Grassland carbon sink measuring system

Abstract

The invention discloses a grassland carbon sink determination system, and relates to the technical field of grassland carbon sink determination, the maximum light energy utilization rate of different grassland types in Inner Mongolia is subdivided, the water stress coefficient is improved by combining the surface water index and rainfall data, the applicability of a CASA model in arid and semi-arid regions is significantly optimized, the NPP estimation error is reduced by 20%-30%, and the accuracy of the CASA model in arid and semi-arid regions is improved. The correlation coefficient of an NEP simulation result and actually measured data is increased to 0.85 or above, and the regional adaptability of carbon sink monitoring is greatly improved.

Description

Technical Field

[0001] This invention relates to the technical field of grassland carbon sequestration measurement, specifically a grassland carbon sequestration measurement system. Background Technology

[0002] Against the backdrop of global climate change, grassland ecosystems, as important terrestrial carbon sinks, require dynamic monitoring and accurate assessment for ecological protection and climate change response.

[0003] However, existing grassland carbon sequestration measurement technologies still face many challenges. Most mainstream carbon sequestration models are based on vegetation parameters in North America or humid regions, and do not fully consider the uniqueness of grasslands in arid and semi-arid regions. For example, grasslands in Inner Mongolia are diverse and account for more than 70% of vegetation cover, but the maximum light energy utilization rate of existing models is not subdivided according to grassland type, resulting in insufficient accuracy in estimating net primary productivity and net ecosystem productivity. In addition, the calculation of water stress coefficient depends on complex soil parameters, making it difficult to achieve high-precision simulation at the regional scale. Summary of the Invention

[0004] The purpose of this invention is to provide a grassland carbon sink measurement system to address the problem that traditional mainstream carbon sink models mentioned above do not fully consider the unique characteristics of grasslands in arid and semi-arid regions.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a grassland carbon sink measurement system, comprising: an extreme climate dataset construction module, a net primary productivity calculation module, a soil respiration rate calculation module, a light energy utilization optimization module, a water stress coefficient optimization module, and a land use and cover type classification accuracy optimization module; Among them, the extreme climate dataset construction module is used to build single and compound extreme climate index datasets in Inner Mongolia; The net primary productivity calculation module is used to estimate the net primary productivity of vegetation using the CASA model; The soil respiration rate calculation module is based on the soil respiration model and is used to calculate the CO2 emissions from soil microbial heterotrophic respiration. The light energy utilization optimization module is used to further refine the CASA model based on the maximum light energy utilization values ​​for different grassland types. The water stress coefficient optimization module is used to calculate the optimized water stress coefficient, which reflects the impact of the available water that vegetation can utilize on light energy utilization. The land use and cover classification accuracy optimization module is used to eliminate the inherent errors in land use and cover classification and the Normalized Difference Vegetation Index (NDVI) data itself.

[0006] In one example, the steps of the extreme climate dataset building module to create a dataset of single and compound extreme climate indices in Inner Mongolia include: S201. First, combining ground observation and field measurement data, and using the correlation coefficient R... 2 The relative error (RE) and Nash efficiency coefficient (NSE) of multi-source remote sensing were compared and analyzed. The accuracy and applicability of the data in Inner Mongolia were then analyzed. The relative uncertainty of different data sources was tested using the triangular hat method, and the data source with the highest accuracy and the lowest uncertainty was selected. S202. Based on the selected precipitation, evapotranspiration, soil water and maximum, minimum and average temperature data, a dataset of single and compound extreme climate indices in Inner Mongolia since 2000 was obtained using threshold identification and standardization methods, and the dataset was validated based on field survey data.

[0007] In one example, the main evaluation parameters used by the extreme climate dataset building module include the correlation coefficient R. 2 The Nash efficiency coefficient (NSE), relative error (RE), and mean absolute error (MAE) are listed below: ; ; ; ; In the formula P si P is the estimated value at time point i. oi The actual value observed at time point i; as well as These represent the average values ​​of the estimated and actual values ​​over the corresponding time periods, respectively; n represents the length of the corresponding time series.

