A vegetation carbon sink prediction method and system based on future scenario simulation
Through future scenario simulation and multi-source data fusion, combined with PLUS and CASA models, the shortcomings in accuracy and long-term prediction capabilities of existing vegetation carbon sink prediction models are solved, and more accurate and reliable vegetation carbon sink prediction is achieved, supporting science Assessment and policy development.
Patent Information
- Application Number
- CN202410978202.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-07-22
AI Technical Summary
The existing vegetation carbon sink prediction model has insufficient accuracy and long-term prediction capabilities, especially for different vegetation types, the land use classification accuracy is limited, and it is difficult to reflect the heterogeneity of ecological processes in the region, and the model's time scale is relatively short. , it is difficult to accurately predict long-term carbon sink changes.
Using a method based on future scenario simulation, the spatial distribution of land use types is simulated through the PLUS model, combined with the CASA model and nuclear normalized vegetation index, and using satellite remote sensing images and meteorological data to calculate vegetation carbon sinks, considering multi-source driver factors and climate Change, calculate the prediction scenarios of various vegetation carbon sinks.
It improves the accuracy and long-term prediction ability of vegetation carbon sink prediction, provides a more solid theoretical basis for vegetation physical and chemical growth characteristics, and can more accurately obtain data on vegetation carbon sink changes, and supports scientific evaluation and policy formulation.
Smart Images

Figure CN118536071B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the fields of remote sensing, geographic information and ecological technology, and in particular to a vegetation carbon sink prediction method and system based on future scenario simulation. Background Art
[0002] As the problem of global warming becomes increasingly serious, mitigation and adaptation to climate change have become major issues that need to be urgently addressed in today's society. Vegetation, as the main source of ecosystem carbon sinks, plays a critical role in the global carbon cycle. Therefore, accurately predicting the changes in carbon sinks of different vegetation types is of great significance for scientifically assessing the carbon balance of terrestrial ecosystems, formulating regional emission reduction policies, and guiding low-carbon development strategies. At present, vegetation carbon sink prediction is mainly based on ecosystem process models, which comprehensively consider the influence of multiple factors such as climate, soil, and vegetation, simulate vegetation photosynthesis, respiration and other processes, and quantitatively evaluate vegetation carbon sinks.
[0003] The current vegetation carbon sink prediction faces some problems and challenges. The accuracy of the model prediction results depends on the reliability of the model algorithm itself and the input parameters. The accuracy of land use classification for different vegetation types is limited, and the reliability of the prediction results depends on the model resolution. However, the resolution of large-scale models is too coarse to reflect the heterogeneity of ecological processes in the region. In addition, the time scale for reliable predictions by the original models is mostly from several years to decades, and the long-term prediction ability of carbon sink changes is still weak. In general, vegetation carbon sink research needs to be further improved in terms of model algorithms, multi-source heterogeneous data fusion, uncertainty analysis, etc. Obtaining long-term high-quality remote sensing satellite images will also provide more powerful data support for prediction models. Summary of the invention
[0004] In view of the deficiencies of the prior art, the present invention provides a vegetation carbon sink prediction method based on future scenario simulation, comprising the following steps:
[0005] Step 1: Collect historical land use data and multi-source driving factors of the study area, simulate the spatial distribution of future land use types under different land use scenarios through the PLUS model, and deduce the land use classification results under different land use scenarios in a certain year in the future;
[0006] Step 2: Collect the monthly mean temperature, monthly total precipitation and monthly total solar radiation data under different future climate scenarios in the study area, and interpolate and resample them to keep the same spatial resolution as the historical land use data collected in step 1;
[0007] Step 3, obtain the historical kernel normalized vegetation index by performing band calculation on the historical satellite remote sensing images, use the historical kernel normalized vegetation index and the meteorological data of the historical meteorological stations to build a relationship model between the kernel normalized vegetation index and meteorological data, and predict the kernel normalized vegetation index of the corresponding forecast year based on the monthly average temperature and monthly total precipitation data under different future climate scenarios collected in step 2;
[0008] Step 4: Based on the CASA model, calculate the net primary productivity of vegetation under multiple future vegetation carbon sink prediction scenarios according to the land use type classification results deduced under the land use scenario in step 1, the monthly average temperature, monthly total precipitation and monthly total solar radiation data under different future climate scenarios in step 2, and the kernel normalized difference vegetation index predicted in step 3;
[0009] Step 5, using the monthly average temperature and monthly total precipitation data under different future climate scenarios collected in step 2 to calculate the predicted soil microbial respiration, combined with the vegetation net primary productivity under multiple future vegetation carbon sink prediction scenarios calculated in step 4, invert the vegetation carbon sink under multiple future scenarios.
