is as follows:
wherein, in the formula (1), F (t) is the fitted 12 month data, and a is the average XCO2B and c are seasonal component coefficients, d is an annual component coefficient, f is a sampling frequency, and t is a sampling interval, wherein the parameter library comprises parameters b and c.
Further, in step S3, a migration learning TCA technique is used to match the spatial fitting value result of step S1 with the fitting value of the point corresponding to step S2, the source data is the spatial fitting value of the entire research area of step S1, and the target data is the fitting value of the point labeled in step S2.
Further, the specific implementation manner of step S4 is;
and (4) bringing the parameters distributed to each point location, namely the parameters b and c distributed in the step S3, into the formula (1) for fitting again, wherein the fitting result is the data product after space-time adjustment.
The invention has the following beneficial effects: the invention provides a carbon dioxide column concentration space-time sequence adjusting method based on machine learning, which can provide a high-quality monthly carbon dioxide concentration data product with high spatial resolution based on greenhouse gas satellite data, a transfer learning technology and time information extracted from original observation data, and make up for data vacancy existing in a greenhouse gas satellite.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a general flow chart of an embodiment of the present invention;
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The invention discloses a carbon dioxide column concentration space-time sequence adjusting method based on machine learning, which comprises the following steps of:
step S1, completing conversion from discrete data points on the space property to a surface through an empirical Bayesian Kriging interpolation (EBK) theory to obtain a space fitting value result;
the invention completes the conversion from discrete points to surfaces on the space property through an empirical Bayesian Kriging interpolation theory. The theoretical method adopted in step S1 is empirical bayesian kriging interpolation to complete the conversion of the carbon dioxide concentration from discrete points to surfaces. The empirical Bayes kriging method is a geostatistical interpolation method, and parameters can be automatically calculated by constructing a subset and a simulation process. The introduced error is illustrated by estimating a basic semi-variant function, generally unstable data can be accurately predicted, and the standard error is more accurate than other kriging methods.
Step S2, constructing a time parameter base, and counting satellite annual XCO through reverse direction2The method comprises the steps of regulating, fitting satellite data with a single pixel effective value in 10 to 12 months by adopting a specific formula, putting obtained parameters into a parameter library, and labeling corresponding point location information in the time parameter library;
the specific method for constructing the time parameter library comprises the following steps:
firstly, counting grid point positions of original satellite observation data with the number of effective months of 10 to 12, then fitting by adopting the following specific formula, and storing fitted parameters b and c into a parameter library. In addition, the point location information of the selected parameter library is labeled, and the fitting specific formula is as follows:
wherein, in the formula (1), F (t) is the fitted 12 month data, and a is the average XCO2B and c are coefficients of seasonal components, d isThe annual component coefficient, f is the sampling frequency (f is 12), and t is the sampling interval. In the step, a parameter library and corresponding marked point positions are obtained, wherein the parameter library mainly comprises parameters b and c.
Step S3, matching the space fitting value result of the step S1 with the fitting value of the corresponding point of the step S2 by adopting a migration learning TCA technology, and distributing the parameters in the time parameter library obtained in the step S2 to all points of the global research area;
according to the invention, the space fitting value result of the step S1 is matched with the fitting value of the point corresponding to the step S2 by adopting a TCA (traffic collision avoidance) migration learning technology, and parameters in the time parameter library of the step S2 are distributed to all points of the global research area. In step S3, the adopted migration learning method is TCA, the adopted source data is spatial interpolation of the whole research area in step S1, and the adopted target data is the fitting value of the labeled point in step S2. This step results in parameters b and c being assigned to each point location of the whole study area.
And step S4, taking each point location as a basic unit, bringing the distributed parameters into a specific formula, and fitting again, wherein the fitting result is the data product after space-time adjustment.
