CN117633539A - A groundwater drought identification method and device for uneven site distribution - Google Patents

Abstract

The application provides a method and a device for identifying drought of groundwater for uneven site distribution, which relate to the technical field of drought management and response and comprise the following steps: acquiring historical monitoring information and position information of a ground water monitoring station point of a target research area, and gridding the target research area; reconstructing the groundwater level sequence of each grid point of the target research area based on a machine learning algorithm and an inverse distance interpolation method; determining a standardized underground water level index sequence of each grid point of a target research area based on a preset standardized precipitation index algorithm and combining a moving average method; based on the run theory, the groundwater drought event is identified, and the groundwater drought event characteristics are determined. The method solves the problems that systematic errors exist in the current regional groundwater level interpolation based on the non-uniform groundwater monitoring site data, and the accuracy of groundwater drought identification is poor.

Description

A groundwater drought identification method and device for uneven site distribution

Technical field

The present application relates to the technical field of drought management and response, specifically, to a groundwater drought identification method and device for uneven site distribution.

Background technique

Groundwater has the characteristics of wide distribution, stable water temperature, good water quality, and easy extraction. It is an important water supply source for many cities in my country where surface water resources are scarce. As climate change and human activities intensify, extreme meteorological and hydrological events occur frequently, and groundwater droughts also occur from time to time. Once a long-lasting groundwater drought event occurs, it will greatly affect the water security of the ecosystem and economic society. Therefore, it is of practical significance to accurately and effectively identify groundwater drought.

As an important part of the hydrological station network, groundwater monitoring stations provide important basic data support in dynamically grasping changes in groundwater levels and water volumes, and promoting optimal allocation and scientific management of groundwater resources. However, the density of groundwater monitoring sites in my country is still small and the spatial distribution is still uneven, resulting in inaccurate regional groundwater monitoring results and an inability to fully understand the status and changing trends of the groundwater system.

A large number of facts show that shallow groundwater levels are closely related to ground elevation. At the same time, affected by the annual distribution of precipitation and water use for irrigation, living and other human activities, changes in shallow groundwater levels have certain regularity and periodicity. Facing the uneven spatial distribution of groundwater monitoring sites, how to make full use of basic site information and actual measurement information to establish a spatiotemporal identification method of groundwater drought is an important prerequisite for accurately and effectively analyzing groundwater drought characteristics and rationally optimizing the allocation of groundwater resources.

Contents of the invention

The purpose of the embodiments of this application is to provide a groundwater drought identification method and device for uneven site distribution, so as to solve the current problems of systematic errors in regional groundwater level interpolation based on uneven groundwater monitoring site data, poor groundwater drought identification accuracy, etc. question.

In the first aspect, a groundwater drought identification method for uneven site distribution is provided. The method can include:

Obtain the groundwater data of the target research area in the historical time period; the groundwater data is obtained by converting the original monitoring groundwater data of the historical time period in the target research area based on the elevation system of the preset historical time point; the groundwater Data include collection location, collection time, groundwater elevation, and groundwater table values;

After gridding the target research area, based on the machine learning algorithm and inverse distance interpolation method, the groundwater level data of each grid point in the historical time period is reconstructed to obtain the area of the target research area. Groundwater level sequence;

Based on the preset standardized precipitation index construction algorithm, standardize the regional groundwater level sequence to obtain a standardized groundwater level index sequence;

Based on the run theory, the standardized groundwater level index sequence is processed to determine the groundwater drought events in the target study area and determine the characteristics of the groundwater drought events.

In a possible implementation, before obtaining the groundwater data of the target study area in the historical time period, the method further includes:

Obtain original monitoring groundwater data for the historical time period in the target research area;

The 3sigma processing algorithm is used to perform outlier processing on the original monitored groundwater data to obtain the groundwater data.

In a possible implementation, the groundwater level data of each grid point in the historical time period is processed to obtain the regional groundwater level sequence of the target research area, including:

The groundwater level data of each grid point in the historical time period is input into the trained XGBoost algorithm model to obtain the regional groundwater level monthly sequence output by the XGBoost algorithm model. The regional groundwater level monthly sequence includes the Groundwater level data for each month during the historical period;

The inverse distance interpolation method is used to calculate the groundwater data corresponding to the n nearest groundwater level monitoring stations in the target research area, and obtain the daily groundwater data corresponding to each grid point;

Based on the daily groundwater data corresponding to each grid point, calculate the monthly moving average water level value, and then subtract the monthly moving average water level value from the daily groundwater level value corresponding to each grid point to obtain the regional groundwater level of the target study area. Intra-monthly series, the intra-monthly series of regional groundwater levels includes intra-monthly fluctuation data of groundwater levels trending on a monthly scale within the historical time period;

The regional groundwater level intermonthly sequence and the data of the corresponding position in the regional groundwater level intramonthly sequence are added to obtain the regional groundwater level sequence of the target research area, and the regional groundwater level sequence includes the historical time period. Daily groundwater level data for each grid point within the grid.

