Grassland carbon sink reserve index premium rate determination method based on time sequence simulation

Abstract

The invention provides a grassland carbon sink reserve index premium rate determination method based on time sequence simulation, and the method comprises the steps: S1, estimating a grassland NPP based on a CASA model, and calculating the grassland vegetation carbon sequestration amount according to the grassland NPP; s2, calculating a grassland vegetation carbon sequestration amount loss rate and a separation carbon sequestration amount fluctuation value according to the grassland vegetation carbon sequestration amount obtained in the step S1; s3, performing panel data stability test on the grassland vegetation carbon sequestration amount and the grassland vegetation carbon sequestration amount loss rate; s4, calculating a carbon sequestration loss rate probability, and determining a pure premium rate by adopting a loss rate expected value method; and S5, carrying out risk division on the research area, and correcting the pure premium rate according to the risk division to obtain an actual premium rate. The method can solve the problems of small pilot range, imperfect product system, undetermined pricing mechanism, unscientific premium rate determination and the like of grassland carbon sink insurance.

Description

Technical Field

[0001] This invention relates to the field of grassland carbon sequestration insurance technology, and in particular to a method for determining grassland carbon sequestration storage index insurance premium rates based on time series simulation. Background Technology

[0002] Compared to forest and wetland carbon sinks, grassland carbon sinks face more risk factors. Various natural disasters, such as droughts, fires, pests and diseases, and sandstorms, can all lead to losses in grassland carbon reserves. In addition to natural disasters, grassland carbon sink resource management and carbon sink project development and trading also face multiple risks, including policy and regulatory fluctuations, market price volatility, and technological advancements. Both grassland carbon sink reserve management and the development of grassland carbon sink projects (with expected emission reductions) face significant risks, necessitating innovative grassland carbon sink insurance products to effectively compensate for carbon losses and meet the risk management needs of grassland carbon sinks.

[0003] Grassland carbon sequestration insurance transfers the multiple risks faced by grassland carbon sequestration through market-based means, protecting against the loss of grassland carbon sequestration reserves and the loss of value of grassland carbon sequestration projects, and playing a role in risk management and financial credit enhancement. On January 21, 2022, China Pacific Insurance Property Insurance Company provided grassland carbon sequestration risk protection to herders in Darhan Muminggan United Banner, Baotou City, achieving the first breakthrough in grassland carbon sequestration insurance. Subsequently, pilot projects were gradually carried out in Ordos City and Chifeng City in Inner Mongolia, Yushu Prefecture in Qinghai Province, and other places, with natural grasslands and artificial pastures as the insured objects, covering an area of ​​more than 20,000 mu and providing risk protection of more than 350,000 yuan.

[0004] However, grassland carbon sequestration insurance currently faces challenges such as a small pilot scope, an incomplete product system, an undetermined pricing mechanism, and unscientific rate determination. The product remains at the pilot and exploratory stage and has not fundamentally solved the key issues hindering the development of grassland carbon sequestration insurance. Summary of the Invention

[0005] To address the technical problems existing in current grassland carbon sequestration insurance, this invention provides a method for determining grassland carbon sequestration index insurance rates based on time series simulation. This invention can solve problems such as small pilot scope, incomplete product system, undetermined pricing mechanism, and unscientific rate determination in grassland carbon sequestration insurance. The specific solution is as follows:

[0006] Step S1: Estimate grassland NPP based on the CASA model, and calculate grassland vegetation carbon sequestration and grassland NEP based on grassland NPP;

[0007] Step S2: Calculate the grassland vegetation carbon sequestration loss rate based on the grassland vegetation carbon sequestration obtained in Step S1, and separate the carbon sequestration fluctuation value.

[0008] Step S3: Perform a panel data stationarity test on grassland vegetation carbon sequestration and grassland vegetation carbon sequestration loss rate.

[0009] Step S4: Calculate the probability of carbon sequestration loss rate and determine the pure insurance premium rate using the expected value method of loss rate;

[0010] Step S5: Divide the research area into risk zones and adjust the pure insurance premium rate according to the risk zone division to obtain the actual insurance premium rate.