The invention relates to a method for fitting each point location by substituting the distributed parameters into a formula (1), wherein the fitting result is a data product after space-time adjustment, and the method comprises the following steps:
in the final product generation, the formula 1 is adopted, the final fitting is carried out by combining the parameters b and c distributed in the step S3, and the required space-time adjustment region interpolation XCO is generated2And (5) producing the product. The final XCO of each point position of the whole research area is obtained in the step2And (5) data products.
It will be understood that modifications and variations can be made by persons skilled in the art in light of the above teachings and all such modifications and variations are intended to be included within the scope of the invention as defined in the appended claims.
Claims (5)
1. A carbon dioxide column concentration space-time sequence adjusting method based on machine learning is characterized by comprising the following steps:
step S1, carrying out interpolation on discrete data points on the spatial property to obtain a spatial fitting value result;
step S2, constructing a time parameter base, and counting the satellite annual XCO in a reverse direction2Fitting satellite data with a single pixel effective value larger than m months, putting the obtained parameters into a parameter library, and labeling corresponding point location information in the time parameter library;
step S3, matching the space fitting value result of the step S1 with the fitting value of the point corresponding to the step S2, and distributing the parameters in the time parameter library obtained in the step S2 to all points of the global research area;
and step S4, fitting the distributed parameters again by taking each point position as a basic unit, wherein the fitting result is the data product after space-time adjustment.
2. The machine learning-based carbon dioxide column concentration space-time sequence filling technique of claim 1, wherein: in step S1, discrete data points on the spatial property are interpolated by an empirical bayesian kriging interpolation method.
3. The machine learning-based carbon dioxide column concentration space-time sequence adjusting method as claimed in claim 1, characterized in that: the specific method for constructing the time parameter library in step S2 is as follows;
firstly, counting grid point positions of original satellite observation data with effective months of 10 to 12, then fitting by adopting the following specific formula, storing fitted parameters b and c into a parameter library, marking point position information of a selected parameter library, wherein the fitted specific formula is as follows:
wherein, in the formula (1), F (t) is the fitted 12 month data, and a is the average XCO2B and c are coefficients of seasonal components, d isThe annual component coefficient, f is the sampling frequency, t is the sampling interval, wherein the parameter library comprises parameters b and c.
4. The machine learning-based carbon dioxide column concentration space-time sequence adjusting method as claimed in claim 1, characterized in that: in step S3, a migration learning TCA technique is used to match the spatial fitting value result of step S1 with the fitting value of the point corresponding to step S2, the source data is the spatial fitting value of the whole research area of step S1, and the target data is the fitting value of the point labeled in step S2.
5. The machine learning-based carbon dioxide column concentration space-time sequence adjusting method according to claim 3, characterized in that: the specific implementation manner of the step S4 is;
and (4) bringing the parameters distributed to each point location, namely the parameters b and c distributed in the step S3, into the formula (1) for fitting again, wherein the fitting result is the data product after space-time adjustment.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110451384.9A CN113204543A (en) | 2021-04-26 | 2021-04-26 | Machine learning-based carbon dioxide column concentration space-time sequence adjustment method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110451384.9A CN113204543A (en) | 2021-04-26 | 2021-04-26 | Machine learning-based carbon dioxide column concentration space-time sequence adjustment method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113204543A true CN113204543A (en) | 2021-08-03 |
Family
ID=77028678
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110451384.9A Pending CN113204543A (en) | 2021-04-26 | 2021-04-26 | Machine learning-based carbon dioxide column concentration space-time sequence adjustment method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113204543A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114878748A (en) * | 2022-05-07 | 2022-08-09 | 国网四川省电力公司电力科学研究院 | CO (carbon monoxide) 2 Method and system for monitoring discharge amount |