In a possible implementation, based on a preset standardized precipitation index construction algorithm, the regional groundwater level sequence is processed to obtain a standardized groundwater level index sequence, including:

Based on the groundwater level data corresponding to each grid point in the regional groundwater level sequence, calculate the multi-year moving average water level value, and then subtract the multi-year moving average water level value from the groundwater level value corresponding to each grid point to obtain the target research area Last year's trend of groundwater level series, said last year's trend of groundwater level series includes groundwater level fluctuation data with the annual trend removed;

After the detrended groundwater level sequence is standardized, the standardized detrended groundwater level sequence is processed with reference to the construction process of the standardized precipitation index to obtain a standardized groundwater level index sequence.

In a possible implementation, based on run theory, the standardized groundwater level index sequence is processed to determine groundwater drought events in the target study area, including:

Based on the run theory, if the standardized groundwater level index value at each time point in the standardized groundwater level index sequence is less than or equal to the preset drought threshold, then the grid point corresponding to the corresponding standardized groundwater level index value is determined as a drought grid point ;

Traverse the target research area, and spatially merge the target drought grid points in the determined drought grid points to obtain drought patches, where the target drought grid points are adjacent drought grid points;

The drought patches at adjacent time nodes are spatially superimposed to determine the overlapping area of the overlapping area, and the drought patches whose overlapping area is greater than the preset area are determined to be the same groundwater drought event and assigned the same drought event. serial number.

In a possible implementation, the method further includes:

For two groundwater drought events with an interval of one month, if there is a spatial intersection between the end time point of the previous groundwater drought event and the start time point of the subsequent groundwater drought event, and the spatial union of the interval months The drought index value is less than the first drought index threshold R0, then the two groundwater drought events will be merged into the same groundwater drought event;

For a groundwater drought event that lasts only one month, if the drought index value is greater than the second drought index threshold R2, the groundwater drought event will be eliminated.

In one possible implementation, characterizing groundwater drought events includes:

Perform feature extraction on the groundwater drought event to obtain characteristics of the groundwater drought event;

The groundwater drought event characteristics include drought start time, drought end time, drought duration, average drought area, maximum drought area, drought coverage area, average drought intensity, maximum drought intensity and drought intensity.

In the second aspect, a groundwater drought identification device for uneven site distribution is provided. The device may include:

The acquisition unit is used to obtain the groundwater data of the target research area in the historical time period; the groundwater data is obtained by converting the original monitoring groundwater data of the historical time period in the target research area based on the elevation system of the preset historical time point. ; the groundwater data includes collection location, collection time, groundwater elevation and groundwater level value;

The processing unit is used to reconstruct the groundwater level data of each grid point in the historical time period based on the machine learning algorithm and the inverse distance interpolation method after gridding the target research area to obtain the Regional groundwater level series for the target study area;

And, based on a preset standardized precipitation index construction algorithm, standardize the regional groundwater level sequence to obtain a standardized groundwater level index sequence;

The determination unit is used to process the standardized groundwater level index sequence based on run theory, determine groundwater drought events in the target study area, and determine characteristics of groundwater drought events.

In a third aspect, an electronic device is provided. The electronic device includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory complete communication with each other through the communication bus;

Memory, used to store computer programs;

The processor is used to implement any of the method steps described in the first aspect when executing a program stored in the memory.

In a fourth aspect, a computer-readable storage medium is provided. A computer program is stored in the computer-readable storage medium. When the computer program is executed by a processor, the method steps described in any one of the above-mentioned first aspects are implemented.

The groundwater drought identification method provided by this application for uneven site distribution needs to obtain the groundwater data of the target study area in the historical time period; the groundwater data is based on the elevation system of the preset historical time point for the historical time period in the target study area. The original monitoring groundwater data is obtained after conversion; the groundwater data includes collection location, collection time, groundwater elevation and groundwater level value; after gridding the target research area, based on the machine learning algorithm and inverse distance interpolation method, the historical time The groundwater level data of each grid point in the segment is reconstructed to obtain the regional groundwater level sequence of the target research area; based on the preset standardized precipitation index construction algorithm, the regional groundwater level sequence is standardized to obtain the standardized groundwater level index sequence; based on Run theory is used to process the standardized groundwater level index sequence, determine groundwater drought events in the target study area, and determine the characteristics of groundwater drought events. This method solves the current problems of systematic errors in regional groundwater level interpolation based on uneven groundwater monitoring site data and poor groundwater drought identification accuracy.

Description of drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required to be used in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present application, therefore This should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without exerting creative efforts.