[0011] The beneficial effects of this invention are: to solve the problems of small pilot scope, imperfect product system, undetermined pricing mechanism, and unscientific insurance premium rate determination in grassland carbon sequestration insurance. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0013] Figure 1 This is a flowchart of the grassland carbon sink storage index insurance premium rate determination method based on time series simulation of the present invention;

[0014] Figure 2 This is a schematic diagram showing the total carbon sequestration, trend value, and fluctuation value of four cities in the Inner Mongolia carbon sink area selected for this invention from 2001 to 2023.

[0015] Figure 3 This is a schematic diagram showing the total carbon sequestration, trend value, and fluctuation value of three cities in the Inner Mongolia carbon sink area selected for this invention from 2001 to 2023. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with specific embodiments and accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of protection.

[0017] Figure 1 This is a flowchart of the grassland carbon sequestration index insurance premium rate determination method based on time series simulation of the present invention. The present invention uses grassland NPP data from 2001 to 2023 in 12 leagues and cities of Inner Mongolia, employs methods such as HP filtering to calculate the carbon sequestration loss rate and fit its distribution, determines the pure premium rate, and then corrects it with multidimensional risk indicators, providing a basis for the design of relevant insurance pricing. Specific solutions include:

[0018] Step S1: Estimate grassland NPP based on the CASA model, and calculate grassland vegetation carbon sequestration based on grassland NPP.

[0019] Specifically, grassland NPP (net primary productivity) is the difference between grassland photosynthetic carbon sequestration and its autotrophic respiration. This invention uses Zhu Wenquan's improved CASA model to simulate vegetation growth and thus correct NPP data.

[0020] Further, the steps for calculating Net Primary Productivity (NPP) using the CASA model in Google Earth Engine (GEE) are as follows: ① Define the Area of ​​Interest (AOI): Select or draw the study area in GEE to define the Area of ​​Interest (AOI). ② Select and import the necessary datasets (including meteorological data, vegetation index data, and land cover data). For meteorological data, import temperature, precipitation, and solar radiation data from datasets such as MODIS or ERA5, which are key inputs to the CASA model. For vegetation index data, use the NDVI or EVI datasets from MODIS, which reflect the photosynthetic activity of vegetation. For land cover data, import the global land cover dataset to help distinguish different types of vegetation and land use. ③ Calculate Photosynthetically Active Radiation (PAR): Calculate or import Photosynthetically Active Radiation (PAR) data, based on solar radiation data and combined with the geographic information of the region. ④ Estimate Absorbed Photosynthetically Active Radiation (APAR): Calculate the absorbed photosynthetically active radiation (APAR) using DVI or EVI data combined with PAR data. This represents the radiation portion actually used by the plant for photosynthesis. ⑤ Calculate Light Use Efficiency (LUE): Set or calculate the light use efficiency (LE) for different vegetation types. This is usually obtained from literature or calculated based on adjustment factors for temperature and moisture conditions. ⑥ Calculate the corrected NPP using the CASA model. The formula for calculating NPP is shown below:

[0021] (1)

[0022] In the formula, NPP represents net primary productivity, and the unit is gC•m. -2 •a -1 APAR represents absorbed photosynthetically active radiation, measured in MJ·m⁻². -2 ε represents photosynthetic efficiency, measured in g C•MJ. -1 .

[0023] Furthermore, in step S1, the method for calculating grassland vegetation carbon sequestration based on grassland NPP is as follows: for every 1.00 kg of dry matter produced by vegetation, 1.63 kg of carbon dioxide can be absorbed, and 1 kg of carbon monoxide contains 0.27 kg of carbon. Considering the 2.2-fold conversion between the dry matter produced by vegetation photosynthesis and net primary productivity (NPP), the carbon sequestration capacity of vegetation is estimated using NPP data.

[0024] (2)

[0025] (3)

[0026] In the formula, Wco2 represents the amount of CO2 fixed per unit area of ​​vegetation during the study period (g·m²). –2 Wc represents the amount of carbon sequestration per unit area of ​​vegetation during the study period (g·m³). –2 N represents the vegetation per unit area NPP (g·m³) during the study period. –2 ).

