A forestry carbon sink prediction method and system based on big data

Abstract

This invention relates to the field of forestry carbon sink prediction technology, specifically a forestry carbon sink prediction method and system based on big data. The method includes: acquiring first data; establishing an objective function and constraints for optimizing wildfire loss deviation based on wildfire sample sequences, biomass volume sequences, biomass growth rate sequences, and air humidity data sequences of the forest area to be assessed; calculating the optimal wildfire loss ratio coefficient using a particle swarm optimization algorithm; and calculating the predicted annual carbon sink volume based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, biomass expansion factor, biomass carbon content, timber density, pre-obtained expected wildfire time series, pre-obtained expected air humidity data sequence, and pre-obtained initial value of predicted annual biomass volume. This invention performs optimization calculations based on the first data to achieve carbon sink volume prediction considering the impact of wildfire losses.

Description

Technical Field

[0001] This invention relates to the field of forestry carbon sink prediction technology, specifically a forestry carbon sink prediction method and system based on big data. Background Technology

[0002] In the field of forestry carbon sink assessment, the initial value of biomass and the change in biomass at the end of the year are usually used as the basis. Based on static parameters such as biomass expansion factor and biomass carbon content, the annual forest carbon sink is calculated according to the principle of carbon conservation. The initial value of biomass and the change in biomass at the end of the year are often obtained through plot surveys. However, since biomass depends on plot surveys, this method cannot predict future forest carbon sinks. To overcome this deficiency, some studies have extrapolated the year-end biomass based on the initial value of biomass and the biomass growth rate. However, this method has the following drawback: it does not fully consider the potential disturbance from wildfires. In arid primary forest areas, seasonal wildfires are frequent. Wildfires cause biomass loss, thus affecting the predicted carbon sink value. Current techniques, when considering wildfire disturbance, often use the wildfire loss rate as a general term, assuming that the wildfire loss rate is a constant obtained through human experience. However, in actual forest environments, the losses caused by wildfires are related not only to the current biomass but also to air humidity. Higher air humidity results in lower losses. Furthermore, in the same forest area, especially in primary forest areas, because wildfires are difficult to extinguish manually, the biomass loss from each fire is often related to the natural endowment and geographical conditions of the forest area, exhibiting certain patterns. Current technology has not yet explored these patterns.

[0003] With the development of big data technology, it has become possible to predict forest carbon sequestration using multidimensional data. Therefore, it is necessary to explore the impact of air humidity on wildfire losses and establish a recursive formula for predicting biomass based on historical data, taking into account the impact of wildfire losses, thereby achieving forest carbon sequestration prediction. Summary of the Invention

[0004] (1) Technical problems to be solved

[0005] The purpose of this invention is to provide a forestry carbon sink prediction method and system based on big data, so as to realize the prediction of forest carbon sink volume considering the impact of wildfire losses.

[0006] (2) Technical solution

[0007] To achieve the above objectives, this invention provides a forestry carbon sink prediction method based on big data, the method comprising the following steps:

[0008] S1, Obtain first data; the first data includes pre-obtained wildfire sample sequences, biomass stock sequences, biomass expansion factors, biomass carbon content, wood density, biomass growth rate sequences, and air humidity data sequences of the forest area to be evaluated.

[0009] S2, establish an objective function and constraints for optimizing the wildfire loss deviation based on the wildfire sample sequence, biomass volume sequence, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated; the optimization variable of the objective function for optimizing the wildfire loss deviation is the wildfire loss ratio coefficient; the wildfire loss ratio coefficient represents the ratio of the wildfire loss ratio to the reciprocal of the air humidity.

[0010] S3, with the objective of minimizing the value of the objective function for optimizing the wildfire loss deviation as the objective and with the constraints as the constraints, the optimal wildfire loss ratio coefficient is calculated using the particle swarm optimization algorithm.

[0011] S4. The predicted annual carbon sink is calculated based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, biomass expansion factor, biomass carbon content, wood density, pre-obtained expected wildfire time series, pre-obtained expected air humidity data series, and pre-obtained initial value of predicted annual biomass.

[0012] Furthermore, the method for obtaining the first data includes:

[0013] Acquire pre-obtained wildfire sample sequences, biomass volume sequences, biomass expansion factors, biomass carbon content, timber density, biomass growth rate sequences, and air humidity data sequences for the forest area to be assessed. The wildfire sample sequences for the forest area to be assessed are the times when wildfires occurred within the forest area to be assessed during the year monitored. The area of ​​the forest area to be assessed is 1 hectare. The times when wildfires occurred within the forest area to be assessed are in daily units. The data in the wildfire sample sequences of the forest area to be assessed are sequentially recorded as the first wildfire time to the Nth wildfire time, where N represents the total number of wildfires occurring within the forest area to be assessed during the year. The biomass volume sequence is the biomass per hectare obtained from sample plot surveys on fixed dates each month. The data in the biomass volume sequence are sequentially recorded as the first biomass volume to the twelfth biomass volume. The sampling times corresponding to the twelfth biomass volume are sequentially recorded as the first to twelfth sampling times; the sampling times are in days; the biomass volume represents the total volume of living tree trunks per hectare of forest area, in cubic meters per hectare; the biomass expansion factor represents the conversion coefficient of the total volume of living tree trunks to the total biomass of living trees, dimensionless; the biomass carbon content represents the proportion of carbon in the weight of the total biomass of living trees; the timber density is in tons per cubic meter; the biomass growth rate sequence represents the average natural growth rate of timber, in cubic meters per day; the air humidity data sequence is a data sequence composed of the average air humidity of each day within the sample year; the wildfire sample sequence, biomass volume sequence, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated are located on the same time coordinate.

[0014] Furthermore, the method for establishing the objective function and constraints for optimizing the wildfire loss deviation based on the wildfire sample sequence, biomass volume sequence, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated includes:

[0015] Obtain the number of sample years Y; denote the Y sample years as the first sample year to the Yth sample year; divide each sample year into 11 intervals, denote them as the first growth interval to the eleventh growth interval, using the first survey time to the twelfth survey time as the dividing point; divide the first growth interval to the eleventh growth interval into the first segment group to the eleventh segment group, using the first wildfire time to the Nth wildfire time as the dividing point; there is no overlap between the first wildfire time to the Nth wildfire time and the first survey time to the twelfth survey time.

[0016] Based on the wildfire sample sequence and biomass sequence of the forest area to be assessed, a biomass recursive equation is established; the biomass recursive equation includes a first biomass recursive equation and a second biomass recursive equation; the first biomass recursive equation is the recursive equation when the number of segments in the i-th segment group in the t-th sample year is greater than 1; the first biomass recursive equation is expressed as:

[0017]

[0018] Among them, u i,j,t This represents the expected biomass of the j-th growth interval in the t-th sample year; i is an integer variable ranging from 1 to 11; j is a variable ranging from 0 to M. i,t The integer variable; t is an integer variable taking values ​​from 1 to Y; when j>0, a i,j,t Δt represents the average biomass growth rate sequence corresponding to the j-th segment within the i-th segment group in the t-th sample year; when j>0, Δt i,j,t M represents the length of the j-th segment within the i-th segment group in the t-th sample year, in days; i,t V represents the number of segments in the i-th segment group in the t-th sample year; i,t w represents the biomass of sample year t; k represents the wildfire loss ratio; when j>0, w i,j,t This represents the average value of the air humidity data sequence corresponding to the j-th segment within the i-th segment group in the t-th sample year.

