A hydrothermal soil humidity coupling regional agricultural drought monitoring method and system

Abstract

The application provides a hydrothermal soil humidity coupling regional agricultural drought monitoring method and system, which comprises the following steps: 1, collecting soil humidity data and meteorological element data of a research area; 2, determining the time scale value with the highest correlation in different seasons by considering the correlation between the accumulated hydrothermal factors and the soil humidity in different time scales by means of the Pearson correlation coefficient method; 3, calculating the lag time of soil moisture to precipitation and temperature by means of the cross wavelet transform method; 4, constructing a hydrothermal soil humidity coupling drought index RADI by means of a binary Copula model; and 5, dividing different drought grades to represent the regional agricultural drought conditions. The application evaluates the reliability and applicability of RADI in representing regional agricultural drought from the time dimension and the space dimension, provides more fine data support for the evolution characteristics of agricultural drought, and has important significance for regional grain production, ecological safety and environmental protection and the like.

Description

Technical Field

[0001] This invention relates to the field of agricultural remote sensing drought monitoring technology, specifically to a method and system for monitoring regional agricultural drought by coupling water, heat, and soil moisture. Background Technology

[0002] Drought is generally considered a slow-onset and complex natural disaster. Agricultural drought, due to its dependence on current meteorological conditions, biophysical characteristics, growth stages, and other factors such as soil properties, and its close relationship with crops, poses a potential threat to food security and sustainable agricultural development. In recent years, regional drought events have shown a trend of wider scope, longer duration, and higher frequency, thus urgently requiring solutions for regional agricultural drought prevention. Many studies have been conducted on quantifying drought, among which drought indices have become an important method and computational foundation for studying drought and its impacts due to their ease of use and ability to capture key information such as the onset, duration, and end of drought.

[0003] Currently, the region is transitioning to a drier climate, and the enhanced land-atmosphere coupling processes associated with persistent soil moisture shortages are exacerbating global warming and anticyclonic circulation anomalies. Nonlinear land-atmosphere interactions involve multiple variables, including meteorological elements such as temperature, precipitation, humidity, and wind speed, as well as soil moisture. These interactions, through processes like soil moisture evaporation, vegetation transpiration, and surface heat flux, influence the occurrence and evolution of natural disasters such as drought. Climate change affects the distribution patterns of global water and heat resources. Precipitation and temperature, as key meteorological elements, directly impact soil moisture replenishment when climate anomalies manifest as insufficient or deficient precipitation. Furthermore, abnormally high temperatures accelerate soil moisture consumption through evaporation, exacerbating soil dryness and triggering drought events. Soil moisture exhibits memory; a potential correlation and hysteresis feedback mechanism exist between precipitation, temperature, and soil moisture. The effects of precipitation and temperature over a period may take several months to eliminate these anomalies. Therefore, by addressing the asynchronous influence of meteorological elements and soil moisture, quantifying the lag correlation between precipitation, temperature, and soil moisture, and constructing a hydrothermal soil humidity coupled drought index, we can provide more refined data support for the evolution characteristics of agricultural drought in Hubei Province, which is of great significance for regional food production, ecological security, and environmental protection. Summary of the Invention

[0004] To address the shortcomings of existing research, this invention discloses a method for monitoring agricultural drought in regions with coupled water, heat, and soil humidity, and establishes a drought index that comprehensively considers the time-delayed coupling relationship between water, heat, and soil humidity.

[0005] To achieve the above objectives, the technical solution of the method of the present invention is as follows:

[0006] A method for monitoring agricultural drought in a region coupled with hydrothermal soil moisture includes the following steps:

[0007] Step 1: Collect soil moisture data and meteorological element data for the study area;

[0008] Step 2: Using the Pearson correlation coefficient method, comprehensively consider the correlation between the cumulative total precipitation, total temperature and total soil moisture at different time scales to determine the time scale value with the highest correlation in different seasons.