[0027] The method for calculating grassland NEP is as follows:

[0028] (1) The soil microbial respiration rate (RH) was calculated using the relational model established by Pei Zhiyong et al. The formula is as follows:

[0029] (4)

[0030] In the formula, T represents the average monthly temperature (°C); P represents the monthly precipitation (mm);

[0031] (2) Calculate grassland NEP. Determine whether the ecosystem of the study area is a carbon source or a carbon sink. Grasslands in areas with NEP > 0 are carbon sink grasslands, and grasslands in areas with NEP < 0 are carbon source grasslands. Grassland net ecosystem productivity (NEP) is obtained by subtracting RH from NPP. The formula is:

[0032] (5)

[0033] In the formula, NEP(x,t) represents the net ecosystem productivity (gC·m³) of vegetation in pixel x in month t. -2 NPP(x,t) represents net primary productivity (g C·m). -2 ), R h (x, t) represents the vegetation of pixel x in month t, and the soil microbial respiration (g C·m³) of pixel x in month t. -2 ).

[0034] Step S2: Calculate the grassland vegetation carbon sequestration loss rate based on the grassland vegetation carbon sequestration obtained in Step S1, and separate the carbon sequestration fluctuation value.

[0035] For details, please refer to Figure 2 Due to natural conditions and human factors, the carbon sequestration per unit area of ​​grassland vegetation varies considerably. (Grassland vegetation carbon sequestration per unit area (Y)) aq It can be separated into the trend value of carbon sequestration per unit of grassland vegetation (Y). tq ), fluctuation value of carbon sequestration per unit of grassland vegetation (Y) wq ) and random values ​​of carbon sequestration per unit of grassland vegetation ( Carbon sequestration per unit area of ​​grassland vegetation (Y) aq ) is represented as:

[0036] (6)

[0037] In the formula, Y aq This refers to the annual carbon sequestration per unit area of ​​grassland vegetation in Inner Mongolia's carbon sink areas (leagues / cities) from 2001 to 2023 (gC·m³). -2 ·a -1 ), Y tq This represents the trend value of unit carbon sequestration by grassland vegetation in Inner Mongolia's carbon sink areas (leagues / cities) from 2001 to 2023 (gC·m³). -2 ·a -1 ), Y wq This refers to the fluctuation value of unit carbon sequestration by grassland vegetation in Inner Mongolia's carbon sink areas (leagues / cities) from 2001 to 2023 (gC·m³). -2 ·a -1 The impact of meteorological and other disaster factors on the fluctuation of unit carbon sequestration in grassland vegetation has both advantages and disadvantages, namely Y. wq Values ​​can be positive or negative. The random value of unit carbon sequestration by grassland vegetation in Inner Mongolia's carbon sink area leagues (cities) from 2001 to 2023 (gC·m³). -2 ·a -1 (This is generally negligible.)

[0038] To determine the carbon sequestration per unit of grassland vegetation (Y) aq The fluctuation value of carbon sequestration per unit of grassland vegetation (Y) was separated. wq First, the trend value of carbon sequestration per unit of grassland vegetation (Y) needs to be removed. tq Regarding the trend value Y of carbon sequestration per unit area of ​​grassland vegetation. tq The calculation also uses HP filtering, which avoids subjective errors in the moving step size. The time series data of grassland vegetation carbon sequestration per unit consists of high-frequency components (fluctuation values) and low-frequency components (trend values). Therefore, HP filtering is used to process the trend yield.

[0039] Y tq Defined as the solution to a problem that minimizes the following formula, which is expressed as follows:

[0040] (7)

[0041] In the formula, min() is the minimum value function, Y t(q+1) Y represents the observed rate of grassland carbon sequestration loss at time q+1. t(q-1) This represents the observed rate of grassland carbon sequestration loss at time q-1. Determine the smoothness of the trend line, set to 100.