[0008] In one example, the structure of the CASA model is expressed as follows: ; In the formula, NPP(x,t), APAR(x,t), and ε(x,t) represent the net primary productivity of vegetation, the photosynthetically active radiation absorbed by vegetation, and the actual light energy utilization rate of pixel x at time t, respectively. The calculation of APAR(x,t) is as follows: ; ; ; In the formula, SOL(x,t) and FPAR(x,t) represent the absorption ratios of total solar radiation and photosynthetically active radiation, respectively; SR is the ratio vegetation index. min and SR max These represent the NDVI percentiles at the 5th and 95th percentiles for certain grassland types, respectively.

[0009] In one example, the soil respiration rate calculation module, based on a soil respiration model, calculates the CO2 emissions from heterotrophic respiration of soil microorganisms, resulting in the soil respiration rate R. s The calculation model is as follows: ; In the formula, R s This represents the monthly soil respiration rate, expressed in kgC / m³. 2 b is the temperature sensitivity coefficient. ;T a is the monthly average temperature in °C; P is the precipitation in cm; Q, f, and k are constants, where Q = 0.05452, f = 1.250, and k = 4.259.

[0010] In one example, soil heterogeneous respiration R in Inner Mongolia H Breathing with the soil S The empirical formula is: ; In the formula, R H This represents the monthly soil heterotrophic respiration rate, expressed in kgC / m³. 2 .

[0011] In one example, the light energy utilization optimization module utilizes MODIS NDVI data products, meteorological data, vegetation type data, and measured net primary productivity (NPP) data for different grasslands. Based on the principle of minimizing error, it establishes a quadratic equation model to optimize the maximum light energy utilization of the three major grassland types in Inner Mongolia, thereby localizing the parameters of the CASA model. ; In the formula, i represents the number of the i-th sample of a certain grassland type, k is the maximum number of samples of a certain grassland type, m is the measured NPP data, and n comes from the basic expression of the CASA model, i.e., n = APAR × T ε1 ×T ε2 ×W ε ε∗ represents the maximum light energy utilization rate (NPP) of a certain grassland type to be determined, and l and μ are the minimum and maximum values ​​of the maximum NPP for each grassland type. Expanding the formula yields a quadratic equation model with the opening facing upwards. After taking the first derivative, the abscissa value corresponding to the minimum error between the measured and simulated NPP can be calculated, which is the maximum NPP ε* of a certain grassland type. The formula is: .

[0012] In one example, the water stress coefficient optimization module incorporates the surface water index LSWI into the CASA model to calculate W. ε(x,t) is used, and the surface moisture index LSWI is calculated using the MODIS surface reflectance product MOD09A1. Then, the precipitation ratio is introduced to further optimize this parameter to obtain the final optimized water stress coefficient W. ε_improved (x,t).

[0013] In one example, the water stress coefficient W ε_improved The specific formula for calculating (x,t) is: ; ; ; ; In the formula, ρ nir and ρ swir Prep and prep represent the surface reflectance in the near-infrared and short-wave infrared bands, respectively. max These represent monthly precipitation and maximum monthly precipitation during the growing season, respectively.

[0014] In one example, the land use and cover classification accuracy optimization module introduces land use and cover classification accuracy, causing the maximum value of the Normalized Difference Vegetation Index (NDVI) to be adjusted accordingly as the classification accuracy changes. Specifically, this includes the following steps: S1001. First, analyze the probability distribution of the maximum value of the Normalized Difference Vegetation Index (NDVI) for different grassland vegetation types; Then, based on the classification accuracy p, within the probability distribution interval... Each pixel of this type is selected internally, thereby eliminating the influence of classification errors; Finally, the probability distribution of the Normalized Difference Vegetation Index (NDVI) is calculated again for the selected pixels. The NDVI corresponding to the 95th percentile of this distribution is the maximum value of the NDVI, and the NDVI corresponding to the 5th percentile is the minimum value of the NDVI. This can eliminate the error caused by noise in remote sensing images to a certain extent.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: This scheme improves the water stress coefficient by subdividing the maximum light energy utilization rate of different grassland types in Inner Mongolia and combining surface water index and precipitation data. It significantly optimizes the applicability of the CASA model in arid and semi-arid regions, reduces the NPP estimation error by 20% to 30%, and increases the correlation coefficient between NEP simulation results and measured data to over 0.85, greatly enhancing the regional adaptability of carbon sink monitoring. Attached Figure Description

