Risk dynamic analysis method for vegetation phenological period in arid region

Abstract

The invention discloses a risk dynamic analysis method for a vegetation phenological period in an arid region, and the method comprises the steps: constructing a spring phenological model and an autumn phenological model in different partitions according to the partitions of different growth season initial periods SOS and growth season final periods EOS of plants in the arid region; respectively calculating vegetation growth risk coefficients of different growth season initial stage SOS partitions and growth season final stage EOS partitions; carrying out weight fusion on the vegetation growth risk coefficients to obtain a long-period vegetation growth risk comprehensive coefficient F in the arid region; calculating an ecological groundwater drought risk coefficient; calculating a vegetation risk dynamic coefficient of the arid region; and evaluating the growth risk of the vegetation in the arid region in the phenological period. According to the method, the drought risk influence coefficient is comprehensively calculated by utilizing representative feature data corresponding to different subareas and phenological areas, and comprehensive analysis of vegetation risk assessment in the arid area is realized by combining the ecological underground water drought risk coefficient on the annual scale.

Description

Technical Field

[0001] This invention relates to the field of ecological analysis, specifically to a method for dynamic risk analysis of vegetation phenology in arid regions. Background Technology

[0002] Currently, research on ecological drought indicators largely focuses on the fusion of different types of drought indicators. For example, atmospheric drought, agricultural drought, and hydrological drought are combined using the Copula model, Gringorten nonparametric joint distribution method, and non-steady-state models to construct a comprehensive drought index. These static index constructions neglect the three effects of drought on ecosystems (lag, accumulation, and legacy), failing to accurately characterize the dynamic response process of ecosystems. Furthermore, existing technologies rarely analyze the impact of drought during different phenological periods on ecological vegetation risk, resulting in a lack of comprehensiveness in research on drought's impact on vegetation ecological risk and inaccurate dynamic risk analysis. Summary of the Invention

[0003] To address the aforementioned shortcomings of existing technologies, this invention provides a method for dynamic risk analysis of vegetation phenology in arid regions.

[0004] To achieve the above-mentioned objectives, the technical solution adopted by this invention is as follows: A method for dynamic risk analysis of vegetation phenology in arid regions is provided, which includes the following steps: S1: Based on the partitioning of the start of the growing season (SOS) and the end of the growing season (EOS) of plants in arid areas, construct spring phenological models and autumn phenological models within different partitions; S2: Using the spring phenological model of the SOS zone at the beginning of the growing season and the autumn phenological model of the EOS zone at the end of the growing season, calculate the vegetation growth risk coefficient of the SOS zone at the beginning of the growing season and the EOS zone at the end of the growing season, respectively. S3: Weighting and fusing the vegetation growth risk coefficients to obtain a comprehensive long-term vegetation growth risk coefficient for arid regions. F ; S4: Construct a comprehensive driving force layer for the ecological effects of groundwater drought, including short-term driving force coefficients. The calculation model and long-term driving force coefficient The optimal calculation model is used for weight fusion to calculate the comprehensive driving force coefficient. The driving force coefficient As input to the ecological threshold Logistic-Sigmoid response, a hierarchical memory-threshold response ecological groundwater drought risk model is obtained, and the ecological groundwater drought risk coefficient is calculated. ; S5: Comprehensive Ecological Groundwater Drought Risk Coefficient Comprehensive coefficient of vegetation growth riskF Calculate the dynamic coefficient of vegetation risk in arid areas Based on the dynamic coefficient of vegetation risk in arid areas The size of the scale is used to assess the growth risk of vegetation phenology in arid regions.

[0005] Furthermore, the partitioning of the plant's SOS at the beginning of different growing seasons and EOS at the end of different growing seasons in step S1 includes: The SOS zone for the beginning of the growing season (31.5°N to 66°N), the EOS zone for the end of the growing season (0°N to 51.5°N), and the EOS zone for the end of the growing season (51.5°N to 66°N).

