Method for predicting influence of different climate changes on shallow groundwater seawater intrusion and application thereof

Abstract

The present application belongs to the technical field of numerical simulation, and discloses a method for predicting the seawater intrusion of shallow groundwater under the influence of different climate changes, comprising the following steps: (1) establishing the hydrogeological concept of the study area, determining the source and sink items, the initial conditions of the groundwater flow field and the salinity field; (2) using the groundwater model GMS to establish the groundwater numerical model and the groundwater solute transport model of the study area and performing identification verification; (3) inputting the climate data after the Cmhyd downscaling into the groundwater solute transport model for simulation, obtaining the predicted groundwater salinity distribution and dynamic change, and analyzing the influence of different climate factors on the seawater intrusion of groundwater. The present application considers the influence of global warming, and the precipitation and temperature change with time instead of being a fixed value, so that the future climate can be better predicted; when simulating the seawater intrusion of shallow groundwater, the changing precipitation and temperature can also comprehensively understand how the climate change influences the degree of the seawater intrusion of shallow groundwater.

Description

Technical Field

[0001] This invention belongs to the field of numerical simulation technology, and in particular to a method and application for predicting the impact of different climate changes on shallow groundwater seawater intrusion. Background Technology

[0002] Future climate change can influence the degree of seawater intrusion into shallow groundwater by altering precipitation and evaporation, with evaporation being represented by maximum and minimum temperatures. Therefore, predicting future precipitation and temperature can help analyze how climate change affects seawater intrusion into groundwater.

[0003] Currently, Cmhyd, a climate model used for hydrological simulation, is a calibration tool that can extract and calibrate data from global and regional climate models. After referencing climate models reflecting global warming, Cmhyd provides different bias correction methods for precipitation and temperature, and downscales overly high-resolution GCM models, providing the downscaled climate data to groundwater solute transport models. The climate data undergoes a series of statistical processing steps, eliminating data that deviates from climate change in the study area, thus providing more realistic climate data for simulating seawater intrusion into shallow groundwater. However, existing techniques often directly change precipitation and evaporation by percentage, or set different climate scenarios based on historical wet, dry, and high-water years, inputting climate data into groundwater solute transport models for impact analysis. While these methods can obtain the impact of climate change on seawater intrusion into shallow groundwater, the changes are fixed values ​​that do not change over time, lacking the rationality and realism of climate condition changes in the study area.

[0004] Therefore, there is an urgent need for one or more new methods to predict the impact of different climate changes on shallow groundwater seawater intrusion. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for predicting the impact of different climate changes on shallow groundwater seawater intrusion and its application.

[0006] The technical solution adopted by this invention to solve its technical problem is:

[0007] A method for predicting the impact of different climate changes on shallow groundwater seawater intrusion, the method comprising the following steps:

[0008] (1) Establish a hydrogeological conceptual model of the study area, determine the boundary range of the study area, and analyze the topography, natural environment and hydrogeological conditions of the study area through literature review and field investigation, so as to generalize the aquifer, groundwater and boundary conditions, and determine the initial conditions of source and sink terms, groundwater flow field and salinity field.

[0009] (2) A numerical model of groundwater flow and a model of groundwater solute transport in the study area were established using the groundwater model GMS. For the groundwater flow numerical model, the simulated groundwater level and the groundwater level observation well data were compared to ensure that the simulated groundwater level value was basically the same as the observed value and that the simulated groundwater flow field was close to the measured groundwater flow field (for groundwater level contour lines with the same value, the error between the simulation result and the measured data did not exceed 3% of the model length). For the groundwater solute transport model, the simulated groundwater salinity and the groundwater salinity observation well data were compared to ensure that the simulated groundwater salinity value was basically the same as the observed value and that the simulated groundwater salinity field was close to the measured groundwater salinity field (for groundwater level contour lines with the same value, the error between the simulation result and the measured data did not exceed 3% of the model length).

[0010] (3) Using multiple GCM models that researchers have used in the study area and can reproduce historical temperature and precipitation well, the climate data downscaled by Cmhyd is input into the groundwater solute transport model for simulation, and the predicted groundwater salinity distribution and dynamic changes are obtained. The influence of different climate factors on groundwater seawater intrusion is analyzed.

[0011] Furthermore, the hydrogeological conceptual model in step (1) specifically includes the following steps:

[0012] 1) Based on the research needs and the hydrogeological conditions of the study area, select complete hydrogeological units to determine the boundary range of the study area;

[0013] 2) Based on the properties, structure, and lithology of aquifers, aquifers are generalized into homogeneous or heterogeneous aquifers, isotropic or anisotropic aquifers, impermeable or poorly permeable aquifers.

[0014] 3) Groundwater is generalized into steady or unsteady flow, unconfined or confined water;

[0015] 4) Based on the geological structure of aquifers and impermeable layers, the characteristics of boundary flow, and the hydraulic connection between water bodies, the boundary conditions are generalized into three categories: the first category is a given head boundary that determines the head at each moment; the second category is a given flow boundary that determines the flow rate per unit area; and the third category is a mixed boundary that considers both head and flow rate.

[0016] Furthermore, the numerical model for groundwater flow in step (2) is as follows:

[0017]

[0018] In the formula: K xx K yy K zz: Permeability coefficients (m / d) in the x, y and z directions, respectively; S: Storage rate (1 / m); h: Water head (m); w: Source and sink terms of the aquifer (1 / d); t: Time.

[0019] Furthermore, the construction steps of the groundwater flow numerical model include the following steps:

[0020] 1) Time Discreteness and Spatial Discreteness

[0021] Based on the established study area boundary, the study area boundary is drawn in GMS. The size and number of model cells are determined according to the size of the study area. The number of model layers is determined according to the hydrogeological conditions and aquifer thickness of the study area. The elevations of the top and bottom plates of each aquifer after model layering and the initial groundwater head are collected, imported into GMS and interpolated. The stress period and simulation start and end times are determined.