[0016] Figure 1This is a schematic diagram of a grassland carbon sequestration measurement system according to the present invention; Figure 2 This is a schematic diagram of the structural equation model of the present invention. Detailed Implementation

[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only 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 making creative efforts are within the scope of protection of the present invention.

[0018] Example 1: Figure 1 and Figure 2 As shown, a grassland carbon sequestration measurement system includes an extreme climate dataset construction module. This module is used to establish single and compound extreme climate index datasets for Inner Mongolia. To obtain high-precision, full-coverage, and high-resolution single and compound extreme climate datasets for Inner Mongolia, such as moisture and temperature, the specific steps are as follows: Step 1: First, combine ground observation and field measurement data, and then use the correlation coefficient R... 2 The relative error (RE) and Nash efficiency coefficient (NSE) of multi-source remote sensing data were compared and analyzed. The accuracy and applicability of the data in Inner Mongolia were then analyzed. The triangular hat method was used to test the relative uncertainty of different data sources, and the data source with the highest accuracy and the lowest uncertainty was selected.

[0019] Step 2: Based on the selected precipitation, evapotranspiration, soil water, maximum, minimum and average temperature data, use threshold identification and standardization methods to obtain a dataset of single and compound extreme climate indices in Inner Mongolia since 2000, and verify it based on field survey data.

[0020] The main evaluation parameters used in the extreme climate dataset construction module include the correlation coefficient R. 2 The Nash efficiency coefficient (NSE), relative error (RE), and mean absolute error (MAE) are listed below: ; ; ; ; In the formula, R 2 , P si P is the estimated value at time point i. oi The actual value observed at time point i; as well as These represent the average values ​​of the estimated and actual values ​​over the corresponding time periods, respectively; n represents the length of the corresponding time series.

[0021] Net ecosystem exchange (NEP) represents net primary productivity (NPP) and soil heterotrophic respiration (R). H The difference between NEP and REP represents the net uptake or net storage of carbon in an ecosystem, thus quantitatively describing the ecosystem's carbon sink function. When NEP > 0, it indicates that the ecosystem is functioning as a carbon sink; conversely, it indicates a carbon source. Accurate estimation of NEP helps in the quantitative analysis of the carbon sequestration status and potential of regional ecosystems. The calculation of NEP generally involves two steps: NPP and REP. H Estimate. This scheme uses the CASA model and the soil respiration model to estimate NPP and R in Inner Mongolia. H Estimate.

[0022] The Net Primary Productivity (NPP) calculation module is used to estimate the NPP of vegetation. The CASA model, as a primary method for estimating vegetation NPP, is widely used in arid and semi-arid regions. This model is a light energy utilization efficiency model built based on vegetation type, NDVI, and meteorological data, primarily determined using photosynthetically active radiation absorbed by vegetation and light energy utilization efficiency. The CASA model structure is expressed as follows: ; In the formula, NPP(x,t), APAR(x,t), and ε(x,t) represent the net primary productivity of vegetation, the photosynthetically active radiation absorbed by vegetation, and the actual light energy utilization rate of pixel x at time t, respectively. APAR(x,t) is calculated as follows: ; ; ; In the formula, SOL(x,t) and FPAR(x,t) represent the absorption ratios of total solar radiation and photosynthetically active radiation, respectively; SR is the ratio vegetation index. min and SR max These represent the NDVI percentiles at the 5th and 95th percentiles for certain grassland types, respectively.