Figure 1 is a schematic flow chart of a groundwater drought identification method for uneven site distribution provided by an embodiment of the present application;

Figure 2 is a schematic flow chart of another groundwater drought identification method for uneven site distribution provided by the embodiment of the present application;

Figure 3 is a schematic diagram comparing prediction data and test data of an optimal model based on the XGBoost algorithm provided by the embodiment of the present application;

Figure 4 is a schematic diagram of the relationship between groundwater level and ground elevation provided by an embodiment of the present application;

Figure 5 is a schematic structural diagram of a groundwater drought identification device for uneven site distribution provided by an embodiment of the present application;

FIG. 6 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all of them. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of this application.

The preferred embodiments of the present application are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described here are only used to illustrate and explain the present application, and are not used to limit the present application. In the absence of conflict, the present application The embodiments and features of the embodiments may be combined with each other.

Figure 1 is a schematic flowchart of a groundwater drought identification method for uneven site distribution provided by an embodiment of the present application. As shown in Figure 1, the method may include:

Step S110: Obtain groundwater data of the target research area in the historical time period.

In the specific implementation, the longitude, latitude, elevation and other location information of groundwater monitoring stations in the historical time period in the target research area are collected, as well as the collection location, collection time, groundwater elevation, groundwater level value and burial depth monitored by the corresponding groundwater monitoring station. Raw groundwater data to obtain raw groundwater data for historical time periods in the target study area.

After that, the collected original monitoring groundwater data is cleaned, specifically:

First, conduct a rationality check on the original groundwater data, including year, month, day, longitude and latitude range, elevation range, etc. Judge and process outliers in groundwater level or burial depth monitoring data, and remove or change obviously erroneous outlier data points.

In a specific embodiment, a 3sigma processing algorithm is used to perform outlier processing on the original groundwater data to obtain groundwater data. That is, when X>μ+3σ or X<μ-3σ, the data are initially judged to be outliers, where μ is the site data mean and σ is the site data standard deviation.

Secondly, the original monitored groundwater data of the historical time period in the target study area is converted based on the elevation system of the preset historical time point, that is, the ground elevation data and groundwater monitoring data are uniformly converted into the ground elevation and groundwater level data of the same elevation system, and we get Groundwater data available for processing.

Step S120: After gridding the target research area, reconstruct the groundwater level data of each grid point in the historical time period based on the machine learning algorithm and the inverse distance interpolation method to obtain the regional groundwater level sequence of the target research area. .

In specific implementation, perform the following steps:

Step 1. After gridding the target research area, select an appropriate supervised learning algorithm by comprehensively considering computer performance, time cost and other factors. Use year, month, longitude, latitude, and ground elevation as input features, and shallow groundwater level as label data to train and filter model parameters. Basic information such as year, month, longitude, latitude, and ground elevation within the study area during the study period is gridded and input into the trained model to calculate the reconstructed monthly scale data of groundwater levels at each grid point. Among them, the supervised learning algorithms that can be selected include multiple linear regression, decision trees, support vector machines, neural networks, etc.

This application uses the distributed gradient enhancement library XGBoost. XGBoost is a kind of gradient boosting tree model. It generates models serially and takes the sum of all models as the output. At the same time, perform a second-order Taylor expansion of the loss function, use the second-order derivative information of the loss function to optimize the loss function, and greedily choose whether to split the node according to whether the loss function decreases. When specifically building the model, grid search and 5-fold cross-validation methods were used to optimize parameters for the maximum depth of the tree and the total number of trees, and the optimal parameters were selected based on the determination coefficient R ^2. The calculation formula for R ² is as follows:

In the formula, y _i is the actual value in the validation set, and y _pred is the model predicted value.

Specifically, the groundwater level data of each grid point in the historical time period is input into the trained XGBoost algorithm model to obtain the monthly series of regional groundwater levels output by the XGBoost algorithm model. The monthly series of regional groundwater levels can include historical time. Groundwater level data for each month in the segment.

Step 2. Use the inverse distance interpolation method to calculate the groundwater data corresponding to the n nearest groundwater level monitoring stations in the target study area, and obtain the daily groundwater data corresponding to each grid point; based on the daily groundwater data corresponding to each grid point Data, calculate the monthly moving average water level value, and then subtract the monthly moving average water level value from the daily groundwater level value corresponding to each grid point to obtain the monthly sequence of regional groundwater levels in the target study area.

Specifically, the moving average method can be used to calculate the moving average water level value corresponding to each grid point, such as the 30-day moving average water level value of each grid point, and then subtract the moving average from the groundwater level value of each grid point, and finally obtain Intra-month fluctuation values of groundwater levels detrended on a monthly scale.

Among them, the calculation formula of the moving average method is as follows:

In the formula, MA _t is the moving average at time point t, X _t is the sequence value at time point t, and n is the size of the moving average window, indicating the number of data points considered when calculating the average. In this step, t is taken as 30 (about 1 month).