[0022] 2) Determine the hydrogeological parameters of the model

[0023] Based on the hydrogeological survey data and pumping test data of the study area, and combined with the topography, geological map and hydrogeological map of the study area, the study area is divided into hydrogeological parameter zones.

[0024] Initial hydrogeological parameters, including horizontal / vertical hydrodynamic conductivity, water storage capacity, water yield, and porosity of weakly permeable layers and aquifers, were determined by empirical hydrogeological parameters and data from research papers on the study area.

[0025] 3) Determine the source and sink items

[0026] (1) Based on the hydrogeological map, combined with the precipitation infiltration recharge coefficient of different topographic and geomorphic units, and taking into account the surface soil quality, underlying surface conditions, and surface water depth, precipitation infiltration and agricultural irrigation zones were divided. The precipitation and agricultural irrigation infiltration recharge in the study area were treated as areal recharge and imported into the groundwater flow numerical model through Recharge Package.

[0027] (2) Based on the data, the starting and ending elevations of the riverbed, the water level and the hydraulic conductivity of the river section in the study area were determined. The infiltration of the river in the study area was treated as a surface recharge and imported into the groundwater flow numerical model through River Package.

[0028] (3) The groundwater extraction volume and the distribution of extraction wells are determined based on data, and appropriate well groups are merged and imported into the groundwater flow numerical model;

[0029] (4) Evaporation as a factor in the loss of shallow groundwater under given climatic conditions

[0030] Evaporation determination: Based on the data, the evaporation rate, groundwater evaporation depth and surface evaporation elevation of the study area were determined, and the data were imported into the groundwater flow numerical model using EVT Package;

[0031] 4) Model identification and validation

[0032] By using national or organizational groundwater monitoring networks or individual field monitoring, the location of observation points is determined and water level data is collected. This data is then imported into the GMS model, the Observation Point function is activated, observation well data is input, and the groundwater flow numerical model is run. Hydrogeological parameters and source-sink values ​​are continuously changed and repeated until the difference between the calculated groundwater level and the observed groundwater level is controlled within 10% of the annual groundwater level fluctuation.

[0033] After running the groundwater flow numerical model, each observation point will display a colored band, with the middle point as the observed value, the upper end as the observed value plus the range value, and the lower end as the observed value minus the range value. The applicability of the model is judged by observing and calculating the degree of fit between the simulated value and the observed value, and the absolute error of the groundwater level value during the simulation of the monitoring point should be less than the actual water level fluctuation of the corresponding simulation period of the monitoring point.

[0034] Furthermore, the groundwater solute transport model in step (2) is calculated using the variable density groundwater flow equation and the solute transport equation, as follows:

[0035] The equation for the flow of variable-density groundwater is as follows:

[0036]

[0037] In the formula:

[0038] ρ: Density of groundwater [ML] -3 ]; ρ0: Density of fresh water at standard temperature [ML] -3 μ: Dynamic viscosity of groundwater [ML] -1 T -1 μ0: Dynamic viscosity of fresh water [ML] -1 T -1 K0: Hydraulic conduction tensor [LT] -1 ]; z: vertical coordinate [L]; h0: freshwater head [L]; S s ′ ,0 Specific water storage ratio [L] -1 ]; t: time [T]; θ: porosity [-]; C: salt concentration [ML] -3 ];q′ S : has density ρ s Source and sink items [T] -1 ];

[0039] The solute transport equation is as follows:

[0040]

[0041] In the formula:

[0042] ρ b Bulk density [ML] -3 ]; Distribution coefficient of type k [L] 3 M -1 ];C k Concentration of species k [ML] -3 D: Hydrodynamic dispersion tensor [L] -2 T -1 Q: Flow rate [LT] -1 ]; Source or sink concentration of species k [ML] -3 ].

[0043] Furthermore, the construction steps of the groundwater solute transport model are as follows:

[0044] Based on the numerical model of groundwater flow, the MT3DMS and SEAWAT modules were added; the Advance package, Dispersion package, and Source / sink mixing package in the MT3DMS module were added, and the VDF package in the SEAWAT module was added. The physicochemical properties of salinity TDS, the solute dispersion, and the molecular dispersion coefficient were adjusted (the initial values ​​of the physicochemical properties of salinity TDS, the solute dispersion, and the molecular dispersion coefficient were determined based on empirical parameters, researchers' surveys of the study area, and research paper data, and were continuously updated) so that the simulated groundwater salinity values ​​were basically close to the measured values ​​(for groundwater level contour lines with the same values, the error between the simulation results and the measured data did not exceed 3% of the model length).

[0045] The measured solute data were imported into the GMS model using the 2D Scatter Data module of GMS and interpolated to obtain the initial concentration. The time and spatial discretization were the same as those of the groundwater flow numerical model. The source and sink terms were the same as those of the groundwater flow numerical model, but salinity source and sink terms were added to the river and ocean source and sink terms to obtain river and ocean salinity (such as data published by the local hydrological bureau, data from research papers by researchers, and empirical parameters). After running the model, the parameters were continuously adjusted to make the simulated groundwater salinity values ​​close to the measured values ​​(for groundwater level contour lines with the same value, the error between the simulation results and the measured data does not exceed 3% of the model length), and the simulated values ​​of groundwater salinity contour lines tend to be consistent with the measured values.

[0046] Using multiple GCM models that researchers have used in the study area and that can reproduce historical temperature and precipitation well, the climate data downscaled by Cmhyd was input into the groundwater solute transport model for simulation, and the predicted groundwater salinity distribution and dynamic changes were obtained. The impact of different climate factors on groundwater seawater intrusion was analyzed.

