Forest carbon reserve and carbon sink monitoring and evaluating method based on multi-source remote sensing technology

Abstract

The invention discloses a forest carbon reserve and carbon sink monitoring and evaluating method based on a multi-source remote sensing technology. The method comprises the following steps: multi-source remote sensing data acquisition: respectively acquiring optical remote sensing data and radar remote sensing data; tree species classification and identification: carrying out tree species classification on the forest in the research area to realize tree species scale space distribution information extraction; for different tree species, biomass model parameters obtained through field investigation are combined, a biomass estimation model based on the tree species is established, model input parameters are biomass model parameters, and vegetation biomass is output; calculating the carbon reserve: calculating the vegetation carbon reserve by utilizing the inverted vegetation biomass and combining the carbon content coefficients of different tree species; estimating the soil carbon reserve, and finally obtaining the total carbon reserve of the forest ecosystem: calculating the carbon reserve variation of the forest ecosystem at different time points based on the multi-source remote sensing data of the long-term sequence. According to the invention, large-range, high-precision and continuous forest carbon reserve and carbon sink monitoring and evaluation considering tree species difference is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ecological environment monitoring, and particularly relates to a forest carbon storage and carbon sink monitoring and evaluation method based on multi-source remote sensing technology. BACKGROUND

[0002] With the increasingly serious global climate change problem, as the largest carbon pool in terrestrial ecosystems, the monitoring and evaluation of forest carbon storage and carbon sink is of great significance for addressing climate change and developing carbon emission reduction strategies. However, the traditional forest carbon storage and carbon sink monitoring and evaluation methods have many defects:

[0003] Ground plot investigation method: by setting up sample plots, measuring the diameter at breast height, height and other parameters of trees, and combining with biomass models to calculate carbon storage. This method has high precision, but has large workload and high cost, and can only obtain carbon storage data of limited sample points, making it difficult to realize large-scale and continuous monitoring; at the same time, in complex terrain and dense vegetation areas, it is difficult to set up sample plots and measure, and the representativeness is limited.

[0004] Model estimation method: using existing carbon storage models such as biomass expansion model and process model, combined with meteorological data, soil data, etc. for estimation; the model estimation method can realize large-scale carbon storage estimation, but the accuracy of the model depends on a large number of parameter inputs and the assumptions of the model itself, and the adaptability to different regions and different vegetation types is poor, and it is difficult to consider the influence of the microstructure of the forest and the difference between tree species on carbon storage.

[0005] Single remote sensing monitoring method: relying only on optical remote sensing or radar remote sensing for monitoring; optical remote sensing can obtain rich surface information of vegetation, but data acquisition is limited in cloudy and foggy weather and during the growing season of vegetation; spaceborne radar remote sensing is sensitive to the structural information of vegetation, but the spatial resolution is relatively low, and the quantitative inversion precision of vegetation type and carbon storage is limited. In addition, these single remote sensing monitoring methods often cannot directly obtain carbon storage and carbon sink information at the tree species scale, while the different growth characteristics and carbon absorption capacity of tree species have important influence on the distribution and change of forest carbon storage and carbon sink. SUMMARY

[0006] In order to overcome the above-mentioned defects of the prior art, the present application proposes a forest carbon storage and carbon sink monitoring and evaluation method based on multi-source remote sensing technology to overcome the defects of the prior art and realize large-scale, high-precision, continuous and tree species difference monitoring and evaluation of forest carbon storage and carbon sink.

[0007] The technical solution adopted by the present application to solve its technical problems is: a forest carbon storage and carbon sink monitoring and evaluation method based on multi-source remote sensing technology, comprising the following steps:

[0008] Multi-source remote sensing data acquisition: optical remote sensing data and radar remote sensing data are acquired respectively;

[0009] Tree species classification and identification: tree information obtained through field investigation is used to construct a tree species classification model based on historical optical remote sensing data and radar remote sensing data, the forest in the study area is classified by tree species, and spatial distribution information of tree species is extracted; the tree information includes tree diameter at breast height, tree height, and tree species type;