[0010] Moreover, the multi-source driving factors in step 1 include elevation, slope, precipitation, temperature, GDP, population density, distance to roads, distance to railways and distance to rivers. The multi-source driving factors are rasterized and resampled. The factors of various land use expansion and driving forces are mined one by one through the random forest algorithm of the land expansion analysis strategy to obtain the land use type. k Cell i Probability of growth And the contribution of driving factors to the changes in various land use types during this period. The specific calculation formula is as follows:
[0011] (1)
[0012] In the formula, is the indicator function of the decision tree set; For vector x No. n The prediction type of the decision tree; d The value is 0 or 1, 0 means that the land use type is changed from other types to k 1 indicates that the land use type is changed from other land use types to land use type k ; M is the total number of decision trees;
[0013] Land use type k The total probability The calculation formula is as follows:
[0014] (2)
[0015] In the formula, Land use type k Cell i The probability of growth; r is a random value between 0 and 1; The land use type is set k Thresholds for generating new land use patches; is the adaptive value; t Indicates the number of iterations; Indicates land use type k Cell i neighborhood effect.
[0016] According to the three land use scenarios of natural development scenario, economic priority scenario and ecological priority scenario, the transfer of different land use types is adjusted accordingly to obtain the land use classification results for a certain year in the future;
[0017] Among them, the natural development scenario continues the land use change trend of historical time and does not adjust the model parameters. The economic priority scenario increases the probability of conversion of other land use types to farmland, forest, and impervious surface. , and reduce the probability of these land use types being converted to other land use types for the natural evolution scenario The ecological priority scenario is to increase the probability of converting other land use types to forests, grasslands, and water bodies. , and reduce the probability of these land use types being converted to other land use types as natural evolution .
[0018] Moreover, in step 3, band operation is performed on the remote sensing image to calculate the kernel normalized vegetation index. The specific calculation formula is as follows:
[0019] (3)
[0020] In the formula, ikB is the kernel normalized difference vegetation index, NIR Represents the near-infrared band data of satellite remote sensing images. Red Represents the red light band data of satellite remote sensing images. is the length scale parameter.
[0021] Select high-accuracy long-term series historical meteorological station data, perform rasterization after thin plate spline interpolation, and resample the data. Align the row and column numbers with the land use data by capturing the grid. Take the temperature and precipitation historical data of the meteorological station as independent variables, and obtain the historical data obtained by band operation. ikBAs the dependent variable, a multiple linear regression model was established to study the relationship between temperature and precipitation data and ikB The relationship between them, the calculation formula of the multiple linear regression model is as follows:
[0022] (4)
[0023] In the formula, Y express ikB , is the residual term, and For temperature data and precipitation data The coefficient of .
[0024] The monthly average temperature and monthly total precipitation data under different future climate scenarios collected in step 2 are input into the model to predict the corresponding forecast year. ikB .
[0025] Furthermore, in step 4, ikB Inverse calculation to obtain the core vegetation coverage FV , and then establish the ratio of effective photosynthetic radiation absorption by vegetation FPAR The linear relationship between:
[0026] (5)
[0027] (6)
[0028] In the formula, FV The pixel binary model is used for calculation and the study area is selected ikB The lower and upper limits of the 95% confidence interval of the value are taken as pure bare soil pixels. and pure vegetation pixels ; and for FPAR The maximum and minimum values of .
