Land surface temperature time series reconstruction method and system considering energy distribution mechanism

Abstract

This application provides a method and system for temporal reconstruction of land surface temperature considering energy distribution mechanisms. The method includes acquiring an energy feature dataset of a target area, wherein the energy feature dataset includes at least one of the following: temporally continuous land surface temperature data, net radiation data, latent heat flux data, sensible heat flux data, albedo data, and snow depth data for each pixel in the target area; performing abrupt change detection on the energy feature dataset based on the energy distribution of each pixel to obtain an identifier matrix corresponding to different stage types, wherein the stage types include snow cover stage, snowmelt stage, and snowless stage; performing regression fitting on the identifier matrix of any stage type to obtain the land surface temperature sequence of the stage type; generating a temporal optimization model based on constraints and the land surface temperature sequences of different stage types; and solving the objective function of the temporal optimization model to obtain the temporal land surface temperature.

Description

Technical Field

[0001] This application relates to the fields of data processing and remote sensing technology, and more specifically, to a method and system for temporal reconstruction of land surface temperature considering energy distribution mechanisms. Background Technology

[0002] Land surface temperature (LST) is a crucial parameter characterizing energy exchange between the Earth and the atmosphere and the thermal state of the Earth's surface. It is widely used in climate change research, drought monitoring, energy balance analysis, and ecological environment assessment. LST not only reflects the thermal condition and energy transfer processes of the Earth's surface but is also a core variable in the Earth's radiation budget and land-atmosphere interactions. However, obtaining and reconstructing LST has long faced challenges in terms of spatiotemporal continuity and physical consistency, especially in high-altitude, frigid, and icy regions, as well as areas with dramatic changes in energy distribution, where the assumptions of traditional models often fail. Summary of the Invention

[0003] In view of this, this application provides a method and system for temporal reconstruction of land surface temperature that takes into account energy distribution mechanisms.

[0004] One aspect of this application provides a method for temporal reconstruction of land surface temperature considering energy distribution mechanisms, comprising: in response to a temperature reconstruction command, acquiring an energy feature dataset of a target area, wherein the energy feature dataset includes at least one of temporally continuous land surface temperature data, net radiation data, latent heat flux data, sensible heat flux data, albedo data, and snow depth data for each pixel in the target area; performing abrupt change detection on the energy feature dataset based on the energy distribution of each pixel to obtain an identifier matrix corresponding to different stage types, wherein the stage types include a snow cover period, a snowmelt period, and a snowless period; performing regression fitting on the data corresponding to the identifier matrix in the energy feature dataset for any type of stage to obtain a land surface temperature sequence for that type of stage; generating a temporal optimization model based on constraints and the land surface temperature sequences of different types of stages; and solving the objective function of the temporal optimization model to obtain the temporal land surface temperature.

[0005] According to an embodiment of this application, mutation detection is performed on the energy feature dataset based on the energy allocation of each pixel in the target region to obtain an identifier matrix corresponding to different stage types. This includes: for each pixel, determining the energy allocation weight sequence of the pixel based on the latent heat flux data and sensible heat flux data of the pixel; classifying the snow depth data of the pixel based on the snow depth threshold to obtain a snow mask sequence characterizing the snow depth classification type of the pixel; and performing mutation detection on the energy feature dataset based on the energy allocation weight sequence and the snow mask sequence to obtain an identifier matrix corresponding to different stage types.

[0006] According to an embodiment of this application, mutation detection is performed on the energy feature dataset based on the energy allocation weight sequence and the snow mask sequence to obtain an identifier matrix corresponding to different stage types. This includes: performing mutation detection on the energy allocation weight sequence to obtain a mutation detection result, wherein the mutation detection result characterizes the location of the turning point where energy mutations occur; dividing the energy allocation weight sequence into multiple stage types and boundary moments between different stage types based on the mutation detection result and the snow depth data; and generating a stage division result matrix based on the boundary moments, according to the mutation detection result and the snow mask sequence, wherein the stage division result matrix includes an identifier matrix for each stage type corresponding to each pixel.

[0007] According to embodiments of this application, the aforementioned energy feature dataset includes data subsets corresponding to different stage types.

[0008] According to an embodiment of this application, regression fitting is performed on the data corresponding to the identifier matrix in the energy feature dataset to obtain the surface temperature sequence of the type stage, including: when the type stage is the snow cover period, regression modeling is performed using the net radiation data, surface temperature data, and latent heat flux data in the data subset corresponding to the snow cover period to obtain an energy-constrained regression model, wherein the energy-constrained regression model is a model dominated by latent heat processes with surface temperature as the dependent variable; the surface temperature sequence of the snow cover period is obtained based on the energy-constrained regression model.

[0009] According to an embodiment of this application, the method further includes: when the above-mentioned type stage is the snowmelt period stage, constructing a piecewise linear regression model using net radiation data from the data subset corresponding to the above-mentioned snowmelt period stage; and obtaining the surface temperature sequence of the above-mentioned snowmelt period stage based on the above-mentioned piecewise linear regression model.

[0010] According to embodiments of this application, the aforementioned energy feature dataset also includes vegetation indices and digital elevation models.

[0011] According to an embodiment of this application, the method further includes: when the above-mentioned type stage is a snowless period, constructing a residual fitting model based on a target regression method using net radiation data, sensible heat flux data, land surface temperature data, vegetation index, albedo data, and digital elevation model from the data subset corresponding to the above-mentioned snowless period, wherein the above-mentioned target regression method is obtained by fusing a geographic weighted regression method and a random forest regression method; and obtaining the land surface temperature sequence of the above-mentioned snowless period based on the above-mentioned residual fitting model.

[0012] According to an embodiment of this application, a time-series optimization model is generated based on constraints and surface temperature sequences of different types of stages, including: for any pixel, constructing an initial optimization function based on continuity constraints and surface temperature sequences of different types of stages; determining an energy residual constraint term based on the conservation of land-atmosphere energy constraints, using net radiation data, latent heat flux data, and sensible heat flux data; and generating a target optimization function corresponding to the pixel based on the initial optimization function and the energy residual constraint term.