[0010] Calculation of vegetation biomass: for different tree species, a biomass estimation model based on tree species is established by combining biomass model parameters obtained through field investigation, the input parameters of the model are biomass model parameters, and the output is vegetation biomass; the biomass model parameters include vegetation index inverted from optical remote sensing and vegetation height inverted from radar remote sensing;

[0011] Calculation of carbon storage: the vegetation carbon storage is calculated by using the inverted vegetation biomass and combining the carbon content coefficients of different tree species; at the same time, the soil carbon storage model is used to estimate the soil carbon storage by combining the vegetation coverage inverted from optical remote sensing, the topographic and geomorphic information inverted from radar remote sensing, and the soil sample data obtained through field investigation, and finally the total carbon storage of the forest ecosystem is obtained;

[0012] Forest carbon sink assessment: the carbon storage change amount of the forest ecosystem at different time points is calculated by combining the Biome-BGC model and the carbon storage estimation results.

[0013] Further, the optical remote sensing data is acquired by using an optical remote sensing satellite to acquire high-resolution band images, and the band images include visible light images and near-infrared images.

[0014] Further, the radar remote sensing data is acquired by using a radar remote sensing satellite, and the radar data used is the synthetic aperture radar of the Sentinel satellite, and the band is VV and VH band.

[0015] Further, the tree species classification model is constructed by the following steps:

[0016] (1) Collect typical tree species sample data in the study area, specifically tree information obtained through field investigation, and record the corresponding positions;

[0017] (2) Obtain samples of optical remote sensing data and radar remote sensing data at the corresponding positions according to the multi-source remote sensing data acquisition step; the optical remote sensing data is used to extract spectral features and texture features; the radar remote sensing data is used to extract backscatter features;

[0018] (3) Construct a multi-dimensional feature vector for the sample points corresponding to the tree positions investigated in the field; the multi-dimensional feature vector is a combination of spectral features, texture features, and backscatter features;

[0019] (4) input the multi-dimensional feature vector into a random forest to build a tree species classification model; the tree species classification model inputs the multi-dimensional feature vector of each sample point and outputs the tree species type.

[0020] Further, the biomass estimation model is constructed according to the following formula:

[0021]

[0022] In the formula:

[0023] B TREE-BSL,i,j,t = biomass of the i-th baseline carbon layer tree species j at the t-th year;

[0024] f j (x1 i,j,t ,x2 i,j,t ,x3 i,j,t ,…) = a regression equation for converting the tree survey factors (x1, x2, x3,...) of the i-th baseline carbon layer tree species j at the t-th year into aboveground biomass; the tree survey factors (x1, x2, x3,...) are tree information;

[0025] R TREE-BSL,i,j = ratio of underground biomass to aboveground biomass of the tree species j;

[0026] N TREE-BSL,i,j,t = number of trees of the i-th baseline carbon layer tree species j at the t-th year; = area of the i-th baseline carbon layer; j = 1, 2, 3,..., tree species in the i-th baseline carbon layer; i = 1, 2, 3,..., the i-th baseline carbon layer; t = 1, 2, 3,..., the number of years since the activity began.

[0027] Further, the vegetation carbon storage is calculated by the following formula:

[0028]

[0029] In the formula, C TREE_BSL,i,j,t = biomass carbon storage of the i-th baseline carbon layer tree species j at the t-th year;

[0030] B TREE_BSL,i,j,t = biomass of the i-th baseline carbon layer tree species j at the t-th year;

[0031] CF TREE_BSL,j = carbon content coefficient, which is the carbon content rate of the biomass of the tree species j;

[0032] = ratio of the molecular weight of CO2 to C.

[0033] Further, the specific method for estimating the soil carbon storage is:

[0034] The soil carbon storage model is established, including a mechanism model and a machine learning model; the mechanism model adopts the Biome-BGC model, which is used to construct a model for outputting a mechanism value of soil carbon storage according to vegetation carbon storage, vegetation coverage, landform information, and soil sample data; the machine learning model adopts an LSTM neural network, which is used to predict a residual value for factors not covered by the mechanism model; the soil carbon storage model outputs soil carbon storage; and the soil carbon storage = soil carbon storage mechanism value + residual value.

[0035] Further, the calculation method of the total carbon storage of the forest ecosystem is total carbon storage = vegetation carbon storage + soil carbon storage.