[0009] Step 3: Based on the time scale values ​​with the highest correlation in different seasons determined in Step 2, filter the cumulative total precipitation and temperature at the time scales with the highest correlation. Calculate the lag time of soil moisture on the cumulative total precipitation and temperature at the time scales with the highest correlation using the cross wavelet transform method.

[0010] Step 4: Construct a hydrothermal soil moisture coupled drought index RADI using a binary Copula model;

[0011] Step 5: Divide the drought into different levels to characterize the regional agricultural drought situation; combine statistical data and relevant drought indices to comprehensively evaluate the reliability and applicability of RADI in characterizing regional agricultural drought from both temporal and spatial dimensions.

[0012] Furthermore, the meteorological data collected in step 1 includes precipitation and temperature, and the three types of data—precipitation, temperature, and soil moisture—are uniformly interpolated to the same spatial and temporal resolution.

[0013] Furthermore, precipitation P is the cumulative total precipitation NP at different time scales. n，m for:

[0014] NP n，m =P n，m-sc+1 +P n，m-sc+2 +…+P n，m (1)

[0015] Temperature T: The cumulative temperature over different time scales (NT) n，m for:

[0016] NT n，m =T n，m-sc+1 +T n，m-sc+2 +…+T n，m (2)

[0017] Soil moisture SM is the sum of accumulated soil moisture at different time scales. n，m for:

[0018] NSM n，m =SM n，m-sc+1 +SMn，m-sc+2 +…+SM n，m (3)

[0019] Where n is the year, m is the month, and sc is the time scale value.

[0020] Furthermore, in step 2, the sample correlation coefficient r is obtained based on the Pearson correlation coefficient method:

[0021]

[0022] Here, X and Y represent two different variables: X represents the total accumulated precipitation or the total accumulated temperature at different time scales, and Y represents the total accumulated soil moisture at different time scales. n is the sample size. and It represents the average of two different variables.

[0023] Furthermore, the lag time of the cumulative total precipitation and the cumulative soil moisture at the time scale with the highest correlation in step 3 is... NP-NSM And the lag time of accumulated temperature and accumulated soil moisture at different time scales. NT-NSM By calculating the cross wavelet transform, the larger cross wavelet energy values ​​in the cross wavelet energy spectrum indicate that the two sets of sequences have a common high-energy region and are significantly correlated with each other. The phase angle represents the local relative phase shift relationship between the two sets of time series, and the lag time is the product of the common period and the phase relationship, thus determining the lag of soil moisture relative to precipitation and high temperature.

[0024] Furthermore, in step 4, the cumulative soil moisture data is used to calculate the soil moisture index SMI using a nonparametric kernel density estimation method, and the standardized precipitation evapotranspiration index SPEI is calculated using the cumulative precipitation and cumulative temperature data after lag correlation calculation. The binary Copula model combines the probability distributions of the cumulative soil moisture data, cumulative precipitation data, and cumulative temperature data to construct the hydrothermal soil moisture coupled drought index RADI.

[0025] Furthermore, given independent observations x1, x2, ..., x... n The cumulative soil moisture (NSM) over a time series is defined as follows: Its probability density function is set to f, and the kernel density estimation formula is:

[0026]

[0027] Where K is the kernel function and h is the bandwidth. This is the scaling kernel function.

[0028] Then, the probability distribution function P of the cumulative soil moisture was calculated using the probability density function of the cumulative soil moisture. SMI ;

[0029] Based on temperature, a potential evapotranspiration formula is established considering the latitude factor, and the temperature lag time (lag) is obtained based on the lag correlation. NT-NSM Estimate the potential evaporation of PET lag ,

[0030] Calculate the difference between monthly precipitation and potential evapotranspiration.

[0031]

[0032] Where i represents the month, P lag PET represents the precipitation calculated with lag. lag This represents the potential evapotranspiration after lag calculation;

[0033] Difference sequence D of potential evapotranspiration lag The cumulative probability value is fitted using a three-parameter Log-logistic probability distribution function to obtain the probability distribution function of the time-delayed hydrothermal coupling data.

