Temporal and spatial dynamic parameterization method and system for vegetation scattering albedo in tau-omega model

Abstract

The application discloses a vegetation scattering albedo spatiotemporal dynamic parameterization method and system in a tau-omega model, wherein the method comprises the following steps: obtaining passive microwave original observation data and auxiliary data, performing data preprocessing and generating corresponding observation reference data; screening effective pixels based on the observation reference data; calculating the corresponding vegetation scattering albedo of each effective pixel on a daily scale to generate a corresponding pixel daily scale sequence; generating a multi-class spatiotemporal dynamic parameter mapping model based on the pixel daily scale sequence; determining an optimal dynamic parameter mapping model from each class of spatiotemporal dynamic parameter mapping model to obtain a target mapping model; and based on the optimal dynamic parameter mapping model, carrying out inversion result evaluation, gain quantification and product output. The application realizes mechanism spatiotemporal dynamic characterization of the vegetation scattering albedo, effectively suppresses system errors caused by static assignment, simultaneously improves microwave SM and VOD inversion accuracy, and optimizes the product to provide high-quality basic data for related research.

Description

Technical Field

[0001] This invention relates to the field of microwave remote sensing inversion technology, and in particular to a spatiotemporal dynamic parameterization of vegetation scattering albedo in a τ–ω model, used to improve the accuracy of microwave SM and VOD inversion. Background Technology

[0002] Surface soil moisture (SM) and vegetation optical depth (VOD) are key remote sensing parameters characterizing terrestrial water processes and vegetation status, and are widely used in fields such as water cycle monitoring and carbon storage analysis. Passive microwave remote sensing is insensitive to clouds and rain, enabling long-term continuous observation. Among them, the L-band (1–2 GHz) has become an effective method for monitoring SM and VOD due to its strong canopy penetration and high revisit frequency. In-orbit L-band satellites such as SMOS and SMAP generally use the zero-order Tau–Omega (τ–ω) radiative transfer model for operational inversion. The vegetation scattering albedo parameter ω is used to characterize the strength of the canopy scattering effect and is a key control quantity affecting the accuracy and stability of SM and VOD inversion, directly influencing their accuracy and stability.

[0003] In existing inversion processes, ω is typically preset as a globally fixed constant or through empirical methods that assign values ​​based on vegetation type. For example, ω is set to approximately 0.06–0.08 for forest areas and 0 for low-lying vegetation areas, and is included in the calculation as a known term during inversion. However, from a mechanistic perspective, canopy properties that affect microwave scattering intensity, such as canopy structure complexity, vegetation water content, and plant morphology, exhibit significant changes with seasonal growth and water stress. Furthermore, ω can even exhibit spatial heterogeneity within the same vegetation type. Summary of the Invention

[0004] The purpose of this invention is to address the problems in existing microwave SM and VOD inversion technologies based on the τ–ω radiative transfer model, where vegetation scattering albedo ω is preset using a globally fixed constant or an empirical method of statically assigning values ​​according to land cover type. This method lacks a mechanistic characterization corresponding to the canopy scattering process and does not explicitly reflect the spatiotemporal dynamic changes of ω, which easily introduces systematic errors, increases inversion uncertainty, and limits inversion accuracy and stability. The invention proposes a parameterization method for vegetation scattering albedo that can improve the accuracy of microwave SM and VOD inversion.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: Firstly, a spatiotemporal dynamic parameterization method for vegetation scattering albedo in a τ–ω model is provided, including the following steps: Acquire observation reference data, which is used to indicate the passive microwave brightness temperature observation value in the target area and within the target time range, and also to indicate the incident angle and observation time corresponding to the passive microwave brightness temperature observation value; Based on the aforementioned observation reference data, effective pixels were selected based on the linear regression relationship between reanalysis soil moisture data and passive microwave brightness temperature observations. Based on the observation values ​​of effective canopy temperature and passive microwave brightness temperature, the vegetation scattering albedo corresponding to each effective pixel is calculated on a daily scale to obtain the corresponding daily-scale pixel sequence. Based on the pixel diurnal scale sequence, a corresponding target mapping model is generated; the target mapping model is used to indicate the mapping relationship between time and / or space and vegetation scattering albedo.

[0006] As one possible implementation method, based on the linear regression relationship between pre-constructed reanalysis soil moisture data and passive microwave brightness temperature observation, pixels with positive regression slopes within a preset incident angle range are considered as valid pixels.

[0007] As one possible implementation method, at the pixel scale, for each incident angle θ of the SMOS satellite, a linear regression relationship is established between the reanalysis soil moisture data and the passive microwave brightness temperature observation values, and the specific formula is as follows: TB i (y,d,θ)=a i,θ +b i,θ SM i ERA5 (y,d); Where i is the pixel number; y is the year; d is the day sequence; TB i (y,d,θ) represents the passive microwave brightness temperature observation value of the i-th pixel under the conditions of year y, day sequence d, and incident angle θ; a i,θ b is the linear regression intercept of the i-th pixel at the incident angle θ; i,θ SM represents the linear regression slope of the i-th pixel at the incident angle θ; i ERA5 (y,d) represents the reanalysis soil moisture data of the i-th pixel under the conditions of year y and day d.

