Quantization method of drought and flood remain effect tradeoff and synergistic effect

Abstract

The invention discloses a drought and flood remain effect tradeoff and synergistic effect quantification method, and belongs to the technical field of vegetation phenology in ecology, and the method comprises the steps: carrying out the spatial partitioning of spring and autumn phenology; constructing a spring and autumn phenology model by adopting a linear mixed effect model; and on the basis of a spring and autumn phenological model, constructing a leaving effect phenological model under different space partitions. According to the method, the core problem of limited explanatory force of a single factor and a single effect in the prior art is comprehensively solved through a refined space partition, a multi-factor integration model and a partition differentiation legacy effect model, and accurate quantification of a drought and flood legacy effect tradeoff and a synergistic effect is realized; and a more systematic and efficient method support is provided for phenological research in the arid region and scientific application in related ecological and climate fields.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of vegetation phenology in ecology, and particularly relates to a quantification method for trade-off and synergistic effect of drought and flood residual effect. BACKGROUND

[0002] Under the background of global climate change acceleration, the traditional cognition believes that the intensification of drought will generally lead to the shortening of vegetation phenology in the northern hemisphere. However, the paradox of "drought intensification-phenology prolongation" challenges this view. Through nearly 40 years of two kinds of remote sensing vegetation phenology data and 319 ground phenology records, two different phenology zones are identified: the phenology synergistic zone (0-51.5 °N) and the phenology trade-off zone (51.5-66 °N), and the dominant role of hydrological factors in the change of spring phenology (SOS) and autumn phenology (EOS) is clarified (SOS accounts for 44.3%, EOS accounts for 29.8%).

[0003] Lag, accumulation and residual effect strengthen the influence of disturbance factors on vegetation phenology, and further affect the structure and function of the ecosystem and the flow of energy and carbon. Under the background of global warming, climate and hydrological processes tend to be frequent, rapid and extreme, and the phenology change under the action of the three effects is more significant, such as the significant advance of plant greening (SOS) and the significant delay of autumn phenology (EOS) in the northern hemisphere. However, considering the complexity of the response of the ecosystem, the explanatory power of a single disturbance factor and a single effect on the change of the vegetation phenology in the arid region is limited. For example, the lag effect of precipitation can only explain the change of vegetation productivity in humid areas, and the drought residual emphasizes the influence of past drought events on the ecosystem. Or a single type of driving factor explains the global vegetation productivity, for example, only considering climate factors to explain the asymmetry of vegetation response, and this single driving or single effect explanation ignores the complex interactions and multiple feedback mechanisms within the arid region ecosystem. SUMMARY

[0004] The purpose of the present application is to provide a quantification method for trade-off and synergistic effect of drought and flood residual effect to solve the problem of limited explanatory power of a single disturbance factor and a single effect on the change of the vegetation phenology in the arid region.

[0005] To achieve the above purpose, the technical solution adopted by the present application is: A quantification method for trade-off and synergistic effect of drought and flood residual effect, comprising the following steps: S1, spatially partitioning spring and autumn phenology; S2, constructing a spring and autumn phenology model using a linear mixed effect model; S3, constructing a residual effect phenology model under different spatial partitions based on the spring-autumn phenology model.

[0006] Further, in S1, the spring-autumn phenology is spatially partitioned, and the spatial partitioning is represented as: The spring phenology SOS I region, corresponding to the dimension of 0-31.5 °N; The spring phenology SOS II region, corresponding to the dimension of 31.5 °N-66 °N; The autumn phenology EOS I region and II region, corresponding to the dimension of 0-51.5 °N; The autumn phenology EOS III region, corresponding to the dimension of 51.5-66 °N.

[0007] Further, in S2, a linear mixed effect model is used to construct the spring-autumn phenology model, and the spring phenology model in the spring-autumn phenology model is represented as:

[0008] In the formula, SOS represents the spring phenology; is the coefficient of the fixed effect; represents the lagged term of the drought index; represents the scaled drought index; represents the scaled flood index; represents the lagged term of the flood index; is the random intercept of the drought index and the flood index; is the random intercept of the year; is the residual.

[0009] Further, in S3, the residual effect phenology model under different spatial partitions includes: The spring phenology SOS I region, the residual effect spring phenology model with the dimension of 0-31.5 °N; The spring phenology SOS II region, the residual effect spring phenology model with the dimension of 31.5 °N-66 °N; The autumn phenology EOS I region and II region, the residual effect autumn phenology model with the dimension of 0-51.5 °N; The autumn phenology EOS III region, the residual effect autumn phenology model with the dimension of 51.5-66 °N.