[0034] Joint distribution function P of water, heat and soil moisture coupling RADI The calculation formula is:

[0035]

[0036] in, and Let P be the marginal cumulative probability distribution function CDF of the variable. SMI and C is the Copula function;

[0037] Copula parameters are obtained through maximum likelihood estimation, and then the minimum AIC value for different dependency structures is used as the best candidate model for fitting variables through the AIC test. The marginal probability distributions of each univariate in the bivariate copula function are as follows: P SMI and

[0038] The Akaike Information Content Criterion (AIC) calculation formula is as follows:

[0039] AIC = 2k - ln L (8)

[0040] Where k represents the number of free parameters; L represents the maximum likelihood function.

[0041] Furthermore, the drought index in step 4 can be obtained by the inverse normal transformation of the joint probability distribution function or the marginal probability distribution function, as shown in the formula:

[0042]

[0043]

[0044]

[0045] in, It is the standard normal distribution function.

[0046] Furthermore, the drought levels of the SMI, SPEI, and RADI drought indices in step 5 are classified as follows:

[0047]

[0048] Among them, D1, D2, D3, D4, and D5 represent extreme drought, severe drought, moderate drought, mild drought, and no obvious drought, respectively.

[0049] On the other hand, the present invention provides a regional agricultural drought monitoring system coupled with hydrothermal soil moisture, comprising:

[0050] Module 1: It is used to collect soil moisture data and meteorological element data in the study area;

[0051] Module 2: It is used to determine the time scale value with the highest correlation in different seasons by comprehensively considering the correlation between the total accumulated precipitation, the total temperature and the total soil moisture at different time scales using the Pearson correlation coefficient method.

[0052] Module 3: It is used to filter the cumulative total precipitation and temperature at the time scale with the highest correlation in different seasons based on the time scale values ​​determined in Module 2. It calculates the lag time of soil moisture on the cumulative total precipitation and temperature at the time scale with the highest correlation using the cross wavelet transform method.

[0053] Module 4: It is used to construct a hydrothermal soil moisture coupled drought index RADI using a binary Copula model;

[0054] Module 5: It is used to classify different drought levels to characterize the regional agricultural drought situation; combining statistical data and relevant drought indices, it comprehensively evaluates the reliability and applicability of RADI in characterizing regional agricultural drought from both temporal and spatial dimensions.

[0055] The water-heat-soil humidity coupled regional agricultural drought monitoring system is used to perform the steps in the above-mentioned water-heat-soil humidity coupled regional agricultural drought monitoring method.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] Under the influence of climate change and human activities, there are complex feedback mechanisms and time lag effects between rainfall, temperature, and soil moisture. This method takes into account multiple drought-related factors and their lag relationships, and proposes and constructs a combined time-lag precipitation, temperature, and soil moisture drought index (RADI) for monitoring and predicting regional agricultural drought. Validated by historical disaster data and crop drought damage statistics, the RADI drought index demonstrates good characterization ability for regional agricultural drought and can provide a reference for more accurate and effective monitoring and prediction of agricultural drought in local areas.

[0058] CopuIa can connect any two or more related variables from arbitrary marginal distributions to flexibly generate their joint distribution. CopuIa accepts nonlinear relationships between variables. Using the CopuIa probabilistic method to express the impact of water, heat and soil moisture on integrated agricultural drought is a beneficial attempt to effectively construct a multivariate drought index. Attached Figure Description

[0059] Figure 1 This is a flowchart of an embodiment of the present disclosure.

[0060] Figure 2 This is the research area of ​​the embodiments of this disclosure.

[0061] Figure 3 Box plots of correlation coefficients between cumulative soil moisture and cumulative precipitation in Hubei Province at different scales, according to embodiments of this disclosure.

[0062] Figure 4 Box plots of correlation coefficients between cumulative soil moisture and cumulative temperature in Hubei Province at different scales, according to embodiments of this disclosure.

[0063] Figure 5 The cross-wavelet energy spectra of NP-NSM and NT-NSM in different agricultural areas of Hubei Province are shown in this embodiment of the present disclosure.