[0013] According to an embodiment of this application, the objective optimization function of the above-mentioned time-series optimization model is solved to obtain the time-series land surface temperature, including: for any pixel, the objective optimization function is solved using the weighted least squares method or the variational method to obtain the land surface detection temperature; the land surface detection temperatures of multiple pixels are combined based on the identification matrix of different stage types to obtain the above-mentioned time-series land surface temperature.

[0014] According to an embodiment of this application, obtaining an energy feature dataset of a target area includes: obtaining an original dataset and cloud mask data of the target area, wherein the original dataset includes hourly surface remote sensing data at a first resolution, original surface temperature data at a second resolution, ground meteorological data, albedo data, and snow depth data, and the ground meteorological data includes temperature, humidity, and net radiation data; filtering the original dataset using the cloud mask data to obtain a filtered original dataset; resampling the surface remote sensing data and original surface temperature data in the filtered original dataset to obtain an intermediate dataset including surface remote sensing data at the target resolution and original surface temperature data; and preprocessing the intermediate dataset to obtain the energy feature dataset, wherein the preprocessing includes correcting slope aspect, the influence of slope on radiation, interpolation recovery when data is missing, and calculating latent heat flux data and sensible heat flux data based on the land-atmosphere energy balance equation.

[0015] Another aspect of this application provides a land surface temperature time-series reconstruction system considering energy allocation mechanisms, comprising: an acquisition module, configured to acquire an energy feature dataset of a target area in response to a temperature reconstruction command, wherein the energy feature dataset includes at least one of time-continuous land surface temperature data, net radiation data, latent heat flux data, sensible heat flux data, albedo data, and snow depth data for each pixel in the target area; a detection module, configured to perform abrupt change detection on the energy feature dataset based on the energy allocation of each pixel to obtain an identifier matrix corresponding to different stage types, wherein the stage types include a snow cover period, a snowmelt period, and a snowless period; a fitting module, configured to perform regression fitting on the data corresponding to the identifier matrix in the energy feature dataset for any type of stage to obtain a land surface temperature sequence for that type of stage; a generation module, configured to generate a time-series optimization model based on constraints and the land surface temperature sequences of different types of stages; and a solution module, configured to solve the objective optimization function of the time-series optimization model to obtain the time-series land surface temperature.

[0016] Another aspect of this application provides an electronic device comprising: one or more processors; and a memory for storing one or more programs, wherein, when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to perform the method as described above.

[0017] Another aspect of this application provides a computer-readable storage medium storing computer-executable instructions that, when executed, are used to implement the method described above.

[0018] Another aspect of this application provides a computer program product comprising computer-executable instructions which, when executed, are used to implement the method described above.

[0019] According to embodiments of this application, abrupt changes in the energy feature dataset are detected based on the energy allocation of each pixel to obtain identifier matrices for the corresponding snow cover period, snowmelt period, and snowless period. Regression fitting is performed on the data in the energy feature dataset corresponding to the identifier matrices to obtain the surface temperature sequence for each type of stage. A time-series optimization model is generated based on constraints and the surface temperature sequences for different types of stages. The objective function of the time-series optimization model is solved to obtain the time-series surface temperature. By introducing an energy allocation mechanism to guide stage division and inter-stage connections, the reconstruction results of surface temperature during energy abrupt changes are significantly enhanced. The method based on stage identification and joint optimization of cross-stage physical constraints can handle abrupt changes in energy allocation caused by seasonal changes in snow cover, thereby improving the accuracy and physical consistency of the time-series reconstruction of surface temperature. Attached Figure Description

[0020] The above and other objects, features and advantages of this application will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0021] Figure 1 An exemplary system architecture for applying a land surface temperature temporal reconstruction method that takes into account energy distribution mechanisms, according to an embodiment of this application, is shown;

[0022] Figure 2 A flowchart of a land surface temperature temporal reconstruction method considering energy distribution mechanisms according to an embodiment of this application is shown;

[0023] Figure 3 A flowchart illustrating the generation process of identifier matrices for different stage types according to embodiments of this application is shown;

[0024] Figure 4 A block diagram of a land surface temperature temporal reconstruction system considering energy distribution mechanisms according to an embodiment of this application is shown;

[0025] Figure 5 A block diagram of an electronic device suitable for implementing the methods described above, according to an embodiment of this application, is shown. Detailed Implementation

[0026] The embodiments of this application will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of this application. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of this application for ease of explanation. However, it will be apparent that one or more embodiments may be implemented without these specific details. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of this application.

[0027] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The terms “comprising,” “including,” etc., as used herein indicate the presence of features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0028] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0029] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0030] In the embodiments of this application, the collection, updating, analysis, processing, use, transmission, provision, disclosure, and storage of data (e.g., including but not limited to user personal information) comply with relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. In particular, necessary measures have been taken to prevent unauthorized access to user personal information data and to safeguard user personal information security and network security.

[0031] In the natural environment, the energy distribution mechanism of the Earth's surface exhibits significant spatiotemporal nonstationarity. With seasonal changes, snowmelt, and variations in surface humidity, the ratio of latent heat to sensible heat fluxes changes significantly, leading to a periodic shift in surface energy processes from "sensible heat dominance" to "latent heat dominance." This change not only affects the temporal smoothness of LST (Low Temperature Series) but also causes systematic biases in conventional time series models based on the stationarity assumption. Currently, common methods for reconstructing LST time series fall into two main categories: one is spatiotemporal interpolation and empirical fitting based on multi-source data, achieving continuity through time weighting and correction for topographic / cloud effects, but it does not explicitly characterize the changes in latent / sensible heat energy distribution, making it prone to systematic biases during snowmelt and freeze-thaw periods; the other is time series segmented modeling based on abrupt change detection (such as piecewise regression and BFAST), which can identify stages but mostly remains at the statistical level, lacking physical constraints on net radiation, latent heat, and sensible heat fluxes, making it difficult to guarantee continuity and interpretability across stages. Therefore, there is an urgent need for an LST time series reconstruction method and system that considers the energy allocation mechanism, introduces energy weight and stage identification linkage, and applies continuity and energy conservation constraints at the stage boundaries to improve the accuracy and stability in non-stationary scenarios such as ice and snow / freeze-thaw.