[0036] Further, the specific method for calculating the carbon storage change amount of the forest ecosystem at different time points is as follows:

[0037] (1) The spatiotemporal scale to be evaluated is determined; the spatiotemporal scale includes a region and a time range;

[0038] (2) The net change amount ΔC of the total carbon storage of the forest ecosystem in the target time period is set; ΔC = C end -C start ;

[0039] (3) According to the time scale and the total carbon storage time series data (C t1 ,C t2 ,...,C tn ) obtained in the above step;

[0040] (4) Collect the key variables of meteorological data covering the target region and time period, including:

[0041] Temperature, including daily mean temperature, maximum temperature, minimum temperature, and accumulated temperature;

[0042] Precipitation, including total amount, intensity, and frequency;

[0043] Solar radiation, specifically photosynthetically active radiation PAR or total radiation;

[0044] Humidity / vapor pressure deficit; wind speed; soil moisture;

[0045] Match the above-processed carbon storage data with the spatiotemporal data to obtain time series meteorological data;

[0046] (5) Use the collected total carbon storage time series data to directly calculate the carbon storage change amount ΔC obs in the target time period: ΔC tend = C tstart ;

[0047] (6) Deduct the carbon loss directly caused by non-meteorological factors in the same period from the carbon storage change amount.

[0048] (7) Establish a meteorological-carbon process relationship model, use Biome-BGC model for modeling, and the trained meteorological-carbon process relationship model is used to simulate the key carbon fluxes driven by meteorological variables; the key carbon fluxes include GPP, Ra, Rh, NPP, NEP; input time series of meteorological data, initial carbon storage, and output: dynamic changes of key carbon fluxes;

[0049] (8) Calculate the simulated final total carbon storage C_sim_t_end;

[0050] (9) Calculate the simulated carbon storage change amount: AC_sim=C_sim_t_end-C_t_start; the final carbon storage change is the result of the integral of the key carbon fluxes output by the meteorological-carbon process relationship model;

[0051] (10) Compare the model-simulated carbon sink amount AC_sim with the benchmark-observed carbon sink amount AC_obs_net after disturbance correction;

[0052] (11) Use statistical indicators to verify and obtain an accurate meteorological-carbon process relationship model; and obtain the final carbon storage change amount according to steps (1)-(10).

[0053] According to the definition of forest carbon sink, the carbon sink amount is equal to the increase of carbon storage, so that the dynamic assessment of the forest carbon sink of the research area is realized.

[0054] Compared with the prior art, the beneficial effects of the present application are:

[0055] Efficient monitoring of large-scale forest carbon storage and carbon sink is realized: through multi-source remote sensing technology, remote sensing data of large-area forests can be obtained in a short time, combined with the method, forest carbon storage and carbon sink monitoring work can be quickly and efficiently carried out, compared with the traditional ground sample survey method, the monitoring efficiency is greatly improved, and the cost is reduced.

[0056] Improve monitoring accuracy and reliability: fuse multi-source remote sensing data, give full play to the advantages of optical remote sensing and radar remote sensing, overcome the limitations of single remote sensing means, and introduce field investigation data for model calibration and verification, so that the estimation results of forest carbon storage and carbon sink are more accurate and reliable.

[0057] Take into account the differences between different tree species in growth characteristics and carbon absorption capacity, the present application fully embodies the fine monitoring of tree species scale in the tree species classification and carbon storage estimation link, and provides more accurate data support for in-depth research on forest carbon cycle process and the development of targeted forest management strategies.

[0058] Dynamic monitoring of forest carbon sink changes: can realize dynamic evaluation of forest carbon sink in combination with the Biome-BGC model, help to timely grasp the carbon sink change trend of the forest ecosystem under climate change and human disturbance, and provide a scientific basis for policy making and emission reduction measure implementation in response to climate change. BRIEF DESCRIPTION OF DRAWINGS

[0059] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description.

[0060] Figure 1 The flow chart of the forest carbon storage and carbon sink monitoring and evaluation method based on multi-source remote sensing technology of the embodiments of the present application. DETAILED DESCRIPTION

[0061] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0062] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application.