[0064] Figure 6 This is a time series diagram of drought index in various agricultural areas of Hubei Province, as per an embodiment of this disclosure.

[0065] Figure 7 This is a spatiotemporal evolution diagram of the 2007 drought event in Hubei Province, as described in this embodiment of the present disclosure. Detailed Implementation

[0066] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0067] like Figure 1 As shown, a method for monitoring agricultural drought in a coupled hydrothermal-soil moisture region includes the following steps:

[0068] Step 1: Collect soil moisture data and meteorological element data for the study area;

[0069] Step 2: Using the Pearson correlation coefficient method, comprehensively consider the correlation between the cumulative total precipitation, total temperature and total soil moisture at different time scales to determine the time scale value with the highest correlation in different seasons.

[0070] Step 3: Based on the time scale values ​​with the highest correlation in different seasons determined in Step 2, filter the cumulative total precipitation and temperature at the time scales with the highest correlation. Calculate the lag time of soil moisture on the cumulative total precipitation and temperature at the time scales with the highest correlation using the cross wavelet transform method.

[0071] Step 4: Construct a hydrothermal soil moisture coupled drought index RADI using a binary Copula model;

[0072] Step 5: Divide the drought into different levels to characterize the regional agricultural drought situation; combine statistical data and relevant drought indices to comprehensively evaluate the reliability and applicability of RADI in characterizing regional agricultural drought from both temporal and spatial dimensions.

[0073] The meteorological data collected in step 1 includes precipitation and temperature. The three types of data, precipitation, temperature and soil moisture, are interpolated to the same spatial and temporal resolution.

[0074] Drought-related variables can be accumulated across multiple time scales to reflect the manifestations of drought in different drought environments and cycles. The main time scales include 3-month (quarterly), 6-month (half-year), and 12-month (annual) time scales.

[0075] Precipitation P is the cumulative total precipitation at different time scales. n，m for:

[0076] NP n，m =P n，m-sc+1 +P n，m-sc+2 +…+Pn，m (1)

[0077] Temperature T: Total accumulated temperature over different time scales (NT) n，m for:

[0078] NT n，m =T n，m-sc+1 +T n，m-sc+2 +…+T n，m (2)

[0079] Soil moisture SM is the sum of accumulated soil moisture at different time scales. n，m for:

[0080] NSM n，m =SM n，m-sc+1 +SM n，m-sc+2 +…+SM n，m (3)

[0081] Where n is the year, m is the month, and sc is the time scale value. To study the correlation between seasonal soil moisture and precipitation and temperature, soil moisture data at a 3-month time scale are accumulated and denoted as NSM3; precipitation and temperature data are accumulated at 3 / 6 / 12-month time scales and denoted as NP3 / NP6 / NP12 and NT3 / NT6 / NT12, respectively.

[0082] The Pearson correlation coefficient comprehensively considers the correlation between accumulated hydrothermal factors and soil moisture at different time scales. The time scale with the highest correlation in different seasons is defined as the product of the covariance of the two variables and their standard deviations. By estimating the sample covariance and standard deviation, the sample correlation coefficient r can be obtained.

[0083]

[0084] Here, X and Y represent two different variables: X represents the total accumulated precipitation or the total accumulated temperature at different time scales, and Y represents the total accumulated soil moisture at different time scales. n is the sample size. and It represents the average of two different variables.

[0085] The lag time of the cumulative total precipitation and the cumulative soil moisture at the time scale with the highest correlation in step 3. NP-NSM And the lag time of accumulated temperature and accumulated soil moisture at different time scales. NT-NSMBy calculating the cross wavelet transform, the larger cross wavelet energy values ​​in the cross wavelet energy spectrum indicate that the two sets of sequences have a common high-energy region and are significantly correlated with each other. The phase angle represents the local relative phase shift relationship between the two sets of time series, and the lag time is the product of the common period and the phase relationship, thus determining the lag of soil moisture relative to precipitation and high temperature.