[0029] Light energy utilization model NPP According to the Monteith equation, the photosynthetically active radiation absorbed by vegetation APAR and actual light energy utilization rate of vegetation in future land use types Calculated, that is:
[0030] (7)
[0031] (8)
[0032] (9)
[0033] In the formula, Represents pixel x exist t The net primary productivity of vegetation in the month, Represents pixel x exist t The photosynthetically active radiation absorbed by the month For pixels x exist t The actual light energy utilization rate of vegetation in a month, Represents pixel x exist t Monthly photosynthetically active radiation, Represents pixel x exist t The proportion of effective photosynthetic radiation absorbed by vegetation in a month, Represents pixel x exist t The total solar radiation in a month. The constant 0.5 represents the proportion of the effective solar radiation that can be used by vegetation at a wavelength of 0.38-0.71μm to the total solar radiation.
[0034] Among them, the actual light energy utilization rate of vegetation High temperature stress coefficient , low temperature stress coefficient , water stress coefficient and the maximum light energy utilization rate of different vegetation The estimation formula is as follows:
[0035] (10)
[0036] (11)
[0037] (12)
[0038] (13)
[0039] In the formula, Indicates the temperature most suitable for vegetation growth; represents the monthly average temperature; When the temperature is less than or equal to -10℃, Take 0; when Than the optimum temperature When the temperature is 10℃ higher or 13℃ lower, equal The optimum temperature Calculated when half of the value; Represents pixel x exist t Actual regional evapotranspiration for the month; Represents pixel x existt The regional potential evapotranspiration for the month is calculated based on the complementary relationship proposed by Boucher. The specific calculation method is as follows:
[0040] (14)
[0041] (15)
[0042] In the formula, Represents pixel x exist t Monthly precipitation, Represents pixel x exist t The net radiation of the surface of the month, Represents pixel x exist t The local potential evaporation in a month is calculated as follows:
[0043] (16)
[0044] (17)
[0045] (18)
[0046] In the formula, For accumulated temperature.
[0047] Moreover, the respiration rate of soil microorganisms in step 5 It is mainly affected by temperature and precipitation, and is calculated according to Pei Zhiyong's model, namely:
[0048] (19)
[0049] In the formula, T represents the monthly mean temperature in the study area, P represents the total monthly precipitation in the study area.
[0050] Vegetation carbon sink through vegetation net primary productivity NPP The difference between the carbon emissions from microbial respiration in the soil and the net ecosystem productivity of vegetation is expressed as follows:
[0051] (20)
[0052] In the formula, NEP It represents the net ecosystem productivity of vegetation in the study area, i.e., the real carbon sink of vegetation; Represents the respiration of soil microorganisms in the study area.
[0053] The present invention also provides a vegetation carbon sink prediction system based on future scenario simulation, which is used to implement the vegetation carbon sink prediction method based on future scenario simulation as described above.
[0054] Furthermore, the method comprises a processor and a memory, wherein the memory is used to store program instructions, and the processor is used to call the stored instructions in the memory to execute the vegetation carbon sink prediction method based on future scenario simulation as described above.
[0055] Alternatively, it includes a readable storage medium having a computer program stored thereon, and when the computer program is executed, the method for predicting vegetation carbon sinks based on future scenario simulation as described above is implemented.
[0056] Compared with the prior art, the present invention has the following advantages:
[0057] The present invention uses satellite remote sensing images and meteorological data as historical data sources, and can accurately and reliably obtain the basic data required for vegetation carbon sink calculation. Compared with directly predicting future vegetation carbon sink results through calculated carbon sink historical data and regression models, the vegetation carbon sink prediction method based on future scenario simulation proposed in the present invention has a solid theoretical basis for the physical and chemical growth characteristics of vegetation, and the calculation results are more accurate. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0059] Figure 1 It is a flow chart of the vegetation carbon sink prediction method according to an embodiment of the present invention.
[0060] Figure 2 It is a spatial distribution map of temperature data in the study area in 2030 according to an embodiment of the present invention.