[0063] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connection" should be understood broadly, for example, it can be fixed connection, or detachable connection, or integrally connected; it can be mechanical connection, or electrical connection; it can be directly connected, or indirectly connected through intermediate medium, or the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances. In addition, the technical features involved in the different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0064] As Figure 1 shown, the forest carbon storage and carbon sink monitoring and evaluation method based on multi-source remote sensing technology of the embodiments of the present application,

[0065] The forest carbon storage and carbon sink monitoring and evaluation method based on multi-source remote sensing technology comprises the following modules:

[0066] Multi-source remote sensing data acquisition: optical remote sensing data and radar remote sensing data are acquired respectively; the optical remote sensing data is collected by using unmanned aerial vehicle ultra-high resolution images; the optical remote sensing data can also be satellite remote sensing data.

[0067] Tree species classification and identification: the tree information obtained by field investigation is used to construct a tree species classification model according to historical optical remote sensing data and radar remote sensing data, the forest in the study area is classified by tree species, and the spatial distribution information of the tree species is extracted; the tree information includes tree diameter at breast height, tree height, and tree species type;

[0068] Calculation of vegetation biomass: for different tree species, a biomass estimation model based on tree species is established by combining the biomass model parameters obtained by field investigation, the input parameters of the model are biomass model parameters, and the output is vegetation biomass; the biomass model parameters include vegetation index and vegetation height inverted by optical remote sensing and radar remote sensing; the vegetation index is NDVI / EVI.

[0069] Calculation of carbon storage: the vegetation carbon storage is calculated by using the inverted vegetation biomass and combining the carbon content coefficients of different tree species; at the same time, the soil carbon storage model is used to estimate the soil carbon storage by combining the vegetation coverage inverted by optical remote sensing, the topographic and geomorphic information inverted by radar remote sensing, and the soil sample data obtained by field investigation, and finally the total carbon storage of the forest ecosystem is obtained;

[0070] Forest carbon sink evaluation: the carbon storage change amount of the forest ecosystem at different time points is calculated by combining the Biome-BGC model and the carbon storage estimation result.

[0071] The optical remote sensing data is obtained by using optical remote sensing satellites to obtain high-resolution band images, and the band images include visible light images and near-infrared images.

[0072] The radar remote sensing data is obtained by using radar remote sensing satellites, and the radar data used is the synthetic aperture radar of the Sentinel satellite, and the wave band is VV and VH.

[0073] The tree species classification model includes the following steps:

[0074] (1) Collect typical tree species sample data in the study area, specifically tree information obtained by field investigation, and record the corresponding position;

[0075] (2) Obtain the samples of optical remote sensing data and radar remote sensing data at the corresponding position according to the multi-source remote sensing data acquisition step; the optical remote sensing data is used to extract spectral features and texture features; the radar remote sensing data is used to extract backscattering features;

[0076] (3) Constructing a multi-dimensional feature vector of the sample point as the position of the tree corresponding to the field survey; the multi-dimensional feature vector is a combination of spectral features, texture features, and backscattering features; 70% of the multi-dimensional feature vector data is used as a training set for training, and 30% of the data is used as a verification set for accuracy verification. The texture feature can be a GLCM index.

[0077] (4) Inputting the multi-dimensional feature vector into a random forest to construct a tree species classification model; the tree species classification model inputs the multi-dimensional feature vector of each sample point and outputs the tree species type; the accuracy of the tree species classification model is verified by the verification set.

[0078] Further, the biomass estimation model is constructed according to the following formula:

[0079]

[0080] In the formula:

[0081] B TREE-BSL,i,j,t = biomass of the i-th baseline carbon layer tree species j at the t-th year;

[0082] f j (x1 i,j,t ,x2 i,j,t ,x3 i,j,t ,…) = regression equation for converting the tree measurement factor (x1, x2, x3,...) of the i-th baseline carbon layer tree species j at the t-th year into aboveground biomass; the tree measurement factor (x1, x2, x3,...) is tree information;

[0083] R TREE-BSL,i,j = ratio of underground biomass to aboveground biomass of tree species j;

[0084] N TREE-BSL,i,j,t = number of trees of the i-th baseline carbon layer tree species j at the t-th year; = area of the i-th baseline carbon layer; j = 1, 2, 3,..., tree species in the i-th baseline carbon layer; i = 1, 2, 3,..., the i-th baseline carbon layer; t = 1, 2, 3,..., the number of years since the start of the activity.