[0086] The cumulative soil moisture data were used to calculate the soil moisture index SMI using a nonparametric kernel density estimation method. The standardized precipitation evapotranspiration index SPEI was calculated using the cumulative precipitation and cumulative temperature data after lag correlation calculation in step 3. The bivariate Copula model was used to combine the probability distributions of the cumulative soil moisture data, cumulative precipitation data, and cumulative temperature data to construct the hydrothermal soil moisture coupled drought index RADI.

[0087] Given independent observations x1, x2, ..., xn with the same distribution. n The cumulative soil moisture (NSM) over a time series is given by the probability density function f. The kernel density estimation formula is defined as follows:

[0088]

[0089] Where K is the kernel function (non-negative, integral of 1, conforming to probability density properties, with a mean of 0), and h is the bandwidth (smoothing parameter). This is the scaling kernel function.

[0090] Then, the probability distribution function P of the cumulative soil moisture was calculated using the probability density function of the cumulative soil moisture. SMI

[0091] Based on temperature lag time NT-NSM Estimate the potential evaporation of PET lag Calculate the difference between monthly precipitation and potential evapotranspiration.

[0092]

[0093] Where i represents the month, P lag PET represents the precipitation calculated with lag. lag This represents the potential evapotranspiration after lag calculation.

[0094] Then, for the difference sequence D lag The cumulative probability value is fitted using a three-parameter Log-logistic probability distribution function to obtain the probability distribution function of the time-delayed hydrothermal coupling data.

[0095] The joint distribution function P of water, heat and soil moisture coupling RADI The calculation formula is:

[0096]

[0097] in, and Let P be the marginal cumulative probability distribution function (CDF) of the variable. SMI and C is the Copula function.

[0098] Six commonly used copulas types in Copula were considered, including Clayton, Frank, Gumbel, Gaussian, and Student t-function types. Copula parameters were obtained through maximum likelihood estimation, and then the minimum AIC value for different dependency structures was used as the best candidate model for fitting the variables through the AIC test. The specific Copula models are shown in Table 1 below:

[0099] Table 1 Copula Model Expression and Model Parameters

[0100]

[0101] Where Φ2 represents the cumulative probability distribution function (CDF) of a bivariate Gaussian distribution with an expected value of 0 and unit variance; θ is the copula parameter to be fitted; u and v are the marginal probability distributions of each univariate in the bivariate distribution, i.e., P SMI and

[0102] The Akaike Information Content Criterion (AIC) calculation formula is as follows:

[0103] AIC=2k-ln L (8)

[0104] Where k represents the number of free parameters; L represents the maximum likelihood function. The smaller the AIC value, the better the fitting effect.

[0105] The drought index can be obtained by the inverse normal transformation of the joint probability distribution function or the marginal probability distribution function, as shown in the formula:

[0106]

[0107]

[0108]

[0109] in, It is the standard normal distribution function.

[0110] Step 5: Classify drought indices into different drought levels; combine statistical data and relevant drought indices to comprehensively evaluate the reliability and applicability of the RADI drought index in characterizing regional agricultural drought from both temporal and spatial dimensions.

[0111] The drought levels of the SMI, SPEI, and RADI drought indices are shown in Table 2 below. D1 represents extreme drought with a drought index less than -2.0, severe drought with an index between -1.50 and -1.99, moderate drought with an index between -1.00 and -1.49, mild drought with an index between -0.50 and -0.99, and drought with an index greater than -0.50, which is considered to indicate no significant drought.

[0112] Table 2. Classification of Drought Index Levels

[0113]

[0114] Example 1

[0115] The steps of the present invention will be described in detail below with specific implementation examples.

[0116] Step 1: Collect data for the study area. Soil moisture data were selected from the 0-7cm depth of ERA5-Land. Meteorological data were selected from CHIRPS satellite precipitation products and ERA5-Land temperature data. To facilitate subsequent experimental research, the interpolation of the three types of data was unified to a spatial resolution of 5km and a temporal resolution of 1 month (1991-2021).