[0042] The difference between the unit carbon sequestration of grassland vegetation and the trend value of unit carbon sequestration of grassland vegetation is used to obtain the fluctuation value Y of unit carbon sequestration of grassland vegetation. wq The formula is as follows:

[0043] (8)

[0044] Grassland vegetation carbon sequestration loss rate (YL) q The fluctuation value of carbon sequestration per unit of grassland vegetation (Y) wq ) as a percentage of trend output (Y) tq The percentage is calculated as follows:

[0045] (9)

[0046] In the formula, YL q The carbon sequestration loss rate (%) of each league (city) in the Inner Mongolia carbon sink area from 2001 to 2023 is given in year q. q A value less than 0 indicates a reduction in grassland vegetation carbon sequestration due to various factors. q The absolute value of represents the loss rate (%).

[0047] Select YL q Samples with values ​​less than 0 are used as samples of carbon sequestration loss in Inner Mongolian grassland vegetation, and their absolute values ​​are denoted as YLR. q The formula is as follows:

[0048] (10)

[0049] In the formula, YLR q The absolute value of the yield reduction rate (%) of grassland vegetation carbon sequestration loss samples in Inner Mongolia carbon sink areas (leagues / cities) from 2001 to 2023.

[0050] Since this invention uses the expected value loss method to determine the pure fee rate, based on the sum of the products of the carbon sequestration loss rate (YLRj) and the corresponding probability (Pj) of the carbon sequestration loss rate under different degrees of grassland, it is necessary to reasonably divide the range of loss rates. According to the main grassland types in Inner Mongolia, which are temperate meadow steppe, temperate typical steppe, and temperate desert steppe, and referring to relevant national forestry carbon sequestration methodologies, and drawing on the equal-interval method for classifying grassland vegetation carbon sequestration loss levels in Fujian Province's "Guidelines on Applying Forestry Carbon Sequestration Compensation Mechanisms to Ecological Restoration in Ecological and Environmental Criminal Cases," all grassland vegetation carbon sequestration loss rates are sorted in ascending order from smallest to largest, and then divided into different levels at equal intervals. Thus, grassland carbon sequestration loss levels are divided into 6 levels: no loss, slight loss, moderate loss, secondary severe loss, and severe loss. This invention obtains the grassland carbon sequestration loss rate YLR for different disaster levels. j When the value is greater than 0, it is considered that there is no loss of carbon sequestration. The intervals for grassland carbon sequestration loss rates are shown in Table 1:

[0051] Table 1. Division of grassland carbon sequestration loss rate ranges

[0052]

[0053] Step S3: Perform panel data stationarity tests on grassland vegetation carbon sequestration and grassland vegetation carbon sequestration loss rate.

[0054] Specifically, non-stationary data can easily lead to spurious regressions, where seemingly significant correlations between variables may actually be due to chance factors such as shared time trends, rendering them meaningless. Furthermore, stationarity is a crucial prerequisite for traditional econometric methods; non-stationary data can cause variance divergence in estimators and statistical inefficiencies, making it impossible to accurately determine the significance of coefficients. To ensure that the statistical analysis results accurately reflect the relationships between grassland vegetation carbon sequestration-related variables in the 12 leagues (cities) of Inner Mongolia and avoid drawing false conclusions, it is necessary to conduct a stationarity test on the panel data of annual grassland vegetation carbon sequestration and grassland vegetation carbon sequestration loss rate. Assuming the original panel data is not stationary, if the p-value for the stationarity test of grassland vegetation carbon sequestration and grassland vegetation carbon sequestration loss rate in the 12 leagues (cities) of Inner Mongolia from 2001 to 2023 is less than 0.01 (i.e., the test is significant), and the unit root test statistic (obtained by performing a t-test on the coefficient δ in the rewritten autoregressive model ΔYt=δYt−1+...) is less than the critical values ​​at different significance levels of 1%, 5%, and 10%, it indicates that the test rejects the null hypothesis, and the original panel data is determined to be stationary. The formula for the panel stationarity test is as follows.

[0055] (11)

[0056] In the formula, y qThe carbon sequestration per unit of grassland vegetation (g C·m -2 ) or the carbon sequestration loss rate per unit of grassland vegetation (%) in the 12 leagues (cities) of Inner Mongolia in the qth year from 2001 to 2023, is the intercept term, y q-1 is the lag term, is the "lagged difference term", that is, the first-order difference of the first-order to (p - 1)th order lag terms. is the coefficient, and its value is determined through time series data and statistical estimation methods. According to the statistical laws of the data, the coefficient value that makes the model have the best fitting effect on the data is found. The specific steps include data preprocessing, model order determination, coefficient estimation and test optimization. is the time trend (if there is no time trend, let = 0), is the disturbance term (independent white noise). By testing whether a key coefficient (δ) is significantly 0, it is indirectly inferred whether Y has a unit root (that is, non-stationary).