[0019] The second recursive equation for biomass is the recursive equation when the number of fragments in the i-th fragment group in the t-th sample year is equal to 1; the second recursive equation for biomass is expressed as:

[0020] u i,1,t =V i,t (1+a i,1,t Δt i,1,t ).

[0021] An objective function for optimizing wildfire loss deviation is established based on the aforementioned bioaccumulation recursive equation; the objective function for optimizing wildfire loss deviation is:

[0022]

[0023] Where f(k) represents the objective function for optimizing the wildfire loss deviation; This indicates that when j takes the value M i,t time u i,j,t The value of .

[0024] Construct constraints, the constraints being:

[0025]

[0026] Furthermore, the method for calculating the predicted annual carbon sink based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, biomass expansion factor, biomass carbon content, wood density, pre-obtained expected wildfire time series, pre-obtained expected air humidity data series, and pre-obtained initial value of predicted annual biomass accumulation includes:

[0027] The year-end biomass volume is calculated based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, pre-obtained expected wildfire time series, pre-obtained expected air humidity data series, and pre-obtained initial value of predicted annual biomass volume.

[0028] The predicted annual carbon sink is calculated based on the year-end biomass, the initial value of the predicted annual biomass, the biomass expansion factor, the biomass carbon content, and the wood density; the formula for calculating the predicted annual carbon sink is as follows:

[0029]

[0030] Among them, C e This represents the projected annual carbon sink, expressed in tons per hectare; ρ represents the projected weight of carbon dioxide absorbed by each hectare of forest area throughout the year; ρ represents the density of the timber. This indicates the year-end biomass, expressed in cubic meters per hectare. B represents the initial value of the predicted annual biomass, in cubic meters per hectare; F represents the biomass expansion factor; and F represents the biomass carbon content.

[0031] Furthermore, the method for calculating the year-end biomass volume based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, pre-obtained expected wildfire time series, pre-obtained expected air humidity data series, and pre-obtained initial value of predicted annual biomass volume includes:

[0032] The process involves acquiring a pre-obtained expected wildfire time series, a pre-obtained expected air humidity data series, and a pre-obtained initial value for the predicted annual biomass. The initial value for the predicted annual biomass is calculated by extrapolating the biomass growth rate series from the twelfth biomass of the sample year closest to the predicted year. The expected wildfire time series represents the times when wildfires occur within the forest area to be assessed during the predicted year, expressed in days. The data in the expected wildfire time series are sequentially recorded as the first predicted wildfire time to the Gth predicted wildfire time, where G represents the time when wildfires occur within the forest area to be assessed during the predicted year. The predicted year is divided into G+1 segments using the first predicted wildfire time to the Gth predicted wildfire time as segmentation points, recorded as the first predicted segment to the G+1 predicted segment. The first predicted wildfire time to the Gth predicted wildfire time do not coincide with the first or last day of the predicted year, and the value of G is greater than 0.

[0033] The year-end biomass volume is calculated using a recursive formula for predicting biomass volume; the recursive formula for predicting biomass volume is:

[0034]

[0035] in, This represents the biomass at the predicted wildfire time l; l is an integer variable ranging from 1 to G+1; when l is 1, for That is, the initial value of the predicted annual biomass; This represents the average value of the biomass growth rate sequence corresponding to the l-th predicted fragment; k represents the length of the l-th predicted segment, in days; * This represents the optimal wildfire loss ratio coefficient; This represents the data value of the day corresponding to the predicted wildfire time in the expected air humidity data series.

[0036] Based on the same inventive concept, this invention also provides a forestry carbon sequestration prediction system based on big data, the system comprising:

[0037] The first data acquisition module is used to acquire the first data; the first data includes pre-obtained wildfire sample sequences, biomass stock sequences, biomass expansion factors, biomass carbon content, wood density, biomass growth rate sequences, and air humidity data sequences of the forest area to be evaluated.

[0038] The optimization model construction module, connected to the first data acquisition module, is used to establish an optimization objective function and constraints for wildfire loss deviation based on the wildfire sample sequence, biomass volume sequence, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated. The optimization variable of the wildfire loss deviation optimization objective function is the wildfire loss proportion coefficient. The wildfire loss proportion coefficient represents the ratio of the wildfire loss proportion to the reciprocal of the air humidity.

[0039] The optimization calculation module, connected to the optimization model construction module, is used to calculate the optimal wildfire loss ratio coefficient using the particle swarm optimization algorithm, with the objective function of minimizing the wildfire loss deviation as the objective and the constraints as the constraints.

[0040] The prediction calculation module, connected to the optimization calculation module, is used to calculate the predicted annual carbon sink based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, biomass expansion factor, biomass carbon content, wood density, pre-obtained expected wildfire time series, pre-obtained expected air humidity data series, and pre-obtained initial value of predicted annual biomass.

[0041] Furthermore, the first data acquisition module includes:

[0042] The data reading module is used to acquire pre-obtained wildfire sample sequences, biomass volume sequences, biomass expansion factors, biomass carbon content, timber density, biomass growth rate sequences, and air humidity data sequences for the forest area to be evaluated. The wildfire sample sequences for the forest area to be evaluated represent the times when wildfires occurred within the forest area to be evaluated during the year monitored. The area of ​​the forest area to be evaluated is 1 hectare. The times when wildfires occurred within the forest area to be evaluated are in daily units. The data in the wildfire sample sequences of the forest area to be evaluated are sequentially recorded as the first wildfire time to the Nth wildfire time, where N represents the total number of wildfires occurring within the forest area to be evaluated during the year. The biomass volume sequence is the biomass per hectare obtained from sample plot surveys on fixed dates each month. The data in the biomass volume sequence are sequentially recorded as the first biomass volume to the twelfth biomass volume. The survey times corresponding to the first to twelfth biomass volumes are sequentially recorded as the first to twelfth survey times; the survey times are in days; the biomass volume represents the total volume of living tree trunks per hectare of forest area, in cubic meters per hectare; the biomass expansion factor represents the conversion coefficient of the total volume of living tree trunks to the total biomass of living trees, dimensionless; the biomass carbon content represents the proportion of carbon in the weight of the total biomass of living trees; the timber density is in tons per cubic meter; the biomass growth rate sequence represents the average natural growth rate of timber, in cubic meters per day; the air humidity data sequence is a data sequence composed of the average daily air humidity within the sample year; the wildfire sample sequence, biomass volume sequence, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated are located on the same time coordinate.

[0043] Furthermore, the optimization model construction module includes:

[0044] The segmentation module is used to obtain the number of sample years Y; the Y sample years are respectively denoted as the first sample year to the Yth sample year; each sample year is divided into 11 intervals with the first survey time to the twelfth survey time as the segmentation point, which are respectively denoted as the first growth interval to the eleventh growth interval; the first growth interval to the eleventh growth interval are respectively divided into the first segment group to the eleventh segment group with the first wildfire time to the Nth wildfire time as the segmentation point; there is no overlap between the first wildfire time to the Nth wildfire time and the first survey time to the twelfth survey time.