[0061] Figure 3 This is the kNDVI spatial distribution map of the study area in 2030 according to the embodiment of the present invention.
[0062] Figure 4 This is the NPP spatial distribution map of the study area in 2030 according to the embodiment of the present invention.
[0063] Figure 5 It is a spatial distribution map of NEP for the natural development-SSP2-4.5 scenario in the study area of the embodiment of the present invention. DETAILED DESCRIPTION
[0064] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in combination with the drawings and embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0065] Example 1
[0066] like Figure 1 As shown, an embodiment of the present invention provides a vegetation carbon sink prediction method based on future scenario simulation, comprising the following steps:
[0067] Step 1: Collect historical land use data and multi-source driving factors in the study area, and simulate the spatial distribution of future land use types under different land use scenarios through the Patch-generating Land Use Simulation (PLUS) model.
[0068] This example uses the land use data of the China Land Cover Dataset (CLCD) from 2000 to 2020 for analysis. Elevation, slope, precipitation, temperature, GDP, population density, distance to roads, distance to railways, and distance to rivers are selected as driving factors, and these multi-source driving factors are unified into a 30-meter spatial resolution through rasterization and resampling. Through the random forest algorithm of the Land Extension Analysis Strategy (LEAS), the factors of various land use expansion and driving forces are mined one by one to obtain the land use type. k Cell i Probability of growth And the contribution of driving factors to the changes in various land use types during this period. The specific calculation formula is as follows:
[0069] (1)
[0070] In the formula, is the indicator function of the decision tree set; For vector x No. n The prediction type of the decision tree; d The value is 0 or 1, 0 means that the land use type is changed from other types to k 1 indicates that the land use type is changed from other land use types to land use type k ; M is the total number of decision trees.
[0071] Land use type kThe total probability The calculation formula is as follows:
[0072] (2)
[0073] In the formula, Land use type k Cell i The probability of growth; r is a random value between 0 and 1; The land use type is set k Thresholds for generating new land use patches; is the adaptive value; t Indicates the number of iterations; Indicates land use type k Cell i neighborhood effect.
[0074] According to the three land use scenarios of natural development scenario (this scenario continues the land use change trend of historical time and does not adjust the model parameters), economic priority scenario (the conversion probability of other land use types to farmland, forest and impervious surface is increased by 30%, and the probability of these land use types being converted to other land use types is reduced to 70% of the natural evolution scenario) and ecological priority scenario (the conversion probability of other land use types to forest, grassland and water body is increased by 30%, and the probability of these land use types being converted to other land use types is reduced to 70% of the natural evolution), the transfer situation of different land use types was adjusted accordingly, and the land use classification results in 2030 were obtained.
[0075] Step 2: Collect the monthly mean temperature, monthly total precipitation, and monthly total solar radiation data under different future climate scenarios in the study area, and interpolate and resample them to maintain the same spatial resolution as the land use data.
[0076] According to the Coupled Model Intercomparison Project (CMIP), which coordinates the design and distribution of global climate models to simulate past, present and future climate, HadGEM3-GC3-LL model data of the sixth assessment report CMIP6 were selected. The monthly mean temperature, monthly total precipitation and monthly total solar radiation data of the two scenarios SSP2-4.5 (moderate forcing scenario, radiative forcing stabilized at ~4.5W / m2 in 2100) and SSP5-8.5 (high forcing scenario, radiative forcing stabilized at ~8.5W / m2 in 2100) were collected, and the data were interpolated and resampled to keep the same spatial resolution as the land use data. The row and column numbers of meteorological data and land use data were aligned by snapping the grid method. Figure 2This is the spatial distribution map of temperature data in the study area in 2030. The low temperature values are concentrated near the mountainous area in the west of the study area, and the high temperature values are concentrated in the east of the study area.
[0077] Step 3: Obtain the historical kernel normalized vegetation index by performing band operations on historical satellite remote sensing images. Use the historical kernel normalized vegetation index and historical meteorological station meteorological data to build a relationship model between the kernel normalized vegetation index and meteorological data. Based on the monthly average temperature and monthly total precipitation data under different future climate scenarios collected in step 2, predict the future change trend of the kernel normalized vegetation index.