[0023] The actual light energy utilization efficiency ε(x,t) is mainly affected by temperature and moisture stress, and the calculation formula is as follows: ; In the formula, T ε1 (x,t),T ε2 (x,t) represents the effect of temperature on light energy utilization; W ε (x,t) reflects the impact of water stress on light energy utilization; ε maxThis represents the maximum light energy utilization rate that vegetation can achieve under ideal conditions.

[0024] The soil respiration rate calculation module, based on a soil respiration model, calculates the CO2 emissions from heterotrophic respiration of soil microorganisms, and the soil respiration rate R. s The calculation model is as follows: ; In the formula, R s This represents the monthly soil respiration rate, expressed in kgC / m³. 2 b is the temperature sensitivity coefficient. ;T a is the monthly average temperature in °C; P is the precipitation in cm; Q, f, and k are constants, where Q = 0.05452, f = 1.250, and k = 4.259.

[0025] Soil heterogeneous respiration in Inner Mongolia H Breathing with the soil S The empirical formula is: ; In the formula, R H This represents the monthly soil heterotrophic respiration rate, expressed in kgC / m³. 2 .

[0026] Example 2: The CASA model was established for vegetation in North America. When applying the model to other regions, the parameters need to be modified for greater accuracy. To more accurately estimate the net primary productivity of different grasslands in Inner Mongolia, this scheme, based on a full consideration of the characteristics of Inner Mongolia, further modifies the CASA model using a light energy utilization optimization module based on the maximum light energy utilization values ​​of different grassland types. The modified model parameters are then calibrated and validated using field measurement data to localize the model.

[0027] Maximum light energy utilization (ε*) is one of the most important input parameters of the CASA model. Using MODIS NDVI data products, meteorological data, vegetation type data, and measured NPP data from different grasslands, a quadratic equation model was established based on the principle of minimizing error to optimize the maximum light energy utilization of the three major grassland types in Inner Mongolia, thus achieving parameter localization of the CASA model. ; In the formula, i represents the number of the i-th sample of a certain grassland type, k is the maximum number of samples of a certain grassland type, m is the measured NPP data, and n comes from the basic expression of the CASA model, i.e., n = APAR × T ε1 ×T ε2 ×W εε∗ represents the maximum light energy utilization rate (NPP) of a certain grassland type to be determined. l and μ are the minimum and maximum values ​​of the maximum NPP for each grassland type. Expanding the formula yields a quadratic equation model with the opening facing upwards. After taking the first derivative, the abscissa value corresponding to the minimum error between the measured and simulated NPP can be calculated, which is the maximum NPP ε* of a certain grassland type. The formula is: .

[0028] Example 3: The water stress coefficient optimization module is used to calculate the optimized water stress coefficient W. ε_improved (x,t). Water stress coefficient W ε (x,t) reflects the impact of available water available to vegetation on light energy utilization efficiency. As available water in the environment increases, W... ε As (x,t) gradually increases, its value in the CASA model ranges from 0.5 to 1 under conditions of extreme drought to very wet conditions. Calculated using a soil water molecule model, it involves numerous complex soil parameters, making the model difficult to implement and its accuracy uncertain. The LSWI (Land Surface Moisture Index) is an indicator describing the water content of vegetation leaves. Considering W... ε (x,t) is directly related to vegetation moisture content, therefore the surface moisture index LSWI is introduced into the CASA model to calculate W. ε The method (x,t) eliminates numerous soil parameters and simplifies the input parameters. Furthermore, considering the profound impact of precipitation on the productivity of the Inner Mongolian ecosystem, this scheme uses the MODIS surface reflectance product MOD09A1 to calculate the surface moisture index LSWI, which not only enhances the accuracy of the data but also improves its resolution. Additionally, the remote sensing data includes topographic information. Then, the precipitation ratio is introduced to further optimize this parameter, resulting in the final optimized water stress coefficient W. ε_improved (x,t), the specific calculation formula is as follows: ; ; ; ; In the formula, ρ nir and ρ swir Prep and prep represent the surface reflectance in the near-infrared and short-wave infrared bands, respectively. max These represent monthly precipitation and maximum monthly precipitation during the growing season, respectively.