[0010] Further, the residual effect spring phenology model of the spring phenology SOS I region with the dimension of 0-31.5 °N is represented as:

[0011] In the formula, represents the difference between the observed value and the predicted value of the spring phenology SOS. This represents the observed value of SOS (Self-Occurring Occurrence) in spring phenology; This represents the predicted value of spring phenology SOS, calculated using the spring phenology model in S2.

[0012] Furthermore, the residual effect spring phenological model for the SOS II region (latitude 31.5°N–66°N) includes three stages: Dormancy induction phase:

[0013] in:

[0014]

[0015] In the formula, This is a state induced by hibernation; Indicates the start date of hibernation induction. Indicates the current calculation time point; Indicates the short photoperiod driving factor; Indicates the low-temperature driving factor; T This is the average daily temperature. DL yes t Daytime sunshine duration; aD , bD and DL crit Parameters for adjusting photoperiod and temperature effects; Hibernation end phase:

[0016] in:

[0017]

[0018] In the formula, Indicates the daily cooling rate; Indicates daily cooling volume; This indicates the moment when dormancy is broken and growth resumes at the beginning of the process after dormancy induction is complete; Shape parameters representing the temperature response during the cooling process; Indicates the optimal temperature for the cooling process; Indicates the critical day length; The slope representing the effect of cooling on photoperiod sensitivity; The critical threshold indicating the cooling state; Indicates the critical temperature; The slope representing the effect of day length on temperature demand; Indicates the current actual length of day; Growth recovery:

[0019] where, represents the accumulation of daily forced rate; represents the daily forced rate; represents the temperature response slope of the forced process.

[0020] Further, the autumnal phenology EOS I area and II area, the dimension of 0~51.5 °N of the residual effect autumnal phenology model, which is expressed as:

[0021]

[0022] where, represents the daily DMP value; represents the temperature upper limit reference value; represents the daily average temperature; represents the temperature response index; represents the photoperiod of the t day; represents the critical photoperiod; represents the starting date the maximum precipitation when iteration starts; represents the daily precipitation; is the critical temperature, which represents the temperature sensitivity threshold value; represents the phenological accumulation threshold value; represents the DMP value of the date x, y and k are used to reflect the participation proportion of temperature, illumination and precipitation respectively.

[0023] Further, the autumnal phenology EOS III area, the dimension of 51.5~66 °N of the residual effect autumnal phenology model, which is expressed as:

[0024]

[0025] where, represents the daily effective low temperature intensity; represents the reference temperature; represents the effective low temperature intensity of the date .

[0026] Further, the autumnal phenology EOS III area, the dimension of 51.5~66 °N of the residual effect autumnal phenology model, which is expressed as:

[0027]

[0028] wherein, denotes the state and rate of leaf discoloration on the tth day; denotes the state and rate of leaf discoloration on the dth day; denotes the photoperiod length.

[0029] The quantitative method for trade-off and synergistic effect of drought and flood residual effects provided by the application has the following beneficial effects: 1. In view of the limitation of the prior art that the phenology zoning is coarse (only synergistic zone and trade-off zone are divided), the present application finely zones the spring and autumn phenology according to latitude (for example, SOS is divided into 0-31.5°N and 31.5-66°N zones, and EOS is divided into 0-51.5°N and 51.5-66°N zones). The phenology feature differences of different latitude regions can be accurately located, the analysis deviation caused by ignoring the region specificity in the traditional zoning is avoided, and an accurate spatial framework is provided for subsequent targeted quantification of residual effects.

[0030] 2. The analysis limitation of single interference factor or single effect is broken through, a linear mixed effect model is used to construct a spring and autumn phenology model, the model synchronously includes the drought and flood indexes and their lag terms and interaction terms, and also considers the random intercepts of pixels and years, can comprehensively capture the comprehensive influence of multiple factors and lag and cumulative effects on the phenology, significantly improves the explanation of the phenology change in the drought area, and solves the problem that the traditional single driving analysis ignores the complex interaction.

