A river organic carbon flux inversion method based on multi-source data and related equipment

Abstract

The application discloses a river organic carbon flux inversion method based on multi-source data and related equipment, and the method comprises the following steps: acquiring multi-source data of a river basin range in a target research area; converting remote sensing images to obtain a normalized difference index; processing the normalized difference water body index and the multi-source data to obtain river parameters of the target research area, and converting the river parameters through a hydraulic geometric relationship to obtain river runoff; based on remote sensing images, a normalized difference chlorophyll index and a normalized difference turbidity index, using a first inversion model to obtain a particulate organic carbon concentration; based on the population quantity and the rainfall, using a second inversion model to obtain a dissolved organic carbon concentration; and converting the river runoff, the particulate organic carbon concentration and the dissolved organic carbon concentration to obtain the river organic carbon flux. The application can significantly improve the accuracy and reliability of monitoring on a high dynamic river, accurately realize river organic carbon flux inversion, and can be widely applied to the technical field of data processing.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a river organic carbon flux inversion method based on multi-source data and related equipment. BACKGROUND

[0002] Rivers are important transport channels connecting the two carbon pools of land and ocean. River ecosystems continuously transport carbon materials in the land and atmosphere to the ocean carbon pool. Therefore, monitoring river organic carbon flux is a core link for understanding global carbon cycle processes, assessing regional ecological environment, and responding to climate change, and has important scientific and practical significance. However, the traditional river organic carbon flux monitoring method usually needs a large amount of measured data collection and calculation. However, the high cost of manpower, material resources and time in the process of obtaining part of the measured data, and the data acquisition process is easily affected by the performance of the measuring equipment and the laying conditions. Secondly, only optical remote sensing images can detect river surface information, and there is a large uncertainty on the high dynamic river. Finally, due to the availability of measured data, it is difficult to carry out quantitative evaluation and historical reconstruction of the organic carbon flux of small and medium-sized rivers which lack historical observation data or are not included in the regular monitoring system. SUMMARY

[0003] The main purpose of the embodiments of the present application is to provide a river organic carbon flux inversion method, device, electronic equipment, storage medium and program product based on multi-source data, which aims to solve at least one problem of the prior art.

[0004] To achieve the above-mentioned purpose, one aspect of an embodiment of the present application provides a river organic carbon flux inversion method based on multi-source data, the method comprising:

[0005] Obtaining multi-source data of the river basin range in the target research area, the multi-source data comprising remote sensing images, digital elevation models, population numbers and rainfall amounts;

[0006] Converting the remote sensing images to obtain normalized difference indices, the normalized difference indices comprising normalized difference water indices, normalized difference chlorophyll indices and normalized difference turbidity indices;

[0007] Processing the normalized difference water indices and the multi-source data to obtain river parameters of the target research area, and then converting the river parameters to obtain river runoff through hydraulic geometric relationships;

[0008] Based on the remote sensing images, the normalized difference chlorophyll indices and the normalized difference turbidity indices, using a preset first inversion model to obtain particulate organic carbon concentrations;

[0009] Based on the population numbers and the rainfall amounts, using a preset second inversion model to obtain dissolved organic carbon concentrations;

[0010] The river organic carbon flux is obtained according to the river runoff, the particulate organic carbon concentration and the dissolved organic carbon concentration.

[0011] In some embodiments, the river parameters include the river width, the river depth and the river flow rate, and the river parameters of the target research area are obtained based on the normalized difference water index and multi-source data processing, including the following steps:

[0012] The river water body of the target research area is determined based on the normalized difference water index, and the river width is quantified by the river water body;

[0013] The river depth is obtained based on the digital elevation model of the range where the river water body is located;

[0014] The river channel slope is determined according to the digital elevation model, and the river flow rate is calculated based on the river channel slope using the Manning formula.

[0015] In some embodiments, the river water body of the target research area is determined based on the normalized difference water index, and the river width is quantified by the river water body, including the following steps:

[0016] Based on the normalized difference water index of each place in the target research area, the river water body is extracted from the remote sensing image through a preset water body threshold;

[0017] The river water body is vectorized, and the profile line is generated at a position perpendicular to the river center line based on the result of the vectorization, and then the river width is quantified.

[0018] In some embodiments, the river depth is obtained based on the digital elevation model of the range where the river water body is located, including the following steps:

[0019] Based on the range corresponding to the river water body, the grid of the river water surface area is extracted from the digital elevation model by mask extraction;

[0020] Based on the grid, the average elevation of each independent river segment unit is obtained as the river depth by using a preset statistical tool.

[0021] In some embodiments, the river parameters include the river width, the river depth and the river flow rate, and the river runoff is obtained by transformation through the hydraulic geometric relationship, including the following steps:

[0022] Based on the preset empirical parameter, an empirical power law relationship between the river parameters and the river runoff is established through the hydraulic geometric relationship;

[0023] Based on the empirical power law relationship, the river runoff is fitted by the product of the river width, the river depth, the river flow rate and the empirical parameter.

[0024] In some embodiments, the particulate organic carbon concentration is obtained by using a preset first inversion model based on remote sensing images, normalized difference chlorophyll index and normalized difference turbidity index, including the following steps:

[0025] The chlorophyll is determined based on the normalized difference chlorophyll index, and the suspended matter is determined based on the normalized difference turbidity index.

[0026] The chlorophyll and the suspended matter are taken as environmental factors.

[0027] The original band reflectivity is extracted from the remote sensing images.

[0028] The environmental factors and the original band reflectivity of all positions in the river basin range are input into the first inversion model, and the particulate organic carbon concentration in the whole region of the river basin range is output.