[0029] Example 4: The land use and cover classification accuracy optimization module is used to eliminate the inherent errors in land use and cover classification and the Normalized Difference Vegetation Index (NDVI) data itself. This scheme introduces land use and cover classification accuracy, so that the maximum value of the NDVI changes accordingly with the classification accuracy. Specifically, it includes the following steps: Step 1: First, analyze the maximum distribution probability of the Normalized Difference Vegetation Index (NDVI) for different grassland vegetation types.

[0030] Step 2: Then, based on the classification accuracy p, within the probability distribution interval... Each pixel of this type is selected internally, thereby eliminating the influence of classification errors.

[0031] Step 3: Finally, calculate the probability distribution of the Normalized Difference Vegetation Index (NDVI) for the selected pixels again. The NDVI corresponding to the 95th percentile of this distribution is the maximum value of the NDVI, and the NDVI corresponding to the 5th percentile is the minimum value of the NDVI. This can eliminate the error caused by noise in remote sensing images to a certain extent.

[0032] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

[0033] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A grassland carbon sequestration measurement system, characterized in that, include: The system includes modules for constructing extreme climate datasets, calculating net primary productivity, calculating soil respiration rate, optimizing light energy utilization, optimizing water stress coefficient, and optimizing land use and cover type classification accuracy. The extreme climate dataset construction module is used to establish single and compound extreme climate index datasets for Inner Mongolia. The net primary productivity calculation module is used to estimate the net primary productivity of vegetation using the CASA model; The soil respiration rate calculation module is based on a soil respiration model and is used to calculate the CO2 emissions from heterotrophic respiration of soil microorganisms. The light energy utilization optimization module is used to further correct the CASA model based on the maximum light energy utilization values ​​of different grassland types. The water stress coefficient optimization module is used to calculate the optimized water stress coefficient, which reflects the impact of the available water that vegetation can utilize on light energy utilization. The land use and cover type classification accuracy optimization module is used to eliminate the inherent errors in land use and cover classification and the Normalized Difference Vegetation Index (NDVI) data itself.

2. The grassland carbon sequestration measurement system as described in claim 1, characterized in that: The steps for establishing single and composite extreme climate index datasets in Inner Mongolia using the extreme climate dataset construction module include: S201. First, combining ground observation and field measurement data, and using the correlation coefficient R... 2 The relative error (RE) and Nash efficiency coefficient (NSE) of multi-source remote sensing were compared and analyzed. The accuracy and applicability of the data in Inner Mongolia were then analyzed. The relative uncertainty of different data sources was tested using the triangular hat method, and the data source with the highest accuracy and the lowest uncertainty was selected. S202. Based on the selected precipitation, evapotranspiration, soil water and maximum, minimum and average temperature data, a dataset of single and compound extreme climate indices in Inner Mongolia since 2000 was obtained using threshold identification and standardization methods, and the dataset was validated based on field survey data.

3. The grassland carbon sequestration measurement system as described in claim 1, characterized in that: The main evaluation parameters used in the extreme climate dataset construction module include the correlation coefficient R. 2 The Nash efficiency coefficient (NSE), relative error (RE), and mean absolute error (MAE) are listed below: ； ； ； ； In the formula P si P is the estimated value at time point i. oi The actual value observed at time point i; as well as These represent the average values ​​of the estimated and actual values ​​over the corresponding time periods, respectively; n represents the length of the corresponding time series.

4. The grassland carbon sequestration measurement system as described in claim 1, characterized in that: The structural expression of the CASA model is as follows: ； In the formula, NPP(x,t), APAR(x,t), and ε(x,t) represent the net primary productivity of vegetation, the photosynthetically active radiation absorbed by vegetation, and the actual light energy utilization rate of pixel x at time t, respectively. The calculation of APAR(x,t) is as follows: ； ； ； In the formula, SOL(x,t) and FPAR(x,t) represent the total solar radiation and the absorption ratio of photosynthetically active radiation, respectively; SR is the ratio vegetation index. min and SR max These represent the NDVI percentiles at the 5th and 95th percentiles for certain grassland types, respectively.