[0008] As one possible implementation method, the specific steps for calculating the vegetation scattering albedo corresponding to each effective pixel on a diurnal scale based on the effective canopy temperature and passive microwave brightness temperature observations, and obtaining the corresponding diurnal sequence of pixels, are as follows: Based on the effective pixels and the date, the effective canopy temperature Tc and the passive microwave brightness temperature observation value TB were paired to obtain the corresponding paired data. Calculate the vegetation scattering albedo for each paired data point based on the incident angle scale; Based on the obtained vegetation scattering albedo, the corresponding pixel diurnal scale sequence is summarized and generated. The formula for calculating vegetation albedo is as follows: ; Where y represents the year; d represents the day sequence; TB i (y,d,θ) represents the brightness temperature value for the corresponding effective pixel, date, and incident angle; T c,i (y,d) represents the effective canopy temperature for the corresponding effective pixels and date.

[0009] As one possible implementation method, based on the pixel-scale sequence, for each effective pixel, the vegetation scattering albedo of multiple years is averaged over the same day sequence to obtain the average albedo corresponding to each day sequence. Based on the average albedo, a corresponding climatological time-dynamic sequence is generated to obtain the corresponding pixel-scale climatological time-dynamic mapping model. Based on the aforementioned climatological time dynamic series, the average albedo of each effective pixel in each study area is averaged according to the study area to obtain a corresponding spatially consistent climatological time dynamic mapping model. Based on the daily pixel scale sequence, the daily pixel scale sequence of each pixel i is averaged over many years over the entire time range to obtain the corresponding time-static spatial dynamic mapping model. The optimal dynamic parameter mapping model is determined from the pixel-scale climatological temporal dynamic mapping model, the spatially consistent climatological temporal dynamic mapping model, and the temporally static spatial dynamic mapping model to obtain the target mapping model.

[0010] As one possible implementation, the pixel-scale climatological temporal dynamic mapping model, the spatially consistent climatological temporal dynamic mapping model, and the temporally static spatial dynamic mapping model are respectively used as mapping models to be evaluated, and the SM temporal consistency improvement magnitude ΔR corresponding to each mapping model to be evaluated is obtained. SM Error reduction magnitude ΔubRMSD SM Improvement magnitude ΔR in VOD space consistency VOD Based on ΔR SM The improvement, ΔubRMSD SM The magnitude of the decrease, and ΔR VOD To assess the extent of improvement, the configuration with the largest combined improvement across the three indicators is selected as the optimal dynamic parameter mapping model.

[0011] Secondly, a spatiotemporal dynamic parameterization system for vegetation scattering albedo in a τ–ω model is provided, including: The data processing module is used to acquire observation reference data, which is used to indicate the passive microwave brightness temperature observation value in the target area and within the target time range, and also to indicate the incident angle and observation time corresponding to the passive microwave brightness temperature observation value. The filtering module is used to filter effective pixels based on the observation reference data and the linear regression relationship between the reanalysis soil moisture data and the passive microwave brightness temperature observation values. The calculation module is used to calculate the vegetation scattering albedo corresponding to each effective pixel on a daily scale based on the effective canopy temperature and passive microwave brightness temperature observations, and obtain the corresponding daily-scale pixel sequence. The mapping module is used to generate a corresponding target mapping model based on the daily-scale sequence of the pixels; the target mapping model is used to indicate the mapping relationship between time and / or space and vegetation scattering albedo.

[0012] As one possible implementation, the screening module uses a pre-constructed linear regression relationship between reanalysis soil moisture data and passive microwave brightness temperature observations to identify pixels with positive regression slopes within a preset incident angle range as valid pixels.

[0013] As one possible implementation, the computing module includes: The pairing unit is used to pair the effective canopy temperature Tc with the passive microwave brightness temperature observation value TB based on the effective pixels and the date, and obtain the corresponding paired data. The computing unit is used to calculate the vegetation scattering albedo of each paired data based on the incident angle scale; The summarization unit is used to summarize and generate the corresponding daily-scale pixel sequence based on the obtained vegetation scattering albedo. The calculation unit calculates the vegetation albedo based on the following formula: ; Where y represents the year; d represents the day sequence; TB i (y,d,θ) represents the brightness temperature value for the corresponding effective pixel, date, and incident angle; T c,i (y,d) represents the effective canopy temperature for the corresponding effective pixels and date.