[0032] The aforementioned related technologies have two main technical defects:

[0033] First, existing methods for reconstructing land surface temperature time series are mostly based on the assumption of stationarity and a single model. They rely on time weighting / topographic cloud correction to make the data continuous, ignoring the abrupt changes in latent heat / sensible heat caused by seasonal changes in snow and ice. This results in large deviations between snowmelt and freeze-thaw periods, discontinuous time series, and poor splicing across stages.

[0034] Second, existing mutation detection segments mostly remain at the statistical level and do not incorporate energy constraints such as net radiation, latent heat, and sensible heat, making it difficult to respond to rapid changes in energy mechanisms. This results in insufficient cross-stage continuity and physical consistency, weak interpretability, and limited robustness.

[0035] In view of this, embodiments of this application provide a method and system for temporal reconstruction of land surface temperature considering energy distribution mechanisms. The method includes acquiring an energy feature dataset of a target area, wherein the energy feature dataset includes at least one of temporally continuous land surface temperature data, net radiation data, latent heat flux data, sensible heat flux data, albedo data, and snow depth data for each pixel in the target area; performing abrupt change detection on the energy feature dataset based on the energy distribution of each pixel to obtain an identifier matrix corresponding to different stage types, wherein the stage types include snow cover stage, snowmelt stage, and snowless stage; performing regression fitting on the data corresponding to the identifier matrix in the energy feature dataset for any type of stage to obtain a land surface temperature sequence for that type of stage; generating a temporal optimization model based on constraints and the land surface temperature sequences of different types of stages; and solving the objective function of the temporal optimization model to obtain the temporal land surface temperature.

[0036] Figure 1 An exemplary system architecture for applying a land surface temperature temporal reconstruction method considering energy distribution mechanisms, according to embodiments of this application, is shown. It should be noted that... Figure 1 The examples shown are merely examples of system architectures that can be applied to the embodiments of this application, in order to help those skilled in the art understand the technical content of this application, but do not mean that the embodiments of this application cannot be used in other devices, systems, environments or scenarios.

[0037] like Figure 1 As shown, the system architecture 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired and / or wireless communication links, etc.

[0038] Users can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 via the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, and / or social media platform software, etc. (for example only).

[0039] The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be various electronic devices with displays and support web browsing, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0040] Server 105 can be a server that provides various services, such as a backend management server that supports websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103 (this is just an example). The backend management server can analyze and process data such as received user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal devices.

[0041] It should be noted that the land temperature time-series reconstruction method considering energy distribution mechanisms provided in this application embodiment can generally be executed by server 105. Correspondingly, the land temperature time-series reconstruction system considering energy distribution mechanisms provided in this application embodiment can generally be set up in server 105. The land temperature time-series reconstruction method considering energy distribution mechanisms provided in this application embodiment can also be executed by a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105. Correspondingly, the land temperature time-series reconstruction system considering energy distribution mechanisms provided in this application embodiment can also be set up in a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105. Alternatively, the land surface temperature time-series reconstruction method considering energy distribution mechanisms provided in this application embodiment can also be executed by the first terminal device 101, the second terminal device 102, or the third terminal device 103, or by other terminal devices different from the first terminal device 101, the second terminal device 102, or the third terminal device 103. Correspondingly, the land surface temperature time-series reconstruction system considering energy distribution mechanisms provided in this application embodiment can also be configured in the first terminal device 101, the second terminal device 102, or the third terminal device 103, or in other terminal devices different from the first terminal device 101, the second terminal device 102, or the third terminal device 103.

[0042] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0043] Figure 2 A flowchart of a land surface temperature temporal reconstruction method considering energy distribution mechanisms according to an embodiment of this application is shown.

[0044] like Figure 2 As shown, the surface temperature time series reconstruction method considering energy distribution mechanisms includes operations S201~S205.

[0045] In operation S201, in response to the temperature reconstruction command, an energy feature dataset of the target area is acquired, wherein the energy feature dataset includes at least one of the following: temporally continuous surface temperature data, net radiation data, latent heat flux data, sensible heat flux data, albedo data, and snow depth data for each pixel in the target area.

[0046] In operation S202, mutation detection is performed on the energy feature dataset based on the energy allocation of each pixel to obtain the identification matrix corresponding to different stage types, including the snow cover period, snow melting period, and snowless period.

[0047] In operation S203, for any type stage's identifier matrix, regression fitting is performed on the data in the energy feature dataset corresponding to the identifier matrix to obtain the surface temperature sequence of the type stage.

[0048] In operation S204, a time-series optimization model is generated based on constraints and surface temperature sequences of different types of stages.

[0049] In operation S205, the objective function of the time series optimization model is solved to obtain the time series surface temperature.

[0050] According to embodiments of this application, the temperature reconstruction command can be input by an operator on an electronic device such as a mobile phone or computer, and the electronic device will respond to the operation to generate a corresponding temperature reconstruction command. The target area can be any geographical location, such as a plot of land in a city.

[0051] According to an embodiment of this application, the database can store the preprocessed energy feature dataset of each region, and in response to the temperature reconstruction command, the energy feature dataset of the target region corresponding to the temperature reconstruction command can be obtained from the database.

[0052] According to an embodiment of this application, since the parameters in the energy feature dataset are obtained based on remote sensing images and other data of the target area, the energy allocation of each pixel can be calculated based on the energy feature dataset of the target area. Then, abrupt change detection is performed on the energy feature dataset based on the energy allocation of each pixel to obtain an identification matrix corresponding to different stage types. The identification matrix is ​​composed of pixel index, time index and matrix elements. Different values ​​of the matrix elements reflect the corresponding stage types.

[0053] According to an embodiment of this application, for any one of the snow cover period, snow melt period, and snowless period, regression fitting is performed on samples in the energy feature dataset corresponding to the identifier matrix of that period (i.e., the data corresponding to the identifier matrix) to form a surface temperature sequence that is highly specific to that period.

[0054] According to the embodiments of this application, after obtaining the surface temperature sequences of various types of stages, multiple surface temperature sequences can be reconstructed into a time-series optimization model of the target area using different constraints. The corresponding time-series surface temperature can be obtained by solving the objective optimization function of the time-series optimization model.