Step 3: Add the regional groundwater level intermonthly series and the regional groundwater level intramonthly series data at corresponding positions to obtain the regional groundwater level series of the target study area.

Among them, the groundwater level series in this region can include daily groundwater level data of each grid point in the historical time period.

Step S130: Standardize the regional groundwater level sequence based on a preset standardized precipitation index construction algorithm to obtain a standardized groundwater level index sequence.

Affected by human factors such as long-term groundwater extraction, groundwater levels may continue to decline, making it difficult to identify groundwater droughts. The moving average method is used to calculate the multi-year moving average water level value of each grid point. Specifically: taking t in the moving average method formula as 1100 (about 3 years), the multi-year moving average water level value of each grid point can be obtained.

After that, based on the multi-year moving average water level value corresponding to each grid point and each groundwater level data in the regional groundwater level sequence, the last year's trend of groundwater level series in the target study area is obtained. Specifically, each groundwater level data in the regional groundwater level series is obtained. (Groundwater level data at each grid point) is subtracted from the multi-year moving average water level value, and finally the groundwater level fluctuation value with the multi-year trend removed is obtained to remove the influence of long-term human activity factors. That is, the groundwater level series with last year's trend can include removing the annual trend. groundwater level fluctuation data.

Afterwards, after standardizing the detrended groundwater level sequence, refer to the construction process of the standardized precipitation index, process the standardized detrended groundwater level sequence to obtain the standardized groundwater level index sequence.

Specifically, the construction process of the standardized groundwater level index is as follows: taking each grid point as the basic spatial unit, and selecting Gamma and Pearson III type as the basic distribution type.

First, the average groundwater level in each decade was fitted, the mean square errors were compared to optimize the distribution, and the maximum likelihood method was used to estimate parameters;

Then, the cumulative distribution of the ten-day average groundwater level is converted into a standard normal distribution, and the standardized groundwater level index is obtained through inverse normalization. Among them, the probability density function of the Gamma distribution is as follows:

In the formula, x is the value of the random variable, α is the shape parameter, β is the scale parameter, is the Gamma function, which is the factorial generalization of the shape parameter α.

The probability density function of the Pearson type III distribution is as follows:

In the formula, x is the value of the random variable, a is the shape parameter, b is the scale parameter, and c is the skew parameter.

The mean square error measures the sum of squares of the differences between model predictions and actual observations, and is calculated as follows:

In the formula, Y _i is the actual value of the i-th observation data point, is the model predicted value of the i-th observed data point, and n is the number of observed data points.

Step S140: Based on the run theory, process the standardized groundwater level index sequence to determine the groundwater drought event in the target research area and determine the characteristics of the groundwater drought event.

In the specific implementation, the preset drought threshold R1 is set with reference to the drought grade classification standard of the standardized precipitation index.

Based on the run theory, the groundwater drought grid points at each time point are screened out, and the n×n grid is used to traverse the study area: if the standardized groundwater level index value at each time point in the standardized groundwater level index sequence is greater than the preset drought threshold R1, then The current grid point corresponding to the corresponding standardized groundwater level index value is determined as a non-drought grid point.

If the standardized groundwater level index value at each time point in the standardized groundwater level index sequence is less than or equal to the preset drought threshold, the current grid point corresponding to the corresponding standardized groundwater level index value is determined as a drought grid point.

At this time, the target research area is traversed. If the corresponding elements of adjacent grid points are still drought grid points, they will be merged to form drought patches, and drought patches that are larger than the preset minimum patch area at each time point will be retained;

Dry patches at adjacent time points are spatially superimposed to determine the overlapping area of the overlapping areas;

If the overlapping area is greater than the preset minimum overlapping area, the drought patches at adjacent time points are determined to belong to a groundwater drought event and given the same drought event number.

Among them, the preset minimum overlap area can be determined according to the following formula:

In the formula, 1 and 0 respectively represent the existence and non-existence of spatial overlap between the current groundwater drought event and the previous groundwater drought event; A _t and A _t-1 respectively represent the current groundwater drought event and the previous groundwater drought event. The _patch area _of Represents the minimum coincidence percentage based on drought event area.

In some embodiments, the method may further include:

For two groundwater drought events with an interval of one month, if there is a spatial intersection between the end time point of the previous groundwater drought event and the start time point of the subsequent groundwater drought event, and the spatial union of the interval months The drought index value is less than the first drought index threshold R0, then the two groundwater drought events will be merged into the same groundwater drought event;

For a groundwater drought event that lasts only one month, if the drought index value is greater than the second drought index threshold R2, the groundwater drought event will be eliminated.

In some embodiments, after determining the groundwater drought event in the target study area, feature extraction of the groundwater drought event can also be performed to obtain the characteristics of the groundwater drought event;

Groundwater drought event characteristics can include drought start time, drought end time, drought duration, average drought area, maximum drought area, drought envelope area, average drought intensity, maximum drought intensity and drought intensity.