[0014] As one possible implementation, the mapping module includes: The first mapping unit is used to take the multi-year average of the vegetation scattering albedo of multiple years for each effective pixel on the same day sequence based on the pixel day-scale sequence, to obtain the average albedo corresponding to each day sequence, and to generate the corresponding climatological time-dynamic sequence based on the average albedo, thereby obtaining the corresponding pixel-scale climatological time-dynamic mapping model. The second mapping unit is used to average the average albedo of each effective pixel in each study area according to the climatological time dynamic sequence, based on the climatological time dynamic sequence, to obtain a corresponding spatially consistent climatological time dynamic mapping model. The third mapping unit is used to take the multi-year average of the daily scale sequence of each pixel i over the entire time range based on the daily scale sequence of the pixels, and obtain the corresponding time-static spatial dynamic mapping model. The evaluation unit is used to determine the optimal dynamic parameter mapping model from the pixel-scale climatological temporal dynamic mapping model, the spatially consistent climatological temporal dynamic mapping model, and the temporally static spatial dynamic mapping model, and obtain the target mapping model.

[0015] This invention, by adopting the above technical solutions, has significant technical effects: This invention utilizes the linear regression relationship between soil moisture and passive microwave brightness temperature (SMWB) to screen effective pixels. Based on the effective canopy temperature and SMWB observation values ​​of the effective pixels, the vegetation scattering albedo of each effective pixel on each date is constructed, obtaining the pixel daily-scale sequence of each effective pixel. By performing statistical analysis on the pixel daily-scale sequence, the corresponding target mapping model is obtained, which is used to provide dynamic vegetation scattering albedo parameters for candidate inversion, effectively improving the accuracy of microwave SM and VOD inversion. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating a spatiotemporal dynamic parameterization method for vegetation scattering albedo in a τ–ω model according to the present invention. Figure 2 This is a schematic diagram of ΔR and ΔubRMSD obtained by spatiotemporal dynamics ω compared to SM inversion based on land cover type, with different reanalysis SMs as references; Figure 3 This is a spatial scatter diagram of the VOD inverted by spatiotemporal dynamics ω and VOD based on land cover type setting ω, respectively, and CCI AGB. Detailed Implementation

[0018] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0019] The values ​​of ω in existing technologies are essentially empirical prior settings, failing to establish a physical connection between ω and canopy structure characteristics, vegetation water content, and its seasonal and water stress responses. Consequently, they cannot explicitly characterize the spatiotemporal dynamic changes of ω. Therefore, when vegetation conditions fluctuate significantly with seasonal changes or under water stress, the fixed or regional constant settings of ω are prone to mismatch with the actual scattering process, introducing systematic biases and limiting the accuracy and stability of SM and VOD inversion results.

[0020] To address the above problems, this application proposes the following embodiments.

[0021] Example 1: A spatiotemporal dynamic parameterization method for vegetation scattering albedo in a τ–ω model, used to improve the accuracy of microwave SM and VOD inversion, such as... Figure 1 As shown, it includes the following steps: S100. Acquire raw passive microwave observation data and auxiliary data, perform data preprocessing, and generate corresponding observation reference data. The raw passive microwave observation data includes: Passive microwave brightness temperature (TB) observation data within the target area and target time range; Observe geometric information, such as the angle of incidence and the observation time.

[0022] Supporting data includes: Data used for quality control and scene screening, such as water and urban masks, frozen soil or snow masks, radio frequency interference markers, and necessary vegetation cover information, etc., can be configured by those skilled in the art according to actual needs, and this specification does not limit them in detail.

[0023] The specific method for preprocessing the raw passive microwave observation data based on the auxiliary data is as follows: Perform projection unification, spatial resampling, and temporal alignment operations on the raw and auxiliary passive microwave observation data to achieve consistency of the spatiotemporal reference of the data. Based on auxiliary data, pixels and observations affected by water bodies, cities, frozen soil or snow, and strong radio frequency interference are removed or masked to generate corresponding observation reference data. That is, the observation reference data is used to indicate the passive microwave brightness temperature observation value in the target area and within the target time range, and is also used to indicate the incident angle and observation time corresponding to the passive microwave brightness temperature observation value. S200. Based on the observed reference data, select valid pixels; Based on the aforementioned observational reference data, effective pixels are selected based on the linear regression relationship between reanalysis soil moisture data and passive microwave brightness temperature observations. In this embodiment, pixels with positive regression slopes within a preset incident angle range are considered effective pixels. These effective pixels satisfy the conditions of "approximately opaque canopy and negligible or significantly reduced soil radiation contribution." In this embodiment, a candidate set is formed based on these effective pixels, and the specific steps are as follows: Using reanalysis soil moisture data as a reference benchmark, the spatiotemporal dimensions of the observation reference data are matched; In this embodiment, the soil moisture reanalysis data used is the surface soil moisture product from the ERA5 reanalysis dataset published by the European Centre for Medium-Range Weather Forecasts, namely ERA5 SM.