[0078] ikB is the kernel normalized difference vegetation index, ikB Compared to NDVI It can solve the saturation effect, complex phenological cycle and seasonal changes in vegetation parameter calculation, and handle mixed pixel problems. It can perform band calculation on historical remote sensing images of Landsat series, Sentinel series and high-resolution satellite series. ikB , the specific calculation formula is as follows:
[0079] (3)
[0080] In the formula, ikB is the kernel normalized difference vegetation index, NIR Represents the near-infrared band data of satellite remote sensing images. Red Represents the red light band data of satellite remote sensing images. is the length scale parameter and can be adjusted by To reflect the nonlinear sensitivity of NDVI to vegetation density, it is usually set hour, ikB The value of .
[0081] Long-term series historical meteorological station data with high accuracy were selected, and then rasterized after thin plate spline interpolation. The data were resampled to a spatial resolution of 30 meters, and the row and column numbers were aligned with the land use data by snapping the grid.
[0082] The temperature and precipitation historical data of the meteorological station are used as independent variables, and the historical data obtained by band operation are ikB As the dependent variable, a multiple linear regression model was established to study the relationship between temperature and precipitation data and ikB The relationship between them. The calculation formula of the multiple linear regression model is as follows:
[0083] (4)
[0084] In the formula, Y express ikB , is the residual term, and For temperature data and precipitation data The coefficient of .
[0085] The meteorological data under the SSP2-4.5 and SSP5-8.5 scenarios in step 2 are input into the model to predict the corresponding forecast years. ikB . Figure 3 The study area in 2030 ikB Spatial distribution map, in which the city center ikB The value is lower.
[0086] Step 4: Based on the CASA model, calculate the net primary productivity of vegetation under various future vegetation carbon sink prediction scenarios according to the land use type classification results deduced under the land use scenario in step 1, the monthly average temperature, monthly total precipitation and monthly total solar radiation data under different future climate scenarios in step 2, and the predicted future kernel normalized difference vegetation index in step 3.
[0087] pass ikB Inverse calculation to obtain the core vegetation coverage FV , and then establish the ratio of effective photosynthetic radiation absorption by vegetation FPAR The linear relationship between:
[0088] (5)
[0089] (6)
[0090] In the formula, FV The pixel binary model is used for calculation and the study area is selected ikB The lower and upper limits of the 95% confidence interval of the value are taken as pure bare soil pixels. and pure vegetation pixels ; and for FPAR The maximum and minimum values of in this embodiment are 0.95 and 0.001 respectively.
[0091] The CASA model is a method for estimating the net primary productivity of vegetation. NPP The light energy utilization model is used to calculate the data required, including the predicted monthly ikB Grid data and the corresponding month of CMIP6 future climate simulation of temperature, precipitation, and total solar radiation meteorological parameters. NPP The photosynthetically active radiation absorbed by vegetation can be used in the Monteith equation APAR and actual light energy utilization of future land use types Calculated, that is:
[0092] (7)
[0093] (8)
[0094] (9)
[0095] In the formula, Represents pixel x exist t Net primary productivity of vegetation , Represents pixel x exist t Photosynthetically active radiation absorbed per month , For pixels x exist t Actual light energy utilization rate of vegetation in January , Represents pixel x exist t Monthly photosynthetically active radiation , Represents pixel x exist t The proportion of effective photosynthetic radiation absorbed by vegetation in a month, Represents pixel x exist t Total solar radiation per month , the constant 0.5 represents the proportion of effective solar radiation that can be utilized by vegetation at a wavelength of 0.38-0.71 μm to the total solar radiation.