[0031] 3. Based on the differences in the zoning phenology driving mechanism, a differentiated residual effect phenology model is constructed (for example, SOSⅡ divides the three-stage model of dormancy induction, and EOS uses the DMP and GDD models in different zones). The differences in the phenology driving of different regions can be adapted, the trade-off and synergistic effect of the drought and flood residual effects can be accurately quantified, and the problems of poor applicability and low quantification precision of the traditional single model are solved. 4. The quantitative method of the present application can provide a scientific tool for predicting the spring and autumn phenology change trend in the drought area under the background of global warming, evaluating the influence of the phenology change on the structure and function of the ecological system and the carbon and energy flow. The quantitative method can provide data support for formulating the ecological protection and climate adaptation strategies in the drought area, and fills the blank of the practical application of the phenology quantification research under the synergistic effect of multiple effects. BRIEF DESCRIPTION OF DRAWINGS

[0032] Figure 1 The flowchart of the quantitative method for trade-off and synergistic effect of drought and flood residual effects of the present embodiment. DETAILED DESCRIPTION

[0033] The specific embodiments of the present application are described below to facilitate the understanding of those skilled in the art of the present application, but it should be clear that the present application is not limited to the scope of the specific embodiments, and for those skilled in the art, all the inventions utilizing the concept of the present application are included in the protection scope as long as the various changes are obvious within the spirit and scope of the present application defined and determined by the appended claims.

[0034] The quantitative method of the trade-off and synergistic effect of drought and flood residual effects in this embodiment selects two phenology data sets that cover the entire arid zone of the Northern Hemisphere. The Global Drought Zone Phenology Dataset (GDPD) from 2001 to 2019 (https: / / doi.org / 10.6084 / m9.figshare.27160602.v2) is based on a new method of high-resolution remote sensing phenology monitoring based on dynamic thresholds, which achieves 88.4% of global arid zone phenology monitoring, 2.2 times higher than MCD12Q2 (39.5%), and accurately captures the spatiotemporal changes of arid zone phenology. The Northern Hemisphere Vegetation Key Phenology Dataset (1982-2022) published by Wu Zhaoyang's team combines high-temporal and high-spatial resolution remote sensing data and ground observations to accurately capture the seasonal changes of vegetation. This dataset includes the Northern Hemisphere annual key phenology dataset (SOS, POS, EOS) extracted from GIMMS NDVI 3G+ data. The data resolution is 8km. The GDPD dataset more suitable for arid areas is used to calibrate the data of Wu Zhaoyang's team through linear regression, and the data is resampled to 0.5° resolution, thus obtaining the spring and autumn phenology (such as SOS, EOS) results of the Northern Hemisphere arid zone from 1982 to 2022. Then, by linear regression on the time series, the slope is calculated to evaluate the trend of spring and autumn phenology.

[0035] Among them, the variables processed include: Climate and hydrological variables; Historical climate data for phenological impact process indicators and phenological extreme event identification analysis were from the ERA5-Land Hourly-ECMWF Climate Reanalysis dataset (https: / / cds.climate.copernicus.eu). The dataset provides global atmospheric reanalysis data from 1 January 1950 to 21 July 2024 with a 3h temporal resolution and a 0.1° spatial resolution. The 3h air temperature T, precipitation P, 4 profiles (0~7cm, 7~28cm, 28~100cm, 100~289cm) soil moisture SM, snow depth SMD, snow water equivalent SW, potential evapotranspiration ETp, vegetation transpiration Tv, soil evaporation Es, water body and ice snow sublimation Eo, canopy interception loss Ei, net radiation Rn, 10 m wind speed, open water and snow evaporation Eo, and runoff data RF (including surface runoff, groundwater runoff, and snowmelt runoff) for the gridded flood event maximum peak threshold method (Isaacson et al., 2024) identification in the arid region of the Northern Hemisphere from 1982 to 2022 were downloaded. The GLDAS-2.2 dataset (https: / / disc.gsfc.nasa.gov / datasets / GLDAS_CLSM025_DA1_D_2.2) provides spatiotemporal variation information of global groundwater storage (GWS) with a 0.25°×0.25° resolution every 3 hours since 2003. Both climate and hydrological variables were resampled to 0.5° to match the spatial resolution of the vegetation phenology data.

[0036] Human activity CO2 emission data; Human activity CO2 emission data The EDGAR v8.0 dataset (https: / / edgar.jrc.ec.europa.eu / dataset_ghg80) records global greenhouse gas emissions and air pollution from human activities from 1970 to 2022, including the power industry, industry, transportation, agriculture, fuel mining, etc. EDGAR provides total national emissions and grid maps with a 0.1° resolution globally, including annual, monthly, and hourly data. This data is mainly used for human activity impact-driven attribution, which is fundamentally different from atmospheric CO2 concentration. Although atmospheric CO2 concentration directly affects vegetation through the “fertilization effect”, this cannot quantify the impact of human activities, so CO2 emission data related to human activities are selected as impact factors and resampled to 0.5°.