[0057] Step S4: Calculate the probability of the carbon sequestration loss rate, and determine the pure premium rate using the expected value method of the loss rate.

[0058] Specifically, use Easyfit software to screen the optimal fit of the probability distribution of the carbon sequestration loss rate of grassland vegetation in the carbon sink areas of Inner Mongolia from 2001 to 2023. Use the goodness-of-fit Anderson Darling (A - D) test to judge whether the sequence data comes from a certain distribution, and use the probability density function (PDF) and cumulative distribution function (CDF) formulas to calculate the probabilities in different drought index intervals, where etc. are all parameters. Assume that the cumulative probability density function of the carbon sequestration index is F(I), set certain index values I0, I1, I2 (I1 < I2), and the probability that the drought index is in the I1 - I2 interval is P(I1 < I < I2) = F(I2) – F(I1).

[0059] The probability density function (PDF) formula of the error distribution is as follows:

[0060] (12)

[0061] The error cumulative distribution function (CDF) formula is as follows:

[0062] (13)

[0063] The probability density function (PDF) formula of the Johnson SB distribution is as follows:

[0064] (14)

[0065] The Johnson SB cumulative distribution function (CDF) formula is as follows:

[0066] (15)

[0067] in, It is the Laplace integral;

[0068] The probability density function (PDF) of the Gen.Extreme Value distribution is as follows:

[0069] (16)

[0070] The formula for the Gen. Extreme Value cumulative distribution function (CDF) is as follows:

[0071] (17)

[0072] The probability density function (PDF) of the Dagum Distribution is as follows:

[0073] (18)

[0074] The cumulative distribution function (CDF) of the Dagum Distribution is as follows:

[0075] (19)

[0076] The probability density function (PDF) of the Gen. Pareto distribution is as follows:

[0077] (20)

[0078] The Gen. Pareto cumulative distribution function (CDF) formula is as follows:

[0079] (twenty one)

[0080] The probability density function (PDF) of the Laplace distribution is as follows:

[0081] (twenty two)

[0082] The Laplace cumulative distribution function (CDF) formula is as follows:

[0083] (twenty three)

[0084] Step S5: Divide the research area into risk zones and adjust the pure insurance premium rate according to the risk zones to obtain the actual insurance premium rate.

[0085] Specifically, risk zoning of the study area includes the selection of risk zoning indicators:

[0086] Due to the different grassland growth environments in various leagues and cities of Inner Mongolia, the carbon sequestration risk of grassland vegetation varies significantly. This invention selects carbon sequestration risk indicators, climate risk indicators, and soil risk indicators to categorize grassland carbon sink risks. The data in the indicator system are standardized, and then the similarity between pure rates is determined using a Euclidean distance model for connection and merging. Systematic clustering (Ward Method) analysis is then used to classify the data. When the Euclidean distance Ward value is set to 10, the seven leagues (cities) of the Inner Mongolia carbon sink area are divided into four categories according to different indicators: high-risk area, second-highest-risk area, medium-risk area, second-lowest-risk area, and low-risk area.

[0087] (1) Carbon sequestration risk: Systematic cluster analysis was used to classify the carbon sequestration risk of grassland vegetation in the carbon sink area into different levels. High carbon sequestration capacity of grassland vegetation also leads to an increased risk of carbon loss. The greater the unit carbon sequestration capacity of grassland vegetation, the higher the carbon sequestration risk. The relative value of the unit carbon sequestration capacity of grassland vegetation in the carbon sink area was used to assess the carbon sequestration risk of grassland vegetation, and the efficiency index was expressed as:

[0088] (twenty four)

[0089] In the formula, The efficiency index for carbon sequestration by grassland vegetation units in Inner Mongolia League (City). Carbon sequestration per unit area of ​​grassland vegetation in Inner Mongolia League (City) This represents the average carbon sequestration per unit area of ​​grassland vegetation in Inner Mongolia. >1, the carbon sequestration per unit area of ​​grassland vegetation in the i League City is higher than that in Inner Mongolia, and the carbon sequestration efficiency is higher.