[0029] The first inversion model is obtained by training a random forest model based on preset training data, and the preset training data is constructed based on the measured particulate organic carbon concentration and the environmental factors corresponding to part of the points in the river basin range.

[0030] In some embodiments, the dissolved organic carbon concentration is obtained by using a preset second inversion model based on the population number and the rainfall, including the following steps:

[0031] The population number is converted into a population density.

[0032] The dissolved organic carbon concentration is obtained by using a multiple linear regression model based on the population density and the rainfall.

[0033] The multiple linear regression model has learned the linear relationship between the dissolved organic carbon concentration and the rainfall and the artificial density in the river basin range in advance.

[0034] To achieve the above object, another aspect of the embodiment of the present application proposes a river organic carbon flux inversion device based on multi-source data, which comprises:

[0035] The first module is configured to acquire multi-source data of a river basin range in a target research area, and the multi-source data comprises remote sensing images, a digital elevation model, a population number and rainfall.

[0036] The second module is configured to convert normalized difference indexes based on the remote sensing images, and the normalized difference indexes comprise a normalized difference water index, a normalized difference chlorophyll index and a normalized difference turbidity index.

[0037] The third module is configured to obtain river parameters of the target research area based on the normalized difference water index and the multi-source data processing, and then convert the river flow rate through a hydraulic geometric relationship.

[0038] The fourth module is configured to obtain the particulate organic carbon concentration by using a preset first inversion model based on the remote sensing image, the normalized difference chlorophyll index, and the normalized difference turbidity index;

[0039] The fifth module is configured to obtain the dissolved organic carbon concentration by using a preset second inversion model based on the population number and the rainfall.

[0040] The sixth module is configured to obtain the river organic carbon flux by converting the river runoff, the particulate organic carbon concentration, and the dissolved organic carbon concentration.

[0041] To achieve the above object, another aspect of the embodiments of the present application provides an electronic device, which comprises a memory and a processor. The memory stores a computer program. The processor implements the foregoing method when executing the computer program.

[0042] To achieve the above object, another aspect of the embodiments of the present application provides a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the foregoing method.

[0043] To achieve the above object, another aspect of the embodiments of the present application provides a computer program product, which comprises a computer program. The computer program is executed by a processor to implement the foregoing method.

[0044] The embodiments of the present invention include at least the following beneficial effects: The present invention provides a method, apparatus, electronic device, storage medium, and program product for river organic carbon flux inversion based on multi-source data. This scheme acquires multi-source data of the river basin in the target study area, including remote sensing images, digital elevation models, population data, and rainfall data; transforms the remote sensing images to obtain normalized difference indices, including normalized difference water index, normalized difference chlorophyll index, and normalized difference turbidity index; processes the normalized difference water index and multi-source data to obtain river parameters of the target study area, and then transforms them into river runoff through hydraulic geometry; inverts particulate organic carbon concentration using a preset first inversion model based on remote sensing images, normalized difference chlorophyll index, and normalized difference turbidity index; inverts dissolved organic carbon concentration using a preset second inversion model based on population data and rainfall; and converts the river organic carbon flux according to the river runoff, particulate organic carbon concentration, and dissolved organic carbon concentration. This invention, through the acquisition of multi-source data including remote sensing imagery, digital elevation models (DEMs), population data, and rainfall data, utilizes publicly available or remotely sensed conventional data. This significantly reduces or even eliminates the expensive and cumbersome field sampling and measurement work required by traditional methods, effectively saving manpower, resources, and time. Furthermore, by fusing DEMs, this invention can obtain parameters related to riverbed topography (such as depth and slope). By incorporating population data, it can indirectly reflect the impact of human activities on dissolved organic carbon. This combination of multi-source data overcomes the limitation that "optical remote sensing imagery alone can only detect river surface information," achieving… A more comprehensive and three-dimensional characterization of the river carbon cycle process can significantly improve the accuracy and reliability of monitoring in highly dynamic rivers. In particular, the embodiments of this invention determine the organic carbon concentration through inversion. By calling up multi-source data from historical periods, the organic carbon flux of that historical period can be obtained through inversion, thus effectively solving the problem of historical reconstruction and providing a feasible technical path for assessing the carbon flux of a large number of unmonitored rivers worldwide. Specifically, through model inversion, this invention can obtain parameters such as particulate organic carbon concentration and runoff that are spatially continuous throughout the entire river basin, thereby more accurately calculating the organic carbon flux at the outlet section of the entire basin and avoiding the errors that may be caused by substituting a point for a whole area. Attached Figure Description

[0045] Figure 1 This is a schematic diagram of an implementation environment for the method of river organic carbon flux inversion based on multi-source data provided in this embodiment of the invention;

[0046] Figure 2 This is a schematic flowchart of a method for inverting river organic carbon flux based on multi-source data provided in an embodiment of the present invention;

[0047] Figure 3 This is a schematic diagram of the process for obtaining river parameters according to an embodiment of the present invention;

[0048] Figure 4 This is a schematic diagram of the unfolding process of step S310 provided in an embodiment of the present invention;

[0049] Figure 5 This is a schematic diagram of the unfolding process of step S320 provided in an embodiment of the present invention;

[0050] Figure 6 This is a schematic diagram of the process of obtaining river runoff through hydraulic geometry transformation provided in an embodiment of the present invention;

[0051] Figure 7 This is a schematic diagram of the unfolding process of step S400 provided in an embodiment of the present invention;

[0052] Figure 8 This is a schematic diagram of the structure of the river organic carbon flux inversion system based on multi-source data provided in an embodiment of the present invention;

[0053] Figure 9 This is a schematic diagram of a river organic carbon flux inversion device based on multi-source data provided in an embodiment of the present invention;

[0054] Figure 10 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in 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. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of this invention; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this invention as detailed in the appended claims.