[0047] Furthermore, step (3) is specifically as follows:

[0048] Using the CMIP5 or CMIP6 climate datasets, future changes in climate variables can be predicted using the global climate model GCM based on greenhouse gas emission drivers. The model also considers the medium radiation forcing scenario RCP4.5 and the high radiation forcing scenario RCP8.5 in the representative concentration pathway RCP, and performs bias correction and downscaling using the CMhyd climate model for hydrological simulation.

[0049] Historically well-performing GCM models within the study area were selected, and bias corrections were performed using different bias correction methods from CMhyd. Each GCM model and each bias correction method were grouped together. The bias-corrected climate data for each group were compared with historical data, and the correlation R and root mean square (RMS) were calculated for each group. The Taylor Skill Score (TSS) is generated, and the TSS of each combination is compared. The precipitation or temperature combination with the highest TSS value in different GCM models is selected as the combination of predicted climate data.

[0050] Based on the project requirements, the prediction time was selected. After the combination with the highest TSS value was predicted using the Cmhyd climate model, climate combinations with climate conditions and extreme climate phenomena that had never occurred in history were removed. The data of the remaining combinations were then input into the groundwater solute transport model to construct groundwater solute transport prediction models under different climate scenarios and analyze the degree of groundwater seawater intrusion under the influence of different climate changes.

[0051] Furthermore, the deviation correction method includes:

[0052] For precipitation bias correction, including method precipitation and temperature distribution maps, linear scaling, precipitation dynamics conversion, and local precipitation intensity scaling;

[0053] For temperature deviation correction, methods include precipitation and temperature mapping, linear scaling, and temperature variance scaling.

[0054] Furthermore, the formula for calculating the Taylor Skill Score (TSS) is as follows:

[0055]

[0056] In the formula:

[0057] R0: The maximum achievable correlation, set to 1;

[0058] R, These are the model's correlation and the normalized root mean square, respectively.

[0059] Where TSS close to 1 indicates that the observed values ​​are closer to the simulated values, and conversely, decreasing towards 0 indicates that the model performance is worse; the model's correlation R and normalized root mean square are also relevant. It was calculated.

[0060] The methods described above are applied to predicting the impacts of different climate changes on shallow groundwater seawater intrusion.

[0061] The advantages and positive effects of this invention are as follows:

[0062] 1. The method of this invention takes into account the impact of global warming on precipitation and temperature. Precipitation and temperature change over time rather than being a fixed value, which can better predict future climate. When simulating shallow groundwater seawater intrusion, the changing precipitation and temperature can also provide a more comprehensive understanding of how climate change affects the degree of shallow groundwater seawater intrusion.

[0063] 2. This invention combines hydrogeological and physical geographical data of the study area to generalize the aquifer and boundary conditions of the model, determine the initial conditions such as source and sink terms, groundwater flow field, and salinity field (considering groundwater salinity transport to assess seawater intrusion), and establish groundwater flow and solute transport models for the study area using GMS, followed by identification and verification. Simultaneously, multiple historically used and well-performing GCM models in the study area are collected, and Cmhyd is used to statistically reduce and correct for precipitation and temperature. The climate scenarios from these multiple GCM models are then input into the groundwater solute transport model for groundwater salinity transport analysis to explore the extent to which climate change affects seawater intrusion.

[0064] 3. The method of this invention can obtain the salinity distribution of groundwater under different climate scenarios, thereby analyzing how climate change affects the degree of seawater intrusion into shallow groundwater; groundwater solute transport models can be constructed using groundwater simulation software such as FEFLOW and VisualMODFLOW; climate change prediction can be performed using software such as SDSM and WRF dynamic models for downscaling and bias correction.

[0065] 4. In the method of this invention, the GCM is represented by mathematical formulas of the main components of the climate system (atmosphere, land surface, ocean and ocean cryosphere) and their interactions. It calculates the equations of the changes of climate components over time and their mutual coupling effects. Since the output results of different GCMs become unstable due to the uncertainty of greenhouse gas emissions, the method of this invention aggregates multiple GCMs that have been used in the study area in the past and have performed well, so as to reduce the output error of each member and improve the prediction effect of future climate. Attached Figure Description

[0066] Figure 1 This is a flowchart of the method of the present invention;

[0067] Figure 2 This is a comparison diagram of simulated water level and measured water level of an observation well in an embodiment of the present invention (where a: observation well Q214010, b: observation well Q217010);

[0068] Figure 3 The above is a heatmap of precipitation TSS values ​​for five GCM models from 1990 to 2005 and their respective bias correction methods in this embodiment of the invention.

[0069] Figure 4 This is a simulation diagram of the total annual precipitation of five GCM models under the RCP4.5 scenario in this embodiment of the invention;

[0070] Figure 5 The groundwater salinity distribution under different precipitation patterns in the embodiments of the present invention (where a: 2050 baseline, b: ACCESS1-0 pattern);

[0071] Figure 6 This is a groundwater salinity distribution map under different evaporation modes in an embodiment of the present invention (where a: baseline in 2050, b: CSRIO-MK3-6-0 temperature mode). Detailed Implementation

[0072] The present invention will be further described below with reference to the embodiments. The following embodiments are descriptive and not limiting, and should not be used to limit the scope of protection of the present invention.

[0073] The various experimental operations involved in the specific embodiments are all conventional techniques in the field. For parts not specifically annotated in this document, those skilled in the art can refer to various commonly used reference books, scientific and technological documents or related instructions and manuals prior to the filing date of this invention to carry out the operations.

[0074] A method for predicting the impact of different climate changes on shallow groundwater seawater intrusion, the method comprising the following steps:

[0075] (1) Establish the hydrogeological concept of the study area, determine the boundary range of the study area, and analyze the topography, natural environment and hydrogeological conditions of the study area through literature review and field investigation. In this way, the aquifer, groundwater and boundary conditions are generalized, and the initial conditions of source and sink terms, groundwater flow field and salinity field are determined.

