Method for identifying spatiotemporal migration of hydrological drought based on distributed hydrological model and barycenter model

Abstract

The application discloses a kind of methods for identifying the spatiotemporal migration of hydrological drought based on distributed hydrological model and gravity center model, comprising the following steps: step 1: collecting and processing long time series data of study area;Step 2: extracting hydrological parameters in the DEM layer of study area;Step 3: obtaining calibrated hydrological model;Step 4, input the hydrological and meteorological observation data of the region in step 1 collated at a certain period, simulate to obtain the runoff of each sub-basin at the corresponding period;Step 5, obtain the drought index sequence of the study area by using the standardized drought index method;Step 6, obtain the spatial distribution of the study area drought by using the inverse distance weighted spatial interpolation method;Step 7, according to the drought index sequence obtained in step 6, each regional comprehensive drought identification is carried out;Step 8, according to the gravity center migration method, the spatiotemporal variation characteristics of drought are analyzed.The application can more accurately capture the change trend of drought event with time and space, and improve the accuracy of drought forecast.

Description

Technical Field

[0001] This invention relates to a method for identifying the spatiotemporal migration of hydrological drought based on distributed hydrological models and centroid models, belonging to the field of hydrological drought characteristic analysis. Background Technology

[0002] Currently, most studies on hydrological drought focus on the relatively simple transmission time between hydrological drought and other types of drought, as well as the temporal or spatial distribution of drought. They analyze the influence relationships between different types of drought and provide an overall understanding of the spatiotemporal distribution attributes of drought. However, these research methods neglect the dynamic nature of drought itself. Furthermore, when studying the spatial distribution characteristics of hydrological drought, they often use spatially unevenly distributed hydrological stations as supporting points for analyzing the spatial distribution characteristics of hydrological drought. The resulting analysis is not comprehensive enough. Because the spatial location of drought is uncertain and frequently migrates, it significantly impacts drought prediction and prevention. Drought migration analysis is generally based on the spatial distribution characteristics of drought at different points in time. Therefore, clarifying the specific spatiotemporal distribution characteristics of hydrological drought and exploring more comprehensive scientific methods for drought migration analysis is of great significance for improving the efficiency of drought relief and disaster mitigation, and for more rationally planning and allocating water resources. Summary of the Invention

[0003] Purpose of the invention: In order to overcome the shortcomings of the existing technology, the present invention provides a method for identifying the spatiotemporal migration of hydrological drought based on a distributed hydrological model and a centroid model. By combining the distributed hydrological model and the centroid model to analyze the spatiotemporal migration of drought characteristics, the method can more accurately capture the changing trend of drought events over time and space, thereby improving the accuracy of drought forecasting.

[0004] Technical Solution: To solve the above-mentioned technical problems, the present invention provides a method for identifying the spatiotemporal migration of hydrological drought based on a distributed hydrological model and a centroid model, comprising the following steps:

[0005] Step 1: Collect and process long-term hydrological and meteorological observation data, land use data, soil data, and DEM data for the study area; hydrological and meteorological observation data include daily minimum temperature, maximum temperature, precipitation, sunshine duration, wind speed, relative humidity, and runoff data;

[0006] Step 2: Extract hydrological parameters from the DEM layer of the study area, mainly including the watershed river network, sub-watershed sub-basins (SUB), and hydrological response units (HRU).

[0007] Step 3: Based on the various data from Step 1 and the sub-basins and response units divided in Step 2, the parameters of the hydrological model are calibrated to obtain the calibrated hydrological model.

[0008] Step 4: Based on the calibration model obtained in Step 3, input the hydrological and meteorological observation data of the region for a certain period of time compiled in Step 1, and simulate the runoff of each sub-basin for the corresponding period of time.

[0009] Step 5: Based on the runoff data obtained in Step 4, the drought index sequence for the study area is derived using the standardized drought index method.

[0010] Step 6: Based on the drought index sequence obtained in Step 5, the spatial distribution of drought in the study area is obtained using the inverse distance weighted spatial interpolation method;

[0011] Step 7: Based on the drought index sequence obtained in Step 6, perform comprehensive drought identification for each region;

[0012] Step 8: Statistically analyze the drought characteristic values ​​obtained in Step 7, and generate spatial distribution maps of different drought frequencies and centroid migration results based on the centroid migration method.