[0056] It is understood that the terms “first,” “second,” etc., used in this invention may be used herein to describe various concepts, but unless specifically stated otherwise, these concepts are not limited by these terms. These terms are used only to distinguish one concept from another. For example, first information may also be referred to as second information without departing from the scope of embodiments of the invention, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to determination” as used herein may be interpreted as “when…” or “when…” or “in response to determination.”

[0057] The terms “at least one,” “multiple,” “each,” “any,” etc., used in this invention, “at least one” includes one, two, or more than two; “multiple” includes two or more than two; “each” refers to each of the corresponding multiple; and “any” refers to any one of the multiple.

[0058] Unless otherwise defined, 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 invention pertains. The terminology used herein is for the purpose of describing embodiments of the invention only and is not intended to limit the invention.

[0059] In related technologies, traditional methods for monitoring river organic carbon flux typically require the collection and calculation of a large amount of measured data. However, the high cost of manpower, resources, and time associated with some measured data, and the data acquisition process being easily constrained by the performance and deployment conditions of the measuring equipment, make it difficult to conduct quantitative assessments and historical reconstructions of organic carbon flux in small and medium-sized rivers that lack historical observation data or are not included in the routine monitoring system.

[0060] In view of this, this invention provides a method and related equipment for river organic carbon flux inversion based on multi-source data. This method acquires multi-source data covering the river basin in the target study area, including remote sensing images, digital elevation models, population data, and rainfall data. Based on the remote sensing images, a normalized difference index (NDI) is obtained, including the normalized difference water index, normalized difference chlorophyll index, and normalized difference turbidity index. Based on the NDI and multi-source data processing, river parameters for the target study area are obtained, and then river runoff is obtained through hydraulic geometry transformation. Based on the remote sensing images, NDI, and turbidity index, a preset first inversion model is used to invert particulate organic carbon concentration. Based on the population data and rainfall data, a preset second inversion model is used to invert dissolved organic carbon concentration. Finally, river organic carbon flux is obtained by converting river runoff, particulate organic carbon concentration, and dissolved organic carbon concentration. This invention, through the acquisition of multi-source data including remote sensing imagery, digital elevation models (DEMs), population data, and rainfall data, utilizes publicly available or remotely sensed conventional data. This significantly reduces or even eliminates the expensive and cumbersome field sampling and measurement work required by traditional methods, effectively saving manpower, resources, and time. Furthermore, by fusing DEMs, this invention can obtain parameters related to riverbed topography (such as depth and slope). By incorporating population data, it can indirectly reflect the impact of human activities on dissolved organic carbon. This combination of multi-source data overcomes the limitation that "optical remote sensing imagery alone can only detect river surface information," achieving… A more comprehensive and three-dimensional characterization of the river carbon cycle process can significantly improve the accuracy and reliability of monitoring in highly dynamic rivers. In particular, the embodiments of this invention determine the organic carbon concentration through inversion. By calling up multi-source data from historical periods, the organic carbon flux of that historical period can be obtained through inversion, thus effectively solving the problem of historical reconstruction and providing a feasible technical path for assessing the carbon flux of a large number of unmonitored rivers worldwide. Specifically, through model inversion, this invention can obtain parameters such as particulate organic carbon concentration and runoff that are spatially continuous throughout the entire river basin, thereby more accurately calculating the organic carbon flux at the outlet section of the entire basin and avoiding the errors that may be caused by substituting a point for a whole area.

[0061] It is understood that the river organic carbon flux inversion method based on multi-source data provided by this invention can be applied to any computer device with data processing and computing capabilities, and this computer device can be various terminals or servers. When the computer device in the embodiments is a server, the server is an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. Optionally, the terminal can be a smartphone, tablet, laptop, or desktop computer, but it is not limited to these.

[0062] like Figure 1 The diagram shown is a schematic representation of an implementation environment provided by an embodiment of the present invention. (Refer to...) Figure 1 The implementation environment includes at least one terminal 102 and a server 101. The terminal 102 and the server 101 can be connected via a network, either wirelessly or via a wired connection, to complete data transmission and exchange.

[0063] Server 101 can be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.

[0064] Additionally, server 101 can also be a node server in a blockchain network. Blockchain is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms.

[0065] Terminal 102 can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, etc., but is not limited to these. Terminal 102 and server 101 can be directly or indirectly connected via wired or wireless communication, and this embodiment of the invention does not impose any limitations.

[0066] For example, based on Figure 1The implementation environment shown in this embodiment of the invention provides a method for inverting river organic carbon flux based on multi-source data. The following description uses the application of this method for inverting river organic carbon flux based on multi-source data in server 101 as an example. It can be understood that this method for inverting river organic carbon flux based on multi-source data can also be applied in terminal 102.

[0067] Reference Figure 2 , Figure 2 This is an optional flowchart of a river organic carbon flux inversion method based on multi-source data provided in this embodiment of the invention. The execution subject of this river organic carbon flux inversion method based on multi-source data can be any of the aforementioned computer devices (including servers or terminals). Figure 2 The method may include, but is not limited to, steps S100 to S600.

[0068] Step S100: Obtain multi-source data on the river basin area within the target study area;

[0069] The multi-source data includes remote sensing images, digital elevation models, population data, and rainfall data.