[0045] A biomass recursive equation construction module, connected to the segmentation module, is used to establish a biomass recursive equation based on the wildfire sample sequence and biomass sequence of the forest area to be evaluated. The biomass recursive equation includes a first biomass recursive equation and a second biomass recursive equation. The first biomass recursive equation is the recursive equation when the number of segments in the i-th segment group in the t-th sample year is greater than 1. The first biomass recursive equation is expressed as:

[0046]

[0047] Among them, u i,j,t This represents the expected biomass of the j-th growth interval in the t-th sample year; i is an integer variable ranging from 1 to 11; j is a variable ranging from 0 to M. i,t The integer variable; t is an integer variable taking values ​​from 1 to Y; when j>0, a i,j,t Δt represents the average biomass growth rate sequence corresponding to the j-th segment within the i-th segment group in the t-th sample year; when j>0, Δt i,j,t M represents the length of the j-th segment within the i-th segment group in the t-th sample year, in days; i,t V represents the number of segments in the i-th segment group in the t-th sample year; i,t w represents the biomass of sample year t; k represents the wildfire loss ratio; when j>0, w i,j,t This represents the average value of the air humidity data sequence corresponding to the j-th segment within the i-th segment group in the t-th sample year.

[0048] The second recursive equation for biomass is the recursive equation when the number of fragments in the i-th fragment group in the t-th sample year is equal to 1; the second recursive equation for biomass is expressed as:

[0049] u i,1,t =V i,t (1+a i,1,t Δt i,1,t ).

[0050] A module for constructing the objective function for optimizing wildfire loss deviation is connected to a module for constructing a recursive equation for bioaccumulation. This module is used to establish the objective function for optimizing wildfire loss deviation based on the bioaccumulation recursive equation. The objective function for optimizing wildfire loss deviation is:

[0051]

[0052] Where f(k) represents the objective function for optimizing the wildfire loss deviation; This indicates that when j takes the value M i,t time u i,j,t The value of .

[0053] The constraint construction module, connected to the wildfire loss deviation optimization objective function construction module, is used to construct constraints, which are as follows:

[0054]

[0055] Furthermore, the prediction calculation module includes:

[0056] The year-end biomass calculation module is used to calculate the year-end biomass based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, pre-obtained expected wildfire time sequence, pre-obtained expected air humidity data sequence, and pre-obtained initial value of predicted annual biomass.

[0057] The module for calculating the predicted annual carbon sink is connected to the module for calculating the year-end biomass volume. It calculates the predicted annual carbon sink volume based on the year-end biomass volume, the initial value of the predicted annual biomass volume, the biomass expansion factor, the biomass carbon content, and the wood density. The formula for calculating the predicted annual carbon sink volume is as follows:

[0058]

[0059] Among them, C e This represents the projected annual carbon sink, expressed in tons per hectare; ρ represents the projected weight of carbon dioxide absorbed by each hectare of forest area throughout the year; ρ represents the density of the timber. This indicates the year-end biomass, expressed in cubic meters per hectare. B represents the initial value of the predicted annual biomass, in cubic meters per hectare; F represents the biomass expansion factor; and F represents the biomass carbon content.

[0060] Furthermore, the year-end biomass calculation module includes:

[0061] The process involves acquiring a pre-obtained expected wildfire time series, a pre-obtained expected air humidity data series, and a pre-obtained initial value for the predicted annual biomass. The initial value for the predicted annual biomass is calculated by extrapolating the biomass growth rate series from the twelfth biomass of the sample year closest to the predicted year. The expected wildfire time series represents the times when wildfires occur within the forest area to be assessed during the predicted year, expressed in days. The data in the expected wildfire time series are sequentially recorded as the first predicted wildfire time to the Gth predicted wildfire time, where G represents the time when wildfires occur within the forest area to be assessed during the predicted year. The predicted year is divided into G+1 segments using the first predicted wildfire time to the Gth predicted wildfire time as segmentation points, recorded as the first predicted segment to the G+1 predicted segment. The first predicted wildfire time to the Gth predicted wildfire time do not coincide with the first or last day of the predicted year, and the value of G is greater than 0.

[0062] The year-end biomass volume is calculated using a recursive formula for predicting biomass volume; the recursive formula for predicting biomass volume is:

[0063]

[0064] in, This represents the biomass at the predicted wildfire time l; l is an integer variable ranging from 1 to G+1; when l is 1, for That is, the initial value of the predicted annual biomass; This represents the average value of the biomass growth rate sequence corresponding to the l-th predicted fragment; k represents the length of the l-th predicted segment, in days; * This represents the optimal wildfire loss ratio coefficient; This represents the data value of the day corresponding to the predicted wildfire time in the expected air humidity data series.

[0065] (3) Beneficial effects

[0066] Compared with the prior art, the beneficial effects of the present invention are:

[0067] 1. Based on the wildfire sample sequence, biomass volume sequence, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated, an objective function and constraints for optimizing wildfire loss deviation were established. The optimal wildfire loss ratio coefficient was calculated using the particle swarm optimization algorithm, thereby quantitatively characterizing the relationship between wildfire loss ratio and air humidity.

[0068] 2. The predicted annual carbon sink is calculated based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, biomass expansion factor, biomass carbon content, wood density, pre-obtained expected wildfire time series, pre-obtained expected air humidity data series, and pre-obtained initial value of predicted annual biomass accumulation, thus taking into account the impact of wildfire loss in the predicted annual carbon sink. Attached Figure Description

[0069] Figure 1 This is a flowchart of a forestry carbon sink prediction method based on big data, according to Embodiment 1 of the present invention.

[0070] Figure 2 This is a schematic diagram of the module composition of a forestry carbon sink prediction system based on big data, according to Embodiment 2 of the present invention. Detailed Implementation

[0071] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0072] Before giving examples, it is necessary to describe the application scenario of the present invention, which is applied to carbon sink prediction in primary forest areas where wildfires are frequent.

[0073] Example 1: As Figure 1 As shown in the figure, this embodiment provides a forestry carbon sink prediction method based on big data. The method includes the following steps:

[0074] S1, Obtain first data; the first data includes pre-obtained wildfire sample sequences, biomass stock sequences, biomass expansion factors, biomass carbon content, wood density, biomass growth rate sequences, and air humidity data sequences of the forest area to be evaluated.

[0075] S2, establish an objective function and constraints for optimizing the wildfire loss deviation based on the wildfire sample sequence, biomass volume sequence, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated; the optimization variable of the objective function for optimizing the wildfire loss deviation is the wildfire loss ratio coefficient; the wildfire loss ratio coefficient represents the ratio of the wildfire loss ratio to the reciprocal of the air humidity.

[0076] S3, with the objective of minimizing the value of the objective function for optimizing the wildfire loss deviation as the objective and with the constraints as the constraints, the optimal wildfire loss ratio coefficient is calculated using the particle swarm optimization algorithm.

[0077] S4. The predicted annual carbon sink is calculated based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, biomass expansion factor, biomass carbon content, wood density, pre-obtained expected wildfire time series, pre-obtained expected air humidity data series, and pre-obtained initial value of predicted annual biomass.

[0078] For example, the wildfire sample sequence, biomass volume sequence, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated reflect the historical biomass volume data of the forest area to be evaluated under wildfire disturbance in the sample year. The biomass expansion factor, biomass carbon content, and timber density are related to the tree species in the forest area. The size of the forest area to be evaluated is 1 hectare.