[0013] Preferably, in step 3, the hydrological model is the SWAT model. The simulation of hydrological processes in the SWAT model is based on the Hydrological Response Unit (HRU), and is achieved by considering the spatiotemporal variability caused by watershed climate and underlying surface factors. The basic water balance equation for its simulation is:

[0014]

[0015] Where: SW t SW0 represents the soil moisture content at the end of the period, in mm; t represents the soil moisture content at the beginning of the period, in mm; R represents the hydrological process time, in days. day Q represents the precipitation on day i, in mm. surf E is the surface runoff on day i, in mm; E is the evaporation on day i, in mm; W seep Q represents the infiltration and lateral flow through the soil layer on day i, in mm. gw The groundwater return flow rate on day i is expressed in mm.

[0016] Preferably, the drought index in step 6 is the Standardized Runoff Index (SRI) and the Standardized Precipitation Index (SPI).

[0017] Preferably, the drought level of the drought index in step 7 is classified according to the "Meteorological Drought Level" (GB / T20481-2006).

[0018] Preferably, in step 8, the characteristic values ​​of drought at each level are statistically analyzed, and the spatiotemporal variation characteristics of the center of gravity migration are analyzed according to the time sequence. The method for calculating the center of gravity is as follows: First, assuming that a certain region is composed of n sub-regions i, then the "center of gravity" of a certain attribute of this region is usually calculated using the following formula:

[0019]

[0020]

[0021] In the formula, X and Y are the longitude and latitude values ​​of the "center of gravity" of a certain attribute in a certain region;

[0022] x i y i Here are the longitude and latitude values ​​of the i-th sub-region center;

[0023] M i Let be the value of a certain attribute in the i-th sub-region.

[0024] Beneficial effects: This invention proposes a method for identifying the spatiotemporal migration of hydrological drought based on a distributed hydrological model and a centroid model. Compared with existing technologies, it has the following advantages:

[0025] (1) Comprehensive analysis capability: By combining distributed hydrological models and centroid models, this method can simultaneously consider the spatial heterogeneity of the hydrological cycle and the spatial distribution of drought characteristics, providing a more comprehensive drought analysis from multiple dimensions. This method not only assesses the severity of drought but also analyzes the spatial distribution and migration paths of drought, providing multi-dimensional information for drought risk assessment.

[0026] (2) High-precision spatial identification: Distributed hydrological models integrate data from multiple sources, including meteorological data, hydrological data, and remote sensing data, which can capture the hydrological response of different sub-regions within the basin, making the identification of drought characteristics more accurate, especially under complex terrain and variable climate conditions.

[0027] In summary, by combining distributed hydrological models and centroid models to analyze the spatiotemporal migration of drought characteristics, this invention can more accurately capture the changing trends of drought events over time and space, improve the accuracy of drought forecasting, and provide drought spatiotemporal migration information to help formulate long-term water resource management and drought response strategies, thus providing more scientific decision support for drought risk management. Attached Figure Description

[0028] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0029] Figure 1 Here is a logical structure block diagram of the method of the present invention:

[0030] Figure 2 The figure shows the spatial distribution and center of gravity shift of drought at different frequencies. In the figure, a, b, c, and d represent the spatial distribution and center of gravity shift of drought during mild drought, moderate drought, severe drought, and extreme drought, respectively. Detailed Implementation

[0031] The invention will now be further described with reference to the accompanying drawings.

[0032] like Figure 1 and Figure 2 As shown, the present invention provides a method for identifying the spatiotemporal migration of hydrological drought based on a distributed hydrological model and a centroid model, which specifically includes the following steps:

[0033] Step 1: Collect and process long-term hydrological and meteorological observation data, land use data, soil data, and DEM data for the study area. Considering that the standardized drought index method requires data on a monthly time scale, the collected long-term hydrological and meteorological information should be on a monthly or smaller time scale (e.g., daily, ten-day). For smaller time scale data such as daily and ten-day data, organize them into monthly data using time aggregation methods.

[0034] Step 2: Input the various data collected in Step 1 into the SWAT hydrological model. The SWAT (Soil and Water Assessment Tool) model is a widely used distributed hydrological model developed with support from the U.S. Department of Agriculture (USDA) to assess hydrological cycling and chemical transport at the watershed scale. Because the SWAT model considers spatial variability, it allows users to divide sub-basins and hydrological response units according to actual watershed characteristics, thereby simulating hydrological processes more accurately. This invention specifically adopts the SWAT hydrological model.

[0035] Since the main consideration is runoff variation, the hydrological process sub-model in the SWAT model is used.

[0036] The SWAT model simulates hydrological processes based on hydrological response units (HRUs), taking into account the spatiotemporal variability caused by factors such as watershed climate and underlying surface. The fundamental water balance equation for this simulation is:

[0037]

[0038] Where: SW t The final soil moisture content (mm);

[0039] SW0 represents the initial soil moisture content (mm);

[0040] t represents the hydrological process time (d);

[0041] R day The precipitation on day i is (mm).