[0055] According to embodiments of this application, abrupt changes in the energy feature dataset are detected based on the energy allocation of each pixel to obtain identifier matrices for the corresponding snow cover period, snowmelt period, and snowless period. Regression fitting is performed on the data in the energy feature dataset corresponding to the identifier matrices to obtain the surface temperature sequence for each type of stage. A time-series optimization model is generated based on constraints and the surface temperature sequences for different types of stages. The objective function of the time-series optimization model is solved to obtain the time-series surface temperature. By introducing an energy allocation mechanism to guide stage division and inter-stage connections, the reconstruction results of surface temperature during energy abrupt changes are significantly enhanced. The method based on stage identification and joint optimization of cross-stage physical constraints can handle abrupt changes in energy allocation caused by seasonal changes in snow cover, thereby improving the accuracy and physical consistency of the time-series reconstruction of surface temperature.

[0056] According to an embodiment of this application, a mutation detection is performed on the energy feature dataset based on the energy allocation of each pixel in the target area to obtain an identifier matrix corresponding to different stage types. This includes: for each pixel, determining the energy allocation weight sequence of the pixel based on the latent heat flux data and sensible heat flux data of the pixel; classifying the snow depth data of the pixel based on the snow depth threshold to obtain a snow mask sequence representing the snow depth classification type of the pixel; and performing mutation detection on the energy feature dataset based on the energy allocation weight sequence and the snow mask sequence to obtain an identifier matrix corresponding to different stage types.

[0057] According to an embodiment of this application, an energy allocation weight function for each pixel is calculated to quantify the relative contribution of latent heat flux data and sensible heat flux data. A larger value indicates that latent heat is dominant. The specific calculation of the surface energy allocation weight (i.e., the energy allocation weight sequence) is shown in formula (1):

[0058] (1)

[0059] Where i represents the i-th pixel, and t represents the t-th time; This represents latent heat flux data (W / m²), indicating the portion of energy consumed through phase change (evaporation, sublimation); The data represents the sensible heat flux (W / m²), indicating the portion of energy transferred via convection. Assign weights to energy, with values ​​ranging from 0 to 1. A larger value indicates that the latent heat process is dominant.

[0060] According to an embodiment of this application, preliminary snow cover identification is performed based on snow depth. This embodiment sets the snow depth threshold to 0.03m, which can be adjusted based on experience in different regions. The snow depth is then considered at each time point. The system determines whether there is snow cover; if the snow depth is above a threshold, the area is considered to have snow cover; otherwise, it is considered to have no snow cover. This process yields a binarized snow cover mask sequence. .

[0061] According to an embodiment of this application, mutation detection is performed on the energy feature dataset based on the energy allocation weight sequence and the snow mask sequence to obtain an identifier matrix corresponding to different stage types.

[0062] Figure 3 A flowchart illustrating the generation process of the identifier matrix for different stage types according to embodiments of this application is shown.

[0063] like Figure 3 As shown, mutation detection is performed on the energy feature dataset based on the energy allocation weight sequence and the snow mask sequence to obtain the identification matrix corresponding to different stage types, including operation S301 to operation S303.

[0064] In operation S301, mutation detection is performed on the energy allocation weight sequence to obtain mutation detection results, where the mutation detection results characterize the position of the inflection point where energy mutation occurs.

[0065] In operation S302, based on mutation detection results and snow depth data, the energy allocation weight sequence is divided into multiple stage types and the boundary moments between different stage types.

[0066] In operation S303, based on the boundary time, a stage division result matrix is ​​generated according to the mutation detection results and the snow cover mask sequence. The stage division result matrix includes an identifier matrix for the stage type corresponding to each pixel.

[0067] According to the embodiments of this application, the formula (1) described above is constructed as follows: Time series (i.e., energy distribution weighted series) are subjected to mutation detection (such as using BFAST or Pettitt test) to automatically identify energy mechanism inflection points and generate mutation detection results.

[0068] According to embodiments of this application, by combining mutation detection results and snow depth information, the time series is divided into three stages: the snow cover stage (stable snow cover, latent heat dominance), the snowmelt stage (rapid changes in energy mechanisms), and the snowless stage (no snow cover, sensible heat dominance). Simultaneously, the boundary moments between the snow cover and snowmelt stages, and between snowmelt and snowless stages, are extracted. , Ultimately, the results of mutation detection and snow masking sequences were used to determine the final outcome. Output stage partitioning result matrix , where i is the cell index, t is the time index, and the matrix elements take values ​​of 1, 2, and 3 to represent the snow cover period, snow melt period, and snowless period, respectively. That is, the identifier matrix for the period type corresponding to each cell.

[0069] According to embodiments of this application, the energy feature dataset includes data subsets corresponding to different stage types.

[0070] According to an embodiment of this application, regression fitting is performed on the data corresponding to the identifier matrix in the energy feature dataset to obtain the surface temperature sequence of the type stage, including: when the type stage is the snow cover period, regression modeling is performed using the net radiation data, surface temperature data, and latent heat flux data in the data subset corresponding to the snow cover period to obtain an energy-constrained regression model, wherein the energy-constrained regression model is a model dominated by latent heat processes with surface temperature as the dependent variable; the surface temperature sequence of the snow cover period is obtained based on the energy-constrained regression model.

[0071] According to embodiments of this application, for the snow accumulation period ( For the sample of =1), i.e., the data subset, with surface temperature as the dependent variable, an energy-constrained regression model dominated by latent heat processes is established. The regression coefficients are solved to obtain the linear mapping relationship between surface temperature and energy terms. The energy-constrained regression model is shown in formula (2):

[0072] (2)

[0073] Where i represents the i-th pixel, and t represents the t-th time; This represents net radiative flux data (W / m²). This represents latent heat flux data (W / m²). This represents near-surface air temperature data (K or °C), i.e., surface temperature data; regression residual term This is used to reflect random disturbances or local differences in observations that are not characterized by the model; a 1,i a 2,i a 3,i Represents the regression coefficient of pixel i; This is the surface temperature sequence during this snow cover period.

[0074] According to an embodiment of this application, the above method further includes: when the type stage is the snowmelt period, constructing a piecewise linear regression model using net radiation data from the data subset corresponding to the snowmelt period; and obtaining the surface temperature sequence of the snowmelt period based on the piecewise linear regression model.