[0117] According to the "Hubei Province High-Standard Farmland Construction Plan (2022-2030)" issued by the Hubei Provincial Department of Agriculture and Rural Affairs, Hubei Province is divided into four major agricultural regions: the Jianghan Plain agricultural region, the Northern Hubei hilly agricultural region, the Eastern Hubei hilly agricultural region, and the Western Hubei mountainous agricultural region. Figure 2 As shown, this method is suitable for monitoring, researching, and analyzing regional agricultural drought in major agricultural areas.

[0118] Step 2: Figure 3 and Figure 4Box plots showing the correlation between NSM and NP / NT at different time scales in spring, summer, autumn, and winter are presented. Generally, a correlation coefficient above 0.7 indicates a very strong correlation, 0.4–0.7 indicates a strong correlation, and 0.2–0.4 indicates a moderate correlation. The results show that the correlation between meteorological elements and cumulative soil moisture varies seasonally. Precipitation and soil moisture are generally positively correlated; temperature and soil moisture are generally negatively correlated. Finally, cumulative precipitation data at a 6-month time scale, cumulative temperature data at a 12-month time scale for spring, and cumulative temperature data at a 6-month time scale for summer, autumn, and winter are used as the basis to further explore the lagged response association between meteorological elements and soil moisture.

[0119] Step 3: Figure 5 This is a cross-wavelet power spectrum of cumulative precipitation / temperature and cumulative soil moisture. The thick black outline indicates that the data passed the red noise test at the 95% significance level. The right arrow (→) indicates that the two data sets are in the same phase and are positively correlated; the left arrow (←) indicates that the two data sets are negatively correlated; the down arrow (↓) indicates a 90° advance, corresponding to a one-quarter time scale; and the up arrow (↑) indicates a 90° lag. To avoid boundary effects and false information from high-frequency wavelet components, the effective spectral values ​​are illustrated within the wavelet influence cone, represented by a thin black line. The main oscillation period of precipitation and soil moisture in Hubei Province shows a common period of 8–16 months, with a higher concentration in the 16-month period. The main oscillation period of temperature and soil moisture also shows a common period of 8–16 months, with a higher concentration in the 8-month period. The phase arrows for precipitation and soil moisture mostly point to the lower right, indicating a time lag of approximately 1 / 16 of a period, meaning that soil moisture lags behind precipitation by less than one month. The phase arrows for temperature and soil moisture in all agricultural areas point to the upper left, indicating a time lag of approximately 1 / 8 of a cycle, or a lag time of less than one month. In summary, in Hubei Province, there is a lag time of about one month between accumulated soil moisture and precipitation and temperature, demonstrating that the time lag effect in Hubei exhibits certain spatiotemporal regional differences.

[0120] Step 4: Use AIC to select the optimal Copula. The smaller the AIC value, the better the fit. As shown in Table 3, the Copula function selection varies from month to month, with the Clayton copula model being the most common choice. A defined binary Copula model is used to construct the joint probability of precipitation, temperature, and soil moisture. The water-heat-soil moisture coupled drought index RADI is obtained through the inverse normal distribution transformation of the joint probability distribution function. This index characterizes the agricultural drought monitoring capabilities of different agricultural areas in Hubei Province, reflecting the time lag effect on data processing and helping to more accurately identify agricultural drought events.

[0121] Table 3. AIC Calculation Table for Different Months under Different Copula Models

[0122]

[0123]

[0124] Step 5: Select monthly drought indices (SMI, SPEI, and RADI) for single points in agricultural areas of Hubei Province from 1991 to 2021 for typical year evaluation, such as... Figure 6 As shown, the red dashed lines represent drought events of different degrees. By comparing the time series trends of different drought indices, RADI can detect more severe drought phenomena than SPEI and SMI. RADI shows a stronger advantage over SMI and SPEI in characterizing short-term and persistent drought.