[0076] (2) A numerical model of groundwater flow and a model of groundwater solute transport in the study area were established using the groundwater model GMS and then identified and verified.

[0077] (3) Using multiple GCM models that have been used and performed well in the study area, the climate data after Cmhyd downscaling was input into the groundwater solute transport model for simulation, the predicted groundwater salinity distribution and dynamic changes were obtained, and the impact of different climate factors on groundwater seawater intrusion was analyzed.

[0078] Preferably, the specific steps are as follows:

[0079] 1. Establish a hydrogeological conceptual model for the study area.

[0080] 1) Determine the boundary of the study area

[0081] Based on the research needs and the hydrogeological conditions of the study area, complete hydrogeological units should be selected as much as possible, and natural water systems (rivers, lakes, etc.), artificial waterways (canals, ditches, etc.), and water-blocking boundaries (streamlines, watersheds, etc.) should be selected as boundary conditions as much as possible, while avoiding artificial boundaries, in order to determine the boundary range of the study area.

[0082] 2) Through literature review and field investigation, analyze the topography, natural environment, hydrogeological conditions and other conditions of the study area, so as to generalize the aquifer, groundwater and boundary conditions.

[0083] Based on the aquifer's properties, structure, and lithology, determine whether the aquifer is homogeneous or heterogeneous, isotropic or anisotropic. If all physical, chemical, and flow properties of the aquifer are constant without obvious gradual changes, it is generally classified as a homogeneous aquifer; otherwise, it is classified as a heterogeneous aquifer. If the aquifer is uniformly distributed and its physical, chemical, and flow properties are consistent in all directions, it is generally classified as isotropic; otherwise, it is classified as anisotropic. For soils and rocks with fine pores and slow groundwater flow, those with a permeability coefficient less than 0.001 m / d are considered impermeable layers, and those with a permeability coefficient between 0.001 and 1 m / d are considered weakly permeable layers. For soils and rocks with larger pores, a permeability coefficient greater than 1 m / d, and an underlying impermeable (or weakly) permeable layer, they are considered as aquifers.

[0084] Determine whether the groundwater is a steady or unsteady flow, or unconfined or confined. If the various motion elements of groundwater at any point in the study area do not change over time, it is generalized as a steady flow; otherwise, it is generalized as an unsteady flow. Based on the groundwater occurrence conditions, if the groundwater is above the first stable impermeable layer, it is generalized as unconfined; if the groundwater is between two impermeable layers, it is generalized as confined water.

[0085] Based on the geological structure of aquifers and impermeable layers, the characteristics of boundary flow, and the hydraulic connection between water bodies, boundary conditions are divided into three categories: given head boundary (first category boundary) which determines the head at each moment, given flow boundary (second category boundary) which determines the flow exchange per unit area, and mixed boundary (third category boundary) which comprehensively considers head and flow on the boundary. One or more boundary conditions are set according to the actual situation.

[0086] 2. Establish a numerical model of groundwater flow in the study area.

[0087] 1) The fundamental differential equation for groundwater movement is actually the groundwater volume balance equation. The partial differential equation for groundwater flow used in constructing the groundwater flow numerical model is:

[0088]

[0089] In the formula:

[0090] K xx K yy K zz : Permeability coefficients (m / d) in the x, y, and z directions, respectively; S: Storage capacity (1 / m); h: Hydraulic head (m); w: Source and sink terms of the aquifer (1 / d); t: Time

[0091] 2) Time and space discretization

[0092] The boundaries of the study area were determined, drawn in GMS, and the number of model layers and mesh generation were established. The collected elevations of the top and bottom plates of each aquifer layer and the initial groundwater head were imported into the software and interpolated. The stress period and simulation start and end times were determined based on the collected data.

[0093] 3) Determine the hydrogeological parameters of the model

[0094] The hydrogeological parameter zoning was based on hydrogeological survey data and pumping test data of the study area, combined with topography, geological maps, and hydrogeological maps of the study area. The initial hydrogeological parameters were determined through data collection and were adjusted during model identification and validation.

[0095] 4) Determine the source and sink terms of the model

[0096] Based on the hydrogeological map and combined with the precipitation infiltration recharge coefficient of different topographic and geomorphic units, the study area was divided into zones, taking into account factors such as surface soil quality, underlying surface conditions, and surface water depth. Precipitation and agricultural irrigation infiltration recharge in the study area were treated as areal recharge and imported into the model using Recharge.

[0097] Rainfall infiltration recharge is calculated using the following formula:

[0098] Q 降 =10 -1 ·a·F·P

[0099] In the formula:

[0100] Q 降 : Precipitation infiltration recharge (10,000 m³ / a); a: Precipitation infiltration recharge coefficient, dimensionless; P: Average annual precipitation in the calculation area (mm); F: Area of ​​the precipitation infiltration calculation area (km²) 2 )

[0101] Based on the data, the starting and ending elevations of the riverbed, the river level, and the hydraulic conductivity of the river sections in the study area were determined. River infiltration in the study area was treated as isal recharge, and the data were imported into the model using the River method.

[0102] River infiltration recharge is calculated using the following formula:

[0103] Q 渗 =K·W·L(R+h) / D

[0104] In the formula:

[0105] Q 渗 River infiltration recharge (m³) 3 K: Hydraulic conductivity coefficient of riverbed (m / d); W: Riverbed width (m); L: River section length (m); R: River water level (m); h: Groundwater level (m); D: Riverbed thickness (m).

[0106] The groundwater extraction volume and distribution of extraction wells were determined based on data, and appropriate well groups were merged and imported into the groundwater flow numerical model.