[0090] Specialization Index: The larger the grassland area, the greater the likelihood of damage from natural disasters. Therefore, the specialization index is used to reflect the risk status of grassland vegetation carbon sequestration. The specialization index is expressed as:

[0091] (25)

[0092] In the formula, This is a specialization index. This refers to the grassland area of ​​city / league (county) in Inner Mongolia. This refers to the total area of ​​grasslands in Inner Mongolia.

[0093] Coefficient of variation for yield per unit area: The smaller the coefficient of variation for carbon sequestration per unit area of ​​grassland vegetation, the lower the risk of carbon sequestration and the more stable the carbon sequestration. The coefficient of variation is expressed as:

[0094] (26)

[0095] In the formula, The coefficient of variation of carbon sequestration yield per unit area of ​​grassland vegetation in League (City) of Inner Mongolia. The standard deviation of carbon sequestration yield per unit area of ​​grassland vegetation in Inner Mongolia (i League (City)). This represents the average carbon sequestration yield per unit area of ​​grassland vegetation in Inner Mongolia's i League (City).

[0096] Disaster loss rate: The probability of grassland vegetation carbon sequestration loss exceeding 10%, 20%, and 30% is used to reflect the disaster loss rate.

[0097] (2) Climate risk

[0098] Water stress: Water affects the normal growth of pasture grasses; the frequency of drought occurrence is selected as an indicator of water stress. The drought index is constructed using the percentage of precipitation anomaly (PA), calculated as follows:

[0099] (27)

[0100] In the formula, PA represents the percentage of precipitation anomalies (%) in the Inner Mongolia carbon sink region leagues (cities) from 2001 to 2023. The average annual precipitation (mm) in the Inner Mongolia carbon sink area leagues (cities) from 2001 to 2023. The annual precipitation (mm) in the leagues and cities of Inner Mongolia carbon sink area from 2001 to 2023 is q. N is taken as 23, and q is the study year, with values ​​of 1, 2, 3, ..., 23.

[0101] Due to the uneven spatial and temporal distribution of precipitation in Inner Mongolia, the risk of grassland drought is incorporated into the climate risk index. The drought index is calculated using the following formula.

[0102] (28)

[0103] In the formula, DI q The drought index (%) for the leagues and cities in the Inner Mongolia carbon sink area from 2001 to 2023.

[0104] (29)

[0105] In the formula, F represents the frequency (%) of drought occurrence in grasslands of Inner Mongolia's carbon sink areas (leagues / cities) from 2001 to 2023, which is the ratio of the number of drought-prone years to the total number of years (N). q A value of 0 or 1 indicates whether a drought has occurred at the weather station.

[0106] Light stress: The cumulative light intensity of each league (city) in Inner Mongolia was used to characterize the impact of light on carbon sequestration per unit of grassland vegetation. The lower the light intensity, the greater the risk of disaster for carbon sequestration per unit of grassland vegetation.

[0107] Temperature stress: Northeast Inner Mongolia is frequently under the influence of low-temperature zones. Low-temperature stress (<15℃) that often occurs in January, February, March, November, and December each year severely affects greening and reduces carbon sequestration per unit of grassland vegetation. Forage grasses thrive between 15 and 25 degrees Celsius. The number of days with unsuitable temperatures was used as a stress indicator. The number of days with daily maximum temperatures above 25℃ in each league and city from 2001 to 2023 was used as a heat damage indicator, and the number of days below 15℃ was used as a freezing damage indicator. The longer the total number of days with heat damage and freezing damage, the more significant the damage to carbon sequestration per unit of grassland vegetation, and the greater the risk to the grassland.

[0108] (3) Soil risk

[0109] Soil respiration significantly impacts carbon sequestration in grassland ecosystems, affecting both efficiency and quantity. This study uses heterotrophic respiration from soils in various leagues and cities of Inner Mongolia as a measure of the soil environment's impact on the carbon sequestration risk per unit of grassland vegetation. Higher soil respiration values ​​indicate greater risk and instability in carbon sequestration.