[0037] Land use data MODIS IGBP land cover dataset (https: / / lpdaac.usgs.gov / products / mcd12q1v006 / ) was used to quantify the impact of land use change (LCC) on phenology, excluding cropland, cropland / natural vegetation mosaic, permanent snow and ice, barren land, and water body regions. Vegetation cover areas were classified into evergreen needleleaf forest (ENF), evergreen broadleaf forest (EBF), deciduous needleleaf forest (DNF), deciduous broadleaf forest (DBF), mixed forest (MF), grassland (GRA), shrub (SHR), cropland (CRO), wetland (WET), and savanna (SAV), which were consistent with the vegetation type classification of FLUXNET2015 eddy correlation (EC) flux sites. The time series of vegetation type change data (LCC) related to human activities were calculated using the difference method and resampled to 0.5° to match the spatial resolution of the vegetation phenology data.

[0038] In order to illustrate the complex synergistic effect of many influencing factors on vegetation phenology in arid regions, remote sensing data was used. Firstly, all data sets were detrended to exclude the influence of long-term trends. Then, time series Pearson partial correlation analysis was used to quantify the impact of climate, hydrology, human activity, and vegetation elements on vegetation phenology change in the northern hemisphere arid region. The partial correlation coefficients between 30 process indicators and 26 extreme event indicators in the time series from 1982 to 2022 and spring and autumn phenology were calculated pixel by pixel, and the factor with the largest absolute correlation value was selected as the dominant phenology influencing factor.

[0039] Specifically, process indicators: 4 types of 29 process indicators were selected for consideration, including 12 climate indicators (CLI), 8 hydrology indicators (HYE), 7 vegetation indicators (VEG), and 3 human activity indicators (HUA), to consider the influence of process indicators on spring and autumn phenology (among them, precipitation P is both a climate indicator and a hydrology indicator, but this does not affect the subsequent research results).

[0040] Among them, the climate index (CLI) includes potential evapotranspiration (ETp), net radiation (Rn), precipitation (P), saturated vapor pressure difference (VPD), average air temperature (Ta), daytime air temperature (Tday), nighttime air temperature (Tnight), 10 m wind speed (U), and cumulative temperature (CT). The hydrological index (HYE) includes snow depth (SD), snowmelt equivalent (SM), precipitation (P), ice and snow sublimation (Eo), soil evaporation (Es), 0-289 cm soil moisture (SW), groundwater storage (GWS), and runoff (RF). The vegetation index (VEG) includes canopy interception loss (Ei), vegetation transpiration (Tv), water use efficiency, potential water use efficiency, evapotranspiration water use efficiency, and total primary productivity (GPP). The human activity index (HUA) includes human activity CO2 emissions (CO2) and land cover change (LCC).

[0041] Among them, the contribution of the vegetation index (VEG) to the spring phenology (SOS) is the residual effect of the last growing season on the spring phenology of the current year.

[0042] Extreme event index: This study selected 6 climate extreme indices (CLI) including: heat wave (HW), humid heat wave (HHW), cold wave (CW), extreme temperature sudden change (RTF) (including: extreme heat→extreme cold, extreme cold→extreme heat), heavy rain (PO) and atmospheric drought (SPEI). Two hydrological extreme indices (HYE) include: flood, drought, and hydrological drought, which is divided into soil drought (SSMI), runoff drought (SRI), groundwater drought (SGI), and snow drought (SMI). The vegetation extreme index (VEG) includes: the peak green-up day of the last growing season (NDVImax_PRE) and the peak green-up day of the growing season (NDVImax_GS). A total of 25 extreme climate, extreme hydrology and extreme vegetation green-up indices (Appendix 3).

[0043] Heat wave (HW) identification: Heat waves are usually identified using the relative threshold d(Tmax_90th)>3d (Wang et al., 2024b), and four indicators are selected to comprehensively evaluate the frequency, intensity, duration and severity of heat waves. Including: the frequency of heat wave events (HWF), i.e. the total number of heat wave occurrences, the total occurrence days of heat wave events (HWO), the maximum duration of heat wave events (HWD), and the cumulative heat of heat wave events (HWC).

[0044] Hot-humid wave (HHW) identification: Daily mean dew point temperature (Twb) exceeds the 95th percentile of the local grid cell and exceeds a minimum threshold of 24 °C. Four indicators are selected to comprehensively assess the frequency, intensity, duration, and severity of heat waves. These include: the frequency of hot-humid wave events (HHWF), the total number of hot-humid wave occurrences, the total duration of hot-humid wave events (HHWO), the maximum duration of hot-humid wave events (HHWD), and the cumulative heat of hot-humid wave events (HHWC).