[0107] Evaporation determination: Based on the data, the evaporation rate, groundwater evaporation depth and surface evaporation elevation of the study area were determined and imported into the model using EVT.

[0108] Evaporation rate is calculated using the following formula:

[0109] E=Q ETM ×(1-D / X)

[0110] In the formula:

[0111] E: Evaporation rate (m³ / d); Q ETM: Maximum evaporation rate (m3 / d); D: Depth of groundwater level from surface evaporation elevation (m); X: Evaporation depth (m).

[0112] 5) Model identification and validation

[0113] Model identification involves continuously adjusting hydrogeological parameters and some source-sink parameters to ensure that the simulated groundwater flow field and the simulated groundwater level dynamics are largely consistent with the actual groundwater flow field and measured groundwater level dynamics, based on the hydrogeological parameters basically conforming to the actual hydrogeological conditions. Model validation further adjusts the parameters based on model identification to ensure that the simulation results are largely consistent with the secondary measured results, thereby improving the model accuracy. The operation is as follows:

[0114] Collect water level and location data from observation points and import them into the model. Activate the Observation Point function, input observation well data, and run the model. Continuously change hydrogeological parameters and repeat the calculation until the difference between the calculated and observed results at each observation point is within an acceptable range. After running the model, each observation point will display a colored band, with the center representing the observed value, the upper end representing the observed value plus the range value, and the lower end representing the observed value minus the range value. If the bar representing the difference between the observed and calculated values ​​is within the calibration confidence range, it is displayed in green; if the bar exceeds the confidence interval but is less than 200%, it is orange; and if it is greater than 200%, it is red. The applicability of the model is judged by observing and calculating the degree of fit between the simulated and observed values. A green observation point indicates a high degree of fit at that location. Avoid selecting monitoring points near the boundary for model identification and verification. Except for these types of monitoring points, at least 60% of the monitoring points should show a good fit between the groundwater dynamic process and the measured results, and the absolute error of the groundwater level value during the simulation at the monitoring point should be less than the actual water level fluctuation during the corresponding simulation period.

[0115] 3. Establish a groundwater solute transport model for the study area.

[0116] Based on the numerical model of groundwater flow, the MT3DMS and SEAWAT modules were added. The Advancement package, Dispersion package, and Source / sinkmixing package from the MT3DMS module were added, and the VDF package from the SEAWAT module was added. The types and physicochemical properties of solutes, their dispersibility, and molecular dispersion coefficients were determined based on available data. The SEAWAT module is designed to simulate three-dimensional variable-density groundwater flow. Variable-density flow (VDF) solves the following form of variable-density groundwater flow equation:

[0117]

[0118] In the formula:

[0119] ρ: Density of groundwater [ML] -3 ]; ρ0: Density of fresh water at standard temperature [ML] -3 μ: Dynamic viscosity of groundwater [ML] -1 T -1 μ0: Dynamic viscosity of fresh water [ML] -1 T -1 K0: Hydraulic conduction tensor [LT] -1 ]; z: vertical coordinate [L]; h0: freshwater head [L]; S s ′ ,0 Specific water storage ratio [L] -1 ]; t: time [T]; θ: porosity [-]; C: salt concentration [ML] -3 ];q′ S : has density ρ s Source and sink items [T] -1 ].

[0120] In SEAWAT, the following form of equation is also required to solve the solute transport equation:

[0121]

[0122] In the formula:

[0123] ρ b Bulk density [ML] -3 ]; Distribution coefficient of type k [L] 3 M -1 ];C k Concentration of species k [ML] -3 D: Hydrodynamic dispersion tensor [L] -2 T -1 Q: Flow rate [LT] -1 ]; Source or sink concentration of species k [ML] -3 ].

[0124] The measured solute data were imported into the model and interpolated using the 2D Scatter Data module of GMS to determine the initial concentration. The results were then verified after running the model.

[0125] 4. Climate Prediction

[0126] For the meteorological data to be predicted, the CMIP5 or CMIP6 climate datasets can be used. Future changes in climate variables can be predicted using global climate models (GCMs) driven by greenhouse gas emissions. GCMs are mathematical expressions of the main climate system components (atmosphere, land surface, ocean, and ocean cryosphere) and their interactions, calculating the equations for the changes in climate components over time and their coupling effects. The outputs of different GCMs become unstable due to the uncertainty of greenhouse gas emissions. Therefore, multiple historically used and well-performing GCMs for the study area are aggregated to reduce the output errors of individual members and improve the prediction accuracy of future climate. For each GCM model, the moderate radiative forcing scenario RCP4.5 and the high radiative forcing scenario RCP8.5 from the representative concentration pathway (RCP) can be considered, as these two RCP scenarios can essentially cover the realistic range of future climate and hydrological changes.

[0127] The excessively high resolution of the original GCM model can lead to significant errors between simulated and observed values. Bias correction and downscaling methods can be implemented using Climate Model data for hydrologic modeling (CMhyd). CMhyd is a correction tool that extracts and corrects data from global and regional climate model aggregations, reducing simulated climate data to a hydrological station scale for better matching with observed data.

[0128] Cmhyd provides different bias correction methods for precipitation and temperature. For precipitation bias correction, methods include Distribution mapping of precipitation and temperature, Linear scaling, Power transformation of precipitation, and Precipitation local intensity scaling; for temperature bias correction, methods include Distribution mapping of precipitation and temperature, Linear scaling, and Variation scaling of temperature. The average bias, root mean square error, and correlation coefficient of multiple models are used to compare the simulation capabilities of various climate models for climate change in the study area, and the relative optimal bias correction method among different GCM models is evaluated. Simultaneously, the Taylor skill score (TSS) can be used to comprehensively consider root mean square error and correlation. One type of Taylor skill score is defined as...