[0042] Q surf Let be the surface runoff of the i-th day (mm);

[0043] E represents the evaporation amount on day i (mm);

[0044] W seep The infiltration and lateral flow rate through the soil layer on day i are (mm).

[0045] Q gw Let be the groundwater return flow rate (mm) on day i.

[0046] Using relative error R e The coefficient of determination (R²) and the Nash-Sutcliffe efficiency coefficient (NSE) are used to evaluate model performance.

[0047] R e R represents the relative error between the simulated value and the measured value. e A value greater than 0 indicates that the simulated value is too large, while a value greater than 0 indicates that the simulated value is too small.

[0048] R² is used to represent the degree of linear correlation between the observed sequence and the simulated sequence. The closer R² is to 1, the stronger the correlation between the simulated value and the measured value.

[0049] NSE is used to measure how well the simulation results fit the observed values. The closer the value is to 1, the better the simulation is.

[0050] R e The formulas for calculating R² and NSE coefficients are as follows:

[0051]

[0052]

[0053]

[0054] In the formula, O i and These represent the measured value and its mean, respectively.

[0055] S i and These represent the simulated value and its mean, respectively.

[0056] n represents the number of measured data points.

[0057] It is generally believed that when the model relative error |R e When | < 20%, coefficient of determination R² > 0.6, and NSE > 0.5, the simulation results of the model are reliable.

[0058] Step 3: Based on the calibration model obtained in Step 2, input the hydrological and meteorological observation data of the region for a certain period of time compiled in Step 1, and simulate the runoff of each sub-basin for the corresponding period of time.

[0059] Step 4: Based on the runoff data of each sub-basin obtained in Step 3, the standardized drought index method is used to deduce the drought index sequence of each type in the study area.

[0060] Step 5: Based on the drought index sequence obtained in Step 4, the spatial distribution of drought in the study area is obtained by using the inverse distance weighted spatial interpolation method.

[0061] Considering the advantages of the Standardized Runoff Index (SRI) in terms of ease of calculation and suitability for temporal and spatial comparisons at multiple scales, this invention selects the SRI index to evaluate hydrological drought. The calculation steps are as follows:

[0062] Assuming that the runoff x over a certain time period follows a Γ distribution, the probability density function f(x) of this distribution is:

[0063]

[0064] In the formula: and These are shape parameters and scale parameters; x, and All are greater than 0; and It can be estimated using the maximum likelihood method.

[0065]

[0066]

[0067]

[0068] The cumulative probability of runoff x over a certain time scale is:

[0069]

[0070] Normalization of the cumulative probability F(x) yields:

[0071]

[0072]

[0073] In the above formula: when F≤0.5, S=-1; when F>0.5, S=1; in addition, c0=2.515517, c1=0.802853, c2=0.010328, d1=1.432788, d2=0.189269, d3=0.001308 are all empirical values.

[0074] Step 6: Based on the drought index sequence obtained in Step 5, identify the comprehensive drought in each region.

[0075] The drought level of the drought index is classified according to the "Meteorological Drought Level" (GB / T20481-2006). See Table 1 below for details:

[0076] Table 1. Classification Criteria for Hydrological Drought Levels

[0077]

[0078] Step 7: Statistically analyze the drought frequency obtained in Step 6, and analyze the spatiotemporal variation characteristics of drought using the centroid migration method.

[0079] The Gravity Migration Method is a method used to analyze and describe the spatial distribution changes and migration trends of objects. It analyzes the migration by calculating the change in the spatial "center of gravity" position of a certain attribute or feature.

[0080] Based on the characteristics of the research period, it was divided into several periods.

[0081] For example, the research period is 59 years from 1961 to 2019, which can be divided into 6 periods according to the years: 1960s (1961-1969), 1970s (1970-1979), 1980s (1980-1989), 1990s (1990-1999), 2000s (2000-2009), and 2010s (2010-2019).

[0082] The centroid coordinates of drought at different times were calculated according to the hydrological drought classification standards (Table 1) for four drought levels: mild drought, moderate drought, severe drought, and extreme drought. The calculation formula is as follows:

[0083]

[0084] The frequency of occurrence P is the ratio of the number of drought events n (SRI≤-0.5) to the total number of months N.

[0085] Before calculating the centroid of drought frequency, the latitude and longitude of the centroids of different sub-basins are obtained, and the hydrological drought index of each sub-basin is classified into levels. Sub-basins corresponding to each drought type are selected based on different levels. Then, the centroid coordinates of different drought frequencies are calculated based on the latitude and longitude of the centroids of each sub-basin. Taking mild drought as an example, the centroid coordinates (X...) of the mild drought level are calculated. 轻 ,Y 轻 ):

[0086]

[0087] In the formula, n is the total number of sub-basins experiencing mild drought;

[0088] x i y i p i Let be the longitude, latitude, and frequency of mild drought at the centroid of the i-th sub-basin.