[0070] For example, in some specific implementations, multi-source data of the study area are collected and preprocessed. This mainly involves collecting river basin extent data, Sentinel-2 series remote sensing images, DEM (Digital Elevation Model) data, population data, and rainfall product images. The images are then cropped using river basin data based on ArcMap software, and cloud cover filtering and cloud masking are performed on the Sentinel-2 series remote sensing images as needed. Note that the Sentinel-2 series remote sensing images selected are of level L2A. If level L1C is selected, atmospheric correction needs to be performed before cloud cover filtering.

[0071] Step S200: Obtain the normalized difference index based on the transformation of remote sensing images;

[0072] Among them, the normalized difference index includes the normalized difference water body index, the normalized difference chlorophyll index, and the normalized difference turbidity index;

[0073] For example, in some specific embodiments, the formula for calculating the Normalized Difference Water Index (MNDWI) can be expressed as:

[0074] MNDWI=(Green - MIR) / (Green + MIR);

[0075] Green and MIR represent the reflectivity of the green and mid-infrared bands, respectively.

[0076] NDCI (Normalized Differential Chlorophyll Index): NDCI = (B5 - B4) / (B5 + B4);

[0077] NDTI (Normalized Differential Turbidity Index): NDTI = (B4 - B3) / (B4 + B3);

[0078] Among them, B3, B4, and B5 represent the green band, red band, and near-infrared band, respectively.

[0079] It should be noted that all parameters used to calculate the normalized difference index were extracted from remote sensing images.

[0080] Step S300: Based on the normalized differential water body index and multi-source data processing, the river parameters of the target study area are obtained, and then the river runoff is obtained through hydraulic geometry transformation.

[0081] It should be noted that river parameters include river width, river depth, and river flow velocity. In some embodiments, such as... Figure 3 As shown, the river parameters of the target study area are obtained based on the Normalized Difference Water Index and multi-source data processing, which may include the following steps: S310, determine the river water body of the target study area based on the Normalized Difference Water Index, and obtain the river width through river water body quantification; S320, obtain the river depth based on the digital elevation model of the river water body area; S330, determine the river slope according to the digital elevation model, and obtain the river flow velocity based on the river slope using the Manning formula.

[0082] Specifically, this invention uses the Normalized Differential Water Index (NDPI) and Digital Elevation Model (DEM) to directly or indirectly calculate the three core hydraulic parameters of river width, depth, and flow velocity. This changes the past reliance on contact-based measurements using shipborne ADCP (Acoustic Doppler Current Profiler) and other equipment, enabling fully automated, contactless measurements based on remote sensing and geographic information system (GIS) technologies. This approach is not only extremely low-cost but also unrestricted by the deployment conditions of measurement equipment. Furthermore, this invention obtains width, depth, and flow velocity systematically and explicitly through physical mechanisms (such as using the Manning formula), giving subsequent runoff calculations based on hydraulic geometry clear physical meaning and higher reliability, directly supporting the accuracy of organic carbon flux calculations.

[0083] It should be noted that in some embodiments, such as Figure 4 As shown, step S310 may include the following steps: S311, extracting river water from remote sensing images based on the normalized differential water index at various locations in the target study area by using a preset water threshold; S312, vectorizing the river water, generating a profile line at a position perpendicular to the river centerline based on the result of the vectorization, and then quantifying the river width.

[0084] Specifically, this invention provides an objective, repeatable, and effective method for quantifying river width by using a "preset water body threshold" and "generating profile lines perpendicular to the river centerline after vectorization processing." Compared to simple pixel counting or visual interpretation, this method is more accurate and is especially suitable for small and medium-sized rivers with frequent width changes, ensuring the accuracy of the basic data source.

[0085] For example, in some specific implementations, the improved Normalized Difference Water Index (MNDWI) can be calculated using Sentinel-2 imagery to extract water body types. River water bodies are extracted and vectorized by setting a threshold, generating profile lines perpendicular to the river centerline, and then the river width is obtained.

[0086] It should be noted that in some embodiments, such as Figure 5 As shown, step S320 may include the following steps: S321, based on the range corresponding to the river water body, extract the raster of the river water surface area from the digital elevation model by means of mask extraction; S322, based on the raster, use a preset statistical tool to obtain the average elevation of each independent river segment unit as the river depth.

[0087] Specifically, the embodiments of the present invention can quickly obtain depth information of a large-scale river area through digital elevation models and methods such as "mask extraction" and "statistical average elevation". It can also serve as an effective estimation method at the regional scale. It provides a feasible alternative to solve the problem of "river depth", a key parameter that is difficult to obtain on a large scale, and is a crucial link in realizing "inversion without measured data".

[0088] For example, in some specific implementations, based on the river water body range of the aforementioned steps, the "mask extraction" method is used to extract the raster of the river water surface area from the DEM, and the "regional statistics" tool is used to calculate the average elevation for each independent river segment cell to determine the river depth.

[0089] For example, in some specific implementations, the river slope can be calculated using a DEM (Digital Elevation Model), and then the river velocity can be obtained using the Manning formula. Manning formula: V = Where V is the river flow velocity, n is the roughness coefficient, R is the hydraulic radius, and S is the slope of the canal (i.e., the river channel slope).

[0090] It should be noted that in some embodiments, such as Figure 6As shown, the process of obtaining river runoff through hydraulic geometry can include the following steps: S340, establishing an empirical power law relationship between river parameters and river runoff based on preset empirical parameters through hydraulic geometry; S350, obtaining river runoff by fitting the product of river width, river depth, river velocity and empirical parameters based on the empirical power law relationship.

[0091] Specifically, this embodiment of the invention links easily obtainable river morphology parameters (width, depth, velocity) with river runoff that is difficult to obtain directly through empirical parameters. This method has a solid physical foundation, efficient and stable calculation process, avoids the complex hydrological model calibration process, and can effectively realize rapid estimation of runoff in a large area and in the context of multiple rivers.