[0045] Cold wave (CW) identification: Cold waves are typically identified using periods where daily temperature is at least five consecutive days below the 10th percentile threshold of the local and seasonally varying threshold. Four indicators are selected to comprehensively assess the frequency, intensity, duration, and severity of cold waves. These include: the frequency of cold wave events (CWF), the total number of cold wave occurrences, the total duration of cold wave events (CWO), the maximum duration of cold wave events (CWD), and the cumulative cold of cold wave events (CWC).

[0046] Extreme temperature flip event (RTF) identification: 5-day mean temperature within 1 standard deviation range from extreme warm (cold) event to extreme cold (warm) event. Three indicators are selected to comprehensively assess the frequency, intensity, and duration of extreme temperature flip events. These include: the frequency of extreme temperature flip events (RTFF), the total number of heat wave occurrences, the intensity of extreme temperature flip events (RTFS), and the duration of extreme temperature flip event transitions (RTFT).

[0047] Heavy precipitation (PO) identification: From the large ensemble climate dataset, extract the rainfall events corresponding to the 150-year return period estimated by the probability limit method test, PO > P_95th.

[0048] Flood (FE) identification: In this study, the peak over threshold (POT) method is used to detect the start and end dates of flood events in order to separate the site and gridded flood events from the ERA5-LandHourly dataset climate dataset and CIMP6 future climate dataset runoff series. To ensure the independence of flood peaks, the method for detecting the start and end dates of flood events follows the concept that the flow at the start or end date should be below a threshold (strictly extreme flood: 95%) and the interval width between the peak and the start or end date should be as short as possible to maintain the independence of two consecutive peaks.

[0049] Drought (D) identification: including meteorological drought and hydrological drought. The hydrological drought includes: soil water drought, runoff drought, groundwater drought and snow drought. Each drought index adopts the standardized precipitation evapotranspiration index (SPEI), the standardized soil moisture index (SSMI), the standardized runoff index (SRI), the standardized groundwater index (SGI) and the standardized snowmelt index (SMI).

[0050] The method is widely used in the determination of the length of the vegetation phenological pre-season and the remote sensing driven attribution. First, the spring and autumn phenological pre-season length of all driving factors is obtained by Pearson partial correlation, and then the partial correlation coefficient of each index in the pre-season length with the spring and autumn phenology is calculated. Taking the spring phenology SOS as an example

[0051]

[0052] In the formula, r j is the Pearson correlation coefficient between the spring and autumn phenology (SOS and EOS) and the process index and extreme event, j is the process index or extreme event factor, r max is r j The maximum absolute value. And through the method of retaining the maximum absolute value, the driving attribution of different types (including: climate, hydrology, human activity and vegetation) after the fusion of the separation index is finally obtained, and the contribution after the synergistic combination is obtained by the method in the literature. Among them, the drought intensification-phenology lengthening paradox is the and , and Superimposed, SGI is the pre-season groundwater drought index.

[0053] The vegetation ecosystem in the arid region which depends on groundwater will be positively affected by the flood recharge of groundwater and the sufficient soil water recharge. Therefore, the potential influence mode of groundwater drought and flood legacy on the spring and autumn phenology of the ecological system in the arid region of the Northern Hemisphere is proposed in the embodiment. The spring and autumn phenology model is constructed by removing the interference factors from the observed values of SOS and EOS which are not affected by extreme events in different phenological partitions. And the theoretical SOS and EOS of all years are predicted according to the model. The final result is used to quantify the trade-off and synergistic effect of drought and flood legacy effect, and the reference Figure 1 , the embodiment includes the following contents: S1, spatially partitioning the spring and autumn phenology; The spring phenology SOSⅠregion corresponds to the dimension of 0~31.5°N; SOSⅡ region of spring phenology, which corresponds to the dimension of 31.5°N~66°N; EOSⅠ region and Ⅱ region of autumn phenology, which corresponds to the dimension of 0~51.5°N; EOSⅢ region of autumn phenology, which corresponds to the dimension of 51.5~66°N.

[0054] S2, a linear mixed effect model is used to construct a spring and autumn phenology model, and a linear mixed effect model (LMM) is used to realize the quantification of the residual effect; Since the construction process of the spring phenology model and the autumn phenology model is completely the same, this embodiment only takes the spring phenology model as an example for description, which is specifically represented as:

[0055] In the formula, SOS represents the spring phenology; is the coefficient of the fixed effect; represents the lag term of the drought index; represents the scaled drought index; represents the scaled flood index; represents the lag term of the flood index; is the random intercept of the drought index and the flood index; is the random intercept of the year; is the residual.