[0075] According to embodiments of this application, the snow melting period stage ( The sample of =2), i.e., the data subset, considers the rapid response characteristics of the energy distribution mechanism during the snowmelt stage near the abrupt change point, using net radiation data R. n Using the core independent variable, a piecewise linear regression model was used to model the relationship between surface temperature and energy, quantifying the change in energy distribution mechanism before and after the abrupt change point. The sample was divided into two segments, and linear regression was performed on each segment separately. The specific piecewise linear regression model is shown in formula (3):

[0076] (3)

[0077] Where i represents the i-th pixel, and t represents the t-th time; b 1,i b 2,i , respectively, represent the regression slopes of pixel i before and after snowmelt, reflecting the intensity of the response of surface temperature to changes in net radiation; c 1,i , c 2,i The corresponding intercepts; This represents net radiation data (W / m²). This represents the critical net radiation value at which the energy mechanism shifts from latent heat dominance to sensible heat dominance. By independently modeling all pixels and applying continuity constraints (such as equality constraints and smoothing penalties) at the discontinuities of each pixel, the surface temperature sequence during the snowmelt period can be obtained. .

[0078] According to an embodiment of this application, the energy feature dataset further includes vegetation indices and digital elevation models; wherein, the above method further includes: when the type stage is a snowless stage, constructing a residual fitting model based on a target regression method using net radiation data, sensible heat flux data, land surface temperature data, vegetation indices, albedo data, and digital elevation models in the data subset corresponding to the snowless stage, wherein the target regression method is obtained by fusing geographically weighted regression and random forest regression methods; and obtaining the land surface temperature sequence for the snowless stage based on the residual fitting model.

[0079] According to embodiments of this application, for the snowless period ( For samples with a resolution of 3, a fusion strategy of Geographically Weighted Regression (GWR) and Random Forest (RF) can be adopted.

[0080] First, the GWR is used to perform a weighted regression of Land Surface Temperature (LST) with energy, meteorology, and surface factors within a local spatial window. The resulting linear estimate is shown in the terms within parentheses of formula (5). Then, the residuals of the GWR are calculated and input into the RF model as features of the nonlinear component. The RF model then fits the residuals, and the output is shown in formula (4). Finally, the linear term is compared with the residual fitting value of formula (4). Adding these together, we obtain the surface temperature sequence for the snowless period. The residual fitting model is shown in equation (5):

[0081] (4)

[0082] (5)

[0083] Where i represents the i-th pixel, and t represents the t-th time; It represents the net radiative flux (W / m²). Data for sensible heat flux (W / m²) This represents near-surface air temperature (K or °C), i.e., surface temperature data. Indicates vegetation index, Represents albedo data. Represents the elevation value determined by the digital elevation model; The local regression coefficients in GWR are represented by spatial weights. Weighted least squares solution, where K is the distance decay kernel function. For bandwidth, Let be the spatial distance between pixel i and its neighboring pixel j; This is the GWR residual term.

[0084] According to embodiments of this application, in the modeling process of the above-mentioned different stage types, by introducing constraints such as energy conservation, physical upper and lower limits and parameter smoothing, the fit robustness and physical consistency are jointly optimized and improved.

[0085] According to an embodiment of this application, a time-series optimization model is generated based on constraints and surface temperature sequences of different types of stages, including: for any pixel, constructing an initial optimization function based on continuity constraints and surface temperature sequences of different types of stages; determining energy residual constraint terms based on the conservation of land-atmosphere energy constraints, using net radiation data, latent heat flux data, and sensible heat flux data; and generating a target optimization function corresponding to the pixel based on the initial optimization function and the energy residual constraint terms.

[0086] According to an embodiment of this application, based on continuity constraints, an initial optimization function as shown in formula (6) is constructed using weighted least squares or variational methods and surface temperature sequences of different types of stages:

[0087] (6)

[0088] Where i represents the i-th pixel, and k represents the stage index. These correspond to the snow accumulation period, snow melting period, and snowless period, respectively. The summation to 2 indicates that only the connection between adjacent periods (1-2, 2-3) is considered. This represents the surface temperature sequence of pixel i at stage k; This represents the boundary time between stage k and stage (k+1). This represents the square of the temperature difference between adjacent stages at the boundary moment, used to constrain numerical continuity. The smaller this value, the smoother the transition between stages. The first-order time gradient representing surface temperature is used to suppress sudden jumps in time series. The smoothness term, representing the time gradient, is used to suppress abrupt changes or oscillations in the time series and ensure that the overall curve is smooth. This represents the smoothing weighting coefficient, which can be determined empirically or through cross-validation (range: 0.01-0.1).

[0089] According to an embodiment of this application, based on the above continuity constraint, in order to satisfy the energy balance relationship of the time series, a ground-atmosphere energy conservation constraint is introduced. The energy residual term reflects the deviation between the actual energy balance and the theoretical balance, and can thus be expressed by formula (7):

[0090] (7)

[0091] Where i represents the i-th pixel, and t represents the t-th time; It represents the net radiative flux (W / m²). This represents latent heat flux data (W / m²). Data for sensible heat flux (W / m²) For surface heat flux (W / m²), the energy feature dataset can include surface heat flux, which can be calculated using empirical formulas, such as... It can be estimated using an energy balance parameterization scheme.

[0092] According to the embodiments of this application, the energy residual constraint term of formula (7) is combined with the initial optimization function obtained by formula (6) based on the continuity constraint to form a unified time series optimization model. The first term ensures the continuity at the stage boundary, the second term ensures the overall smoothness of the sequence, and the third term introduces energy budget constraints to improve physical consistency. The objective optimization function of the time series optimization model is shown in formula (8):

[0093] (8)

[0094] in, This represents the time gradient of Earth's surface temperature, i.e., the rate of temperature change. The smoothing constraint term, representing the time gradient, is used to suppress abrupt changes or oscillations in the time series and ensure the overall curve is smooth. The constraint terms representing the energy residual items are obtained from formula (7); as well as These are all weighting coefficients, which control the time series smoothness and the energy conservation strength, respectively. They can be determined through empirical settings or cross-validation (value range 0.01-0.1).

[0095] According to an embodiment of this application, the objective optimization function of the time-series optimization model is solved to obtain the time-series land surface temperature, including: for any pixel, the objective optimization function is solved using the weighted least squares method or the variational method to obtain the land surface detection temperature; the land surface detection temperatures of multiple pixels are combined based on the identifier matrix of different stage types to obtain the time-series land surface temperature.