[0129]

[0130] In the formula:

[0131] R0: The maximum achievable correlation, set to 1;

[0132] R, These are the model's correlation and the normalized root mean square, respectively.

[0133] A TSS close to 1 indicates that the observed values ​​are closer to the simulated values, while a decrease towards 0 indicates poorer model performance.

[0134] By comparing the TSS values ​​of different GCM models and bias correction methods, several precipitation and temperature combinations with higher TSS values ​​were selected as the predicted climate data combinations. After the Cmhyd climate model made climate predictions for these combinations, climate combinations that clearly did not conform to the climate change patterns in the study area were removed. The data of the remaining combinations were then input into groundwater solute transport models to construct groundwater solute transport prediction models under different climate scenarios, and to analyze the degree of groundwater-seawater intrusion under different climate change influences.

[0135] The combination with the highest TSS from each GCM model is selected. The output precipitation data can be directly input into the GMS model to construct a groundwater solute transport prediction model. The output temperature data is used to calculate evaporation using the Hargreaves empirical formula and then input into the GMS model. The groundwater solute transport prediction model is based on the established groundwater solute transport model, but the simulation time is changed (to be the same as or part of the climate prediction time). The hydrogeological parameters, source and sink terms, etc., of the groundwater solute transport prediction model remain unchanged.

[0136] Since different combinations of GCM models and bias correction methods produce different precipitation and temperature outputs, they can be used as different climate scenarios to predict the impact of climate change on seawater intrusion into shallow groundwater. The impact of seawater intrusion on groundwater is determined based on the groundwater salinity distribution simulated by the groundwater solute transport prediction model, including the movement of the 1 g / L isoline (the boundary between fresh and saltwater), changes in salt concentration in coastal areas, and the size of the area delineated for salinity ranges (e.g., 3–5 g / L).

[0137] The specific preparation and detection methods are as follows:

[0138] A method for predicting the impact of different climate changes on seawater intrusion into shallow groundwater in the Mekong Delta can be used as follows: Figure 1 As shown, the method includes the following steps:

[0139] (1) Establish the hydrogeological concept of the study area, determine the boundary range of the study area, and analyze the topography, natural environment and hydrogeological conditions of the study area through literature review and field investigation. In this way, the aquifer, groundwater and boundary conditions are generalized, and the initial conditions of source and sink terms, groundwater flow field and salinity field are determined.

[0140] (2) A numerical model of groundwater flow and a model of groundwater solute transport in the study area were established using the groundwater model GMS and then identified and verified.

[0141] (3) Using multiple GCM models that have been used and performed well in the study area, the climate data after Cmhyd downscaling was input into the groundwater solute transport model for simulation, the predicted groundwater salinity distribution and dynamic changes were obtained, and the impact of different climate factors on groundwater seawater intrusion was analyzed.

[0142] Preferably, the specific steps are as follows:

[0143] 1. Construction of a hydrogeological conceptual model for the study area

[0144] 1) Determine the scope of the study area: The study area is defined as the central alluvial plain of the Mekong Delta. The study area is roughly triangular in shape, with the Mekong River as its northeastern boundary, the Basak River as its southwestern boundary, and the South China Sea as its eastern boundary.

[0145] 2) Aquifer: The Holocene aquifer near the surface is the subject of this study. The upper part of the aquifer is divided into the Q2 weakly permeable layer based on its hydrogeological structure, and the lower part is the qh aquifer.

[0146] 2. Construction of Numerical Model for Groundwater Flow in the Study Area

[0147] The partial differential variance of groundwater flow used in GMS software to construct a numerical model of groundwater flow is:

[0148]

[0149] In the formula: K xx K yy K zz : Permeability coefficients (m / d) in the x, y and z directions, respectively; S: Storage rate (1 / m); h: Water head (m); w: Source and sink terms of the aquifer (1 / d); t: Time.

[0150] 1) Spatial and temporal discretization: The study area was divided into 151 rows and 179 columns, with 5 vertical model layers. The cell resolution was 1km × 1km. The model simulation period was from 2012 to 2021, with each stress period lasting 3 months. The model identification period was from 2012 to 2016, and the model validation period was from 2017 to 2021.

[0151] 2) Aquifer and Boundary Condition Generalization: The Holocene aquifer near the surface is used as the research object. The upper part of the aquifer is divided into the Q2 weakly permeable layer according to the hydrogeological structure, and the lower part is the qh aquifer. To refine the hydrogeological parameters of the model, the Q2 weakly permeable layer is divided into 2 layers vertically, each with a thickness of about 10m, for a total of 20m; the qh aquifer is divided into 3 layers, each with a thickness of about 10m, for a total of 30m.

[0152] Lateral boundary: Since the depth of the Mekong River, Basak River and the South China Sea adjacent to the study area is generally no more than 20m, the Q2 weakly permeable layer is regarded as having hydraulic connection with the above three boundaries and is generalized as a given head boundary; the lower qh aquifer is recharged by the lateral aquifer and is generalized as a flow boundary.

[0153] Vertical boundary: The top of the model is the Earth's surface as the upper boundary, which exchanges water with the outside world; the bottom of the model is a low-permeability, weakly permeable layer, which is generalized as a flow-free boundary.

[0154] 3) Determination and Zoning of Hydrogeological Parameters: The initial values ​​of hydrogeological parameters are determined based on empirical values ​​and collected data. The basic hydrogeological parameters include horizontal hydraulic conductivity (8-35 m / d), vertical hydraulic conductivity (4-35 m / d), water storage capacity (0.000001-0.000141 / m), specific yield (0.0009-0.01), and porosity (0.3), etc., and are zoned according to the data (e.g., zoning based on provincial boundaries).