[0006] Furthermore, the method for constructing the spring phenological model for the SOS zone at the beginning of the growing season (31.5°N–66°N) is as follows: The SOS at the beginning of the growing season is divided into three stages: the dormancy induction stage and the growth recovery stage. The dormancy induction phase consists of short daytime cycles. and low temperature Triggered when the daytime cycle is short and low temperature Cumulative value of the product The hibernation induction process ends when the set threshold is reached; the accumulated value is then recorded. The computational model serves as a spring phenological model for the dormancy-inducing phase: ; in, t 0 represents the start time of the hibernation induction phase. t The duration of the dormancy induction phase. The short day cycle representing the start and end time of the dormancy induction phase. and low temperature Summing the products This is the temperature correlation coefficient. This represents the daily average temperature threshold for the dormancy induction phase. T This represents the average daily temperature during the dormancy induction phase. For the duration of sunshine, This represents the critical value for daylight duration during the dormancy induction phase. After dormancy induction is completed and dormancy is lifted, the vegetation growth recovery phase begins, representing parallel cooling and forced processes, respectively, with daily cooling rate... S C Defined as daily cooling volume R C The accumulation, and the daily forced rate R f Based on photoperiod and daily cooling rate S CThe decision is that increasing the photoperiod will reduce the daily forced rate. R f And daily cooling rate S C An increase in this will increase the daily forced rate. R f ; will accumulate forced rate S f The computational model serves as a spring phenological model for the growth and recovery phase; ; in, This marks the start of the growth and recovery phase. for, The correlation coefficient for cooling temperature. This represents the daily average temperature threshold during the growth and recovery phase. The average daily temperature during the growth and recovery phase. This is the critical value for the daily cooling rate. The correlation coefficient for daily cooling rate. This serves as a reference for daylight duration after half of the growth recovery phase has been completed. The correlation coefficient for sunshine duration. This serves as a reference for the average daily temperature after half of the growth and recovery phase has been completed. The correlation coefficient is the daily average temperature.

[0007] Furthermore, the method for constructing the autumn phenological model for the EOS region at the end of the growing season (0–51.5°N) is as follows: Increased temperature and precipitation at the end of the growing season lead to a delay in the end-of-growing-season EOS; hydrothermal interactions affect autumn phenological changes in a nonlinear manner. By establishing a nonlinear function to extend the original units to include temperature, photoperiod, and precipitation accumulation units, the autumn phenological model is as follows: ; in, This refers to the period of vegetation growth at the end of the growing season. The duration of the end of the growing season. This represents the vegetation status coefficient at the end of the growing season. This refers to the average daily reference temperature at the end of the growing season. This refers to the real-time temperature at the end of the growing season. Real-time photoperiod at the end of the growing season. This is the critical value for photoperiod at the end of the growing season. This represents the maximum rainfall at the end of the growing season. This represents real-time precipitation at the end of the growing season. This represents the critical value for the average daily temperature at the end of the growing season. These are the participation ratios of temperature, sunlight, and precipitation, respectively. This represents the cumulative value of the vegetation status coefficient at the end of the growing season.

[0008] Furthermore, the autumn phenological models for the EOS region at the end of the growing season in the range of 51.5–66°N include the GDD model and the DM model. The GDD model is specifically as follows: ; in, This represents the average daily temperature difference at the end of the growing season. The DM model is a cold-day autumn phenological model sensitive to photoperiod. It assumes that both temperature and shortened photoperiod can trigger the onset of autumn phenology, and leaf senescence rate... It is temperature-controlled and interacts with photoperiod; .

[0009] Furthermore, the vegetation growth risk factor includes: For the SOS zone at the beginning of the growing season (31.5°N–66°N), the vegetation growth risk coefficient during the dormancy induction phase. for , The cumulative value set Threshold, vegetation growth risk coefficient during the growth and recovery phase for , The cumulative forced rate threshold is set; For EOS zones at the end of the growing season (0–51.5°N and 51.5–66°N), the vegetation growth risk coefficient of EOS at the end of the growing season was calculated. for , The threshold value for the cumulative value of the vegetation status coefficient is set.