[0096] According to the embodiments of this application, the above objective optimization function is solved using the weighted least squares method or the variational method to obtain the optimal time series solution, as shown in formula (9):

[0097] (9)

[0098] in, Represents the surface temperature sequence for stage k; The comprehensive objective optimization function corresponding to formula (8); This represents the final time-series land surface temperature result for pixel i. The optimized results from each stage are combined using a stage indicator matrix to obtain the complete time-series land surface temperature.

[0099] According to embodiments of this application, by introducing continuity and energy balance constraints at stage boundaries, a smooth transition of surface temperature changes across different stages is ensured. A joint objective function is constructed, and least-squares solutions are performed by integrating time smoothing, gradient constraints, and energy residual terms to suppress noise and boundary jumps. Furthermore, Kalman or Bayesian fusion is employed for temporal optimization to generate a seamless, physically consistent, and stable continuous surface temperature sequence across stages.

[0100] According to embodiments of this application, the obtained time-series land surface temperature is structured using a spatial gridded organization and time-series management approach. Specifically, this includes storing the time-series land surface temperature in a multi-dimensional matrix according to spatial coordinates and time indices, forming a unified data structure that supports rapid indexing and interpolation access at different spatial resolutions. Based on business needs and product application scenarios, the land surface temperature time-series results are standardized and synthesized. Time averages, extreme values, or comprehensive indices are calculated using time windows as units to generate synthetic images or thematic maps at daily, weekly, and monthly scales. Simultaneously, it supports exporting rasterized data formats conforming to industry standards (such as GeoTIFF or NetCDF) for seamless integration with other land surface process models or business systems. A standardized land surface temperature product is generated, containing temporally continuous and spatially consistent land surface temperature information, which can be directly applied to subsequent business scenarios such as land surface energy balance simulation, freeze-thaw disaster monitoring, snow cover and climate process analysis, and ecological environment assessment. This product possesses cross-stage consistency and physical interpretability, providing a unified data foundation for high spatiotemporal resolution land surface energy process research.

[0101] The reconstruction method proposed in this embodiment identifies surface stages using energy indicators and introduces continuity and energy constraints at stage boundaries for joint optimization, achieving physically consistent temporal reconstruction and significantly improving the reconstruction accuracy and cross-stage continuity for snowmelt and freeze-thaw periods. Building upon this, by integrating energy weights and continuity constraints into the temporal reconstruction of surface temperature, dynamic energy distribution process modeling is introduced based on stage identification. This physically driven approach enhances the connection between stages, effectively improving the physical consistency and temporal robustness of the reconstruction results for energy abrupt change stages.

[0102] According to an embodiment of this application, obtaining an energy feature dataset of a target area includes: obtaining an original dataset and cloud mask data of the target area, wherein the original dataset includes hourly surface remote sensing data at a first resolution, original surface temperature data at a second resolution, ground meteorological data, albedo data, and snow depth data, and the ground meteorological data includes temperature, humidity, and net radiation data; filtering the original dataset using cloud mask data to obtain a filtered original dataset; resampling the surface remote sensing data and original surface temperature data in the filtered original dataset to obtain an intermediate dataset including surface remote sensing data at the target resolution and original surface temperature data; and preprocessing the intermediate dataset to obtain an energy feature dataset, wherein the preprocessing includes correcting slope aspect, the influence of slope on radiation, interpolation recovery when data is missing, and calculating latent heat flux data and sensible heat flux data based on the land-atmosphere energy balance equation.

[0103] According to the embodiments of this application, multi-source, multi-scale remote sensing and meteorological data supporting the target area are obtained. LST data mainly comes from geostationary meteorological satellites, such as hourly surface brightness temperature data (4km spatial resolution, 1-hour temporal resolution) provided by FY-4A, i.e., surface remote sensing data, as well as medium-resolution raw surface temperature data from MODIS (1km resolution). Ground meteorological measurement data includes key variables such as air temperature, wind speed, humidity, and net radiation, sourced from regional automatic weather station data. In addition, it is necessary to acquire Digital Elevation Model (DEM), MODIS albedo data, snow depth data, NDVI vegetation index data, cloud mask data, and FY-4A cloud detection data. All relevant remote sensing and meteorological data can be obtained through publicly available data platforms (such as satellite meteorological centers).

[0104] The original dataset was filtered using cloud masking data to obtain the filtered original dataset. Temporal matching and spatial resampling were performed on FY-4A and MODIS data to unify them to a 1 km spatial resolution and hourly / daily time scales. A terrain correction algorithm was used to correct the impact of slope aspect and gradient on radiation. Low-confidence pixels were removed based on cloud detection results, and missing data was recovered using temporal interpolation. Finally, non-target surfaces (such as water bodies or bare rock) were removed using an NDVI threshold mask. Latent heat flux and sensible heat flux were calculated using a dual-source energy balance model based on the land-atmosphere energy balance equation, combined with surface temperature, meteorological elements, and radiation parameters. The final result is a spatially consistent, temporally continuous multi-source energy feature dataset.

[0105] Figure 4 A block diagram of a land surface temperature temporal reconstruction system considering energy distribution mechanisms according to an embodiment of this application is shown.

[0106] like Figure 4 As shown, the land surface temperature time series reconstruction system 400 considering the energy distribution mechanism includes an acquisition module 410, a detection module 420, a fitting module 430, a generation module 440, and a solution module 450.

[0107] The acquisition module 410 is used to acquire an energy feature dataset of the target area in response to a temperature reconstruction command. The energy feature dataset includes at least one of the following: temporally continuous surface temperature data, net radiation data, latent heat flux data, sensible heat flux data, albedo data, and snow depth data for each pixel in the target area.

[0108] The detection module 420 is used to perform mutation detection on the energy feature dataset based on the energy distribution of each pixel, and obtain the identification matrix corresponding to different stage types, including the snow cover period stage, the snow melting period stage, and the snowless period stage.

[0109] The fitting module 430 is used to perform regression fitting on the data corresponding to the identification matrix in the energy feature dataset for any type stage, so as to obtain the surface temperature sequence of the type stage.

[0110] The generation module 440 is used to generate a time series optimization model based on constraints and surface temperature sequences of different types of stages.