[0085] The calculation of the vegetation carbon storage is specifically calculated by the following formula:

[0086]

[0087] In the formula, C TREE_BSL,i,j,t = biomass carbon storage of the i-th baseline carbon layer tree species j at the t-th year;

[0088] B TREE_BSL,i,j,t = biomass of the i-th baseline carbon layer tree species j at the t-th year;

[0089] CFTREE_BSL,j = the carbon content coefficient of the biomass of tree species j, i.e., the carbon content of the biomass of tree species j; for example, the carbon content coefficient of coniferous forests is 0.45-0.50, and the carbon content coefficient of broad-leaved forests is 0.48-0.53.

[0090] = the ratio of the molecular weight of CO2 to the molecular weight of C.

[0091] The specific method for estimating the soil carbon storage is as follows:

[0092] The soil carbon storage model is established, including a mechanism model and a machine learning model; the mechanism model uses the Biome-BGC model to construct a model for outputting a mechanism value of the soil carbon storage according to the vegetation carbon storage, the vegetation coverage, the topographic information, and the soil sample data; the machine learning model uses an LSTM neural network to predict a residual value for factors not covered by the mechanism model; the soil carbon storage model outputs the soil carbon storage; and the soil carbon storage = the mechanism value of the soil carbon storage + the residual value.

[0093] The calculation method for the total carbon storage of the forest ecosystem is total carbon storage = vegetation carbon storage + soil carbon storage.

[0094] Forest carbon sink assessment: based on long-term sequence multi-source remote sensing data, the growth changes of different tree species are analyzed, and the carbon storage estimation results are combined to calculate the carbon storage change amount of the forest ecosystem at different time points. According to the definition of forest carbon sink, the carbon sink amount is equal to the increase amount of the carbon storage, thereby realizing the dynamic assessment of the forest carbon sink of the research region.

[0095] The specific method for calculating the carbon storage change amount of the forest ecosystem at different time points is as follows:

[0096] (1) defining the spatiotemporal scale of the assessment; the spatiotemporal scale includes the region and the time range;

[0097] (2) setting the net change amount ΔC of the total carbon storage of the forest ecosystem in the target time period; ΔC = C end -C start ;

[0098] (3) obtaining the time sequence data (C t1 , C t2 ,..., C tn ) of the total carbon storage according to the time scale and the above steps;

[0099] (4) collecting key variables of meteorological data covering the target region and time period, and the key variables include:

[0100] temperature, including daily mean temperature, maximum temperature, minimum temperature, and accumulated temperature;

[0101] precipitation, including total amount, intensity, and frequency;

[0102] Solar radiation, in particular photosynthetically active radiation (PAR) or total radiation;

[0103] Humidity / vapor pressure deficit (VPD); wind speed; soil moisture; (data sources: weather station observations, reanalysis data (e.g. ERA5, MERRA-2), remote sensing products, regional climate model outputs.

[0104] Temporal and spatial matching of the above-mentioned treatment and carbon storage data to obtain time series of meteorological data;

[0105] (5) Using the collected total carbon storage time series data, directly calculate the carbon storage change amount of the target time period: ΔC obs = C tend -C tstart .

[0106] (6) Deduct the carbon loss directly caused by non-meteorological factors (mainly human disturbance and natural disturbance) in the same period from the carbon storage change amount;

[0107] (7) Establish a meteorological-carbon process relationship model, use the Biome-BGC model for modeling, and the trained meteorological-carbon process relationship model is used to simulate how meteorological variables drive the key carbon fluxes of GPP (gross primary productivity, carbon sequestration), Ra (autotrophic respiration, carbon consumption), Rh (heterotrophic respiration, carbon consumption), NPP (net primary productivity, NPP=GPP-Ra), NEP (net ecosystem productivity, NEP=NPP-Rh); Input time series of meteorological data, initial carbon storage, output: dynamic changes of GPP, NPP, Ra, Rh, NEP;

[0108] (8) Calculate the simulated end-of-period total carbon storage C_sim_t_end;

[0109] (9) Calculate the simulated carbon storage change amount (i.e. the carbon sink amount evaluated by the model): ΔC_sim=C_sim_t_end-C_t_start; The final carbon storage change is the result of the integral of the key carbon fluxes output by the meteorological-carbon process relationship model;

[0110] (10) Compare the model-simulated carbon sink amount ΔC_sim with the disturbance-corrected baseline observed carbon sink amount ΔC_obs_net;

[0111] (11) Use statistical indicators to verify and obtain an accurate meteorological-carbon process relationship model; and obtain the final carbon storage change amount according to steps (1)-(10).