[0079] Based on the wildfire sample sequences, biomass volume sequences, biomass growth rate sequences, and air humidity data sequences of the forest area to be evaluated, an objective function and constraints for optimizing wildfire loss deviation are established. The optimization variable of the objective function is the wildfire loss proportion coefficient. This coefficient represents the ratio of the wildfire loss proportion to the reciprocal of the air humidity. Air humidity refers to relative humidity, ranging from 0 to 1. For the same forest area, without human intervention, the wildfire loss proportion is approximately inversely proportional to air humidity. Higher air humidity results in a lower wildfire loss proportion. The wildfire loss proportion represents the percentage of biomass loss caused by a single wildfire.

[0080] The optimal wildfire loss ratio coefficient was calculated to be 0.2276 using the particle swarm optimization algorithm. This optimal wildfire loss ratio coefficient reflects the quantitative relationship between the wildfire loss ratio and air humidity. The objective function for optimizing the wildfire loss deviation reflects the sum of the absolute deviations between the measured biomass sequence and the values ​​calculated by the recursive model based on the wildfire loss ratio. Therefore, the final optimal wildfire loss ratio coefficient best reflects the dynamic changes in biomass under wildfire disturbance. It is worth noting that the optimal wildfire loss ratio coefficient is related to factors such as the natural topography and tree species of the forest area to be evaluated; therefore, the optimal wildfire loss ratio coefficient will vary for different forest areas to be evaluated.

[0081] Finally, the predicted annual carbon sink was calculated based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, biomass expansion factor, biomass carbon content, timber density, pre-obtained expected wildfire time series, pre-obtained expected air humidity data series, and pre-obtained initial value of predicted annual biomass. The predicted annual carbon sink represents the total weight of carbon dioxide expected to be absorbed by the forest area to be assessed in a given predicted year. Since the area of ​​the forest area to be assessed is 1 hectare, the predicted annual carbon sink is simply the total weight of carbon dioxide expected to be absorbed by 1 hectare of the forest area to be assessed in a given predicted year. The calculated predicted annual carbon sink is 2.17 tons per hectare.

[0082] Furthermore, the method for obtaining the first data includes:

[0083] Acquire pre-obtained wildfire sample sequences, biomass volume sequences, biomass expansion factors, biomass carbon content, timber density, biomass growth rate sequences, and air humidity data sequences for the forest area to be assessed. The wildfire sample sequences for the forest area to be assessed are the times when wildfires occurred within the forest area to be assessed during the year monitored. The area of ​​the forest area to be assessed is 1 hectare. The times when wildfires occurred within the forest area to be assessed are in daily units. The data in the wildfire sample sequences of the forest area to be assessed are sequentially recorded as the first wildfire time to the Nth wildfire time, where N represents the total number of wildfires occurring within the forest area to be assessed during the year. The biomass volume sequence is the biomass per hectare obtained from sample plot surveys on fixed dates each month. The data in the biomass volume sequence are sequentially recorded as the first biomass volume to the twelfth biomass volume. The sampling times corresponding to the twelfth biomass volume are sequentially recorded as the first to twelfth sampling times; the sampling times are in days; the biomass volume represents the total volume of living tree trunks per hectare of forest area, in cubic meters per hectare; the biomass expansion factor represents the conversion coefficient of the total volume of living tree trunks to the total biomass of living trees, dimensionless; the biomass carbon content represents the proportion of carbon in the weight of the total biomass of living trees; the timber density is in tons per cubic meter; the biomass growth rate sequence represents the average natural growth rate of timber, in cubic meters per day; the air humidity data sequence is a data sequence composed of the average air humidity of each day within the sample year; the wildfire sample sequence, biomass volume sequence, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated are located on the same time coordinate.

[0084] For example, the wildfire sample sequence for the forest area to be assessed includes a 24-year wildfire sample sequence, obtained by reading historical wildfire archives. The biomass volume sequence is the biomass volume per hectare obtained through plot surveys on the 10th of each month. The biomass expansion factor is related to tree species; since the forest area to be assessed is a fir forest, the biomass expansion factor is taken as the typical value of 1.25 for mature fir forests. The biomass carbon content is also related to tree species, taking a typical value of 0.491. The timber density is 0.38 tons per cubic meter. The biomass growth rate sequence represents the average natural growth rate of timber, which is related to season and tree species, and is obtained through historical data statistics. The air humidity data sequence is obtained through historical monitoring data.

[0085] Furthermore, the method for establishing the objective function and constraints for optimizing the wildfire loss deviation based on the wildfire sample sequence, biomass volume sequence, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated includes:

[0086] Obtain the number of sample years Y; denote the Y sample years as the first sample year to the Yth sample year; divide each sample year into 11 intervals, denote them as the first growth interval to the eleventh growth interval, using the first survey time to the twelfth survey time as the dividing point; divide the first growth interval to the eleventh growth interval into the first segment group to the eleventh segment group, using the first wildfire time to the Nth wildfire time as the dividing point; there is no overlap between the first wildfire time to the Nth wildfire time and the first survey time to the twelfth survey time.

[0087] Based on the wildfire sample sequence and biomass sequence of the forest area to be assessed, a biomass recursive equation is established; the biomass recursive equation includes a first biomass recursive equation and a second biomass recursive equation; the first biomass recursive equation is the recursive equation when the number of segments in the i-th segment group in the t-th sample year is greater than 1; the first biomass recursive equation is expressed as:

[0088]

[0089] Among them, u i,j,t This represents the expected biomass of the j-th growth interval in the t-th sample year; i is an integer variable ranging from 1 to 11; j is a variable ranging from 0 to M. i,t The integer variable; t is an integer variable taking values ​​from 1 to Y; when j>0, a i,j,t Δt represents the average biomass growth rate sequence corresponding to the j-th segment within the i-th segment group in the t-th sample year; when j>0, Δt i,j,t M represents the length of the j-th segment within the i-th segment group in the t-th sample year, in days; i,t V represents the number of segments in the i-th segment group in the t-th sample year; i,t w represents the biomass of sample year t; k represents the wildfire loss ratio; when j>0, w i,j,t This represents the average value of the air humidity data sequence corresponding to the j-th segment within the i-th segment group in the t-th sample year.

[0090] The second recursive equation for biomass is the recursive equation when the number of fragments in the i-th fragment group in the t-th sample year is equal to 1; the second recursive equation for biomass is expressed as:

[0091] u i,1,t =V i,t (1+a i,1,t Δt i,1,t ).

[0092] An objective function for optimizing wildfire loss deviation is established based on the aforementioned bioaccumulation recursive equation; the objective function for optimizing wildfire loss deviation is:

[0093]

[0094] Where f(k) represents the objective function for optimizing the wildfire loss deviation; This indicates that when j takes the value M i,t time u i,j,t The value of .

[0095] Construct constraints, the constraints being:

[0096]

[0097] For example, taking the first sample year as an example, each sample year is divided into 11 intervals, denoted as the first to eleventh growth intervals, using the first to twelfth survey dates as dividing points. Since the first to twelfth survey dates are January 10th, February 10th, March 10th, April 10th, May 10th, June 10th, July 10th, August 10th, September 10th, October 10th, November 10th, and December 10th respectively, the first growth interval is January 10th to February 9th, the second growth interval is February 10th to March 9th, the third growth interval is March 10th to April 9th, and so on. It is worth noting that, to ensure consistency in the statistical scale of the segmented data, the beginning of each growth interval is the survey date, and the end of each growth interval is the day before the survey date. In the first sample year, five wildfires occurred in the forest area to be assessed, on February 18, March 13, March 27, April 7, and April 25. That is, the times of the first to fifth wildfires were February 18, March 13, March 27, April 7, and April 25. There was no overlap between the times of the first to fifth wildfires and the times of the first to twelfth surveys. If overlap occurred in practice, the survey times were postponed. Further calculations can be made to obtain M. 1,1 To M 11,1 The numbers are 1, 2, 4, 2, 1, 1, 1, 1, 1, 1, 1.