[0024] At the pixel scale, linear regression relationships were established between reanalysis soil moisture data and passive microwave brightness temperature observations for each incident angle θ of the SMOS satellite, as shown in the following formulas: TB i (y,d,θ)=a i,θ +b i,θ SM i ERA5 (y,d); Where i is the pixel number; y is the year; d is the day sequence; TB i (y,d,θ) represents the passive microwave brightness temperature observation value of the i-th pixel under the conditions of year y, day sequence d, and incident angle θ; a i,θ b is the linear regression intercept of the i-th pixel at the incident angle θ; i,θ SM represents the linear regression slope of the i-th pixel at the incident angle θ; i ERA5 (y,d) represents the ERA5 reanalysis surface soil moisture data of the i-th pixel under the conditions of year y and day d; In this embodiment, the brightness temperature observation is multi-incident angle observation data. Those skilled in the art can set the incident angle range according to actual needs. In this embodiment, the incident angle range is 20°–55°.

[0025] The principle of cross-angle consistency is used as the screening criterion: only when a pixel is exposed to all incident angles between 20° and 55° does the regression slope b...i,θ If all values ​​are positive, the pixel is determined to be a valid pixel that meets the requirements; if the slope corresponding to any incident angle does not meet the positive value condition, the pixel is directly discarded.

[0026] For each valid pixel selected, its spatial grid position, valid year and date range, and the number of valid observations at each incident angle are recorded to construct a candidate set.

[0027] S300. Calculate the vegetation scattering albedo of each effective pixel on a daily scale and generate the corresponding daily-scale pixel sequence. In this embodiment, the specific steps for calculating vegetation scattering albedo ω based on simplified radiative transfer relationships and generating pixel diurnal scale sequences are as follows: S310. Based on the effective pixels and the date, perform data pairing between the effective canopy temperature Tc and the passive microwave brightness temperature observation value TB to obtain the corresponding paired data. Since the effective pixels satisfy the condition of "the canopy is approximately opaque and the contribution of soil radiation is negligible", the vegetation transmittance γ can be made approximately 0, thereby establishing a direct relationship between the canopy scattering albedo ω and the passive microwave brightness temperature observation value TB and the effective canopy temperature Tc, that is, simplifying the radiation transmission relationship. Based on the simplified radiative transfer relationship, the effective canopy temperature Tc and the passive microwave brightness temperature observation value TB are paired according to the effective pixels and the date to obtain the corresponding paired data.

[0028] In this embodiment, the effective canopy temperature Tc can be obtained by ERA5 reanalysis of the temperature field.

[0029] S320. Calculate the vegetation scattering albedo ω for each paired data based on the incident angle scale; For each valid pixel i, each date (y,d), and incident angle θ, the value of ω at the incident angle scale is calculated based on the simplified radiative transfer relation, as shown in the following formula: ; Where y represents the year; d represents the day sequence; TB i (y,d,θ) represents the brightness temperature value for the corresponding effective pixel, date, and incident angle; T c,i (y,d) represents the effective canopy temperature for the corresponding effective pixels and date.

[0030] S330. Based on the obtained vegetation scattering albedo ω, robustly summarize and generate a daily pixel-scale sequence. Based on the vegetation scattering albedo ω at the incident angle scale, to reduce the impact of single-angle inversion noise, multiple incident angle ω values ​​for the same pixel and the same date are robustly summarized to generate a pixel-scale daily sequence. The specific formula is as follows: ω i (y,d)=medianθ∈[20°,55°] (ω i (y,d,θ)); In this embodiment, the robust aggregation method is to take the median.

[0031] S400. Based on the pixel daily scale sequence, generate multiple spatiotemporal dynamic parameter mapping models; That is, based on the pixel daily-scale sequence, multi-year statistical summaries are performed, and a pixel-scale climatological temporal dynamic mapping model, a spatially consistent climatological temporal dynamic mapping model, and a temporally static spatial dynamic mapping model are generated. Specifically: The pixel-scale climatological temporal dynamic mapping model refers to the process of averaging the vegetation scattering albedo ω over multiple years for each pixel i on the same day d to obtain the average albedo corresponding to each day. Based on the average albedo The climatological time dynamics series that retains seasonal variation but suppresses diurnal stochastic fluctuations are obtained, as shown in the following formula: ; Where, ω i (y,d) represents the daily scale ω value of pixel i in year y and day order d; N is the number of years included in the statistics.

[0032] A spatially consistent climatological temporal dynamic mapping model refers to, based on the aforementioned pixel-scale climatological temporal dynamics, mapping the average albedo of each effective pixel in each study area according to the study region. By performing regional averaging, we obtain the spatially consistent vegetation scattering albedo ω that varies only with the seasons. c (d), the specific formula is as follows: ; Where M is the number of valid pixels.

[0033] The time-static spatial dynamic mapping model refers to averaging the daily-scale sequence of each pixel i over the entire time period to obtain the static albedo of each effective pixel. This static albedo can stabilize spatial differences and does not change over time with the seasons. The specific formula is as follows: ; Where ω i (t) represents the daily scale sequence of pixel i at time step t.