5. The grassland carbon sequestration measurement system as described in claim 1, characterized in that: The soil respiration rate calculation module, based on a soil respiration model, calculates the CO2 emissions from heterotrophic respiration of soil microorganisms, and the soil respiration rate R. s The calculation model is as follows: ； In the formula, R s This represents the monthly soil respiration rate, expressed in kgC / m³. 2 b is the temperature sensitivity coefficient. ;T a is the monthly average temperature in °C; P is the precipitation in cm; Q, f, and k are constants, where Q = 0.05452, f = 1.250, and k = 4.

259.

6. The grassland carbon sequestration measurement system as described in claim 5, characterized in that: Soil heterogeneous respiration in Inner Mongolia H Breathing with the soil S The empirical formula is: ； In the formula, R H This represents the monthly soil heterotrophic respiration rate, expressed in kgC / m³. 2 .

7. The grassland carbon sequestration measurement system as described in claim 1, characterized in that: The light energy utilization optimization module utilizes MODIS NDVI data products, meteorological data, vegetation type data, and measured net primary productivity (NPP) data for different grasslands. Based on the principle of minimizing error, it establishes a quadratic equation model to optimize the maximum light energy utilization of the three major grassland types in Inner Mongolia, thereby localizing the parameters of the CASA model. ； In the formula, i represents the number of the i-th sample of a certain grassland type, k is the maximum number of samples of a certain grassland type, m is the measured NPP data, and n comes from the basic expression of the CASA model, i.e., n = APAR × T ε1 ×T ε2 ×W ε ε∗ represents the maximum light energy utilization rate (NPP) of a certain grassland type to be determined, and l and μ are the minimum and maximum values ​​of the maximum NPP for each grassland type. Expanding the formula yields a quadratic equation model with the opening facing upwards. After taking the first derivative, the abscissa value corresponding to the minimum error between the measured and simulated NPP can be calculated, which is the maximum NPP ε* of a certain grassland type. The formula is: 。 8. The grassland carbon sequestration measurement system as described in claim 1, characterized in that: The water stress coefficient optimization module incorporates the surface water index LSWI into the CASA model to calculate W. ε (x,t) is used, and the surface moisture index LSWI is calculated using the MODIS surface reflectance product MOD09A1. Then, the precipitation ratio is introduced to further optimize this parameter to obtain the final optimized water stress coefficient W. ε_improved (x,t).

9. The grassland carbon sequestration measurement system as described in claim 8, characterized in that: The water stress coefficient W ε_improved The specific formula for calculating (x,t) is: ； ； ； ； In the formula, ρ nir and ρ swir Prep and prep represent the surface reflectance in the near-infrared and short-wave infrared bands, respectively. max These represent monthly precipitation and maximum monthly precipitation during the growing season, respectively.

10. The grassland carbon sequestration measurement system as described in claim 1, characterized in that: The land use and cover classification accuracy optimization module incorporates land use and cover classification accuracy, causing the maximum value of the Normalized Difference Vegetation Index (NDVI) to be adjusted accordingly as the classification accuracy changes. Specifically, this includes the following steps: S1001. First, analyze the probability distribution of the maximum value of the Normalized Difference Vegetation Index (NDVI) for different grassland vegetation types; S1002. Then, based on the classification accuracy p, within the probability distribution interval... Each pixel of this type is selected internally, thereby eliminating the influence of classification errors; S1003. Finally, the probability distribution of the Normalized Difference Vegetation Index (NDVI) is calculated again for the selected pixels. The NDVI corresponding to the 95th percentile of this distribution is the maximum value of the NDVI, and the NDVI corresponding to the 5th percentile is the minimum value of the NDVI, thereby eliminating the error caused by remote sensing image noise to a certain extent.

Images