CN115271265A (en) * | 2022-09-27 | 2022-11-01 | 四川中电启明星信息技术有限公司 | Electric energy carbon flow analysis method and system based on carbon satellite data |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111881569A (en) * | 2020-07-24 | 2020-11-03 | 中国科学院大气物理研究所 | Inversion method and device for carbon dioxide column concentration, storage medium and electronic equipment |
CN112597651A (en) * | 2020-12-22 | 2021-04-02 | 武汉大学 | CO inversion based on OCO-2 data and WRF-STILT model2Method and system for background field concentration |
-
2021
- 2021-04-26 CN CN202110451384.9A patent/CN113204543A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111881569A (en) * | 2020-07-24 | 2020-11-03 | 中国科学院大气物理研究所 | Inversion method and device for carbon dioxide column concentration, storage medium and electronic equipment |
CN112597651A (en) * | 2020-12-22 | 2021-04-02 | 武汉大学 | CO inversion based on OCO-2 data and WRF-STILT model2Method and system for background field concentration |
Non-Patent Citations (1)
Title |
---|
XIN MA: "A Regional Spatiotemporal Downscaling Method for CO2 Columns", 《IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114878748A (en) * | 2022-05-07 | 2022-08-09 | 国网四川省电力公司电力科学研究院 | CO (carbon monoxide) 2 Method and system for monitoring discharge amount |
CN115271265A (en) * | 2022-09-27 | 2022-11-01 | 四川中电启明星信息技术有限公司 | Electric energy carbon flow analysis method and system based on carbon satellite data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Schuh et al. | Quantifying the impact of atmospheric transport uncertainty on CO2 surface flux estimates | |
Ehret et al. | MERLIN: A French-German space lidar mission dedicated to atmospheric methane | |
Basu et al. | The impact of transport model differences on CO 2 surface flux estimates from OCO-2 retrievals of column average CO 2 | |
Winkler et al. | Earth system models underestimate carbon fixation by plants in the high latitudes | |
CN109142679B (en) | Forest soil nutrient space prediction method based on artificial neural network kriging interpolation | |
CN113204543A (en) | Machine learning-based carbon dioxide column concentration space-time sequence adjustment method | |
Liu et al. | Bare‐Earth DEM generation in urban areas for flood inundation simulation using global digital elevation models | |
Nicely et al. | A machine learning examination of hydroxyl radical differences among model simulations for CCMI-1 | |
Engelen et al. | Global observations of the carbon budget: 1. Expected satellite capabilities for emission spectroscopy in the EOS and NPOESS eras | |
CN113324923B (en) | Remote sensing water quality inversion method combining space-time fusion and deep learning | |
CN114114324A (en) | Atmospheric CO for space-borne laser radar and high-speed spectrometer2Concentration cooperative inversion method | |
CN105738915A (en) | Three-dimensional radar measuring method and three-dimensional radar measuring device | |
CN108267725B (en) | CO of multi-wavelength laser radar2Concentration hierarchical inversion method and system | |
Fu et al. | Evaluation of ocean biogeochemistry and carbon cycling in CMIP earth system models with the international ocean model benchmarking (IOMB) software System | |
Philip et al. | OCO‐2 satellite‐imposed constraints on terrestrial biospheric CO2 fluxes over South Asia | |
Sweeney et al. | Statistical challenges in estimating past climate changes | |
CN105046046A (en) | Ensemble Kalman filter localization method | |
CN103197303A (en) | Earth surface two-direction reflection characteristic retrieval method and earth surface two-direction reflection characteristic retrieval system based on multiple sensors | |
Wang et al. | Combining XCO2 measurements derived from SCIAMACHY and GOSAT for potentially generating global CO2 maps with high spatiotemporal resolution | |
Gong et al. | Evaluating the monthly and interannual variation of net primary production in response to climate in Wuhan during 2001 to 2010 | |
Li et al. | Generating daily high-resolution and full-coverage XCO2 across China from 2015 to 2020 based on OCO-2 and CAMS data | |
CN105469391A (en) | Cloud shadow detection method and cloud shadow detection system | |
Parard et al. | Remote sensing the sea surface CO 2 of the Baltic Sea using the SOMLO methodology | |
Sun et al. | Midwest US croplands determine model divergence in North American carbon fluxes | |
CN107491588B (en) | Inversion method and system of temperature and pressure profile |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210803 |