[0155] 4) Identify source and sink components: The Holocene aquifer in the study area exchanges water with the external environment. Its recharge includes precipitation, river infiltration, agricultural irrigation recharge, and ocean exchange; its discharge includes evaporation and groundwater extraction.

[0156] Precipitation: Precipitation data were obtained from Mekong River Commission hydrological stations (900–1800 mm / year); the initial precipitation infiltration recharge coefficient was determined from data (0.05–0.15) and adjusted during model identification and verification; precipitation zones were determined by surface type, precipitation contour lines, etc.

[0157] River infiltration: River water level references hydrological station data near the river; initial riverbed hydraulic conductivity is determined from data and adjusted during model identification and verification.

[0158] Agricultural irrigation supply: Agricultural irrigation volume was determined based on data (6000m³). 3 / acre / year); Irrigation zones are the same as precipitation zones.

[0159] Ocean exchange: The amount of water exchanged in the ocean is derived from a numerical model of groundwater flow.

[0160] Evaporation: Evaporation requires determining the evaporation rate, evaporation depth, and surface evaporation elevation. The evaporation rate is obtained from the Mekong River Commission's hydrological stations (800–1200 mm); the evaporation depth is set at 3 m, and the surface evaporation elevation is the same as the ground elevation; evaporation zones are determined by surface type, surface elevation, etc.

[0161] Groundwater extraction: Groundwater extraction volume and distribution of groundwater extraction wells were obtained from data (Holocene aquifer extraction volume was 18,000 m³). 3 / d).

[0162] 5) Model Identification and Verification: Adjust the model's hydrogeological parameters and source / sink terms (including manual adjustment and automatic parameter adjustment via the GMS model's built-in PEST function), determine a water level error (generally 10% of the annual difference in groundwater level), and ensure that the model's simulated groundwater level is as close as possible to the measured groundwater level within the error range, reflecting the actual flow field characteristics, such as... Figure 2 As shown.

[0163] 3. Construction of a groundwater solute transport model in the study area

[0164] SEAWAT is designed to simulate three-dimensional variable-density groundwater flow. The variable-density flow (VDF) process in SEAWAT solves the following form of variable-density groundwater flow equation:

[0165]

[0166] Where: ρ: density of groundwater [ML] -3 ]; ρ0: Density of fresh water at standard temperature [ML] -3 μ: Dynamic viscosity of groundwater [ML] -1 T -1 μ0: Dynamic viscosity of fresh water [ML] -1 T -1 K0: Hydraulic conduction tensor [LT] -1 ]; z: vertical coordinate [L] h0: freshwater head [L]; S s ′ ,0 Specific water storage ratio [L] -1 ]; t: time [T]; θ: porosity [-]; C: salt concentration [ML] -3 ];q′ S : has density ρ s Source and sink items [T] -1 ].

[0167] Assume the density of fresh water (ρ = 0) is 1000 kg / m³. -3 The maximum density of saline water (ρ) s =35g / L) is 1025kg m -3It is assumed that groundwater density is only related to salt concentration, and therefore they have a linear relationship.

[0168] In SEAWAT, the following form of equation is also required to solve the solute transport equation:

[0169]

[0170] In the formula: ρ b Bulk density [ML] -3 ]; Distribution coefficient of type k [L] 3 M -1 ];C k Concentration of species k [ML] -3 D: Hydrodynamic dispersion tensor [L] -2 T -1 Q: Flow rate [LT] -1 ]; Source or sink concentration of species k [ML] -3 ].

[0171] 1) Generalization of aquifer and boundary conditions:

[0172] Aquifer generalization: Based on the lithology of the aquifer medium and the groundwater occurrence conditions, aquifers are divided into an upper weakly permeable layer and a lower aquifer. The upper weakly permeable layer is mainly composed of clay and silt, while the lower aquifer is mainly composed of fine sand, gravel, and pebbles. The various hydrogeological parameters of the aquifer vary spatially, thus it is generally classified as a heterogeneous anisotropic aquifer.

[0173] Boundary condition generalization: Lateral boundary: The lateral boundary of the upper weakly permeable layer of the model is generalized as a given head boundary. The lateral boundary of the lower aquifer of the model is generalized as a flow boundary. Vertical boundary: In the vertical direction, the land surface is used as the upper boundary. Through the upper boundary model, water exchange with the outside world can occur through rainfall, evaporation, and surface water bodies. The lower boundary is generalized as a non-flow boundary and an impermeable base.

[0174] 2) Solute source and sink: Ocean salinity is constant at 35 g / L, river salinity is obtained through hydrological stations, and aquifer salinity is given a constant salinity of 0–9 g / L through lateral recharge.

[0175] 3) Hydrogeological parameters: Same as the groundwater flow numerical model, but with the addition of aquifer longitudinal dispersion (1m), lateral dispersion and vertical dispersion (0.1), and molecular dispersion coefficient (8.64×10⁻⁶). -5 m 2 d -1 ).

[0176] 4) Spatial and temporal discretization: Same as groundwater flow numerical model.

[0177] 5) Initial concentration in the study area: Using salinity TDS as the target solute, the groundwater salinity distribution in 2011 was established by collecting data and set as the initial concentration of the groundwater solute transport model.