[0098] For the first growth interval, since M 1,1 =1, therefore the second recursive equation for biomass is adopted:

[0099] u 1,1,1 =V 1,1 (1+a 1,1,1 Δt 1,1,1 ).

[0100] For the second growth interval, since M 2,1 =2, therefore, using the first recursive equation for biomass, we get:

[0101]

[0102] Similarly, the same algorithm is applied to the third to eleventh growth intervals. The same algorithm is also applied to the second to the 24th sample year.

[0103] Finally, the objective function and constraints for optimizing the wildfire loss deviation are established:

[0104]

[0105] Furthermore, the method for calculating the predicted annual carbon sink based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, biomass expansion factor, biomass carbon content, wood density, pre-obtained expected wildfire time series, pre-obtained expected air humidity data series, and pre-obtained initial value of predicted annual biomass accumulation includes:

[0106] The year-end biomass volume is calculated based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, pre-obtained expected wildfire time series, pre-obtained expected air humidity data series, and pre-obtained initial value of predicted annual biomass volume.

[0107] The predicted annual carbon sink is calculated based on the year-end biomass, the initial value of the predicted annual biomass, the biomass expansion factor, the biomass carbon content, and the wood density; the formula for calculating the predicted annual carbon sink is as follows:

[0108]

[0109] Among them, C e This represents the projected annual carbon sink, expressed in tons per hectare; ρ represents the projected weight of carbon dioxide absorbed by each hectare of forest area throughout the year; ρ represents the density of the timber. This indicates the year-end biomass, expressed in cubic meters per hectare. B represents the initial value of the predicted annual biomass, in cubic meters per hectare; F represents the biomass expansion factor; and F represents the biomass carbon content.

[0110] For example, the principle behind the formula for calculating the predicted annual carbon sink is the conservation of carbon.

[0111] Furthermore, the method for calculating the year-end biomass volume based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, pre-obtained expected wildfire time series, pre-obtained expected air humidity data series, and pre-obtained initial value of predicted annual biomass volume includes:

[0112] The process involves acquiring a pre-obtained expected wildfire time series, a pre-obtained expected air humidity data series, and a pre-obtained initial value for the predicted annual biomass. The initial value for the predicted annual biomass is calculated by extrapolating the biomass growth rate series from the twelfth biomass of the sample year closest to the predicted year. The expected wildfire time series represents the times when wildfires occur within the forest area to be assessed during the predicted year, expressed in days. The data in the expected wildfire time series are sequentially recorded as the first predicted wildfire time to the Gth predicted wildfire time, where G represents the time when wildfires occur within the forest area to be assessed during the predicted year. The predicted year is divided into G+1 segments using the first predicted wildfire time to the Gth predicted wildfire time as segmentation points, recorded as the first predicted segment to the G+1 predicted segment. The first predicted wildfire time to the Gth predicted wildfire time do not coincide with the first or last day of the predicted year, and the value of G is greater than 0.

[0113] The year-end biomass volume is calculated using a recursive formula for predicting biomass volume; the recursive formula for predicting biomass volume is:

[0114]

[0115] in, This represents the biomass at the predicted wildfire time l; l is an integer variable ranging from 1 to G+1; when l is 1, for That is, the initial value of the predicted annual biomass; This represents the average value of the biomass growth rate sequence corresponding to the l-th predicted fragment; k represents the length of the l-th predicted segment, in days; * This represents the optimal wildfire loss ratio coefficient; This represents the data value of the day corresponding to the predicted wildfire time in the expected air humidity data series.

[0116] For example, the method for obtaining the expected wildfire time series is as follows: First, based on the wildfire sample sequence of the forest area to be assessed in the sample year and the pre-obtained annual climate data of the sample year, a correlation analysis algorithm is used to calculate the correlation between the wildfire sample sequence and the annual climate data of the forest area to be assessed. Then, based on a regression analysis algorithm, the annual climate data of the predicted year is calculated using the annual climate data of the sample year and data information from the climate database. The annual climate data of the predicted year includes temperature anomalies, precipitation anomalies, drought indicators, etc. Finally, the expected wildfire time series is calculated based on the annual climate data of the predicted year and the correlation between the wildfire sample sequence and the annual climate data of the forest area to be assessed. The expected air humidity data series is obtained through the annual climate data of the sample year and a seasonal prediction algorithm. The initial value of the predicted year's biomass is calculated by extrapolating the twelfth biomass of the sample year closest to the predicted year using the biomass growth rate sequence.

[0117] Example 2: Based on the same inventive concept, such as Figure 2 As shown, this embodiment also provides a forestry carbon sink prediction system based on big data, the system comprising:

[0118] The first data acquisition module is used to acquire the first data; the first data includes pre-obtained wildfire sample sequences, biomass stock sequences, biomass expansion factors, biomass carbon content, wood density, biomass growth rate sequences, and air humidity data sequences of the forest area to be evaluated.

[0119] The optimization model construction module, connected to the first data acquisition module, is used to establish an optimization objective function and constraints for wildfire loss deviation based on the wildfire sample sequence, biomass volume sequence, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated. The optimization variable of the wildfire loss deviation optimization objective function is the wildfire loss proportion coefficient. The wildfire loss proportion coefficient represents the ratio of the wildfire loss proportion to the reciprocal of the air humidity.

[0120] The optimization calculation module, connected to the optimization model construction module, is used to calculate the optimal wildfire loss ratio coefficient using the particle swarm optimization algorithm, with the objective function of minimizing the wildfire loss deviation as the objective and the constraints as the constraints.

[0121] The prediction calculation module, connected to the optimization calculation module, is used to calculate the predicted annual carbon sink based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, biomass expansion factor, biomass carbon content, wood density, pre-obtained expected wildfire time series, pre-obtained expected air humidity data series, and pre-obtained initial value of predicted annual biomass.

[0122] Furthermore, the first data acquisition module includes:

[0123] The data reading module is used to acquire pre-obtained wildfire sample sequences, biomass volume sequences, biomass expansion factors, biomass carbon content, timber density, biomass growth rate sequences, and air humidity data sequences for the forest area to be evaluated. The wildfire sample sequences for the forest area to be evaluated represent the times when wildfires occurred within the forest area to be evaluated during the year monitored. The area of ​​the forest area to be evaluated is 1 hectare. The times when wildfires occurred within the forest area to be evaluated are in daily units. The data in the wildfire sample sequences of the forest area to be evaluated are sequentially recorded as the first wildfire time to the Nth wildfire time, where N represents the total number of wildfires occurring within the forest area to be evaluated during the year. The biomass volume sequence is the biomass per hectare obtained from sample plot surveys on fixed dates each month. The data in the biomass volume sequence are sequentially recorded as the first biomass volume to the twelfth biomass volume. The survey times corresponding to the first to twelfth biomass volumes are sequentially recorded as the first to twelfth survey times; the survey times are in days; the biomass volume represents the total volume of living tree trunks per hectare of forest area, in cubic meters per hectare; the biomass expansion factor represents the conversion coefficient of the total volume of living tree trunks to the total biomass of living trees, dimensionless; the biomass carbon content represents the proportion of carbon in the weight of the total biomass of living trees; the timber density is in tons per cubic meter; the biomass growth rate sequence represents the average natural growth rate of timber, in cubic meters per day; the air humidity data sequence is a data sequence composed of the average daily air humidity within the sample year; the wildfire sample sequence, biomass volume sequence, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated are located on the same time coordinate.