[0056] S3, a residual effect phenology model under different spatial partitions is constructed based on the spring and autumn phenology model; In order to eliminate the interference of drought and flood events, this embodiment calculates the difference between the observation and the prediction affected by the extreme event through the spring and autumn phenology model in S2. The observation of SOS and EOS not affected by the extreme event is used to construct the spring and autumn phenology model, and the theoretical SOS and EOS of all years are predicted according to the model, and the final result is used to quantify the interactive residual effect. Mainly including: (1) neither flood nor groundwater drought; (2) only flood; (3) only groundwater drought; (4) both flood and groundwater drought, and quantifying the SOS and EOS response under the conditions of mild (75 th ~90 th ), moderate (90 th ~95 th ), severe (95 th ~99 th ), extreme (>99 th ) flood and mild [-1, -0.5], moderate [-1.5, -1], severe [-2, -1.5], extreme [-∞, -2] groundwater drought.

[0057] Specifically, the embodiment includes the following for the phenological models of the legacy effect in different spatial zones: Spring phenology SOS I zone, the legacy effect spring phenology model in the dimension of 0-31.5 °N; Spring phenology SOS II zone, the legacy effect spring phenology model in the dimension of 31.5 °N-66 °N; Autumn phenology EOS I and II zones, the legacy effect autumn phenology model in the dimension of 0-51.5 °N; Autumn phenology EOS III zone, the legacy effect autumn phenology model in the dimension of 51.5-66 °N.

[0058] In a specific embodiment, the spring phenology SOS I zone, the legacy effect spring phenology model in the dimension of 0-31.5 °N is represented as: The existing spring phenology model based on tropical regions is blank, which has an adverse effect on the prediction of phenological productivity under future climate change. Therefore, the embodiment uses long-term SOS in the past forty years as the input data of the model. The main basis is that the SOS of plants usually changes slightly between adjacent years. By calculating the SOS difference between all consecutive years , and then identifying whether a drought and flood event has occurred in these years. Since the SOS current is set as the baseline value unaffected by drought, when drought and flood occur before the SOS current , the sample is removed. In the embodiment, the method is only applied to the tropical and subtropical regions without spring phenology model.

[0059] Inspired by the study of the legacy effect of spring phenology drought, for the following regions with spring and autumn phenology models, the spring and autumn phenology difference is obtained by and .

[0060] In the formula, represents the difference between the observed value and the predicted value of spring phenology SOS; represents the observed value of spring phenology SOS; represents the predicted value of spring phenology SOS, which is calculated by the spring phenology model in S2.

[0061] In a specific embodiment, the spring phenology SOS II zone, the legacy effect spring phenology model in the dimension of 31.5 °N-66 °N is represented as: In the spring phenology model DORMPHOT, the algorithm introduces the effect of photoperiod on dormancy and refines the algorithm into three stages: dormancy induction, dormancy release, and growth recovery Dormancy induction stage: The dormancy induction process is induced by short day length (DR P) and low temperature (DR T ) triggers, when the cumulative value of the product of DR P and DR T reaches a certain threshold value, the dormancy induction process ends (DS>Dcrit):

[0062] where:

[0063]

[0064] where, is the state of dormancy induction; denotes the start date of dormancy induction, denotes the current calculation time point; denotes the short-day photoperiod driving factor; denotes the low temperature driving factor; T is the daily average temperature, DL is the t daylight length of the day; aD , bD and DL crit is the parameter that adjusts the photoperiod and temperature effects; dormancy release and growth recovery start after the completion of dormancy induction (DS>Dcrit) t d , respectively, with the parallel cooling and forced processes; dormancy release phase:

[0065] where:

[0066]

[0067] where, denotes the daily cooling rate; denotes the daily cooling amount; denotes the start time of dormancy release and growth recovery upon completion of dormancy induction; denotes the temperature response shape parameter of the cooling process; denotes the optimum temperature of the cooling process; denotes the critical day length; denotes the slope of the cooling effect on photoperiod sensitivity; denotes the critical threshold of the cooling state; denotes the critical temperature; denotes the slope of the day length effect on temperature requirements; denotes the current actual day length; Growth recovery:

[0068] where, denotes the daily forced rate of accumulation; denotes the daily forced rate; denotes the temperature response slope of the forcing process.