[0010] Furthermore, the hierarchical memory-threshold response ecological groundwater drought risk model is as follows: ; in, The ecological threshold Logistic response sensitivity coefficient This represents the annual ecological groundwater drought risk coefficient. This represents the propagation threshold from the atmospheric drought precipitation evapotranspiration index (SPEI) to the groundwater level index (SGI). These represent the combined driving force coefficients. 90% and 10%.

[0011] The beneficial effects of this invention are as follows: This invention divides the phenological zones of vegetation into zones, uses the characteristic data of the phenological zones of different zones to affect the vegetation growth status for growth status analysis, and uses the representative characteristic data corresponding to different zones and phenological zones to comprehensively calculate the drought risk impact coefficient, combined with the ecological groundwater drought risk coefficient on an annual scale, to achieve a comprehensive analysis of vegetation risk assessment in arid areas. Attached Figure Description

[0012] Picture 1 This is a flowchart of a method for dynamic risk analysis of vegetation phenology in arid regions. Detailed Implementation

[0013] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0014] like Picture 1 As shown, a method for dynamic risk analysis of vegetation phenology in arid regions includes the following steps: S1: Based on the partitioning of the start of the growing season (SOS) and the end of the growing season (EOS) of plants in arid areas, construct spring phenological models and autumn phenological models within different partitions; The partitioning of the plant's SOS at the beginning of different growing seasons and EOS at the end of different growing seasons in step S1 includes: The method for constructing the spring phenological model for the SOS zone at the beginning of the growing season (31.5°N–66°N) is as follows: The SOS at the beginning of the growing season is divided into three stages: the dormancy induction stage and the growth recovery stage. The dormancy induction phase consists of short daytime cycles. and low temperature Triggered when the daytime cycle is short and low temperature Cumulative value of the product The hibernation induction process ends when the set threshold is reached; the accumulated value is then recorded. The computational model serves as a spring phenological model for the dormancy-inducing phase: ; in, t 0 represents the start time of the hibernation induction phase. t The duration of the dormancy induction phase. The short day cycle representing the start and end time of the dormancy induction phase. and low temperature Summing the products This is the temperature correlation coefficient. This represents the daily average temperature threshold for the dormancy induction phase. T This represents the average daily temperature during the dormancy induction phase. For the duration of sunshine, This represents the critical value for daylight duration during the dormancy induction phase. After dormancy induction is completed and dormancy is lifted, the vegetation growth recovery phase begins, representing parallel cooling and forced processes, respectively, with daily cooling rate... S C Defined as daily cooling volume R C The accumulation, and the daily forced rate R f Based on photoperiod and daily cooling rate S C The decision is that increasing the photoperiod will reduce the daily forced rate. R f And daily cooling rate S C An increase in this will increase the daily forced rate. R f ; will accumulate forced rate S f The computational model serves as a spring phenological model for the growth and recovery phase; ; in, This marks the start of the growth and recovery phase. for, The correlation coefficient for cooling temperature. This represents the daily average temperature threshold during the growth and recovery phase. The average daily temperature during the growth and recovery phase. This is the critical value for the daily cooling rate. The correlation coefficient for daily cooling rate. This serves as a reference for daylight duration after half of the growth recovery phase has been completed. The correlation coefficient for sunshine duration. This serves as a reference for the average daily temperature after half of the growth and recovery phase has been completed. The correlation coefficient is the daily average temperature. The method for constructing the autumn phenological model for the EOS zone at the end of the growing season (0–51.5°N) is as follows: Increased temperature and precipitation at the end of the growing season lead to a delay in the end-of-growing-season EOS; hydrothermal interactions affect autumn phenological changes in a nonlinear manner. By establishing a nonlinear function to extend the original units to include temperature, photoperiod, and precipitation accumulation units, the autumn phenological model is as follows: ; in, This refers to the period of vegetation growth at the end of the growing season. The duration of the end of the growing season. This represents the vegetation status coefficient at the end of the growing season. This refers to the average daily reference temperature at the end of the growing season. This refers to the real-time temperature at the end of the growing season. Real-time photoperiod at the end of the growing season. This is the critical value for photoperiod at the end of the growing season. This represents the maximum rainfall at the end of the growing season. This represents real-time precipitation at the end of the growing season. This represents the critical value for the average daily temperature at the end of the growing season. These are the participation ratios of temperature, sunlight, and precipitation, respectively. This represents the cumulative value of the vegetation state coefficient at the end of the growing season. The autumn phenological models for the EOS zone at the end of the growing season, ranging from 51.5 to 66°N, include the GDD model and the DM model. The GDD model is specifically as follows: ; in, This represents the average daily temperature difference at the end of the growing season. The DM model is a cold-day autumn phenological model sensitive to photoperiod. It assumes that both temperature and shortened photoperiod can trigger the onset of autumn phenology, and leaf senescence rate... It is temperature-controlled and interacts with photoperiod; .