[0124] Furthermore, the optimization model construction module includes:

[0125] The segmentation module is used to obtain the number of sample years Y; the Y sample years are respectively denoted as the first sample year to the Yth sample year; each sample year is divided into 11 intervals with the first survey time to the twelfth survey time as the segmentation point, which are respectively denoted as the first growth interval to the eleventh growth interval; the first growth interval to the eleventh growth interval are respectively divided into the first segment group to the eleventh segment group with the first wildfire time to the Nth wildfire time as the segmentation point; there is no overlap between the first wildfire time to the Nth wildfire time and the first survey time to the twelfth survey time.

[0126] A biomass recursive equation construction module, connected to the segmentation module, is used to establish a biomass recursive equation based on the wildfire sample sequence and biomass sequence of the forest area to be evaluated. The biomass recursive equation includes a first biomass recursive equation and a second biomass recursive equation. The first biomass recursive equation is the recursive equation when the number of segments in the i-th segment group in the t-th sample year is greater than 1. The first biomass recursive equation is expressed as:

[0127]

[0128] Among them, u i,j,t This represents the expected biomass of the j-th growth interval in the t-th sample year; i is an integer variable ranging from 1 to 11; j is a variable ranging from 0 to M. i,t The integer variable; t is an integer variable taking values ​​from 1 to Y; when j>0, a i,j,t Δt represents the average biomass growth rate sequence corresponding to the j-th segment within the i-th segment group in the t-th sample year; when j>0, Δt i,j,t M represents the length of the j-th segment within the i-th segment group in the t-th sample year, in days; i,t V represents the number of segments in the i-th segment group in the t-th sample year; i,t w represents the biomass of sample year t; k represents the wildfire loss ratio; when j>0, w i,j,t This represents the average value of the air humidity data sequence corresponding to the j-th segment within the i-th segment group in the t-th sample year.

[0129] The second recursive equation for biomass is the recursive equation when the number of fragments in the i-th fragment group in the t-th sample year is equal to 1; the second recursive equation for biomass is expressed as:

[0130] u i,1,t =V i,t (1+a i,1,t Δt i,1,t ).

[0131] A module for constructing the objective function for optimizing wildfire loss deviation is connected to a module for constructing a recursive equation for bioaccumulation. This module is used to establish the objective function for optimizing wildfire loss deviation based on the bioaccumulation recursive equation. The objective function for optimizing wildfire loss deviation is:

[0132]

[0133] Where f(k) represents the objective function for optimizing the wildfire loss deviation; This indicates that when j takes the value M i,t time u i,j,t The value of .

[0134] The constraint construction module, connected to the wildfire loss deviation optimization objective function construction module, is used to construct constraints, which are as follows:

[0135]

[0136] Furthermore, the prediction calculation module includes:

[0137] The year-end biomass calculation module is used to calculate the year-end biomass based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, pre-obtained expected wildfire time sequence, pre-obtained expected air humidity data sequence, and pre-obtained initial value of predicted annual biomass.

[0138] The module for calculating the predicted annual carbon sink is connected to the module for calculating the year-end biomass volume. It calculates the predicted annual carbon sink volume based on the year-end biomass volume, the initial value of the predicted annual biomass volume, the biomass expansion factor, the biomass carbon content, and the wood density. The formula for calculating the predicted annual carbon sink volume is as follows:

[0139]

[0140] Among them, C e This represents the projected annual carbon sink, expressed in tons per hectare; ρ represents the projected weight of carbon dioxide absorbed by each hectare of forest area throughout the year; ρ represents the density of the timber. This indicates the year-end biomass, expressed in cubic meters per hectare. B represents the initial value of the predicted annual biomass, in cubic meters per hectare; F represents the biomass expansion factor; and F represents the biomass carbon content.

[0141] Furthermore, the year-end biomass calculation module includes:

[0142] The process involves acquiring a pre-obtained expected wildfire time series, a pre-obtained expected air humidity data series, and a pre-obtained initial value for the predicted annual biomass. The initial value for the predicted annual biomass is calculated by extrapolating the biomass growth rate series from the twelfth biomass of the sample year closest to the predicted year. The expected wildfire time series represents the times when wildfires occur within the forest area to be assessed during the predicted year, expressed in days. The data in the expected wildfire time series are sequentially recorded as the first predicted wildfire time to the Gth predicted wildfire time, where G represents the time when wildfires occur within the forest area to be assessed during the predicted year. The predicted year is divided into G+1 segments using the first predicted wildfire time to the Gth predicted wildfire time as segmentation points, recorded as the first predicted segment to the G+1 predicted segment. The first predicted wildfire time to the Gth predicted wildfire time do not coincide with the first or last day of the predicted year, and the value of G is greater than 0.

[0143] The year-end biomass volume is calculated using a recursive formula for predicting biomass volume; the recursive formula for predicting biomass volume is:

[0144]

[0145] in, This represents the biomass at the predicted wildfire time l; l is an integer variable ranging from 1 to G+1; when l is 1, for That is, the initial value of the predicted annual biomass; This represents the average value of the biomass growth rate sequence corresponding to the l-th predicted fragment; k represents the length of the l-th predicted segment, in days; * This represents the optimal wildfire loss ratio coefficient; This represents the data value of the day corresponding to the predicted wildfire time in the expected air humidity data series.