[0069] In one embodiment, the autumnal phenology EOS Zone I and Zone II, the legacy effect autumnal phenology model with dimension 0~51.5°N; The original DMP model was extended by introducing an exponential function to consider temperature, photoperiod, and precipitation accumulation. The new DMP model was mainly based on two assumptions: (i) the increase of temperature and precipitation leads to the delay of EOS, and (ii) the interaction of water and heat affects the autumnal phenology change in a nonlinear way. Then, the original unit was extended to the unit of temperature, photoperiod, and precipitation accumulation by establishing a nonlinear function. The DMP model identifies the EOS date when the cumulative DMP(t) reaches a threshold (Y crit , arbitrary unit), which is expressed as:

[0070]

[0071] where, denotes the daily DMP value; denotes the upper temperature reference value; denotes the daily mean temperature; denotes the temperature response exponent; denotes the photoperiod on day t; denotes the critical photoperiod; denotes the maximum precipitation at the start date when the iteration starts; denotes the daily precipitation; is the critical temperature, which denotes the temperature sensitivity threshold; denotes the phenological accumulation threshold; denotes the DMP value on day x, y, and k are used to reflect the participation proportion of temperature, light, and precipitation, respectively.

[0072] In one embodiment, the autumnal phenology EOS Zone III, the legacy effect autumnal phenology model with dimension 51.5~66°N; Two classic autumnal phenology models were used: the GDD model and a photoperiod-sensitive degree-day autumnal phenology model (DM model). CDD(t) was defined as the daily mean cooling temperature (T t ) below the temperature threshold (Tth ) is expressed as:

[0073]

[0074] where, represents the effective low temperature intensity of each day; represents the reference temperature; represents the date of the effective low temperature intensity.

[0075] Another expression is that the entire process starts from the d0 day, when the daily temperature is lower than the critical temperature ( T b ), and ends on the EOS date, when the GDD (t) reaches the required coldness Y crit . The DM model assumes that both temperature and shortened photoperiod can trigger the start of autumn phenology. The daily leaf senescence rate DM (t) is controlled by temperature and interacts with the daily photoperiod, which is expressed as:

[0076]

[0077] where, represents the state and rate of leaf discoloration on the tth day; represents the state and rate of leaf discoloration on the dth day; represents the photoperiod length.

[0078] Model calibration and evaluation of the reconstructed residual effect phenology model under different spatial partitions include the following: Based on the functional type classification scheme for annual plants, the vegetation in the arid region of the Northern Hemisphere is classified. Then, based on the model construction, the spring and autumn phenology algorithm is parameterized, and the spatial distribution of the main homogeneous pixels is used, which correspond to different vegetation types, to divide the remote sensing phenology dataset, and finally obtain the phenology dataset at the same vegetation community level. The particle swarm optimization (PSO) algorithm (Chen et al., 2024) is used to optimize the parameters of DORMPHOT, DM, and DMP algorithms, and a hybrid function considering multiple evaluation indicators is used as the objective function to find the global optimal parameters. The parameter optimization data comes from the phenology and flux site data.

[0079]

[0080] where, R 2The coefficients are the coefficient of determination, NSE is the Nash-Sutcliffe efficiency coefficient, and RMSE is the root mean square error. The coefficients preceding each term in the formula are used to adjust the weights of different evaluation metrics. The smaller the objective function value, the closer the algorithm's simulated values ​​are to the observed values.

[0081] Using the coefficient of determination (R²) 2 The performance of the actual evapotranspiration model is evaluated using the Nash–Sutcliffe coefficient (NSE), root mean square error (RMSE), and Akaike criterion (AIC), as detailed below.

[0082]

[0083]

[0084]

[0085]

[0086] In the formula, M i and G i These are the model's predicted value and the actual value, respectively, with subscripts. i For the first i One sample, N The total number of data points for the research period. and These are the mean values ​​of the model's predicted values ​​and the actual values, respectively. k It refers to the number of parameters.

[0087] Although specific embodiments of the invention have been described in detail with reference to the accompanying drawings, this should not be construed as limiting the scope of protection of this patent. Various modifications and variations that can be made by a person skilled in the art without inventive effort within the scope described in the claims still fall within the scope of protection of this patent.

Claims

1. A quantitative method for weighing and synergistic effects of drought and flood legacy effects, characterized in that, Includes the following steps: S1. Spatial division of spring and autumn phenology; S2. A spring and autumn phenological model was constructed using a linear mixed-effects model. S3. Based on the Spring and Autumn Phenological Model, construct phenological models of residual effects under different spatial zones.