[0015] S2: Using the spring phenological model of the SOS zone at the beginning of the growing season and the autumn phenological model of the EOS zone at the end of the growing season, calculate the vegetation growth risk coefficient of the SOS zone at the beginning of the growing season and the EOS zone at the end of the growing season, respectively. For the SOS zone at the beginning of the growing season (31.5°N–66°N), the vegetation growth risk coefficient during the dormancy induction phase. for , The cumulative value set Threshold, vegetation growth risk coefficient during the growth and recovery phase for , The cumulative forced rate threshold is set; For EOS zones at the end of the growing season (0–51.5°N and 51.5–66°N), the vegetation growth risk coefficient of EOS at the end of the growing season was calculated. for , The threshold value for the cumulative value of the vegetation state coefficient is set; S3: Weighting and fusing the vegetation growth risk coefficients to obtain a comprehensive long-term vegetation growth risk coefficient for arid regions. F ; , These are the weights for assessing vegetation growth risk; S4: Construct a comprehensive driving force layer for the ecological effects of groundwater drought, including short-term driving force coefficients. The calculation model and long-term driving force coefficient The optimal calculation model is used for weight fusion to calculate the comprehensive driving force coefficient. The driving force coefficient As input to the ecological threshold Logistic-Sigmoid response, a hierarchical memory-threshold response ecological groundwater drought risk model is obtained, and the ecological groundwater drought risk coefficient is calculated. ; The hierarchical memory-threshold response ecological groundwater drought risk model is as follows: ; in, The ecological threshold Logistic response sensitivity coefficient This represents the annual ecological groundwater drought risk coefficient. This represents the propagation threshold from the atmospheric drought precipitation evapotranspiration index (SPEI) to the groundwater level index (SGI). These represent the combined driving force coefficients. 90% and 10%.

[0016] The method for constructing the hierarchical memory-threshold response ecological groundwater drought risk model in this embodiment is as follows: The lag effect of groundwater drought on vegetation and cumulative effect The short-term driving force coefficient is obtained. The computational model.

[0017] ; in, t This embodiment uses the monthly sliding window as the unit for short-term driving research. t Take the 12 months of the current year, with the monthly sliding window unit being 1 month. t ), i For a time-lag index, it shifts backwards from the current month to each previous historical month (e.g., i =0 represents the current month ( t ), i =1 represents last month ( t− 1), and so on;n The sliding window size is for the past 12 months (in this example, it represents the past 12 months). The normalized vegetation index outlier for the current month Normalized Value Added (NWV) average deviation, The average normalized vegetation index for a given month across multiple years, representing long-term historical data or a sliding window average across different months. NDVI Values ​​(for example, you can calculate each month over the past 10 or 20 years) NDVI The mean, for each month from 2003 to 2022 in this study. NDVI (mean) The impact of groundwater drought on the normalized vegetation index NDVI The hysteresis coefficient of the influence ranges from 0 to 1. This refers to the groundwater drought index for last month. The impact of groundwater drought on the normalized vegetation index NDVI The cumulative intensity coefficient of the influence ranges from 0 to 1. For the first time in history i Monthly groundwater drought index i =0-11 months, These are the short-term driving force weights of the lag effect and the short-term driving force weights of the cumulative effect, respectively. At that time, groundwater drought NDVI The impact has not diminished; it is continuous. The impact will diminish rapidly, with only the current month's groundwater drought affecting the area. NDVI The impact at other times is almost zero; The impact will gradually diminish.