[0146] It should be noted that the specific methods by which each module performs operations in the system described in the above embodiments have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0147] Finally, it should be noted that although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A forestry carbon sink prediction method based on big data, characterized in that, The method includes the following steps: S1, Obtain the first data; the first data includes the pre-obtained wildfire sample sequence, biomass volume sequence, biomass expansion factor, biomass carbon content, wood density, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated. S2, establish an objective function and constraints for optimizing wildfire loss deviation based on the wildfire sample sequence, biomass volume sequence, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated; the optimization variable of the objective function for optimizing wildfire loss deviation is the wildfire loss ratio coefficient; the wildfire loss ratio coefficient represents the ratio of the wildfire loss ratio to the reciprocal of the air humidity. S3, with the objective of minimizing the value of the objective function for optimizing the wildfire loss deviation as the objective and with the constraints as the constraints, the optimal wildfire loss ratio coefficient is calculated using the particle swarm optimization algorithm. S4. The predicted annual carbon sink is calculated based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, biomass expansion factor, biomass carbon content, wood density, the pre-obtained expected wildfire time series, the pre-obtained expected air humidity data series, and the pre-obtained initial value of the predicted annual biomass. The method for obtaining the first data includes: Acquire pre-obtained wildfire sample sequences, biomass volume sequences, biomass expansion factor, biomass carbon content, timber density, biomass growth rate sequences, and air humidity data sequences for the forest area to be assessed. The wildfire sample sequences for the forest area to be assessed represent the times when wildfires occurred within the monitored sample year within the forest area to be assessed. The area of ​​the forest area to be assessed is 1 hectare. The times when wildfires occurred within the forest area to be assessed are in daily units. The data in the wildfire sample sequences of the forest area to be assessed are sequentially recorded as the first wildfire time to the second... N Wildfire moments; among them N This indicates the total number of wildfires occurring within the forest area to be assessed during the sample year; the biomass sequence is the biomass per hectare obtained from plot surveys on fixed dates each month; the data in the biomass sequence are sequentially recorded as the first biomass to the twelfth biomass; the plot survey times corresponding to the first to the twelfth biomass are sequentially recorded as the first survey time to the twelfth survey time; the plot survey time is in days; the biomass represents the total volume of standing tree trunks per hectare of forest area, in cubic meters per hectare; the biomass... The volume expansion factor represents the conversion coefficient between the total volume of the standing tree trunk and the total biomass of the standing tree, and is dimensionless; the biomass carbon content represents the proportion of carbon in the weight of the total biomass of the standing tree; the unit of wood density is tons per cubic meter; the biomass growth rate sequence represents the average natural growth rate of wood, and is in cubic meters per day; the air humidity data sequence is a data sequence composed of the average air humidity of each day within the sample year; the wildfire sample sequence, biomass volume sequence, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated are located on the same time coordinate. The method for establishing the objective function and constraints for optimizing wildfire loss deviation based on the wildfire sample sequence, biomass volume sequence, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated includes: Obtain the number of sample years Y ;Will Y The sample years are respectively denoted as the first sample year to the second sample year. Y Sample year; each sample year is divided into 11 intervals, from the first survey time to the twelfth survey time, and is denoted as the first growth interval to the eleventh growth interval; the first wildfire time to the eleventh survey time is used as the dividing point; N Using the time of the wildfire as a dividing point, the first to eleventh growth intervals are respectively divided into the first segment group to the eleventh segment group; the first wildfire time to the eleventh... N The timing of the wildfire did not overlap with the timings of the first to twelfth surveys. Based on the wildfire sample sequence and biomass sequence of the forest area to be assessed, a recursive equation for biomass volume is established; the recursive equation for biomass volume includes a first recursive equation for biomass volume and a second recursive equation for biomass volume; the first recursive equation for biomass volume is the... t Sample year The recursive equation for the number of fragments in a fragment group being greater than 1; the first recursive equation for bioaccumulation is expressed as: ； in, Indicates the first t Sample year The first growth interval Expected biomass; For integer variables with values ​​from 1 to 11, it is the index number of the fragment group, growth interval, and bioaccumulation. The value is 0 to The integer variable is the index number of the fragment and the expected biomass. t The value is 1 to Y Integer variables; when hour, Indicates the first t Sample year The first in the fragment group The average value of the corresponding biomass growth rate sequence within each segment; when hour, Indicates the first t Sample year The first in the fragment group The length of each segment, in days; Indicates the first t Sample year The number of segments within a segment group; Indicates the first t Sample year Biomass; Indicates the proportion of wildfire losses; when hour, Indicates the first t Sample year The first in the fragment group The average value of the air humidity data sequence corresponding to each segment; The second recursive equation for biomass is the first... t Sample year The recursive equation for the number of fragments in a fragment group equal to 1; the second recursive equation for bioaccumulation is expressed as: ； An objective function for optimizing wildfire loss deviation is established based on the aforementioned recursive equation for bioaccumulation; the objective function for optimizing wildfire loss deviation is: ； in, This represents the objective function for optimizing the wildfire loss deviation. Indicates when Values hour The possible values ​​of ; Construct constraints, the constraints being: 。 2. The forestry carbon sink prediction method based on big data as described in claim 1, characterized in that, The method for calculating the predicted annual carbon sink based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, biomass expansion factor, biomass carbon content, wood density, pre-obtained expected wildfire time series, pre-obtained expected air humidity data series, and pre-obtained initial value of predicted annual biomass accumulation includes: The year-end biomass volume is calculated based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, pre-obtained expected wildfire time sequence, pre-obtained expected air humidity data sequence, and pre-obtained initial value of predicted annual biomass volume. The predicted annual carbon sink is calculated based on the year-end biomass, the initial value of the predicted annual biomass, the biomass expansion factor, the biomass carbon content, and the wood density; the formula for calculating the predicted annual carbon sink is as follows: ； in, This indicates the projected annual carbon sink, expressed in tons per hectare, and represents the projected weight of carbon dioxide absorbed by each hectare of forest area throughout the year. Indicates wood density; This indicates the year-end biomass, expressed in cubic meters per hectare. This represents the initial value of the predicted annual biomass, expressed in cubic meters per hectare. Indicates biomass expansion factor; This indicates the carbon content of biomass.

3. The forestry carbon sink prediction method based on big data as described in claim 2, characterized in that, The method for calculating the year-end biomass volume based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, pre-obtained expected wildfire time series, pre-obtained expected air humidity data series, and pre-obtained initial value of predicted annual biomass volume includes: Obtain the pre-obtained expected wildfire time series, the pre-obtained expected air humidity data series, and the pre-obtained initial value of the predicted annual biomass. The initial value of the predicted annual biomass is calculated by extrapolating the biomass growth rate series from the twelfth biomass of the sample year closest to the predicted year. The expected wildfire time series refers to the times when wildfires will occur within the forest area to be assessed within the predicted year, in days. The data in the expected wildfire time series are sequentially recorded as the first predicted wildfire time to the second predicted wildfire time. G Predicting the timing of wildfires; among them G This indicates the predicted time of wildfire occurrence within the forest area to be assessed during the year; the time from the first predicted wildfire time to the [previous time]... G The predicted wildfire time is used as the dividing point to divide the predicted year into segments. G +1 segment, denoted as the first predicted segment to the... G +1 prediction segment; the first predicted wildfire time to the... G The predicted times of wildfires do not coincide with the first or last day of the predicted year; G The value of is greater than 0; The year-end biomass volume is calculated using a recursive formula for predicting biomass volume; the recursive formula for predicting biomass volume is: ； in, Indicates the first Predict biomass at the moment of wildfire; The value is 1 to G +1 integer variable; when When the value is 1, for That is, the initial value of the predicted annual biomass; Indicates the first The average value of the biomass growth rate sequence corresponding to the predicted fragment; Indicates the first Predict the length of the segment, in days; This represents the optimal wildfire loss ratio coefficient; This indicates the expected air humidity data series and the first... The data values ​​for the predicted wildfire time corresponding to the day.