[0125] According to relevant literature, Hubei Province experienced significant drought from April 25 to May 20, early June, September 1 to October 25, and early November 2007. The spatiotemporal distribution results of the RADI drought index during these periods were selected. Figure 7 The data shows that from April to June 2007, extreme drought occurred in northwestern Hubei Province, and the drought area migrated from April to June, consistent with historical records. In May 2007, moderate to severe drought occurred in parts of central and eastern Hubei Province, primarily the Jianghan Plain. By June, severe drought covered almost the entire Jianghan Plain, persisting until October and November 2007, with another extreme drought occurring in early November, the main affected area matching historical drought records. By measuring both the size of the drought monitoring area and the severity of the drought, the RADI drought index effectively captures the occurrence of different drought events and is consistent with changes in agricultural drought.

[0126] Example 2

[0127] This embodiment provides a regional agricultural drought monitoring system coupled with hydrothermal and soil moisture, including:

[0128] Module 1: It is used to collect soil moisture data and meteorological element data in the study area;

[0129] Module 2: It is used to determine the time scale value with the highest correlation in different seasons by comprehensively considering the correlation between the total accumulated precipitation, the total temperature and the total soil moisture at different time scales using the Pearson correlation coefficient method.

[0130] Module 3: It is used to filter the cumulative total precipitation and temperature at the time scale with the highest correlation in different seasons based on the time scale values ​​determined in Module 2. It calculates the lag time of soil moisture on the cumulative total precipitation and temperature at the time scale with the highest correlation using the cross wavelet transform method.

[0131] Module 4: It is used to construct a hydrothermal soil moisture coupled drought index RADI using a binary Copula model;

[0132] Module 5: It is used to classify different drought levels to characterize the regional agricultural drought situation; combining statistical data and relevant drought indices, it comprehensively evaluates the reliability and applicability of RADI in characterizing regional agricultural drought from both temporal and spatial dimensions.

[0133] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0134] Obviously, those skilled in the art can make various modifications and variations to the embodiments of the present invention without departing from the spirit and scope of the embodiments of the present invention. Thus, if these modifications and variations to the embodiments of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention also intends to include these modifications and variations.

[0135] All other parts not described in detail are existing technologies.

Claims

1. A method for monitoring regional agricultural drought coupled with water, heat, and soil moisture, characterized in that, Includes the following steps: Step 1: Collect soil moisture data and meteorological element data for the study area; Step 2: Using the Pearson correlation coefficient method, comprehensively consider the correlation between the cumulative total precipitation, total temperature and total soil moisture at different time scales to determine the time scale value with the highest correlation in different seasons. Step 3: Based on the time scale values ​​with the highest correlation in different seasons determined in Step 2, filter the cumulative total precipitation and temperature at the time scales with the highest correlation. Calculate the lag time of soil moisture on the cumulative total precipitation and temperature at the time scales with the highest correlation using the cross wavelet transform method. Step 4: Construct a hydrothermal soil moisture coupled drought index RADI using a binary Copula model; including: calculating the soil moisture index SMI from cumulative soil moisture data using a nonparametric kernel density estimation method, calculating the standardized precipitation evapotranspiration index SPEI from cumulative precipitation and temperature data after lag correlation calculation, and constructing the hydrothermal soil moisture coupled drought index RADI by combining the probability distributions of cumulative soil moisture data, cumulative precipitation data, and cumulative temperature data using a binary Copula model. Given independent observations that have the same distribution , , ..., That is, the cumulative soil moisture NSM over time series, whose probability density function is set as follows: The formula for estimating nuclear density is: （5） in, For kernel function, For bandwidth, For scaling kernel functions; Then, the probability distribution function of cumulative soil moisture is calculated using the probability density function of cumulative soil moisture. ; Based on temperature, a potential evapotranspiration formula is established considering the latitude factor, and the temperature lag time is obtained based on the lag correlation. Estimate potential evaporation , Calculate the difference between monthly precipitation and potential evapotranspiration. : （6） in, Indicates the month. This indicates the precipitation after lag calculation. This represents the potential evapotranspiration after lag calculation; Difference sequence of potential evapotranspiration The cumulative probability value is fitted using a three-parameter Log-logistic probability distribution function to obtain the probability distribution function of the time-delayed hydrothermal coupling data. ; Joint distribution function of water, heat and soil moisture coupling The calculation formula is: （7） in, and The marginal cumulative probability distribution function (CDF) of the variables is expressed as follows: and , It is a Copula function; The Copula parameters are obtained through maximum likelihood estimation, and then the minimum AIC value for different dependency structures is used as the best candidate model for fitting variables through the AIC test. The marginal probability distributions of each univariate in the bivariate Copula function are as follows: and ; The Akaike Information Content Criterion (AIC) calculation formula is as follows: （8） in, Indicates the number of free parameters; Represents the maximum likelihood function; Step 5: Divide into different drought levels to characterize the regional agricultural drought situation; combine statistical data with soil moisture index (SMI) and standardized precipitation evapotranspiration index (SPEI) to comprehensively evaluate the reliability and applicability of RADI in characterizing regional agricultural drought from both temporal and spatial dimensions.