[0111] The solver module 450 is used to solve the objective function of the time series optimization model to obtain the time series surface temperature.

[0112] According to embodiments of this application, abrupt changes in the energy feature dataset are detected based on the energy allocation of each pixel to obtain identifier matrices for the corresponding snow cover period, snowmelt period, and snowless period. Regression fitting is performed on the data in the energy feature dataset corresponding to the identifier matrices to obtain the surface temperature sequence for each type of stage. A time-series optimization model is generated based on constraints and the surface temperature sequences for different types of stages. The objective function of the time-series optimization model is solved to obtain the time-series surface temperature. By introducing an energy allocation mechanism to guide stage division and inter-stage connections, the reconstruction results of surface temperature during energy abrupt changes are significantly enhanced. The method based on stage identification and joint optimization of cross-stage physical constraints can handle abrupt changes in energy allocation caused by seasonal changes in snow cover, thereby improving the accuracy and physical consistency of the time-series reconstruction of surface temperature.

[0113] Any one or more of the modules, submodules, units, and subunits according to the embodiments of this application, or at least part of the functions of any one or more of them, can be implemented in one module. Any one or more of the modules, submodules, units, and subunits according to the embodiments of this application can be implemented by dividing them into multiple modules. Any one or more of the modules, submodules, units, and subunits according to the embodiments of this application can be at least partially implemented as hardware circuits, such as field-programmable gate arrays (FPGAs), programmable logic arrays (PLAs), systems-on-a-chip, systems-on-a-substrate, systems-on-package, application-specific integrated circuits (ASICs), or implemented by hardware or firmware in any other reasonable manner by integrating or packaging circuits, or implemented in any one of software, hardware, and firmware, or in a suitable combination of any of these. Alternatively, one or more of the modules, submodules, units, and subunits according to the embodiments of this application can be at least partially implemented as computer program modules, which, when run, can perform corresponding functions.

[0114] For example, any multiple of the acquisition module 410, detection module 420, fitting module 430, generation module 440, and solving module 450 can be combined into one module / unit / subunit, or any one of these modules / units / subunits can be split into multiple modules / units / subunits. Alternatively, at least part of the functionality of one or more of these modules / units / subunits can be combined with at least part of the functionality of other modules / units / subunits and implemented in one module / unit / subunit. According to embodiments of this application, at least one of the acquisition module 410, detection module 420, fitting module 430, generation module 440, and solving module 450 can be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or any other reasonable means of integrating or packaging the circuitry, or implemented in software, hardware, or firmware, or in any suitable combination of any of these three implementation methods. Alternatively, at least one of the acquisition module 410, detection module 420, fitting module 430, generation module 440, and solution module 450 may be at least partially implemented as a computer program module, which can perform corresponding functions when the computer program module is run.

[0115] It should be noted that the surface temperature time series reconstruction system part considering energy distribution mechanism in the embodiments of this application corresponds to the surface temperature time series reconstruction method part considering energy distribution mechanism in the embodiments of this application. For a detailed description of the surface temperature time series reconstruction system part considering energy distribution mechanism, please refer to the surface temperature time series reconstruction method part considering energy distribution mechanism, which will not be repeated here.

[0116] Figure 5 A block diagram of an electronic device suitable for implementing the methods described above, according to an embodiment of this application, is shown. Figure 5 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0117] like Figure 5 As shown, an electronic device 500 according to an embodiment of this application includes a processor 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage portion 508 into a random access memory (RAM) 503. The processor 501 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 501 may also include onboard memory for caching purposes. The processor 501 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of this application.

[0118] RAM 503 stores various programs and data required for the operation of electronic device 500. Processor 501, ROM 502, and RAM 503 are interconnected via bus 504. Processor 501 executes various operations of the method flow according to embodiments of this application by executing programs in ROM 502 and / or RAM 503. It should be noted that the programs may also be stored in one or more memories other than ROM 502 and RAM 503. Processor 501 may also execute various operations of the method flow according to embodiments of this application by executing programs stored in said one or more memories.

[0119] According to embodiments of this application, the electronic device 500 may further include an input / output (I / O) interface 505, which is also connected to a bus 504. The electronic device 500 may also include one or more of the following components connected to the input / output (I / O) interface 505: an input section 506 including a keyboard, mouse, etc.; an output section 507 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the input / output (I / O) interface 505 as needed. A removable medium 511, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 510 as needed so that computer programs read from it can be installed into the storage section 508 as needed.

[0120] According to embodiments of this application, the method flow according to embodiments of this application can be implemented as a computer software program. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable storage medium, the computer program containing program code for performing the methods shown in the flowchart. In such embodiments, the computer program can be downloaded and installed from a network via communication section 509, and / or installed from removable medium 511. When the computer program is executed by processor 501, it performs the functions defined in the system of embodiments of this application. According to embodiments of this application, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0121] This application also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of this application.

[0122] According to embodiments of this application, the computer-readable storage medium can be a non-volatile computer-readable storage medium. Examples include, but are not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0123] For example, according to embodiments of this application, a computer-readable storage medium may include the ROM 502 and / or RAM 503 described above and / or one or more memories other than ROM 502 and RAM 503.

[0124] Embodiments of this application also include a computer program product comprising a computer program containing program code for performing the methods provided in the embodiments of this application. When the computer program product is run on an electronic device, the program code is used to enable the electronic device to implement the methods provided in the embodiments of this application.

[0125] When the computer program is executed by the processor 501, it performs the functions defined in the system / apparatus of this application embodiment. According to the embodiments of this application, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0126] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and may be downloaded and installed via the communication section 509, and / or installed from a removable medium 511. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0127] According to embodiments of this application, program code for executing the computer programs provided in the embodiments of this application can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0128] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions. Those skilled in the art will understand that the features described in the various embodiments of this application can be combined and / or combined in various ways, even if such combinations are not explicitly described in this application. In particular, without departing from the spirit and teachings of this application, the features described in the various embodiments of this application can be combined and / or combined in various ways. All such combinations and / or combinations fall within the scope of this application.

[0129] The embodiments of this application have been described above. However, these embodiments are merely illustrative and not intended to limit the scope of this application. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. This application does not depart from its scope, and those skilled in the art can make various substitutions and modifications, all of which should fall within the scope of this application.