Among them, the drought duration is the entire time from the beginning to the end of a groundwater drought event; the average drought area is the average area occupied at each time point during the duration of a groundwater drought event; the maximum drought area is the duration of a groundwater drought event The maximum value of the area occupied by each time point in The cumulative value of the drought index; the average drought intensity is the ratio of the drought intensity of a groundwater drought event to the drought duration; the maximum drought intensity is the minimum value of the drought index at each grid point at each time point within the duration of a drought event.

In one example, as shown in Figure 2 is a flow diagram of another groundwater drought identification method with uneven site distribution, the method may include:

(1) Select the Huaibei Plain as the research area, collect the "China Geological Environment Monitoring Groundwater Level Yearbook" from 2005 to 2020, and screen shallow groundwater data in the four provinces of Henan, Shandong, Anhui, and Jiangsu. Since all groundwater monitoring sites have relatively detailed address, elevation, groundwater level or burial depth information, but only some groundwater monitoring sites have latitude and longitude information, geocoding services provided by platforms such as Google Maps are used to convert the address information of groundwater monitoring sites into WGS84 latitude and longitude coordinate information.

(2) Conduct a rationality check on the basic information of groundwater monitoring sites and the original data of monitoring information, including year, month, day, longitude and latitude range, elevation range, etc. Use the 3sigma method to screen outliers in groundwater level or burial depth monitoring data, and perform judgment and processing one by one to remove or change obviously erroneous outlier data points. The ground elevation data and groundwater monitoring data are uniformly converted into the ground elevation and groundwater level data of the 1985 Yellow Sea elevation system.

(3) Based on the XGBoost algorithm, reconstruct the monthly series of regional groundwater levels, using year, month, longitude, latitude, and ground elevation as input features, using shallow groundwater levels as label data, and using grid search and 5-fold cross-validation methods. , perform parameter optimization on the maximum depth of the tree and the total number of trees. The maximum depth of the tree is selected in the range of 1 to 10, and the total number of trees is selected in the range of 100 to 1000. The parameter optimization process is shown in Table 1. By comparing the test results of 100 sets of parameters, it can be seen that when the maximum depth is 7 and the total number of trees is 300, the model has the best effect. The prediction data and test data of the optimal model are compared as shown in Figure 3. The basic information such as year, month, longitude, latitude, and ground elevation within the study area during the study period is gridded (spatial resolution is 0.25° × 0.25°), input into the trained model, and the weight of each grid point is calculated. Monthly data on groundwater levels of the structure.

Table 1 XGBoost algorithm parameter selection

(4) For each grid point in the study area, search for the 8 nearest shallow groundwater level monitoring stations, and use the inverse distance interpolation method to calculate the groundwater level at each grid point. On this basis, the moving average method is used to calculate the 30-day moving average water level value of each grid point, and then the groundwater level value of each grid point is subtracted from the moving average value to finally obtain the monthly fluctuation value of the groundwater level trend on a monthly scale.

(5) Add the intermonthly series of regional groundwater levels reconstructed based on the machine learning algorithm and the intramonthly series of regional groundwater levels reconstructed based on the inverse distance interpolation method to calculate the reconstructed regional groundwater level series.

(6) Use the moving average method to calculate the moving average water level value of each grid point for 1100 days (about 3 years), and then subtract the moving average value from the groundwater level value of each grid point to obtain the groundwater level fluctuation value minus the multi-year trend. The detrended groundwater level time series was standardized using the Min-Max standardization method. Gamma and Pearson type III were selected as the basic distribution types, and the standardized groundwater level averages of each grid point in each decade were fitted. The distribution was optimized based on the fitted mean square error and the maximum likelihood method was used to estimate the parameters. The cumulative frequency distribution of the mean water level is transformed into a standard normal distribution, and the standardized groundwater level index is obtained by inverse normalization.

(7) Referring to the "Meteorological Drought Level" (GB/T 20481-2017) standardized precipitation index to divide the boundaries between no drought and light drought, set the threshold R1 to -0.5, and rely on the run theory to screen out the groundwater drought grid points at each time point, and Use a 3×3 grid to traverse the study area. If the corresponding elements of adjacent grid points are still drought grid points, they will be merged to form drought patches. Refer to the "Regional Drought Process Monitoring and Assessment Method" (QX/T597-2021), and only retain Dry patches larger than 10 grids at each time point in the study area (approximately 5% of the study area). Drought patches at adjacent time points are spatially superimposed. Refer to the "Regional Drought Process Monitoring and Assessment Method" (QX/T597-2021) and related literature, and set α to 1% and β to 20%. , if the overlapping area is greater than the minimum overlapping area, they are considered to belong to the same drought event and given the same drought event number.