2. The method for monitoring regional agricultural drought coupled with hydrothermal soil moisture according to claim 1, characterized in that, The meteorological data collected in step 1 includes precipitation and temperature. The three types of data—precipitation, temperature, and soil moisture—are interpolated to the same spatial and temporal resolution.

3. The method for monitoring regional agricultural drought coupled with hydrothermal soil moisture according to claim 1, characterized in that, Precipitation P The sum of accumulated precipitation at different time scales for: （1） The sum of temperatures accumulated at different time scales (T) for: （2） Soil moisture SM Total accumulated soil moisture at different time scales for: （3） in, n For the year, m For months, sc This is a time-scale value.

4. The method for monitoring regional agricultural drought coupled with hydrothermal soil moisture according to claim 1, characterized in that, In step 2, the sample correlation coefficient r is obtained based on the Pearson correlation coefficient method: （4） in, and Representing two different variables, This represents the total accumulated precipitation at different time scales or the total accumulated temperature at different time scales. This represents the cumulative soil moisture at different time scales, where n is the sample size. and It represents the average of two different variables.

5. The method for monitoring regional agricultural drought coupled with hydrothermal soil moisture according to claim 1, characterized in that, In step 4, the drought index is obtained through the inverse normal distribution transformation of the joint probability distribution function or the marginal probability distribution function, as shown in the formula: （9） （10） （11） in, It is the standard normal distribution function.

6. The method for monitoring regional agricultural drought coupled with hydrothermal soil moisture according to claim 1, characterized in that, The drought levels of the SMI, SPEI, and RADI drought indices in step 5 are classified as follows: Among them, D1, D2, D3, D4, and D5 represent extreme drought, severe drought, moderate drought, mild drought, and no obvious drought, respectively.

7. A regional agricultural drought monitoring system coupled with hydrothermal soil moisture, characterized in that, include: Module 1: It is used to collect soil moisture data and meteorological element data in the study area; Module 2: It is used to determine the time scale value with the highest correlation in different seasons by comprehensively considering the correlation between the total accumulated precipitation, the total temperature and the total soil moisture at different time scales using the Pearson correlation coefficient method. Module 3: It is used to filter the cumulative total precipitation and temperature at the time scale with the highest correlation in different seasons based on the time scale values ​​determined in Module 2. It calculates the lag time of soil moisture on the cumulative total precipitation and temperature at the time scale with the highest correlation using the cross wavelet transform method. Module 4: It is used to construct a hydrothermal soil moisture coupled drought index RADI using a binary Copula model; Module 5: It is used to classify different drought levels to characterize the regional agricultural drought situation; combining statistical data and relevant drought indices, it comprehensively evaluates the reliability and applicability of RADI in characterizing regional agricultural drought from both temporal and spatial dimensions. The hydrothermal-soil humidity coupled regional agricultural drought monitoring system is used to perform the steps in the hydrothermal-soil humidity coupled regional agricultural drought monitoring method as described in any one of claims 1-6.

Images