[0110] These indicators include both positive and negative indicators. The indicator system contains two negative indicators: the efficiency index and the temperature stress index. That is, scores on these two indicators are negatively correlated with the amount of carbon sequestration per unit of grassland vegetation. The remaining indicators are all positive indicators, with scores positively correlated with the amount of carbon sequestration per unit of grassland vegetation, and they are standardized. The specific calculation formulas are as follows: The standardization formulas for the positive and negative indicators are:

[0111] (30)

[0112] (31)

[0113] In the formula, Z ij X represents the standardized value. ij Let represent the variable value of the j-th secondary indicator under the i-th primary indicator, min{.} denotes the minimum function, and max{.} denotes the maximum function.

[0114] This study uses the expected value loss method to determine the net insurance premium rate. The net insurance premium rate is determined based on the sum of the products of the carbon sequestration loss rate and the probability of the corresponding carbon sequestration loss rate under different levels of grassland carbon sequestration loss. The calculation formula is as follows:

[0115] (32)

[0116] R represents the net insurance premium rate (%); E (loss) represents the average loss rate (%). To ensure the proportion is accurate, a value of 100% is generally used. The expected relative yield (gC·m) -2 ·a -1 ); P j The probability (%) corresponding to the carbon sequestration loss rate at different disaster levels is determined by different distribution models; YLR j Let represent the carbon sequestration loss rate of grassland under different disaster levels, where j represents different disaster levels, j=1,2,3,......

[0117] Since the actual insurance premium rate is related to the risk index of declining grassland vegetation carbon sequestration, and also involves safety factors, operating expenses, and expected surplus rates, the actual insurance premium rate is obtained by adjusting the pure insurance premium rate.

[0118] (33)

[0119] In the formula, The net premium rate (%) is adjusted for each league (city), R is the net insurance premium rate (%) for each league (city), D1 is the safety factor, with a value of 15%, D2 is the predetermined surplus rate, with a value of 5%, D3 is the operating expense, with a value of 20%, and D4 is the regional risk factor, with a value of 0.8 for low-risk areas, 0.9 for the second lowest risk areas, 1.0 for medium-risk areas, 1.1 for the second highest risk areas, and 1.2 for high-risk areas.

[0120] The grassland carbon sink storage index insurance premium rate determination method based on time series simulation of this invention can solve the problems of small pilot scope, imperfect product system, undetermined pricing mechanism and unscientific insurance premium rate determination in grassland carbon sink insurance.

[0121] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort, such as modifications to the technical solutions described in the following embodiments or equivalent substitutions of some technical features, are within the scope of protection of the present invention.

Claims

1. A method for determining grassland carbon sink storage index insurance premium rates based on time series simulation, characterized in that, Includes the following steps: Step S1: Estimate grassland NPP based on the CASA model, and calculate grassland vegetation carbon sequestration and grassland NEP based on grassland NPP; Step S2: Calculate the grassland vegetation carbon sequestration loss rate based on the grassland vegetation carbon sequestration obtained in Step S1, and separate the carbon sequestration fluctuation value. Step S3: Perform a panel data stationarity test on grassland vegetation carbon sequestration and grassland vegetation carbon sequestration loss rate. Step S4: Calculate the probability of carbon sequestration loss rate and determine the pure insurance premium rate using the expected value of the loss rate method; Step S5: Divide the research area into risk zones and adjust the pure insurance premium rate according to the risk zones to obtain the actual insurance premium rate.

2. The method for determining grassland carbon sink storage index insurance premium rates based on time series simulation according to claim 1, characterized in that, Step S1, which involves estimating grassland NPP based on the CASA model, includes: Map the research area in Google Earth Engine and define the region of interest; Import a dataset containing meteorological data, vegetation index data, and land cover data; Calculate photosynthetically active radiation based on meteorological data and regional geographic information; Based on the imported dataset and photosynthetically active radiation, estimate the absorbed photosynthetically active radiation. The grassland NPP is calculated using the CASA model. The grassland NPP calculation formula is as follows: In the formula, NPP represents net primary productivity, and the unit is gC•m. -2 •a -1 APAR represents absorbed photosynthetically active radiation, measured in MJ·m⁻². -2 ε represents photosynthetic efficiency, measured in g C•MJ. -1 .