[0089] The formula for calculating the center of gravity migration distance in different years is as follows:

[0090]

[0091]

[0092] Among them, D s-k Indicates the distance the center of gravity has moved over different years;

[0093] A s-k Indicates the angle of shift of the center of gravity over different years;

[0094] s and k represent two different years; (X) s ,Y s ) and (X k ,Y k Let X and K represent the spatial geographic coordinates of the centroids of different drought levels in year s and year k, respectively. C is a constant with a value of 111.111, which is the coefficient for converting geographic coordinate units (1°) to planar distances (km). C*(X) s -X k ), C*(Y s -Y k The numbers () represent the actual distances traveled along the longitude and latitude of different drought centroid frequencies from year k to year s, respectively. The calculation methods for the decadal centroid and its migration distance are similar. The spatial distribution and centroid migration results for different drought frequencies are shown below. Figure 2 As shown.

[0095] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for identifying the spatiotemporal migration of hydrological drought based on a distributed hydrological model and a centroid model, characterized in that, Includes the following steps: Step 1: Collect and process long-term hydrological and meteorological observation data, land use data, soil data, and DEM data for the study area; hydrological and meteorological observation data include daily minimum temperature, maximum temperature, precipitation, sunshine duration, wind speed, relative humidity, and runoff data; Step 2: Extract hydrological parameters from the DEM layer of the study area, including the watershed network, sub-watershed sub-basins (SUB), and hydrological response units (HRU). Step 3: Based on the various data from Step 1 and the sub-basins and response units divided in Step 2, the parameters of the hydrological model are calibrated to obtain the calibrated hydrological model. Step 4: Based on the calibration model obtained in Step 3, input the hydrological and meteorological observation data of the region for a certain period of time compiled in Step 1, and simulate the runoff of each sub-basin for the corresponding period of time. Step 5: Based on the runoff data obtained in Step 4, the drought index sequence for the study area is derived using the standardized drought index method. Step 6: Based on the drought index sequence obtained in Step 5, the spatial distribution of drought in the study area is obtained using the inverse distance weighted spatial interpolation method; Step 7: Based on the drought index sequence obtained in Step 6, perform comprehensive drought identification for each region; Step 8: Statistically analyze the drought characteristic values ​​obtained in Step 7, and generate spatial distribution maps of different drought frequencies and centroid migration results based on the centroid migration method.

2. The method for identifying the spatiotemporal migration of hydrological drought based on a distributed hydrological model and a centroid model according to claim 1, characterized in that: In step 3, the hydrological model is the SWAT model. The simulation of hydrological processes in the SWAT model is based on the hydrological response unit (HRU), which is achieved by considering the spatiotemporal variability caused by watershed climate and underlying surface factors. The basic water balance equation for the simulation is: Where: SW t SW0 represents the soil moisture content at the end of the period, in mm; t represents the soil moisture content at the beginning of the period, in mm; R represents the hydrological process time, in days. day Q represents the precipitation on day i, in mm. surf The surface runoff of the i-th day is expressed in mm. E represents the evaporation on day i, in mm; W seep Q represents the infiltration and lateral flow through the soil layer on day i, in mm. gw The groundwater return flow rate on day i is expressed in mm.

3. The method for identifying the spatiotemporal migration of hydrological drought based on a distributed hydrological model and a centroid model according to claim 1, characterized in that: The drought index mentioned in step 6 is the Standardized Runoff Index (SRI) and the Standardized Precipitation Index (SPI).

4. The method for identifying the spatiotemporal migration of hydrological drought based on a distributed hydrological model and a centroid model according to claim 1, characterized in that: In step 7, the drought level of the drought index is classified according to the "Meteorological Drought Level" (GB / T20481-2006).

5. The method for identifying the spatiotemporal migration of hydrological drought based on a distributed hydrological model and a centroid model according to claim 1, characterized in that: In step 8, the characteristic values ​​of drought at each level are statistically analyzed, and the spatiotemporal variation characteristics of the center of gravity migration are analyzed according to the time sequence. The method for calculating the center of gravity is as follows: First, assuming that a certain region is composed of n sub-regions i, then the "center of gravity" of a certain attribute of this region is usually calculated using the following formula: In the formula, X and Y are the longitude and latitude values ​​of the "center of gravity" of a certain attribute in a certain region; x i y i Here are the longitude and latitude values ​​of the i-th sub-region center; M i Let be the value of a certain attribute of the i-th sub-region.

Images