[0034] S500. Determine the optimal dynamic parameter mapping model from various spatiotemporal dynamic parameter mapping models to obtain the target mapping model; Various spatiotemporal dynamic parameter mapping models are embedded into the τ–ω inversion model to generate corresponding SM and VOD inversion results. The inversion effects of various spatiotemporal dynamic parameter mapping models are compared and evaluated using external calibration data. The optimal scheme with the largest comprehensive improvement is selected. The following evaluation steps are performed on various spatiotemporal dynamic parameter mapping models as the mapping models to be evaluated: S510. Perform calibration on the elements of the mapping model to be evaluated and calculate the comparison index; In this embodiment, the calibration objects are divided into SM calibration and VOD calibration, specifically: SM calibration and index calculation: Using ERA5 SM as reference data, inverted SM data under each ω configuration are matched according to time series. The time correlation coefficient R and the root mean square error ubRMSD between SM and ERA5 SM under each configuration are calculated; the difference between these and the corresponding indices of the baseline scheme is then used to obtain the improvement magnitude ΔR in SM time series consistency. SM And the magnitude of error reduction ΔubRMSD SM ; VOD calibration and index calculation: Independent canopy height (CH) data were selected as the spatial reference. Pixel-by-pixel matching and inversion were used to calculate the spatial consistency index between VOD and CH for each ω configuration, such as the spatial correlation coefficient R, to evaluate the degree of improvement in VOD spatial representation capability brought about by different ω configurations. The difference between this and the spatial correlation coefficient of the baseline scheme was then calculated to obtain the VOD spatial consistency improvement magnitude ΔR. VOD .

[0035] The baseline scheme is a pre-set scheme based on the existing empirical method of assigning values ​​according to vegetation type zoning.

[0036] S520. Based on the comparison indicators, the optimal dynamic parameter mapping model is determined by comprehensive criteria. Comparing the gains of each spatiotemporal dynamic parameter mapping model relative to the baseline scheme, ΔR is given priority. SM The improvement, ΔubRMSD SM The magnitude of the decrease, and ΔR VOD The extent of improvement was determined. The configuration with the largest overall improvement across the three indicators was selected as the optimal dynamic parameter mapping model. In this embodiment, the improvement rates of the above three-phase indicators are summed to obtain the corresponding comprehensive improvement rate. In actual use, those skilled in the art can set their own evaluation method for the comprehensive improvement rate, such as weighted summation. This specification does not limit it in detail.

[0037] In this embodiment, the optimal solution is the pixel-scale climatological time dynamic parameter mapping model.

[0038] In practical applications, the optimal dynamic parameter mapping model can also be used to evaluate inversion results, quantify gains, and output products.

[0039] S600. Based on the optimal dynamic parameter mapping model, the inversion gain of the optimal dynamic parameter mapping model is verified through multi-source reference data, and the final SM and VOD inversion products and accuracy improvement indicators are output. The specific steps are as follows: S610. Based on the optimal dynamic parameter mapping model, perform a full-process inversion to generate the target inversion product; Based on the optimal dynamic parameter mapping model, it is embedded into the τ–ω radiative transfer inversion model, and a full-process inversion is performed for the target area and target time period to generate SM and VOD inversion products under this configuration.

[0040] S620. Based on the target inversion product, further quantify its improvement effect relative to the static baseline scheme; Based on the target inversion product, the inversion gain of the optimal dynamic parameter mapping model is verified using multi-source reference data, and its improvement effect relative to the static ω baseline scheme is further quantified. Specifically: Pixel-by-pixel robust evaluation of SM: In addition to ERA5, to enhance the robustness of the evaluation results, reference SM data from different sources can be introduced to carry out pixel-by-pixel evaluation, such as multiple reanalysis SMs including ERA5-land, GLDAS, GEOS-5 FP and MEERA-2, to construct a reference benchmark library for SM evaluation. The time series of SM and multi-source reference SM are inverted by matching the optimal configuration at the pixel scale, and the time correlation coefficient R and the root mean square error of bias removal ubRMSD are calculated pixel by pixel. The difference between R and ubRMSD obtained based on the optimal dynamic parameter mapping model and the corresponding index of the baseline scheme based on the land cover type ω is used to obtain the pixel-by-pixel accuracy improvement magnitude ΔR. SM With ΔubRMSD SM Furthermore, an improved amplitude spatial distribution map was plotted to identify the spatial heterogeneity characteristics of the inversion gain.

[0041] VOD spatial improvement evaluation: External biomass data such as ESA, CCI, and AGB were selected as spatial control benchmarks, and their spatial grids were matched with the optimal configuration for VOD inversion. The spatial correlation coefficient between the inverted VOD and AGB is calculated, and this value is compared with the R obtained based on the land cover type setting ω to quantify the improvement of VOD's ability to characterize the spatial pattern of biomass.

[0042] S630. Based on the target inversion product and the improvement effect of the quantification, output the inversion product and the index difference product, specifically: The inversion products are the target time period SM inversion products and VOD inversion products under the optimal dynamic parameter mapping model; The index difference products include SM pixel-by-pixel ΔR and ΔubRMSD spatial distribution maps and VOD and AGB spatial correlation comparison reports. This example uses the tropical forest regions of South America and Africa as the research object. Taking the SMOS satellite as an example, the brightness temperature observations are multi-incident angle observation data, with an incident angle range of 20°–55°.