(8) Set the threshold R2 as -1.0 and R0 as 0 for the division of light drought and moderate drought based on the standardized precipitation index of "Meteorological Drought Levels" (GB/T 20481-2017).

For two drought events with an interval of one month, if there is a spatial intersection between the last drought time point of the previous drought event and the first drought time point of the subsequent drought event, and in the space of the interval month If the union drought index value is less than 0, these two drought events will be merged into one drought event. For drought events that last only one month, if the index value is greater than -1.0, the drought event will be eliminated.

(9) Relying on the run theory, extract the characteristics of groundwater drought events, including start time, end time, duration, average area, maximum area, coverage area, average intensity, maximum intensity, intensity, etc.

Since the shallow groundwater level has a good correlation with the ground elevation, as shown in Figure 4, the identification method of this application is illustrated by reproducing the relationship between the shallow groundwater level Water table (vertical axis) and the ground elevation DEM (horizontal axis). The advantages of It can be seen from the figure that according to the relationship between measured data, the vast majority of shallow groundwater levels are lower than the ground elevation (below the 1:1 line), but the range of groundwater levels calculated by the inverse distance interpolation method is large, and there are a large number of groundwater levels higher than The ground elevation is unreasonable. The calculation results of the XGBoost algorithm are basically within the coverage of the measured data, reflecting the objective laws but insufficiently reflecting the time fluctuation. The method in this paper absorbs the advantages of the XGBoost algorithm and the inverse distance interpolation method at the same time. It can not only correctly analyze the relationship between shallow groundwater levels and ground elevation, longitude and latitude, year, and month, but also reflect the medium and long-term changes of shallow groundwater levels over time. It can better reflect the short-term fluctuation of shallow groundwater level over time. From Figure 4, most of the calculation results of this application are within the coverage of the measured data. Compared with the XGBoost algorithm, the calculation results of this application have a wider coverage of the measured data and are more volatile, which reflects the method of this article. The advantages.

On the basis of reconstructing the shallow groundwater level sequence of the Huaibei Plain, the spatiotemporal identification of groundwater drought in the Huaibei Plain was carried out. The results of the spatiotemporal identification of groundwater drought in the Huaibei Plain from 2005 to 2020 are shown in Table 2.

Table 2

groundwater drought Number of sessions Duration (number of ten days) Coverage area (number of grids) intensity Drought duration>=0 ten days 72 588 2741 15328 Drought duration>=3 decades 37 536 2174 14795 Drought lasts >= 6 decades twenty four 487 1846 14300 Drought duration>=9 ten days twenty one 468 1755 14139 Drought duration>=18 days 12 363 1283 12391 Drought duration>=36 days 4 185 613 7678

After drought spatial and temporal identification analysis, a total of 37 groundwater drought events lasting more than 30 years in the Huaibei Plain were obtained. According to the "China Flood and Drought Disaster Prevention Bulletin", "China Climate Bulletin", "China Meteorological Disaster Yearbook", "China Meteorological Yearbook" and other literature, A total of 26 actual drought events related to "Huaibei", "Huanghuai", "Huanghuai Winter Wheat Area", "Henan", "Shandong", "Anhui", "Jiangsu" and other regions were obtained (mostly pointing to low precipitation, Soil moisture loss, crop yield reduction and other meteorological and agricultural droughts); using actual drought events as the comparison benchmark, the groundwater drought events identified through time and space can completely cover the occurrence period of 17 actual drought events (accounting for 65.4%), and partially cover 5 events. The occurrence period of actual drought events (accounting for 19.2%) was not covered by the occurrence period of 4 actual drought events (accounting for 15.4%). The identified groundwater drought events reflected the occurrence of actual drought events well.

Corresponding to the above method, embodiments of the present application also provide a groundwater drought identification device for uneven site distribution. As shown in Figure 5, the device includes:

The acquisition unit 510 is used to obtain the groundwater data of the target research area in the historical time period; the groundwater data is converted from the original monitoring groundwater data in the historical time period in the target research area based on the elevation system of the preset historical time point. Obtained; the groundwater data includes collection location, collection time, groundwater elevation and groundwater level value;

The processing unit 520 is configured to, after gridding the target research area, reconstruct the groundwater level data of each grid point in the historical time period based on the machine learning algorithm and the inverse distance interpolation method to obtain the Describe the regional groundwater level sequence of the target study area;

And, based on a preset standardized precipitation index construction algorithm, standardize the regional groundwater level sequence to obtain a standardized groundwater level index sequence;

The determination unit 530 is configured to process the standardized groundwater level index sequence based on the run theory, determine the groundwater drought event in the target study area, and determine the characteristics of the groundwater drought event.

The functions of each functional unit of the groundwater drought identification device for uneven site distribution provided by the above embodiments of the present application can be realized through the above method steps. Therefore, the groundwater drought identification for uneven site distribution provided by the embodiment of the present application The specific working processes and beneficial effects of each unit in the device will not be described again here.