2. The method for quantifying the trade-off and synergistic effects of drought and flood legacy effects according to claim 1, characterized in that, In S1, the phenological phenomena of spring and autumn are spatially divided into regions, and these regions are represented as follows: Spring phenology zone SOS I, corresponding to latitudes of 0–31.5°N; Spring phenology zone SOS II, corresponding to latitudes of 31.5°N to 66°N; The autumn phenological zones EOS I and II correspond to latitudes of 0–51.5°N; The autumn phenological zone EOSⅢ corresponds to latitudes of 51.5~66°N.

3. The method for quantifying the trade-off and synergistic effects of drought and flood legacy effects according to claim 1, characterized in that, In S2, a linear mixed-effects model is used to construct the spring and autumn phenological model. The spring phenological model in this model is expressed as follows: In the formula, Indicates spring phenology; The coefficients are for fixed effects; Indicates the lagged term of drought indicators; This represents a scaled-down drought index; This represents the scaled flood index; Represents the lag term of flood indicators; For the random intercepts of drought and flood indices; The random intercept for the year; This is the residual.

4. The method for quantifying the trade-off and synergistic effects of drought and flood legacy effects according to claim 1, characterized in that, In S3, the residual effect phenological models under different spatial partitions include: Spring phenology in SOS I region, with residual effects of spring phenology model at latitudes of 0–31.5°N; Spring phenology SOS II region, residual effect spring phenology model at latitudes 31.5°N~66°N; Autumn phenology models of residual effects in EOS I and II regions with latitudes of 0–51.5°N; Autumn phenology EOSⅢ zone, residual effect autumn phenology model at latitude 51.5~66°N.

5. The method for quantifying the trade-off and synergistic effects of drought and flood legacy effects according to claim 4, characterized in that, The residual effect spring phenological model for the SOS I region at latitudes 0–31.5°N is expressed as follows: In the formula, This indicates the difference between observed and predicted values ​​of spring phenological SOS. This represents the observed value of SOS (Self-Occurring Occurrence) in spring phenology; This represents the predicted value of spring phenology SOS, calculated using the spring phenology model in S2.

6. The method for quantifying the trade-off and synergistic effects of drought and flood legacy effects according to claim 4, characterized in that, The residual effect spring phenological model for the SOS II region (latitude 31.5°N–66°N) includes three stages: Dormancy induction phase: in: In the formula, This is a state induced by hibernation; Indicates the start date of hibernation induction. Indicates the current calculation time point; Indicates the short photoperiod driving factor; Indicates the low-temperature driving factor; T This is the average daily temperature. DL yes t Daytime sunshine duration; aD , bD and DL crit Parameters for adjusting photoperiod and temperature effects; Hibernation end phase: in: In the formula, Indicates the daily cooling rate; Indicates daily cooling volume; This indicates the moment when dormancy is broken and growth resumes at the beginning of the process after dormancy induction is complete; Shape parameters representing the temperature response during the cooling process; Indicates the optimal temperature for the cooling process; Indicates the critical day length; The slope representing the effect of cooling on photoperiod sensitivity; The critical threshold indicating the cooling state; Indicates the critical temperature; The slope representing the effect of day length on temperature demand; Indicates the current actual length of day; Growth recovery: In the formula, This represents the cumulative daily forced rate. Indicates the daily mandatory rate; This represents the slope of the temperature response during a forced process.

7. The method for quantifying the trade-off and synergistic effects of drought and flood legacy effects according to claim 4, characterized in that, The residual autumn phenological model for EOS I and II regions at latitudes of 0–51.5°N is expressed as follows: In the formula, This represents the daily DMP value; This indicates the upper limit reference value for temperature; Indicates the average daily temperature; Indicates the temperature response index; This represents the light period on day t. Indicates the critical optical period; Indicates from the start date The maximum precipitation at the start of the iteration; This indicates the amount of precipitation on that day; The critical temperature represents the temperature sensitivity threshold. Indicates the threshold for phenological accumulation; Indicates date The DMP value; x, y, and k are used to reflect the participation ratio of temperature, light, and precipitation, respectively.

8. The method for quantifying the trade-off and synergistic effects of drought and flood legacy effects according to claim 7, characterized in that, The residual autumn phenological model for the EOSⅢ region at latitudes 51.5–66°N is expressed as follows: In the formula, Indicates the daily effective low temperature intensity; Indicates the reference temperature; Indicates date Effective low-temperature strength.

9. The method for quantifying the trade-off and synergistic effects of drought and flood legacy effects according to claim 7, characterized in that, The residual autumn phenological model for the EOSⅢ region at latitudes 51.5–66°N is expressed as follows: In the formula, This represents the state and rate of leaf discoloration on day t; This indicates the state and rate of leaf discoloration on day d; It represents the length of the optical period.

Images