[0092] For example, in some specific implementations, an empirical power-law relationship between flow velocity (V), river depth (D), river width (W), and flow rate (Q) can be established through hydraulic geometry. The fitting formula for river runoff is:

[0093] Q=a*W*D*V

[0094] Where Q, W, D and V are runoff, river width, river depth and flow velocity, respectively; a is an empirical parameter.

[0095] Step S400: Based on remote sensing images, normalized differential chlorophyll index and normalized differential turbidity index, the particulate organic carbon concentration is obtained by inversion using a preset first inversion model.

[0096] It should be noted that in some embodiments, such as Figure 7 As shown, step S400 may include the following steps: S410, determining chlorophyll based on the normalized differential chlorophyll index, and determining suspended matter based on the normalized differential turbidity index; S420, using chlorophyll and suspended matter as environmental factors; S430, extracting raw band reflectance from remote sensing images; S440, inputting the environmental factors and raw band reflectance of all locations within the river basin into the first inversion model, and outputting the spatial distribution of particulate organic carbon concentration across the entire river basin; wherein, the first inversion model is obtained by training a random forest model based on preset training data, and the preset training data is constructed based on the measured particulate organic carbon concentration and environmental factors corresponding to some points within the river basin.

[0097] Specifically, this invention introduces chlorophyll and suspended matter as environmental factors and utilizes a random forest model to capture the complex nonlinear relationship between POC and multispectral features and environmental factors. The core advantage of this invention is that it only requires limited field measurements at representative locations within the watershed to train the model, which can then be successfully extended to the entire watershed, achieving spatially comprehensive POC concentration retrieval. This significantly reduces the reliance on large-scale, intensive field water quality sampling, representing a key technological innovation to address the high cost issue. Furthermore, this invention incorporates chlorophyll and suspended matter information through the Normalized Difference Index (NDCI, NDTI), which are direct environmental factors affecting POC concentration. This makes the retrieval mechanism clearer, rather than a simple "black box" operation, thus improving the interpretability and reliability of the results.

[0098] For example, in some specific implementations, the particulate organic carbon (POC) concentration inversion model uses Sentinel-2 satellite remote sensing data combined with environmental factors such as chlorophyll and suspended matter to construct a machine learning model to invert particulate organic carbon concentration, and combines it with measured data for training and verification.

[0099] Chlorophyll was retrieved using the red and near-infrared bands of Sentinel-2 via the NDCI (Normalized Differential Chlorophyll Index), while suspended matter was retrieved via the NDTI (Normalized Differential Turbidity Index).

[0100] NDCI (Normalized Differential Chlorophyll Index): NDCI = (B5 - B4) / (B5 + B4)

[0101] NDTI (Normalized Differential Turbidity Index): NDTI = (B4 - B3) / (B4 + B3)

[0102] Among them, B3, B4, and B5 represent the green band, red band, and near-infrared band, respectively.

[0103] Input: raw band reflectance, NDCI, NDTI of Sentinel-2 satellite remote sensing data, and actual sampled POC concentration (actual sampled POC concentration is used for model training and validation).

[0104] Model: Random Forest model;

[0105] Output: The trained model can be used to invert POC concentration based on the original band reflectance, NDCI, and NDTI to obtain a spatial distribution map of POC concentration (including POC concentration at various locations in the river basin).

[0106] Step S500: Based on population size and rainfall, the dissolved organic carbon concentration is obtained by using a preset second inversion model.

[0107] It should be noted that in some embodiments, step S500 may include the following steps: converting the population size into population density; using a multiple linear regression model to invert the dissolved organic carbon concentration based on the population density and rainfall; wherein the multiple linear regression model has been pre-fitted and learned to have a linear relationship between dissolved organic carbon concentration and rainfall and artificial density in the river basin.

[0108] Specifically, this invention proposes to utilize two readily available macroscopic indicators—population density (representing sources such as human activity and sewage discharge) and rainfall (representing natural sources such as surface erosion and leaching)—from publicly available databases, to invert DOC concentration using a multiple linear regression model. This provides a novel, low-cost, and physically interpretable technical approach for DOC concentration estimation. Furthermore, since population and rainfall data typically have long-term series and globally covered records, the method described in this invention is particularly suitable for DOC assessment and historical DOC concentration reconstruction in watersheds lacking data.

[0109] For example, in some specific implementations, a dissolved organic carbon (DOC) concentration inversion model is used to obtain population data of a river basin to calculate population density, and then combine it with rainfall product images to fit the dissolved organic carbon concentration of the river.

[0110] Since the concentration of dissolved organic carbon in rivers has a significant linear relationship with rainfall and population density in the river basin, this study requires the preparation of measured DOC concentration, rainfall, and population density, and the use of a multiple linear regression model to invert the concentration of dissolved organic carbon in rivers.

[0111] The mathematical model fitted by the multiple linear regression model is: DOC = β0 + β1 × (population density) + β2 × (rainfall) + ε; β0 is the constant term of the regression model, which represents the baseline value of the predicted DOC concentration when all independent variables (population density and rainfall) are zero; β1 and β2 represent the regression coefficients of population density and rainfall; ε is the error term / residual.

[0112] Step S600: The river organic carbon flux is obtained by converting the river runoff, particulate organic carbon concentration, and dissolved organic carbon concentration.