Claims

1. A method for temporal reconstruction of land surface temperature considering energy distribution mechanisms, characterized in that, include: In response to a temperature reconstruction command, an energy feature dataset of the target area is acquired, wherein the energy feature dataset includes at least one of the following: temporally continuous surface temperature data, net radiation data, latent heat flux data, sensible heat flux data, albedo data, and snow depth data for each pixel in the target area. Based on the energy allocation of each pixel, mutation detection is performed on the energy feature dataset to obtain an identifier matrix corresponding to different stage types, wherein the stage types include snow cover stage, snow melting stage and snowless stage; For any type of stage, regression fitting is performed on the data corresponding to the identification matrix in the energy feature dataset to obtain the surface temperature sequence of the type of stage. Based on the constraints and the surface temperature sequences of different types of stages, a time-series optimization model is generated; The objective function of the time-series optimization model is solved to obtain the time-series surface temperature.

2. The method according to claim 1, characterized in that, Based on the energy distribution of each pixel in the target region, mutation detection is performed on the energy feature dataset to obtain an identifier matrix corresponding to different stage types, including: For each pixel, an energy allocation weight sequence for that pixel is determined based on the latent heat flux data and sensible heat flux data of that pixel; The snow depth data of the pixels are classified based on the snow depth threshold to obtain a snow mask sequence that represents the snow depth classification type of the pixels. Based on the energy allocation weight sequence and the snow mask sequence, mutation detection is performed on the energy feature dataset to obtain the identifier matrix corresponding to different stage types.

3. The method according to claim 2, characterized in that, Based on the energy allocation weight sequence and the snow mask sequence, mutation detection is performed on the energy feature dataset to obtain an identifier matrix corresponding to different stage types, including: The energy allocation weight sequence is subjected to mutation detection to obtain mutation detection results, wherein the mutation detection results characterize the position of the inflection point where energy mutation occurs; Based on the mutation detection results and the snow depth data, the energy allocation weight sequence is divided into multiple stage types and the boundary moments between different stage types; Based on the boundary time, a stage division result matrix is ​​generated according to the mutation detection result and the snow cover mask sequence, wherein the stage division result matrix includes an identifier matrix for the stage type corresponding to each pixel.

4. The method according to claim 1, characterized in that, The energy feature dataset includes data subsets corresponding to different stage types; Specifically, regression fitting is performed on the data corresponding to the identifier matrix in the energy feature dataset to obtain the surface temperature sequence of the type stage, including: When the type stage is the snow cover period, regression modeling is performed using net radiation data, surface temperature data, and latent heat flux data from the data subset corresponding to the snow cover period to obtain an energy-constrained regression model. The energy-constrained regression model is a model dominated by latent heat processes, with surface temperature as the dependent variable. The surface temperature sequence for the snow cover period is obtained based on the energy-constrained regression model.

5. The method according to claim 4, characterized in that, Also includes: When the type stage is the snowmelt period, a piecewise linear regression model is constructed using the net radiation data from the data subset corresponding to the snowmelt period. The surface temperature sequence during the snowmelt period was obtained based on the piecewise linear regression model.

6. The method according to claim 4, characterized in that, The energy feature dataset also includes vegetation indices and digital elevation models; The method further includes: When the type stage is a snowless period, a residual fitting model is constructed based on the target regression method using net radiation data, sensible heat flux data, surface temperature data, vegetation index, albedo data and digital elevation model from the data subset corresponding to the snowless period. The target regression method is obtained by fusing the geographic weighted regression method and the random forest regression method. The surface temperature sequence for the snowless period is obtained based on the residual fitting model.

7. The method according to claim 1, characterized in that, Based on the constraints and surface temperature sequences for different types of stages, a time-series optimization model is generated, including: For any pixel, an initial optimization function is constructed based on the continuity constraint and the surface temperature sequence of different types of stages; Based on the constraint of energy conservation between the Earth and the atmosphere, the energy residual constraint term is determined according to the net radiation data, latent heat flux data, and sensible heat flux data. Based on the initial optimization function and the energy residual constraint term, a target optimization function corresponding to the pixel is generated.

8. The method according to claim 1 or 7, characterized in that, Solving the objective function of the time-series optimization model yields the time-series land surface temperature, including: For any pixel, the objective optimization function is solved using the weighted least squares method or the variational method to obtain the surface temperature; The time-series surface temperature is obtained by combining the surface temperatures of multiple pixels based on the identifier matrix of different stage types.

9. The method according to claim 1, characterized in that, Obtain the energy feature dataset of the target region, including: Obtain the raw dataset and cloud mask data of the target area. The raw dataset includes hourly satellite remote sensing data at a first resolution, raw surface temperature data at a second resolution, ground meteorological data, albedo data, and snow depth data. The ground meteorological data includes temperature, humidity, and net radiation data. The original dataset is filtered using the cloud mask data to obtain the filtered original dataset; The surface remote sensing data and raw surface temperature data in the filtered raw dataset are resampled to obtain an intermediate dataset that includes surface remote sensing data and raw surface temperature data at the target resolution. The intermediate dataset is preprocessed to obtain the energy feature dataset. The preprocessing includes correcting the aspect, the effect of slope on radiation, interpolation recovery when data is missing, and calculating latent heat flux data and sensible heat flux data based on the land-atmosphere energy balance equation.

10. A temporal reconstruction system for land surface temperature considering energy distribution mechanisms, characterized in that, include: The acquisition module is used to acquire an energy feature dataset of the target area in response to a temperature reconstruction command. The energy feature dataset includes at least one of the following: temporally continuous surface temperature data, net radiation data, latent heat flux data, sensible heat flux data, albedo data, and snow depth data for each pixel in the target area. The detection module is used to perform mutation detection on the energy feature dataset based on the energy allocation of each pixel to obtain an identifier matrix corresponding to different stage types, wherein the stage types include snow cover stage, snow melting stage and snowless stage; The fitting module is used to perform regression fitting on the data corresponding to the identification matrix in the energy feature dataset for any type of stage, so as to obtain the surface temperature sequence of the type of stage. The generation module is used to generate time-series optimization models based on constraints and surface temperature sequences of different types of stages. The solution module is used to solve the objective function of the time series optimization model to obtain the time series surface temperature.

Images