3. The method for determining grassland carbon sink storage index insurance premium rates based on time series simulation according to claim 1, characterized in that, Step S1, which calculates grassland vegetation carbon sequestration and grassland NEP based on grassland NPP, includes: Calculate the carbon sequestration of grassland vegetation: In the formula, Wco2 represents the amount of CO2 fixed by vegetation per unit area during the study period; Wc represents the amount of carbon sequestrated by vegetation per unit area during the study period; and N represents the NPP of grassland per unit area during the study period. Calculate soil microbial respiration rate (RH): In the formula, T represents the average monthly temperature; P represents the monthly precipitation; Calculate grassland NEP: In the formula, NEP(x,t) represents the net ecosystem productivity of pixel x in month t, NPP(x,t) represents the net primary productivity, and R h( x, t) represents the soil microbial respiration of pixel x in month t.

4. The method for determining grassland carbon sink storage index insurance premium rates based on time series simulation according to claim 1, characterized in that, Step S2 includes: Step S21: Separate the fluctuation value of grassland vegetation carbon sequestration per unit from the grassland vegetation carbon sequestration per unit: In the formula, Y aq Y is the carbon sequestration per unit area of ​​grassland vegetation; wq Y is the fluctuation value of carbon sequestration per unit of grassland vegetation; tq The trend value of carbon sequestration per unit of grassland vegetation is defined as: In the formula, min() is the minimum value function, Y t(q+1) Y represents the observed rate of grassland carbon sequestration loss at time q+1. t(q-1) This represents the observed rate of grassland carbon sequestration loss at time q-1; Step S22: Calculate the grassland vegetation carbon sequestration loss rate based on the percentage of the fluctuation value of grassland vegetation unit carbon sequestration to the trend value of grassland vegetation unit carbon sequestration. In the formula, YL q The carbon sequestration loss rate of grassland vegetation.

5. The method for determining grassland carbon sink storage index insurance premium rates based on time series simulation according to claim 1, characterized in that, The formula for panel data stationarity testing of grassland vegetation carbon sequestration and grassland vegetation carbon sequestration loss rate in step S3 is as follows: In the formula, y q This represents the amount of carbon sequestration per unit of grassland vegetation in year q. g C·m -2 Or, the carbon sequestration loss rate per unit of grassland vegetation (%). For the intercept term, y q-1 It is a lagged term. This is a "lagging difference term".

6. The method for determining grassland carbon sink storage index insurance premium rates based on time series simulation according to claim 1, characterized in that, Step S4 includes: The optimal fit of the probability distribution of grassland vegetation carbon sequestration loss rate was screened using Easyfit software. The goodness-of-fit Anderson-Darling test was used to determine the probability distribution of grassland vegetation carbon sequestration loss rate, and the probability of grassland vegetation carbon sequestration loss rate was calculated based on different models. The pure insurance premium rate is determined by summing the products of the carbon sequestration loss rate and the probability of the corresponding carbon sequestration loss rate under different disaster levels in grasslands. The calculation formula is as follows: R represents the pure insurance premium rate; E (loss) represents the average loss rate; P represents the expected relative output. j The probabilities corresponding to the carbon sequestration loss rates at different disaster levels are determined by different distribution models; YLR j Let represent the carbon sequestration loss rate of grassland under different disaster levels, where j represents different disaster levels, j=1,2,3,......

7. The method for determining grassland carbon sink storage index insurance premium rates based on time series simulation according to claim 1, characterized in that, Step S5, risk delineation of the study area includes: Carbon sequestration risk indicators, climate risk indicators, and soil risk indicators were selected to classify the risks in the study area; Among them, carbon sequestration risk indicators include efficiency index, specialization index, yield variation coefficient, and disaster loss rate; climate risk indicators include water stress, light stress, and temperature stress; and soil risk indicators include soil respiration.

Images