[0043] like Figure 2 As shown, based on the optimal dynamic parameter mapping model proposed in this method, compared with the scheme that sets ω based on land cover type, ΔR and ΔubRMSD are significantly improved for most pixels in the tropical forest regions of South America and Africa, and the corresponding ubRMSD is reduced, especially in Africa, where the ubRMSD of most pixels is reduced by 0.04m. 3 / m 3 above; like Figure 3 As shown, by comparing the spatial correlation between the VOD retrieved under the optimal dynamic parameter mapping model proposed in this paper and the scheme based on setting ω according to land cover type with the CCI AGB, it can be found that the scheme proposed in this paper can improve the correlation between VOD and AGB from 0.26 to 0.53. This invention achieves a synergistic improvement in the accuracy of SM and VOD inversion through independent data calibration and optimal configuration screening. The optimized product can better support application scenarios such as carbon cycle monitoring, water cycle assessment, and drought and forest ecological process diagnosis, providing higher-quality basic data for ecosystem and climate change research.

[0044] In summary, this invention overcomes the limitations of existing technologies that use globally fixed or land-type-based values ​​for ω, proposing a spatiotemporal dynamic parameterization scheme for ω. This scheme can characterize the seasonal evolution and spatial differences of ω at the pixel scale and embed it as a unified, reusable parameter into the τ–ω radiative transfer inversion process, effectively suppressing systematic errors caused by static assignment. The ω parameterization method of this invention has a clear physical mechanism, based on diurnal ω estimation driven by passive microwave brightness temperature (TB) observations, and then uses multi-year statistical summaries to form parameterized results such as pixel-level climatological temporal dynamics, regionally consistent climatological temporal dynamics, and temporal static spatial dynamics, ensuring that the characterization of ω changes is consistent with canopy physical processes. This invention achieves a synergistic improvement in the accuracy of SM and VOD inversions through independent data calibration and optimal configuration selection. SM inversion shows improved temporal correlation and reduced ubRMSD; VOD inversion shows enhanced spatial consistency with reference data such as canopy height and AGB. The optimized product can better support application scenarios such as carbon cycle monitoring, water cycle assessment, and drought and forest ecological process diagnosis, providing higher-quality basic data for ecosystem and climate change research.

[0045] Example 2: A spatiotemporal dynamic parameterization system for vegetation scattering albedo in a τ–ω model, comprising: The data processing module is used to acquire observation reference data, which is used to indicate the passive microwave brightness temperature observation value in the target area and within the target time range, and also to indicate the incident angle and observation time corresponding to the passive microwave brightness temperature observation value. The filtering module is used to filter effective pixels based on the observation reference data and the linear regression relationship between the reanalysis soil moisture data and the passive microwave brightness temperature observation values. The calculation module is used to calculate the vegetation scattering albedo corresponding to each effective pixel on a daily scale based on the effective canopy temperature and passive microwave brightness temperature observations, and obtain the corresponding daily-scale pixel sequence. The mapping module is used to generate a corresponding target mapping model based on the daily-scale sequence of the pixels; the target mapping model is used to indicate the mapping relationship between time and / or space and vegetation scattering albedo.

[0046] In this embodiment, the screening module uses a pre-constructed linear regression relationship between reanalysis soil moisture data and passive microwave brightness temperature observation to select pixels with positive regression slopes within a preset incident angle range as valid pixels.

[0047] In this embodiment, the computing module includes: The pairing unit is used to pair the effective canopy temperature Tc with the passive microwave brightness temperature observation value TB based on the effective pixels and the date, and obtain the corresponding paired data. The computing unit is used to calculate the vegetation scattering albedo of each paired data based on the incident angle scale; The summarization unit is used to summarize and generate the corresponding daily-scale pixel sequence based on the obtained vegetation scattering albedo. The calculation unit calculates the vegetation albedo based on the following formula: ; Where y represents the year; d represents the day sequence; TB i (y,d,θ) represents the brightness temperature value for the corresponding effective pixel, date, and incident angle; T c,i (y,d) represents the effective canopy temperature for the corresponding effective pixels and date.

[0048] In this embodiment, the mapping module includes: The first mapping unit is used to take the multi-year average of the vegetation scattering albedo of multiple years for each effective pixel on the same day sequence based on the pixel day-scale sequence, to obtain the average albedo corresponding to each day sequence, and to generate the corresponding climatological time-dynamic sequence based on the average albedo, thereby obtaining the corresponding pixel-scale climatological time-dynamic mapping model. The second mapping unit is used to average the average albedo of each effective pixel in each study area according to the climatological time dynamic sequence, based on the climatological time dynamic sequence, to obtain a corresponding spatially consistent climatological time dynamic mapping model. The third mapping unit is used to take the multi-year average of the daily scale sequence of each pixel i over the entire time range based on the daily scale sequence of the pixels, and obtain the corresponding time-static spatial dynamic mapping model. The evaluation unit is used to determine the optimal dynamic parameter mapping model from the pixel-scale climatological temporal dynamic mapping model, the spatially consistent climatological temporal dynamic mapping model, and the temporally static spatial dynamic mapping model, and obtain the target mapping model.