[0096] Among them, the actual light energy utilization rate of vegetation High temperature stress coefficient , low temperature stress coefficient , water stress coefficient and the maximum light energy utilization rate of different vegetation The estimation formula is as follows:
[0097] (10)
[0098] (11)
[0099] (12)
[0100] (13)
[0101] In the formula, Indicates the temperature most suitable for vegetation growth; represents the monthly average temperature; When the temperature is less than or equal to -10℃, Take 0; when Than optimum temperature When the temperature is 10℃ higher or 13℃ lower, equal The optimum temperature Calculated when half of the value; Represents pixel x exist t Monthly actual regional evapotranspiration (mm); Represents pixel x exist t The regional potential evapotranspiration (mm) for each month is calculated based on the complementary relationship proposed by Boucher. The specific calculation method is as follows: (14)
[0102] (15)
[0103] In the formula, Represents pixel x exist t Monthly precipitation (mm), Represents pixel x exist t Net surface radiation in the month (mm), Represents pixel x exist t Monthly local potential evaporation (mm), calculated as follows:
[0104] (16)
[0105] (17)
[0106] (18)
[0107] In the formula, For accumulated temperature.
[0108] In the present invention, only three land use types with vegetation carbon fixation types, namely farmland, forest and grassland, need to be considered for calculating vegetation carbon sinks. Figure 4 The study area in 2030 NPP Spatial distribution map, where low values are mainly concentrated near cities, and high values are distributed in the west of the study area, with the maximum value of 248gC / (m 2 ·moon).
[0109] Step 5, using the monthly average temperature and monthly total precipitation data under different future climate scenarios collected in step 2 to calculate the predicted soil microbial respiration, combined with the vegetation net primary productivity under multiple future vegetation carbon sink prediction scenarios calculated in step 4, invert the vegetation carbon sink under multiple future scenarios.
[0110] Respiration of soil microorganisms It is mainly affected by temperature and precipitation, and is calculated according to Pei Zhiyong's model, namely:
[0111] (19)
[0112] In the formula, T represents the monthly average temperature in the study area (℃), P represents the total monthly precipitation in the study area (mm).
[0113] Vegetation carbon sink through vegetation net primary productivity NPP The difference between the carbon emissions from microbial respiration in the soil and the net ecosystem productivity of vegetation is expressed as follows:
[0114] (20)
[0115] In the formula, NEP It represents the net ecosystem productivity of vegetation in the study area, that is, the real carbon sink of vegetation ; Represents the respiration of soil microorganisms in the study area .
[0116] Combining land use scenarios and climate model scenarios, the following six different scenarios of vegetation carbon sinks were obtained (see Table 1).
[0117] Table 1 Prediction of vegetation carbon sink under different scenarios
[0118]
[0119] Figure 5 This is a comparison of the natural development-SSP2-4.5 scenario in Table 1. NEP The calculation results are displayed, and the overall spatial distribution and Figure 4 In NPP The spatial distribution is similar, where values less than 0 represent carbon sources, values greater than 0 represent carbon sinks, and the maximum value is 184 gC / (m 2 ·month), with a minimum value of -59 gC / (m 2 ·moon).
[0120] Example 2
[0121] Based on the same inventive concept, the present invention also provides a vegetation carbon sink prediction system based on future scenario simulation, including a processor and a memory, the memory is used to store program instructions, and the processor is used to call the program instructions in the memory to execute a vegetation carbon sink prediction method based on future scenario simulation as described above.
[0122] Example 3
[0123] Based on the same inventive concept, the present invention also provides a vegetation carbon sink prediction system based on future scenario simulation, including a readable storage medium, on which a computer program is stored. When the computer program is executed, a vegetation carbon sink prediction method based on future scenario simulation as described above is implemented.
[0124] In specific implementation, the method proposed in the technical solution of the present invention can be implemented by technical personnel in this field using computer software technology to realize automatic operation process. System devices for implementing the method, such as computer-readable storage media storing the corresponding computer program of the technical solution of the present invention and computer equipment running the corresponding computer program, should also be within the scope of protection of the present invention.
[0125] The specific embodiments described herein are merely examples of the spirit of the present invention. Those skilled in the art may make various modifications or additions to the specific implementation cases described, or replace them with similar methods, without departing from the spirit of the present invention or exceeding the scope defined by the appended claims.