[0018] Fusion lag effect and cumulative effect This represents the ecosystem's rapid response to environmental stresses during the current growing season.

[0019] Short-term driving force weights of lag effects The calculation method is as follows: ; Short-term driving force weights of cumulative effects The calculation method is as follows: ; Hysteresis strength coefficient The calculation method is as follows: Obtain the driving factors for each month within the monthly sliding window. The driving factors include the precipitation evapotranspiration index. SPEI Soil moisture index SSMI Runoff drought index SRIGroundwater drought index (SGI) and flood index FE Using driving factors as variables x ; Calculate the lag within the monthly sliding window. i Normalized Difference Vegetation Index for one month With variables x Pearson correlation coefficient Filter within the monthly sliding window n Pearson correlation coefficient corresponding to each lag month The maximum value is used as the hysteresis strength coefficient. ; ; in, For the lagging first i Variables over a month; Cumulative strength coefficient The calculation method is as follows: Get the variables within each lag month in the monthly sliding window x extremum Calculate the extreme values With Normalized Difference Vegetation Index Pearson correlation coefficient Filter within the monthly sliding window n Pearson correlation coefficient corresponding to each lag month The maximum value is used as the cumulative intensity coefficient. ; .

[0020] Obtain the groundwater drought index on a historical annual scale , y For historical years, construct long-term driving force coefficients The computational model; ; in, The residual effect memory decay coefficient; The larger the value, the better the system's understanding of past states. The stronger the memory, the better for the present. The weaker the state response, the stronger the memory and the weaker the response, mainly represented by forest ecosystems; The smaller the value, the better the system's response to the current situation. The more sensitive the response to the state, the shorter the "memory" of history, exhibiting weak memory and strong response, mainly represented by grassland ecosystems.

[0021] Based on the set legacy effect memory decay coefficient The range of values Within the range of values, select several residual effect memory decay coefficients, with a step size of 0.05. Input these residual effect memory decay coefficients into the long-term driving force coefficient. In the computational model, the predicted values ​​of the long-term driving force coefficient are calculated under different values ​​of the legacy effect memory decay coefficient. , m This is the numbering of the memory decay coefficient for the legacy effect; Calculated M Predicted values ​​of long-term driving force coefficients Mean square error ,get M Predicted values Corresponding mean square error ; ; in, This represents the true value of the long-term driving force coefficient on a historical year scale. filter M Predicted values Corresponding mean square error minimum value , minimum value The corresponding residual effect memory decay coefficient is taken as the optimal value. ; to the optimal value Input long-term driving force coefficient In the calculation model, the long-term driving force coefficient is obtained. The optimal computational model; .

[0022] short-term driving force coefficient The calculation model and long-term driving force coefficient The optimal computational model is used for weight fusion; ; in, These are the weights for short-term and long-term driving forces, respectively. This is the residual strength coefficient, with a value ranging from 0 to 1. This is the comprehensive driving force coefficient.

[0023] Residual strength coefficient The calculation method is as follows: Obtain the extreme values ​​of the driving factors for each year within the annual sliding window as variables. , z Calculate the variable for the year number within the year sliding window. Spring phenological indicators for each year within the annual sliding window SOS Pearson correlation coefficient between The Pearson correlation coefficient for each year within the annual sliding window. The maximum value is used as the residual strength coefficient. ; .