[0113] For example, in some specific embodiments, the river organic carbon flux is calculated based on the river runoff and the river organic carbon concentration. The river organic carbon flux is the mass of particulate organic carbon flowing through a certain area within a certain period. The river's particulate organic carbon flux and dissolved organic carbon flux can be calculated based on the particulate organic carbon concentration, dissolved organic carbon concentration, and river runoff obtained from the above steps.

[0114] To explain in detail the principle of the technical solution of the present invention, the overall process of the present invention will be described below with reference to some specific embodiments. It is easy to understand that the following is an explanation of the technical principle of the present invention and should not be regarded as a limitation of the present invention.

[0115] Given the shortcomings of existing technologies, constructing a river organic carbon flux model based on multi-source remote sensing data to obtain the organic carbon flux of small and medium-sized rivers is a problem that needs to be addressed by existing technologies.

[0116] In view of this, this invention proposes an optimized algorithm for monitoring river organic carbon flux, addressing the limitations of existing methods for acquiring river organic carbon flux based on remote sensing and measured data, such as limitations in data acquisition, reliance on single river surface information, and the lack of measured data for small and medium-sized rivers. This algorithm fully utilizes the advantages of multi-source remote sensing to retrieve particulate organic carbon concentration, dissolved organic carbon concentration, and river runoff, effectively improving the detection efficiency and flexibility of river organic carbon flux. It is particularly suitable for monitoring local, large-scale, and sustainable changes in river organic carbon flux, solving several key problems of existing technologies.

[0117] like Figure 8 As shown, the technical solution of this invention can be implemented through the following process:

[0118] Multi-source data for the study area were collected and preprocessed, primarily including river basin extents, Sentinel-2 series remote sensing images, DEM data, population data, and rainfall product images. River basin data was cropped using ArcMap software, and cloud cover filtering and cloud masking were performed on the Sentinel-2 series remote sensing images as needed. Note that L2A level Sentinel-2 series remote sensing images were selected; L1C level images require atmospheric correction before cloud cover filtering.

[0119] River runoff was calculated using optical remote sensing and a digital imagery (DEM). First, for river width, the improved Normalized Difference Water Index (MNDWI) was calculated using Sentinel-2 imagery to extract water body types. River water was extracted and vectorized by setting a threshold, generating a profile line perpendicular to the river centerline, and then the river width was obtained. Second, for river depth, based on the river water extent from the previous steps, a "mask extraction" method was used to extract the raster of the river surface area from the DEM. The "regional statistics" tool was used to calculate the average elevation for each independent river segment cell. Finally, for river velocity, the river slope was calculated using the DEM, and then the Manning formula was used to obtain the river velocity. Ultimately, hydraulic geometry established an empirical power-law relationship between velocity (V), river depth (D), river width (W), and flow rate (Q).

[0120] 1) The formula for calculating MNDWI can be expressed as:

[0121] MNDWI = (Green - MIR) / (Green + MIR)

[0122] Green and MIR represent the reflectivity of the green and mid-infrared bands, respectively.

[0123] 2) Manning's formula: V = ;

[0124] Where V is the river flow velocity, n is the roughness coefficient, R is the hydraulic radius, and S is the slope of the canal (i.e., the river channel slope).

[0125] 3) The fitting formula for river runoff is:

[0126] Q=a*W*D*V

[0127] Where Q, W, D and V are runoff, river width, river depth and flow velocity, respectively; a is an empirical parameter.

[0128] The particulate organic carbon (POC) concentration inversion model uses Sentinel-2 satellite remote sensing data combined with environmental factors such as chlorophyll and suspended matter to construct a machine learning model to invert particulate organic carbon concentration, and combines it with measured data for training and validation.

[0129] Chlorophyll was retrieved using the red and near-infrared bands of Sentinel-2 via the NDCI (Normalized Differential Chlorophyll Index), while suspended matter was retrieved via the NDTI (Normalized Differential Turbidity Index).

[0130] NDCI (Normalized Differential Chlorophyll Index): NDCI = (B5 - B4) / (B5 + B4)

[0131] NDTI (Normalized Differential Turbidity Index): NDTI = (B4 - B3) / (B4 + B3)

[0132] Among them, B3, B4, and B5 represent the green band, red band, and near-infrared band, respectively.

[0133] Input: raw band reflectance, NDCI, NDTI, and actual sampled POC concentration from Sentinel-2 satellite remote sensing data;

[0134] Model: Random Forest model;

[0135] Output: Spatial distribution map of POC concentration.

[0136] The dissolved organic carbon (DOC) concentration inversion model obtains population data from river basins to calculate population density, and then combines rainfall product images to fit the dissolved organic carbon concentration of the river.

[0137] Since the concentration of dissolved organic carbon in rivers has a significant linear relationship with rainfall and population density in the river basin, this study requires the preparation of measured DOC concentration, rainfall, and population density, and the use of a multiple linear regression model to invert the concentration of dissolved organic carbon in rivers.

[0138] Model: DOC = β0 + β1 × (population density) + β2 × (rainfall) + ε.

[0139] River organic carbon flux is calculated based on river runoff and river organic carbon concentration. River organic carbon flux is the mass of particulate organic carbon flowing through a certain area within a specific period. The particulate organic carbon concentration, dissolved organic carbon concentration, and river runoff obtained from the above steps are used to calculate the river's particulate organic carbon flux and dissolved organic carbon flux.