Claims
1. A vegetation carbon sink prediction method based on future scenario simulation, characterized in that: The following steps are involved: Step 1: Collect historical land use data and multi-source driving factors of the study area, simulate the spatial distribution of future land use types under different land use scenarios through the PLUS model, and deduce the land use classification results under different land use scenarios in a certain year in the future; The multi-source driving factors include elevation, slope, precipitation, temperature, GDP, population density, distance to roads, distance to railways and distance to rivers. The multi-source driving factors are rasterized and resampled. The random forest algorithm of the land expansion analysis strategy is used to explore the factors of various land use expansion and driving forces one by one to obtain the land use type. k Cell i Probability of growth , the specific calculation formula is as follows: (1) In the formula, is the indicator function of the decision tree set; For vector x No. n The prediction type of the decision tree; d The value is 0 or 1, 0 means that the land use type is changed from other types to k 1 indicates that the land use type is changed from other land use types to land use type k ; M is the total number of decision trees; Land use type k The total probability The calculation formula is as follows: (2) In the formula, Land use type k Cell i The probability of growth; r is a random value between 0 and 1; The land use type is set k Thresholds for generating new land use patches; is the adaptive value; t Indicates the number of iterations; Indicates land use type k Cell i Neighborhood effects; Step 2: Collect the monthly mean temperature, monthly total precipitation and monthly total solar radiation data under different future climate scenarios in the study area, and interpolate and resample them to keep the same spatial resolution as the historical land use data collected in step 1; Step 3, obtain the historical kernel normalized vegetation index by performing band calculation on the historical satellite remote sensing images, use the historical kernel normalized vegetation index and the meteorological data of the historical meteorological stations to build a relationship model between the kernel normalized vegetation index and meteorological data, and predict the kernel normalized vegetation index of the corresponding forecast year based on the monthly average temperature and monthly total precipitation data under different future climate scenarios collected in step 2; Step 4: Based on the CASA model, calculate the net primary productivity of vegetation under multiple future vegetation carbon sink prediction scenarios according to the land use type classification results deduced under the land use scenario in step 1, the monthly average temperature, monthly total precipitation and monthly total solar radiation data under different future climate scenarios in step 2, and the kernel normalized difference vegetation index predicted in step 3; Step 5, using the monthly average temperature and monthly total precipitation data under different future climate scenarios collected in step 2 to calculate the predicted soil microbial respiration, combined with the vegetation net primary productivity under multiple future vegetation carbon sink prediction scenarios calculated in step 4, invert the vegetation carbon sink under multiple future scenarios.
2. A vegetation carbon sink prediction method based on future scenario simulation as claimed in claim 1, characterized in that: In step 1, according to the three land use scenarios of natural development scenario, economic priority scenario and ecological priority scenario, the transfer of different land use types is adjusted accordingly to obtain the land use classification results for a certain year in the future; Among them, the natural development scenario continues the land use change trend of historical time and does not adjust the model parameters. The economic priority scenario increases the probability of conversion of other land use types to farmland, forest, and impervious surface. , and reduce the probability of these land use types being converted to other land use types for the natural evolution scenario The ecological priority scenario is to increase the probability of converting other land use types to forests, grasslands, and water bodies. , and reduce the probability of these land use types being converted to other land use types as natural evolution .
3. The vegetation carbon sink prediction method based on future scenario simulation according to claim 1, characterized in that: In step 3, band operations are performed on the remote sensing image to calculate the kernel normalized vegetation index. The specific calculation formula is as follows: (3) In the formula, ikB is the kernel normalized difference vegetation index, NIR Represents the near-infrared band data of satellite remote sensing images. Red Represents the red light band data of satellite remote sensing images. is the length scale parameter; Select high-accuracy long-term series historical meteorological station data, perform rasterization after thin plate spline interpolation, and resample the data. Align the row and column numbers with the land use data by capturing the grid. Take the temperature and precipitation historical data of the meteorological station as independent variables, and obtain the historical data obtained by band operation. ikB As the dependent variable, a multiple linear regression model was established to study the relationship between temperature and precipitation data and ikB The relationship between them, the calculation formula of the multiple linear regression model is as follows: (4) In the formula, Y express ikB , is the residual term, and For temperature data and precipitation data The coefficient of The monthly average temperature and monthly total precipitation data under different future climate scenarios collected in step 2 are input into the model to predict the corresponding forecast year. ikB .