[0112] The above merely illustrates the specific embodiments of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of the changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for monitoring and evaluating forest carbon storage and carbon sink based on multi-source remote sensing technology, characterized in that, The method comprises the following steps: Multi-source remote sensing data acquisition: optical remote sensing data and radar remote sensing data are acquired respectively; Tree species classification and identification: tree information obtained through field investigation is used to construct a tree species classification model according to historical optical remote sensing data and radar remote sensing data, the forest in the study area is classified by tree species, and spatial distribution information of the tree species is extracted; the tree information includes tree diameter at breast height, tree height and tree species type; Calculation of vegetation biomass: for different tree species, a biomass estimation model based on tree species is established by combining biomass model parameters obtained through field investigation, the input parameter of the model is a biomass model parameter, and the biomass model parameter includes a vegetation index inversed from optical remote sensing and a vegetation height inversed from radar remote sensing, and the output is vegetation biomass. The vegetation index is, for example, NDVI / EVI. Calculation of carbon storage: the vegetation carbon storage is calculated by using the inversed vegetation biomass and combining carbon content coefficients of different tree species; meanwhile, a soil carbon storage model is used to estimate soil carbon storage by combining vegetation coverage inversed from optical remote sensing, topographic information inversed from radar remote sensing and soil sample data obtained through field investigation, and finally the total carbon storage of the forest ecosystem is obtained; Forest carbon sink evaluation: the carbon storage change amount of the forest ecosystem at different time points is calculated by combining the Biome-BGC model and the carbon storage estimation result. 2.The method for monitoring and evaluating forest carbon storage and carbon sink based on multi-source remote sensing technology according to claim 1, characterized in that, The optical remote sensing data is acquired by using an optical remote sensing satellite to acquire high-resolution band images, and the band images include visible light images and near-infrared images. 3.The method of claim 1, wherein the method further comprises: The radar remote sensing data is acquired by using a radar remote sensing satellite, and the radar data used is synthetic aperture radar of a Sentinel satellite, and the band is VV and VH band. 4.The method for monitoring and evaluating forest carbon storage and carbon sink based on multi-source remote sensing technology according to claim 1, characterized in that, The tree species classification model comprises the following steps: (1) collecting typical tree species sample data in the study area, specifically tree information obtained through field investigation and recording corresponding positions; (2) acquiring samples of optical remote sensing data and radar remote sensing data of the corresponding positions according to the multi-source remote sensing data acquisition step; the optical remote sensing data is used to extract spectral features and texture features; the radar remote sensing data is used to extract backscatter features; (3) constructing a multi-dimensional feature vector of the sample points as the corresponding tree positions investigated in the field; the multi-dimensional feature vector is a combination of spectral features, texture features and backscatter features; (4) inputting the multi-dimensional feature vector into a random forest to construct a tree species classification model; the input of the tree species classification model is the multi-dimensional feature vector of each sample point, and the output is the tree species type. 5.The method for monitoring and evaluating forest carbon storage and carbon sink based on multi-source remote sensing technology according to claim 1, characterized in that, The biomass estimation model is constructed according to the following formula: B TREE-BSL,i,j,t = f j (x1 i,j,t ,x2 i,j,t ,x3 i,j,t ,…)*(1+R TREE-BSL,i,j )*N TREE-BSL,i,j,t *A TREE-BSL,i ; In the formula: B TREE-BSL,i,j,t = biomass of the i-th baseline carbon layer tree species j at year t; f j (x1 i,j,t ,x2 i,j,t ,x3 i,j,t ,…) = the regression equation that converts the dendrometric factors (x1, x2, x3, …) of tree species j in the baseline carbon layer i in year t into aboveground biomass; the dendrometric factors (x1, x2, x3, …) are tree information; R TREE-BSL,i,j = ratio of belowground biomass / aboveground biomass for species j; N TREE-BSL,i,j,t = number of trees of species j in the ith baseline carbon layer at time t; A TREE-BSL,i = area of the ith baseline carbon layer; j = 1, 2, 3,... species in the ith baseline carbon layer; i = 1, 2, 3,... the ith baseline carbon layer; t = 1, 2, 3,... years since the activity began. 