[0024] Comprehensive driving force coefficient As input to the ecological threshold Logistic-Sigmoid response, a hierarchical memory-threshold response ecological groundwater drought risk model is constructed. ; in, The ecological threshold Logistic response sensitivity coefficient This represents the annual ecological groundwater drought risk coefficient. This represents the propagation threshold from the atmospheric drought precipitation evapotranspiration index (SPEI) to the groundwater level index (SGI). These represent the combined driving force coefficients. 90% and 10%.

[0025] S5: Comprehensive Ecological Groundwater Drought Risk Coefficient Comprehensive coefficient of vegetation growth risk F Calculate the dynamic coefficient of vegetation risk in arid areas Based on the dynamic coefficient of vegetation risk in arid areas The size of the risk coefficient is used to assess the growth risk of vegetation during phenological periods in arid regions. (Arid region vegetation risk dynamic coefficient) The larger the value, the greater the risk of vegetation phenological growth in arid regions, and vice versa.

[0026] This invention divides vegetation into phenological zones, analyzes the growth status by using characteristic data of the phenological zones in different zones that affect vegetation growth, and calculates the drought risk impact coefficient by using representative characteristic data corresponding to different zones and phenological zones. Combined with the annual ecological groundwater drought risk coefficient, this invention achieves a comprehensive analysis of vegetation risk assessment in arid areas.

Claims

1. A method for dynamic risk analysis of vegetation phenology in arid regions, characterized in that, Includes the following steps: S1: Based on the partitioning of the start of the growing season (SOS) and the end of the growing season (EOS) of plants in arid areas, construct spring phenological models and autumn phenological models within different partitions; S2: Using the spring phenological model of the SOS zone at the beginning of the growing season and the autumn phenological model of the EOS zone at the end of the growing season, calculate the vegetation growth risk coefficient of the SOS zone at the beginning of the growing season and the EOS zone at the end of the growing season, respectively. S3: Weighting and fusing the vegetation growth risk coefficients to obtain a comprehensive long-term vegetation growth risk coefficient for arid regions. F ; S4: Construct a comprehensive driving force layer for the ecological effects of groundwater drought, including short-term driving force coefficients. The calculation model and long-term driving force coefficient The optimal calculation model is used for weight fusion to calculate the comprehensive driving force coefficient. The driving force coefficient As input to the ecological threshold Logistic-Sigmoid response, a hierarchical memory-threshold response ecological groundwater drought risk model is obtained, and the ecological groundwater drought risk coefficient is calculated. ; S5: Comprehensive Ecological Groundwater Drought Risk Coefficient Comprehensive coefficient of vegetation growth risk F Calculate the dynamic coefficient of vegetation risk in arid areas Based on the dynamic coefficient of vegetation risk in arid areas The size of the scale is used to assess the growth risk of vegetation phenology in arid regions.

2. The method for dynamic risk analysis of vegetation phenology in arid regions according to claim 1, characterized in that, The partitioning of the plant's SOS at the beginning of the growing season and EOS at the end of the growing season in step S1 includes: The SOS zone for the beginning of the growing season (31.5°N to 66°N), the EOS zone for the end of the growing season (0°N to 51.5°N), and the EOS zone for the end of the growing season (51.5°N to 66°N).