[0049] Specifically, the evaluation unit is used to take the pixel-scale climatological temporal dynamic mapping model, the spatially consistent climatological temporal dynamic mapping model, and the temporally static spatial dynamic mapping model as mapping models to be evaluated, and obtain the SM temporal consistency improvement magnitude ΔR corresponding to each mapping model to be evaluated. SM Error reduction magnitude ΔubRMSD SM Improvement magnitude ΔR in VOD space consistency VOD Based on ΔR SM The improvement, ΔubRMSD SM The magnitude of the decrease, and ΔR VOD To assess the extent of improvement, the configuration with the largest combined improvement across the three indicators is selected as the optimal dynamic parameter mapping model.

[0050] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0051] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0052] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0053] This invention is described with reference to flowchart illustrations and / or block diagrams of the method, terminal device (system), and computer program product according to the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0054] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0055] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0056] It should be noted that: The phrase "an embodiment" or "an embodiment" used in this specification means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention. Therefore, the phrase "an embodiment" or "an embodiment" appearing in various places throughout the specification does not necessarily refer to the same embodiment.

[0057] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the invention.

[0058] Furthermore, it should be noted that the shapes and names of the parts and components described in the specific embodiments described in this specification may differ. All equivalent or simple variations made to the structure, features, and principles described in this patent concept are included within the protection scope of this patent. Those skilled in the art to which this invention pertains may make various modifications or additions to the described specific embodiments or use similar methods to replace them, as long as they do not depart from the structure of this invention or exceed the scope defined in these claims, they should all fall within the protection scope of this invention.

Claims

1. A spatiotemporal dynamic parameterization method for vegetation scattering albedo in a τ–ω model, characterized in that, Includes the following steps: Acquire observation reference data, which is used to indicate the passive microwave brightness temperature observation value in the target area and within the target time range, and also to indicate the incident angle and observation time corresponding to the passive microwave brightness temperature observation value; Based on the aforementioned observation reference data, effective pixels were selected based on the linear regression relationship between reanalysis soil moisture data and passive microwave brightness temperature observations. Based on the observation values ​​of effective canopy temperature and passive microwave brightness temperature, the vegetation scattering albedo corresponding to each effective pixel is calculated on a daily scale to obtain the corresponding daily-scale pixel sequence. Based on the pixel daily scale sequence, a corresponding target mapping model is generated; The target mapping model is used to indicate the mapping relationship between time and / or space and vegetation scattering albedo.

2. The spatiotemporal dynamic parameterization method for vegetation scattering albedo in the τ–ω model according to claim 1, characterized in that: Based on the pre-constructed linear regression relationship between reanalysis soil moisture data and passive microwave brightness temperature observation, pixels with positive regression slopes within a preset incident angle range are considered as valid pixels.

3. The spatiotemporal dynamic parameterization method for vegetation scattering albedo in the τ–ω model according to claim 2, characterized in that: At the effective pixel scale, linear regression relationships were established between reanalysis soil moisture data and passive microwave brightness temperature observations for each incident angle θ of the SMOS satellite, as shown in the following formulas: TB i (y,d,θ)=a i,θ +b i,θ ·SM i ERA5 (y,d); Where i is the pixel number; y is the year; d is the day sequence; TB i (y,d,θ) represents the passive microwave brightness temperature observation value of the i-th pixel under the conditions of year y, day sequence d, and incident angle θ; a i,θ b is the linear regression intercept of the i-th pixel at the incident angle θ; i,θ SM represents the linear regression slope of the i-th pixel at the incident angle θ; i ERA5 (y,d) represents the reanalysis soil moisture data of the i-th pixel under the conditions of year y and day d.

4. The spatiotemporal dynamic parameterization method for vegetation scattering albedo in the τ–ω model according to any one of claims 1 to 3, characterized in that, Based on the observations of effective canopy temperature and passive microwave brightness temperature, the specific steps for calculating the vegetation scattering albedo corresponding to each effective pixel on a diurnal scale and obtaining the corresponding diurnal sequence are as follows: Based on the effective pixels and the date, the effective canopy temperature Tc and the passive microwave brightness temperature observation value TB were paired to obtain the corresponding paired data. Calculate the vegetation scattering albedo for each paired data point based on the incident angle scale; Based on the obtained vegetation scattering albedo, the corresponding pixel diurnal scale sequence is summarized and generated. The formula for calculating vegetation albedo is as follows: ； Where y represents the year; d represents the day sequence; TB i (y,d,θ) represents the brightness temperature value for the corresponding effective pixel, date, and incident angle; T c,i (y,d) represents the effective canopy temperature for the corresponding effective pixels and date.