The embodiment of the present application also provides an electronic device, as shown in Figure 6 , including a processor 610, a communication interface 620, a memory 630, and a communication bus 640. The processor 610, the communication interface 620, and the memory 630 communicate through the communication bus 640. complete mutual communication.

Memory 630, used to store computer programs;

The processor 610 is used to execute the program stored on the memory 630 to implement the following steps:

Obtain the groundwater data of the target research area in the historical time period; the groundwater data is obtained by converting the original monitoring groundwater data of the historical time period in the target research area based on the elevation system of the preset historical time point; the groundwater Data include collection location, collection time, groundwater elevation, and groundwater table values;

After gridding the target research area, based on the machine learning algorithm and inverse distance interpolation method, the groundwater level data of each grid point in the historical time period is reconstructed to obtain the area of the target research area. Groundwater level sequence;

Based on the preset standardized precipitation index construction algorithm, standardize the regional groundwater level sequence to obtain a standardized groundwater level index sequence;

Based on the run theory, the standardized groundwater level index sequence is processed to determine the groundwater drought events in the target study area and determine the characteristics of the groundwater drought events.

The communication bus mentioned above may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into address bus, data bus, control bus, etc. For ease of presentation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

The communication interface is used for communication between the above-mentioned electronic devices and other devices.

The memory may include random access memory (Random Access Memory, RAM) or non-volatile memory (Non-Volatile Memory, NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located far away from the aforementioned processor.

The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (Network Processor, NP), etc.; it can also be a digital signal processor (Digital SignalProcessing, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

Since the problem-solving implementation and beneficial effects of each component of the electronic device in the above embodiments can be achieved by referring to the steps in the embodiment shown in FIG. 1, therefore, the specific working process and beneficial effects of the electronic device provided by the embodiments of the present application are The effect will not be described in detail here.

In yet another embodiment provided by this application, a computer-readable storage medium is also provided. The computer-readable storage medium stores instructions, which when run on a computer, cause the computer to execute any one of the above embodiments. The described groundwater drought identification method for uneven site distribution.

In yet another embodiment provided by the present application, a computer program product containing instructions is also provided, which when run on a computer causes the computer to perform the groundwater analysis for uneven site distribution described in any of the above embodiments. Drought identification methods.

Those skilled in the art should understand that the embodiments in the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the embodiments of the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment that combines software and hardware aspects. Furthermore, embodiments of the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. .

The embodiments of the present application are described with reference to flowcharts and/or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present application. It will be understood that each process and/or block in the flowchart illustrations and/or block diagrams, and combinations of processes and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions executed by the processor of the computer or other programmable data processing device produce a use A device for realizing the functions specified in a process or processes in a flowchart and/or a block or blocks in a block diagram.

These computer program instructions may also be stored in a computer-readable memory that causes a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including the instruction means, the instructions The device implements the functions specified in a process or processes in the flowchart and/or in a block or blocks in the block diagram.

These computer program instructions may also be loaded onto a computer or other programmable data processing device, causing a series of operating steps to be performed on the computer or other programmable device to produce computer-implemented processing, thereby executing on the computer or other programmable device. Instructions provide steps for implementing the functions specified in a process or processes of a flowchart diagram and/or a block or blocks of a block diagram.

Although the preferred embodiments among the embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once the basic inventive concepts are understood. Therefore, the appended application is intended to be construed to include the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the application.

Obviously, those skilled in the art can make various changes and modifications to the embodiments of the present application without departing from the spirit and scope of the embodiments of the present application. In this way, if these modifications and variations of the embodiments of the present application fall within the scope of the embodiments of the present application and equivalent technologies, then the embodiments of the present application are also intended to include these modifications and variations.

Claims (10)