6.The method for monitoring and evaluating forest carbon storage and carbon sink based on multi-source remote sensing technology according to claim 1, characterized in that, The vegetation carbon storage is calculated by the following formula: where C TREE_BSL,i,j,t = the biomass carbon stock of the i-th baseline carbon layer of tree species j at year t; B TREE_BSL,i,j,t = biomass of the i-th baseline carbon layer tree species j at year t; CF TREE_BSL,j = carbon content coefficient, i.e. the carbon content rate of the biomass of tree species j; The molecular weight ratio of C. 7.The forest carbon stock and carbon sink monitoring and assessment method based on multi-source remote sensing technology according to claim 1, characterized in that, The specific method for estimating the soil carbon storage is: A soil carbon storage model is established, including a mechanism model and a machine learning model; the mechanism model adopts the Biome-BGC model, which is used to construct a model for outputting a mechanism value of soil carbon storage according to vegetation carbon storage, vegetation coverage, topographic information and soil sample data; the machine learning model adopts an LSTM neural network, which is used to predict a residual value for factors not covered by the mechanism model; the soil carbon storage model outputs the soil carbon storage; and the soil carbon storage = the mechanism value of soil carbon storage + the residual value. 8.The forest carbon storage and carbon sink monitoring and evaluation method based on multi-source remote sensing technology according to claim 1, characterized in that, The method for calculating the total carbon storage of a forest ecosystem is total carbon storage = vegetation carbon storage + soil carbon storage. 9.The forest carbon storage and carbon sink monitoring and evaluation method based on multi-source remote sensing technology according to claim 1, characterized in that, The specific method for calculating the carbon storage change of a forest ecosystem at different time points is as follows: (1) defining the spatiotemporal scale of the assessment; the spatiotemporal scale includes the region and the time range; (2) setting a net change amount AC of the total carbon storage of the forest ecosystem in a target time period; AC = C end -C start ; (3) Obtain the total carbon storage time series data (C t1 ,C t2 ,...,C tn ); (4) collecting key variables of meteorological data covering the target region and time period, including: temperature, including daily mean temperature, maximum temperature, minimum temperature, accumulated temperature; precipitation, including total amount, intensity, frequency; solar radiation, specifically photosynthetically active radiation PAR or total radiation; humidity / vapor pressure deficit; wind speed; soil moisture; spatiotemporal matching of the above-mentioned processing and carbon storage data to obtain time series meteorological data; (5) Using the total carbon storage time series data collected, directly calculate the carbon storage change amount of the target time period: AC obs = C tend -C tstart ; (6) deducting the carbon loss directly caused by non-meteorological factors in the same period from the carbon storage change; (7) establishing a meteorological-carbon process relationship model, using the Biome-BGC model for modeling, and using the trained meteorological-carbon process relationship model to simulate meteorological variables driving key carbon fluxes; the key carbon fluxes include GPP, Ra, Rh, NPP, NEP; input time series meteorological data, initial carbon storage, and output: dynamic changes of key carbon fluxes; (8) calculating the simulated end-of-period total carbon storage C_sim_t_end; (9) calculating the simulated carbon storage change: ΔC_sim = C_sim_t_end - C_t_start; the final carbon storage change is the result of integrating the key carbon fluxes output by the meteorological-carbon process relationship model; (10) comparing the model-simulated carbon sink amount ΔC_sim with the benchmark-observed carbon sink amount ΔC_obs_net after disturbance correction; (11) using statistical indicators to verify and obtain an accurate meteorological-carbon process relationship model; and obtaining the final carbon storage change according to steps (1)-(10).

Images