4. The vegetation carbon sink prediction method based on future scenario simulation according to claim 1, characterized in that: In step 4, ikB Inverse calculation to obtain the core vegetation coverage FV , and then establish the ratio of effective photosynthetic radiation absorption by vegetation FPAR The linear relationship between: (5) (6) In the formula, FV The pixel binary model is used for calculation and the study area is selected ikB The lower and upper limits of the 95% confidence interval of the value are taken as pure bare soil pixels. and pure vegetation pixels ; and for FPAR The maximum and minimum values of .
5. A vegetation carbon sink prediction method based on future scenario simulation as claimed in claim 4, characterized in that: In the light energy utilization model in step 4 NPP According to the Monteith equation, the photosynthetically active radiation absorbed by vegetation APAR and actual light energy utilization rate of vegetation in future land use types Calculated, that is: (7) (8) (9) In the formula, Represents pixel x exist t The net primary productivity of vegetation in the month, Represents pixel x exist t The photosynthetically active radiation absorbed by the month For pixels x exist t The actual light energy utilization rate of vegetation in a month, Represents pixel x exist t Monthly photosynthetically active radiation, Represents pixel x exist t The proportion of effective photosynthetic radiation absorbed by vegetation in a month, Represents pixel x exist t Total solar radiation for the month.
6. A vegetation carbon sink prediction method based on future scenario simulation as claimed in claim 5, characterized in that:Actual light energy utilization rate of vegetation in step 4 High temperature stress coefficient , low temperature stress coefficient , water stress coefficient and the maximum light energy utilization rate of different vegetation The estimation formula is as follows: (10) (11) (12) (13) In the formula, Indicates the temperature most suitable for vegetation growth; represents the monthly average temperature; When the temperature is less than or equal to -10℃, Take 0; when Than optimum temperature When the temperature is 10℃ higher or 13℃ lower, equal The optimum temperature Calculated when half of the value; Represents pixel x exist t Actual regional evapotranspiration for the month; Represents pixel x exist t The regional potential evapotranspiration for the month is calculated based on the complementary relationship proposed by Boucher. The specific calculation method is as follows: (14) (15) In the formula, Represents pixel x exist t Monthly precipitation, Represents pixel x exist t The net radiation of the surface of the month, Represents pixel x exist t The local potential evaporation in a month is calculated as follows: (16) (17) (18) In the formula, For accumulated temperature.
7. The vegetation carbon sink prediction method based on future scenario simulation according to claim 1, characterized in that: Respiration of soil microorganisms in step 5 It is mainly affected by temperature and precipitation, and is calculated according to Pei Zhiyong's model, namely: (19) In the formula, T represents the monthly mean temperature in the study area, P represents the total monthly precipitation in the study area; Vegetation carbon sink through vegetation net primary productivity NPP The difference between the carbon emissions from microbial respiration in the soil and the net ecosystem productivity of vegetation is expressed as follows: (20) In the formula, NEP It represents the net ecosystem productivity of vegetation in the study area, i.e., the real carbon sink of vegetation; Represents the respiration of soil microorganisms in the study area.
8. A vegetation carbon sink prediction system based on future scenario simulation, characterized in that: It includes a processor and a memory, the memory is used to store program instructions, and the processor is used to call the program instructions in the memory to execute a vegetation carbon sink prediction method based on future scenario simulation as described in any one of claims 1-7.
9. A vegetation carbon sink prediction system based on future scenario simulation, characterized in that: It includes a readable storage medium, on which a computer program is stored. When the computer program is executed, a vegetation carbon sink prediction method based on future scenario simulation as described in any one of claims 1 to 7 is implemented.