1. A groundwater drought identification method oriented to uneven site distribution, characterized in that the method includes: Obtain the groundwater data of the target research area in the historical time period; the groundwater data is obtained by converting the original monitoring groundwater data of the historical time period in the target research area based on the elevation system of the preset historical time point; the groundwater Data include collection location, collection time, groundwater elevation, and groundwater table values; After gridding the target research area, based on the machine learning algorithm and inverse distance interpolation method, the groundwater level data of each grid point in the historical time period is reconstructed to obtain the area of the target research area. Groundwater level sequence; Based on the preset standardized precipitation index construction algorithm, standardize the regional groundwater level sequence to obtain a standardized groundwater level index sequence; Based on the run theory, the standardized groundwater level index sequence is processed to determine the groundwater drought events in the target study area and determine the characteristics of the groundwater drought events. 2. The method according to claim 1, characterized in that before obtaining the groundwater data of the target study area in the historical time period, the method further includes: Obtain original monitoring groundwater data for the historical time period in the target research area; The 3sigma processing algorithm is used to perform outlier processing on the original monitored groundwater data to obtain the groundwater data. 3. The method of claim 1, characterized in that, based on a machine learning algorithm and an inverse distance interpolation method, the groundwater level data of each grid point in the historical time period is reconstructed to obtain the target research area. The regional groundwater level series includes: The groundwater level data of each grid point in the historical time period is input into the trained XGBoost algorithm model to obtain the regional groundwater level monthly sequence output by the XGBoost algorithm model. The regional groundwater level monthly sequence includes the Groundwater level data for each month during the historical period; The inverse distance interpolation method is used to calculate the groundwater data corresponding to the n nearest groundwater level monitoring stations in the target research area, and obtain the daily groundwater data corresponding to each grid point; Based on the daily groundwater data corresponding to each grid point, calculate the monthly moving average water level value, and then subtract the monthly moving average water level value from the daily groundwater level value corresponding to each grid point to obtain the regional groundwater level of the target study area. Intra-monthly series, the intra-monthly series of regional groundwater levels includes intra-monthly fluctuation data of groundwater levels trending on a monthly scale within the historical time period; The regional groundwater level intermonthly sequence and the data of the corresponding position in the regional groundwater level intramonthly sequence are added to obtain the regional groundwater level sequence of the target research area, and the regional groundwater level sequence includes the historical time period. Daily groundwater level data for each grid point within the grid. 4. The method of claim 1, characterized in that, based on a preset standardized precipitation index construction algorithm, the regional groundwater level sequence is standardized to obtain a standardized groundwater level index sequence, including: Based on the groundwater level data corresponding to each grid point in the regional groundwater level sequence, calculate the multi-year moving average water level value, and then subtract the multi-year moving average water level value from the groundwater level value corresponding to each grid point to obtain the target research area Last year's trend of groundwater level series, said last year's trend of groundwater level series includes groundwater level fluctuation data with the annual trend removed; After standardizing the groundwater level sequence of last year's trend, the standardized groundwater level sequence of last year's trend is processed with reference to the construction process of the standardized precipitation index to obtain a standardized groundwater level index sequence. 5. The method of claim 1, characterized in that, based on run theory, the standardized groundwater level index sequence is processed to determine groundwater drought events in the target study area, including: Based on the run theory, if the standardized groundwater level index value at each time point in the standardized groundwater level index sequence is less than or equal to the preset drought threshold, then the grid point corresponding to the corresponding standardized groundwater level index value is determined as a drought grid point ; Traverse the target research area, and spatially merge the target drought grid points in the determined drought grid points to obtain drought patches, where the target drought grid points are adjacent drought grid points; The drought patches at adjacent time nodes are spatially superimposed to determine the overlapping area of the overlapping area, and the drought patches whose overlapping area is greater than the preset area are determined to be the same groundwater drought event and assigned the same drought event. serial number. 6. The method of claim 1, further comprising: For two groundwater drought events with an interval of one month, if there is a spatial intersection between the end time point of the previous groundwater drought event and the start time point of the subsequent groundwater drought event, and the spatial union of the interval months The drought index value is less than the first drought index threshold, then the two groundwater drought events will be merged into the same groundwater drought event; For a groundwater drought event that lasts only one month, if the drought index value is greater than the second drought index threshold, the groundwater drought event will be eliminated. 7. The method of claim 1, wherein determining characteristics of groundwater drought events includes: Perform feature extraction on the groundwater drought event to obtain characteristics of the groundwater drought event; The groundwater drought event characteristics include drought start time, drought end time, drought duration, average drought area, maximum drought area, drought coverage area, average drought intensity, maximum drought intensity and drought intensity. 8. A groundwater drought identification device for uneven site distribution, characterized in that the device includes: The acquisition unit is used to obtain the groundwater data of the target research area in the historical time period; the groundwater data is obtained by converting the original monitoring groundwater data of the historical time period in the target research area based on the elevation system of the preset historical time point. ; the groundwater data includes collection location, collection time, groundwater elevation and groundwater level value; The processing unit is used to reconstruct the groundwater level data of each grid point in the historical time period based on the machine learning algorithm and the inverse distance interpolation method after gridding the target research area to obtain the Regional groundwater level series for the target study area; And, based on a preset standardized precipitation index construction algorithm, standardize the regional groundwater level sequence to obtain a standardized groundwater level index sequence; The determination unit is used to process the standardized groundwater level index sequence based on run theory, determine groundwater drought events in the target study area, and determine characteristics of groundwater drought events. 9. An electronic device, characterized in that the electronic device includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete communication with each other through the communication bus; Memory, used to store computer programs; A processor configured to implement the method described in any one of claims 1-7 when executing a program stored on the memory. 10. A computer-readable storage medium, characterized in that a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the method of any one of claims 1-7 is implemented.