[0140] It should be noted that while traditional methods for monitoring river organic carbon flux use remote sensing imagery to retrieve some parameters, they fail to leverage the advantages of multi-source remote sensing imagery. Some key parameters still rely on measured data, resulting in high data acquisition costs and susceptibility to limitations imposed by the performance and deployment of measurement equipment. In contrast, this invention utilizes multi-source remote sensing data, including optical remote sensing imagery, DEM, and product imagery, to obtain river organic carbon flux using a fitting formula. This fully leverages machine learning algorithms based on remote sensing imagery to acquire multi-time-period river organic carbon flux information, improving application efficiency. Furthermore, traditional methods for calculating river runoff only use optical imagery to obtain river width information and then directly calculate runoff based on simple cross-sectional information, failing to fully consider the complexity of river cross-sections and leading to significant calculation errors. To address this, this invention uses high-precision DEM data combined with Sentinel-2 series remote sensing imagery to create three-dimensional river data and then obtains river runoff, improving the accuracy of river runoff retrieval, especially for complex and long rivers.

[0141] In summary, this invention proposes using multi-source data to construct a three-dimensional model of a river, and then combining this model with river flow velocity to calculate river runoff. The technical solution of this invention reduces the influence of measured river data, and the use of high-resolution DEMs allows for the analysis of cross-sections of small and medium-sized rivers, improving application scope and computational efficiency. Furthermore, this invention adaptively uses multi-source product data to invert river organic carbon concentration: since the principal component influencing factors of particulate organic carbon and dissolved organic carbon in rivers differ, appropriate product images or statistical data are selected based on their characteristics to invert their concentrations, ultimately enabling accurate acquisition of river organic carbon flux.

[0142] like Figure 9 As shown, this embodiment of the invention also provides a river organic carbon flux inversion device 900 based on multi-source data, which can implement the above-described method. This device may include:

[0143] The first module 910 is used to acquire multi-source data on the river basin area in the target study area. The multi-source data includes remote sensing images, digital elevation models, population data, and rainfall data.

[0144] The second module 920 is used to obtain the normalized difference index based on the transformation of remote sensing images. The normalized difference index includes the normalized difference water index, the normalized difference chlorophyll index, and the normalized difference turbidity index.

[0145] The third module 930 is used to obtain river parameters of the target study area based on the normalized differential water body index and multi-source data processing, and then obtain the river runoff through hydraulic geometric relationship transformation.

[0146] The fourth module 940 is used to retrieve the particulate organic carbon concentration based on remote sensing images, normalized differential chlorophyll index, and normalized differential turbidity index using a preset first inversion model.

[0147] The fifth module 950 is used to invert the dissolved organic carbon concentration based on population size and rainfall using a preset second inversion model;

[0148] Module 6, 960, is used to convert river organic carbon flux based on river runoff, particulate organic carbon concentration, and dissolved organic carbon concentration.

[0149] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0150] This invention also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described above. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0151] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0152] like Figure 10 As shown, Figure 10 The hardware structure of an electronic device 1000 according to another embodiment is illustrated. The electronic device 1000 includes:

[0153] The processor 1001 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (aSIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present invention.

[0154] The memory 1002 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RaM). The memory 1002 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1002 and is called and executed by the processor 1001.

[0155] Input / output interface 1003 is used to implement information input and output;

[0156] The communication interface 1004 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0157] Bus 1005 transmits information between various components of the device (e.g., processor 1001, memory 1002, input / output interface 1003, and communication interface 1004);

[0158] The processor 1001, memory 1002, input / output interface 1003 and communication interface 1004 are connected to each other within the device via bus 1005.

[0159] The electronic device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0160] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0161] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0162] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0163] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0164] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0165] The present invention provides a method, apparatus, electronic device, storage medium, and program product for river organic carbon flux inversion based on multi-source data. This method acquires multi-source data covering the river basin area in the target study area, including remote sensing imagery, digital elevation models, population data, and rainfall data. Based on the remote sensing imagery, it obtains normalized difference indices, including normalized difference water index, normalized difference chlorophyll index, and normalized difference turbidity index. Based on the normalized difference water index and multi-source data processing, it obtains river parameters for the target study area, and then transforms these parameters into river runoff through hydraulic geometry. Based on the remote sensing imagery, normalized difference chlorophyll index, and normalized difference turbidity index, it inverts particulate organic carbon concentration using a preset first inversion model. Based on the population data and rainfall data, it inverts dissolved organic carbon concentration using a preset second inversion model. Finally, it converts the river organic carbon flux based on the river runoff, particulate organic carbon concentration, and dissolved organic carbon concentration. This invention, through the acquisition of multi-source data including remote sensing imagery, digital elevation models (DEMs), population data, and rainfall data, utilizes publicly available or remotely sensed conventional data. This significantly reduces or even eliminates the expensive and cumbersome field sampling and measurement work required by traditional methods, effectively saving manpower, resources, and time. Furthermore, by fusing DEMs, this invention can obtain parameters related to riverbed topography (such as depth and slope). By incorporating population data, it can indirectly reflect the impact of human activities on dissolved organic carbon. This combination of multi-source data overcomes the limitation that "optical remote sensing imagery alone can only detect river surface information," achieving… A more comprehensive and three-dimensional characterization of the river carbon cycle process can significantly improve the accuracy and reliability of monitoring in highly dynamic rivers. In particular, the embodiments of this invention determine the organic carbon concentration through inversion. By calling up multi-source data from historical periods, the organic carbon flux of that historical period can be obtained through inversion, thus effectively solving the problem of historical reconstruction and providing a feasible technical path for assessing the carbon flux of a large number of unmonitored rivers worldwide. Specifically, through model inversion, this invention can obtain parameters such as particulate organic carbon concentration and runoff that are spatially continuous throughout the entire river basin, thereby more accurately calculating the organic carbon flux at the outlet section of the entire basin and avoiding the errors that may be caused by substituting a point for a whole area.