5. The spatiotemporal dynamic parameterization method for vegetation scattering albedo in the τ–ω model according to any one of claims 1 to 3, characterized in that: Based on the pixel-scale sequence, for each effective pixel, the vegetation scattering albedo of multiple years is averaged over the same day sequence to obtain the average albedo corresponding to each day sequence. Based on the average albedo, the corresponding climatological time dynamic sequence is generated to obtain the corresponding pixel-scale climatological time dynamic mapping model. Based on the aforementioned climatological time dynamic series, the average albedo of each effective pixel in each study area is averaged according to the study area to obtain a corresponding spatially consistent climatological time dynamic mapping model. Based on the daily pixel scale sequence, the daily pixel scale sequence of each pixel i is averaged over many years over the entire time range to obtain the corresponding time-static spatial dynamic mapping model. The optimal dynamic parameter mapping model is determined from the pixel-scale climatological temporal dynamic mapping model, the spatially consistent climatological temporal dynamic mapping model, and the temporally static spatial dynamic mapping model to obtain the target mapping model.

6. The spatiotemporal dynamic parameterization method for vegetation scattering albedo in the τ–ω model according to claim 5, characterized in that: The pixel-scale climatological temporal dynamic mapping model, the spatially consistent climatological temporal dynamic mapping model, and the temporally static spatial dynamic mapping model are respectively used as mapping models to be evaluated, and the SM temporal consistency improvement magnitude ΔR corresponding to each mapping model to be evaluated is obtained. SM Error reduction magnitude ΔubRMSD SM Improvement magnitude ΔR in VOD space consistency VOD Based on ΔR SM The improvement, ΔubRMSD SM The magnitude of the decrease, and ΔR VOD To assess the extent of improvement, the configuration with the largest combined improvement across the three indicators is selected as the optimal dynamic parameter mapping model.

7. A spatiotemporal dynamic parameterization system for vegetation scattering albedo in a τ–ω model, characterized in that, include: The data processing module is used to acquire observation reference data, which is used to indicate the passive microwave brightness temperature observation value in the target area and within the target time range, and also to indicate the incident angle and observation time corresponding to the passive microwave brightness temperature observation value. The filtering module is used to filter effective pixels based on the observation reference data and the linear regression relationship between the reanalysis soil moisture data and the passive microwave brightness temperature observation values. The calculation module is used to calculate the vegetation scattering albedo corresponding to each effective pixel on a daily scale based on the effective canopy temperature and passive microwave brightness temperature observations, and obtain the corresponding daily-scale pixel sequence. The mapping module is used to generate a corresponding target mapping model based on the daily scale sequence of the pixels; The target mapping model is used to indicate the mapping relationship between time and / or space and vegetation scattering albedo.

8. The spatiotemporal dynamic parameterization system for vegetation scattering albedo in the τ–ω model according to claim 7, characterized in that: The screening module is based on the linear regression relationship between pre-constructed reanalysis soil moisture data and passive microwave brightness temperature observation. Pixels with positive regression slopes within a preset incident angle range are selected as valid pixels.

9. The spatiotemporal dynamic parameterization system for vegetation scattering albedo in the τ–ω model according to claim 7 or 8, characterized in that, The computing module includes: The pairing unit is used to pair the effective canopy temperature Tc with the passive microwave brightness temperature observation value TB based on the effective pixels and the date, and obtain the corresponding paired data. The computing unit is used to calculate the vegetation scattering albedo of each paired data based on the incident angle scale; The summarization unit is used to summarize and generate the corresponding daily-scale pixel sequence based on the obtained vegetation scattering albedo. The calculation unit calculates the vegetation albedo based on the following formula: ； Where y represents the year; d represents the day sequence; TB i (y,d,θ) represents the brightness temperature value for the corresponding effective pixel, date, and incident angle; T c,i (y,d) represents the effective canopy temperature for the corresponding effective pixels and date.

10. The spatiotemporal dynamic parameterization method for vegetation scattering albedo in the τ–ω model according to claim 7 or 8, characterized in that: The mapping module includes: The first mapping unit is used to take the multi-year average of the vegetation scattering albedo of multiple years for each effective pixel on the same day sequence based on the pixel day-scale sequence, to obtain the average albedo corresponding to each day sequence, and to generate the corresponding climatological time-dynamic sequence based on the average albedo, thereby obtaining the corresponding pixel-scale climatological time-dynamic mapping model. The second mapping unit is used to average the average albedo of each effective pixel in each study area according to the climatological time dynamic sequence, based on the climatological time dynamic sequence, to obtain a corresponding spatially consistent climatological time dynamic mapping model. The third mapping unit is used to take the multi-year average of the daily scale sequence of each pixel i over the entire time range based on the daily scale sequence of the pixels, and obtain the corresponding time-static spatial dynamic mapping model. The evaluation unit is used to determine the optimal dynamic parameter mapping model from the pixel-scale climatological temporal dynamic mapping model, the spatially consistent climatological temporal dynamic mapping model, and the temporally static spatial dynamic mapping model, and obtain the target mapping model.

Images