3. The method for dynamic risk analysis of vegetation phenology in arid regions according to claim 2, characterized in that, The method for constructing the spring phenological model for the SOS zone at the beginning of the growing season in the range of 31.5°N to 66°N is as follows: The SOS at the beginning of the growing season is divided into three stages: the dormancy induction stage and the growth recovery stage. The dormancy induction phase consists of short daytime cycles. and low temperature Triggered when the daytime cycle is short and low temperature Cumulative value of the product The hibernation induction process ends when the set threshold is reached; the accumulated value is then recorded. The computational model serves as a spring phenological model for the dormancy-inducing phase: ； in, t 0 represents the start time of the hibernation induction phase. t The duration of the dormancy induction phase. The short day cycle representing the start and end time of the dormancy induction phase. and low temperature Summing the products This is the temperature correlation coefficient. This represents the daily average temperature threshold for the dormancy induction phase. T This represents the average daily temperature during the dormancy induction phase. For the duration of sunshine, This represents the critical value for daylight duration during the dormancy induction phase. After dormancy induction is completed and dormancy is lifted, the vegetation growth recovery phase begins, representing parallel cooling and forced processes, respectively, with daily cooling rate... S C Defined as daily cooling volume R C The accumulation, and the daily forced rate R f Based on photoperiod and daily cooling rate S C The decision is that increasing the photoperiod will reduce the daily forced rate. R f And daily cooling rate S C An increase in this will increase the daily forced rate. R f ; will accumulate forced rate S f The computational model serves as a spring phenological model for the growth and recovery phase; ； in, This marks the start of the growth and recovery phase. for, The correlation coefficient for cooling temperature. This represents the daily average temperature threshold during the growth and recovery phase. The average daily temperature during the growth and recovery phase. This is the critical value for the daily cooling rate. The correlation coefficient for daily cooling rate. This serves as a reference for daylight duration after half of the growth recovery phase has been completed. The correlation coefficient for sunshine duration. This serves as a reference for the average daily temperature after half of the growth and recovery phase has been completed. The correlation coefficient is the daily average temperature.

4. The method for dynamic risk analysis of vegetation phenology in arid regions according to claim 3, characterized in that, The method for constructing the autumn phenological model for the EOS region at the end of the growing season (0–51.5°N) is as follows: Increased temperature and precipitation at the end of the growing season lead to a delay in the end-of-growing-season EOS; hydrothermal interactions affect autumn phenological changes in a nonlinear manner. By establishing a nonlinear function to extend the original units to include temperature, photoperiod, and precipitation accumulation units, the autumn phenological model is as follows: ； in, This refers to the period of vegetation growth at the end of the growing season. The duration of the end of the growing season. This represents the vegetation status coefficient at the end of the growing season. This refers to the average daily reference temperature at the end of the growing season. This refers to the real-time temperature at the end of the growing season. Real-time photoperiod at the end of the growing season. This is the critical value for photoperiod at the end of the growing season. This represents the maximum rainfall at the end of the growing season. This represents real-time precipitation at the end of the growing season. This represents the critical value for the average daily temperature at the end of the growing season. These are the participation ratios of temperature, sunlight, and precipitation, respectively. This represents the cumulative value of the vegetation status coefficient at the end of the growing season.

5. The method for dynamic risk analysis of vegetation phenology in arid regions according to claim 4, characterized in that, The autumn phenological models for the EOS zone at the end of the growing season in the range of 51.5–66°N include the GDD model and the DM model. The GDD model is specifically as follows: ； in, This represents the average daily temperature difference at the end of the growing season. The DM model is a cold-day autumn phenological model sensitive to photoperiod. It assumes that both temperature and shortened photoperiod can trigger the onset of autumn phenology, and leaf senescence rate... It is temperature-controlled and interacts with photoperiod; 。 6. The method for dynamic risk analysis of vegetation phenology in arid regions according to claim 5, characterized in that, The vegetation growth risk factor includes: For the SOS zone at the beginning of the growing season (31.5°N–66°N), the vegetation growth risk coefficient during the dormancy induction phase. for , The cumulative value set Threshold, vegetation growth risk coefficient during the growth and recovery phase for , The cumulative forced rate threshold is set; For EOS zones at the end of the growing season (0–51.5°N and 51.5–66°N), the vegetation growth risk coefficient of EOS at the end of the growing season was calculated. for , The threshold value for the cumulative value of the vegetation status coefficient is set.

7. The method for dynamic risk analysis of vegetation phenology in arid regions according to claim 6, characterized in that, The hierarchical memory-threshold response ecological groundwater drought risk model is as follows: ； in, The ecological threshold Logistic response sensitivity coefficient This represents the annual ecological groundwater drought risk coefficient. This represents the propagation threshold from the atmospheric drought precipitation evapotranspiration index (SPEI) to the groundwater level index (SGI). These represent the combined driving force coefficients. 90% and 10%.

Images