[0166] The embodiments described in this invention are for the purpose of more clearly illustrating the technical solutions of the embodiments of this invention, and do not constitute a limitation on the technical solutions provided by the embodiments of this invention. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this invention are also applicable to similar technical problems.

[0167] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present invention, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0168] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0169] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0170] The preferred embodiments of the present invention have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present invention. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and spirit of the present invention should be within the scope of the claims of the present invention.

Claims

1. A method for inverting river organic carbon flux based on multi-source data, characterized in that, The method includes the following steps: Acquire multi-source data on the river basin area in the target study area, including remote sensing images, digital elevation models, population data, and rainfall data. The normalized difference index is obtained based on the transformation of the remote sensing image. The normalized difference index includes the normalized difference water index, the normalized difference chlorophyll index, and the normalized difference turbidity index. Based on the normalized differential water body index and the multi-source data processing, the river parameters of the target study area are obtained, and then the river runoff is obtained through hydraulic geometry transformation. The river parameters include river width, river depth, and river flow velocity. The process of obtaining the river parameters for the target study area based on the normalized differential water index and the multi-source data processing includes the following steps: The river water body in the target study area is determined based on the normalized differential water body index, and the river width is obtained by quantifying the river water body. The river depth is obtained statistically based on the digital elevation model of the area where the river body is located; The river slope is determined based on the digital elevation model, and the river flow velocity is calculated using the Manning formula based on the river slope. Based on the remote sensing image, the normalized differential chlorophyll index, and the normalized differential turbidity index, the particulate organic carbon concentration is obtained by inversion using a preset first inversion model. Based on the population size and rainfall, the dissolved organic carbon concentration is obtained by using a preset second inversion model. River organic carbon flux is obtained by converting the river runoff, the particulate organic carbon concentration, and the dissolved organic carbon concentration.

2. The method according to claim 1, characterized in that, The process of determining the river water body in the target study area based on the normalized differential water body index, and obtaining the river width through the quantification of the river water body, includes the following steps: Based on the normalized differential water body index at various locations in the target study area, river water bodies are extracted from the remote sensing images by using a preset water body threshold. The river water body is vectorized, and a profile line is generated at a position perpendicular to the river centerline based on the result of the vectorization process, thereby quantifying the river width.

3. The method according to claim 1, characterized in that, The method for obtaining the river depth based on the digital elevation model of the river's location includes the following steps: Based on the range corresponding to the river water body, the raster of the river water surface area is extracted from the digital elevation model by mask extraction. Based on the grid, the average elevation of each independent river segment unit is obtained using a preset statistical tool and is taken as the river depth.

4. The method according to claim 1, characterized in that, The river parameters include river width, river depth, and river flow velocity. The process of converting these parameters through hydraulic geometry to obtain the river runoff includes the following steps: Based on preset empirical parameters, an empirical power-law relationship between the river parameters and the river runoff is established through the hydraulic geometry. Based on the empirical power law relationship, the river runoff is obtained by fitting the product of the river width, the river depth, the river velocity, and the empirical parameter.

5. The method according to claim 1, characterized in that, The step of obtaining particulate organic carbon concentration by inverting the remote sensing image, the normalized differential chlorophyll index, and the normalized differential turbidity index using a preset first inversion model includes the following steps: Chlorophyll content is determined based on the normalized differential chlorophyll index, and suspended matter is determined based on the normalized differential turbidity index. The chlorophyll and the suspended matter are considered as environmental factors. Extract the original band reflectance from the remote sensing image; The environmental factors and the original band reflectance at all locations within the river basin are input into the first inversion model, and the spatial distribution of the particulate organic carbon concentration across the entire river basin is output. The first inversion model is obtained by training a random forest model based on preset training data, which is constructed based on the measured particulate organic carbon concentration at some points in the river basin and the environmental factors.

6. The method according to claim 1, characterized in that, The process of obtaining dissolved organic carbon concentration based on the population size and rainfall using a preset second inversion model includes the following steps: Convert the population size into population density; Based on the population density and the rainfall, the dissolved organic carbon concentration was obtained by inversion using a multiple linear regression model. The multiple linear regression model was pre-fitted and learned to have a linear relationship between dissolved organic carbon concentration and rainfall and population density in the river basin.

7. A river organic carbon flux inversion device based on multi-source data, characterized in that, The device includes: The first module is used to acquire multi-source data on the river basin area in the target study area, including remote sensing images, digital elevation models, population data, and rainfall data. The second module is used to obtain the normalized difference index based on the remote sensing image, the normalized difference index including the normalized difference water index, the normalized difference chlorophyll index and the normalized difference turbidity index. The third module is used to obtain the river parameters of the target study area based on the normalized differential water body index and the multi-source data processing, and then obtain the river runoff through hydraulic geometry transformation. The river parameters include river width, river depth, and river flow velocity. When the third module performs the step of obtaining the river parameters of the target study area based on the normalized differential water index and the multi-source data processing, it performs the following operations: The river water body in the target study area is determined based on the normalized differential water body index, and the river width is obtained by quantifying the river water body. The river depth is obtained statistically based on the digital elevation model of the area where the river body is located; The river slope is determined based on the digital elevation model, and the river flow velocity is calculated using the Manning formula based on the river slope. The fourth module is used to retrieve the particulate organic carbon concentration based on the remote sensing image, the normalized differential chlorophyll index, and the normalized differential turbidity index using a preset first inversion model. The fifth module is used to obtain the dissolved organic carbon concentration by using a preset second inversion model based on the population size and the rainfall. The sixth module is used to convert the river organic carbon flux based on the river runoff, the particulate organic carbon concentration, and the dissolved organic carbon concentration.

8. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 6.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 6.

Images