4. A forestry carbon sequestration prediction system based on big data, characterized in that, The system includes: The first data acquisition module is used to acquire the first data; the first data includes the pre-obtained wildfire sample sequence, biomass volume sequence, biomass expansion factor, biomass carbon content, wood density, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated. The optimization model construction module, connected to the first data acquisition module, is used to establish an optimization objective function and constraints for wildfire loss deviation based on the wildfire sample sequence, biomass volume sequence, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated; the optimization variable of the wildfire loss deviation optimization objective function is the wildfire loss ratio coefficient; the wildfire loss ratio coefficient represents the ratio of the wildfire loss ratio to the reciprocal of the air humidity. The optimization calculation module is connected to the optimization model construction module. It is used to calculate the optimal wildfire loss ratio coefficient by using the particle swarm optimization algorithm with the objective function of minimizing the wildfire loss deviation as the objective and the constraints as the constraints. The prediction calculation module, connected to the optimization calculation module, is used to calculate the predicted annual carbon sink based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, biomass expansion factor, biomass carbon content, wood density, pre-obtained expected wildfire time series, pre-obtained expected air humidity data series, and pre-obtained initial value of predicted annual biomass. The first data acquisition module includes: The data reading module is used to acquire pre-obtained wildfire sample sequences, biomass volume sequences, biomass expansion factors, biomass carbon content, timber density, biomass growth rate sequences, and air humidity data sequences for the forest area to be evaluated. The wildfire sample sequences for the forest area to be evaluated represent the times when wildfires occurred within the monitored sample year within the forest area to be evaluated. The area of ​​the forest area to be evaluated is 1 hectare. The times when wildfires occurred within the forest area to be evaluated are in daily units. The data in the wildfire sample sequences of the forest area to be evaluated are sequentially recorded as the first wildfire time to the next... N Wildfire moments; among them N This indicates the total number of wildfires occurring within the forest area to be assessed during the sample year; the biomass sequence is the biomass per hectare obtained from plot surveys on fixed dates each month; the data in the biomass sequence are sequentially recorded as the first biomass to the twelfth biomass; the plot survey times corresponding to the first to the twelfth biomass are sequentially recorded as the first survey time to the twelfth survey time; the plot survey time is in days; the biomass represents the total volume of standing tree trunks per hectare of forest area, in cubic meters per hectare; the biomass... The volume expansion factor represents the conversion coefficient between the total volume of the standing tree trunk and the total biomass of the standing tree, and is dimensionless; the biomass carbon content represents the proportion of carbon in the weight of the total biomass of the standing tree; the unit of wood density is tons per cubic meter; the biomass growth rate sequence represents the average natural growth rate of wood, and is in cubic meters per day; the air humidity data sequence is a data sequence composed of the average air humidity of each day within the sample year; the wildfire sample sequence, biomass volume sequence, biomass growth rate sequence, and air humidity data sequence of the forest area to be evaluated are located on the same time coordinate. The optimization model construction module includes: The segmentation module is used to obtain the number of sample years. Y ;Will Y The sample years are respectively denoted as the first sample year to the second sample year. Y Sample year; each sample year is divided into 11 intervals, from the first survey time to the twelfth survey time, and is denoted as the first growth interval to the eleventh growth interval; the first wildfire time to the eleventh survey time is used as the dividing point; N Using the time of the wildfire as a dividing point, the first to eleventh growth intervals are respectively divided into the first segment group to the eleventh segment group; the first wildfire time to the eleventh... N The timing of the wildfire did not overlap with the timings of the first to twelfth surveys. A biomass recursive equation construction module, connected to the segmentation module, is used to establish a biomass recursive equation based on the wildfire sample sequence and biomass sequence of the forest area to be assessed. The biomass recursive equation includes a first biomass recursive equation and a second biomass recursive equation. The first biomass recursive equation is the... t Sample year The recursive equation for the number of fragments in a fragment group being greater than 1; the first recursive equation for bioaccumulation is expressed as: ； in, Indicates the first t Sample year The first growth interval Expected biomass; For integer variables with values ​​from 1 to 11, it is the index number of the fragment group, growth interval, and bioaccumulation. The value is 0 to The integer variable is the index number of the fragment and the expected biomass. t The value is 1 to Y Integer variables; when hour, Indicates the first t Sample year The first in the fragment group The average value of the corresponding biomass growth rate sequence within each segment; when hour, Indicates the first t Sample year The first in the fragment group The length of each segment, in days; Indicates the first t Sample year The number of segments within a segment group; Indicates the first t Sample year Biomass; Indicates the proportion of wildfire losses; when hour, Indicates the first t Sample year The first in the fragment group The average value of the air humidity data sequence corresponding to each segment; The second recursive equation for biomass is the first... t Sample year The recursive equation for the number of fragments in a fragment group equal to 1; the second recursive equation for bioaccumulation is expressed as: ； A module for constructing the objective function for optimizing wildfire loss deviation is connected to a module for constructing a recursive equation for bioaccumulation. This module is used to establish the objective function for optimizing wildfire loss deviation based on the bioaccumulation recursive equation. The objective function for optimizing wildfire loss deviation is: ； in, This represents the objective function for optimizing the wildfire loss deviation. Indicates when Values hour The possible values ​​of ; The constraint construction module, connected to the wildfire loss deviation optimization objective function construction module, is used to construct constraints, which are as follows: 。 5. The forestry carbon sequestration prediction system based on big data as described in claim 4, characterized in that, The prediction calculation module includes: The year-end biomass calculation module is used to calculate the year-end biomass based on the optimal wildfire loss ratio coefficient, biomass growth rate sequence, pre-obtained expected wildfire time sequence, pre-obtained expected air humidity data sequence, and pre-obtained predicted annual biomass initial value. The module for calculating the predicted annual carbon sink is connected to the module for calculating the year-end biomass volume. It calculates the predicted annual carbon sink volume based on the year-end biomass volume, the initial value of the predicted annual biomass volume, the biomass expansion factor, the biomass carbon content, and the wood density. The formula for calculating the predicted annual carbon sink volume is as follows: ； in, This indicates the projected annual carbon sink, expressed in tons per hectare, and represents the projected weight of carbon dioxide absorbed by each hectare of forest area throughout the year. Indicates wood density; This indicates the year-end biomass, expressed in cubic meters per hectare. This represents the initial value of the predicted annual biomass, expressed in cubic meters per hectare. Indicates biomass expansion factor; This indicates the carbon content of biomass.

6. The forestry carbon sink prediction system based on big data as described in claim 5, characterized in that, The year-end biomass calculation module includes: Obtain the pre-obtained expected wildfire time series, the pre-obtained expected air humidity data series, and the pre-obtained initial value of the predicted annual biomass. The initial value of the predicted annual biomass is calculated by extrapolating the biomass growth rate series from the twelfth biomass of the sample year closest to the predicted year. The expected wildfire time series refers to the times when wildfires will occur within the forest area to be assessed within the predicted year, in days. The data in the expected wildfire time series are sequentially recorded as the first predicted wildfire time to the second predicted wildfire time. G Predicting the timing of wildfires; among them G This indicates the predicted time of wildfire occurrence within the forest area to be assessed during the year; the time from the first predicted wildfire time to the [previous time]... G The predicted wildfire time is used as the dividing point to divide the predicted year into segments. G +1 segment, denoted as the first predicted segment to the... G +1 prediction segment; the first predicted wildfire time to the... G The predicted times of wildfires do not coincide with the first or last day of the predicted year; G The value of is greater than 0; The year-end biomass volume is calculated using a recursive formula for predicting biomass volume; the recursive formula for predicting biomass volume is: ； in, Indicates the first Predict biomass at the moment of wildfire; The value is 1 to G +1 integer variable; when When the value is 1, for That is, the initial value of the predicted annual biomass; Indicates the first The average value of the biomass growth rate sequence corresponding to the predicted fragment; Indicates the first Predict the length of the segment, in days; This represents the optimal wildfire loss ratio coefficient; This indicates the expected air humidity data series and the first... The data values ​​for the predicted wildfire time corresponding to the day.

Images