[0178] 6) Model recognition and verification:

[0179] By continuously adjusting hydrogeological parameters and some source-sink parameters, and ensuring that the hydrogeological parameters basically conform to the actual hydrogeological conditions, the simulated values ​​of groundwater salinity isopleths are largely consistent with the measured values, and the simulated dynamic changes in groundwater salinity are largely consistent with the measured dynamic changes in groundwater salinity. Model validation, based on model identification, involves further adjusting parameters to ensure that the simulation results are largely consistent with the secondary measured results, thereby improving model accuracy. The operation is as follows:

[0180] Collect water level and location data from observation points and import them into the model. Activate the Observation Point function, input observation well data, and run the model. Continuously change hydrogeological parameters and repeat the calculation until the difference between the calculated and observed results at each observation point is within an acceptable range. After running the model, each observation point will display a colored band, with the center representing the observed value, the upper end representing the observed value plus the range value, and the lower end representing the observed value minus the range value. If the bar representing the difference between the observed and calculated values ​​is within the calibration confidence range, it is displayed in green; if the bar exceeds the confidence interval but is less than 200%, it is orange; and if it is greater than 200%, it is red. The applicability of the model is judged by observing and calculating the degree of fit between the simulated and observed values. A green observation point indicates a high degree of fit at that location. Avoid selecting monitoring points near the boundary for model identification and verification. Except for these types of monitoring points, at least 60% of the monitoring points should show a good fit between the dynamic process of groundwater salinity and the measured results, and the error between the simulated and measured groundwater salinity values ​​should generally not exceed 0.5–1 g / L.

[0181] 4. Climate Prediction

[0182] 1) Data preparation: Historical climate data (including precipitation and temperature) for the study area were collected; historically well-performing GCM models for the study area were collected, with ACCESS1-0, CCSM4, CSIRO-MK3-6-0, MPI-ESM-LR, and NorESM1-M selected for this study; RCP scenarios were selected, with RCP4.5 and RCP8.5 chosen for this study; the bias correction model used was the CMhyd climate model (which includes four precipitation bias correction methods (e.g., 1, 2, 3, 4) and three temperature bias correction methods (e.g., 5, 6, 7), where evaporation can be calculated using the Hargreaves empirical formula for the highest and lowest temperatures.

[0183] 2) Comparison of Climate Data Statistics: Historical precipitation and temperature data from each GCM model were statistically reduced using different bias correction methods under the RCP4.5 and RCP8.5 scenarios (e.g., temperature—ACCESS1-0—RCP4.5—1 or precipitation—ACCESS1-0—RCP4.5—1), and compared with historical observation data using TSS to create a heat map, such as... Figure 3 As shown, the group with the highest TSS among different GCM modes was selected. In this study, the following modes were selected: ACCESS1-0 (Linear scaling, 0.819), CCSM4 (Linear scaling, 0.791), CSIRO-MK3-6-0 (Power transformation of precipitation, 0.661), MPI-ESM-LR (Linear scaling, 0.81), and NorESM1-M (Linear scaling, 0.81).

[0184] 3) Future Precipitation and Temperature Trends: After selecting the group with the highest TSS among different GCM models, future precipitation and temperature were predicted in CMhyd. The prediction period selected in this study is from 2022 to 2050. After the prediction was completed, GCM models with outliers were removed. For example, in this study, CSIRO-MK3-6-0 showed an annual precipitation of 4000 mm under the RCP4.5 scenario. Figure 4 As shown, precipitation from this GCM model is not considered in the groundwater solute transport model. Temperature prediction follows the same procedure.

[0185] 5. Construction of a prediction model for groundwater solute transport

[0186] 1) Generalization of aquifer and boundary conditions:

[0187] 2) Aquifer Generalization: Based on the lithology of the aquifer medium and the groundwater occurrence conditions, aquifers are divided into an upper weakly permeable layer and a lower aquifer. The upper weakly permeable layer is mainly composed of clay and silt, while the lower aquifer is mainly composed of fine sand, gravel, and pebbles. The various hydrogeological parameters of the aquifer vary spatially, thus it is generally classified as a heterogeneous anisotropic aquifer.

[0188] Boundary condition generalization: Lateral boundary: The lateral boundary of the upper weakly permeable layer of the model is generalized as a given head boundary. The lateral boundary of the lower aquifer of the model is generalized as a flow boundary. Vertical boundary: In the vertical direction, the land surface is used as the upper boundary. Through the upper boundary model, water exchange with the outside world can occur through rainfall, evaporation, and surface water bodies. The lower boundary is generalized as a non-flow boundary and an impermeable base.

[0189] 2) Solute source and sink terms: Same as groundwater solute transport model

[0190] The ocean salinity is constant at 35 g / L, the river salinity is obtained through hydrological stations, and the aquifer salinity is given a constant salinity of 0–9 g / L through lateral recharge.

[0191] 3) Hydrogeological parameters:

[0192] Similar to the groundwater flow numerical model, but with added longitudinal aquifer dispersion (1m), lateral and vertical dispersion (0.1), and molecular dispersion coefficient (8.64×10⁻⁶). -5 m 2 d -1 ).

[0193] 4) Spatial and temporal discretization: Spatial discretization is the same as that of the groundwater solute transport model. Temporal discretization is determined based on the climate prediction time. In this study, the groundwater solute transport prediction simulation time is from 2022 to 2050.

[0194] 5) Initial concentration in the study area: The groundwater salinity distribution in 2021, as determined by the groundwater solute transport model, was used as the initial salinity distribution for the groundwater solute transport prediction model.

[0195] 6) Setting different climate scenarios: The predicted minimum and maximum annual rainfall were selected from the ACCESS1-0 model and CCSM4 rainfall model under the RCP8.5 scenario, respectively, because the predicted data of these two GCM models under the RCP8.5 scenario have almost no outliers. The predicted minimum and maximum temperatures were selected from the CSRIO-MK3-6-0 GCM model under the RCP8.5 scenario, because it corresponds to a very significant temporal variation in evaporation calculated by Hargreaves, which can better highlight the impact of evaporation variation on groundwater solute transport.

[0196] 7) Predicting changes in groundwater salinity:

[0197] First, the change in groundwater salinity is calculated under the baseline condition, which is equivalent to the groundwater solute transport model using the source and sink terms in 2021 for prediction and simulation, with the simulation period from 2022 to 2050.

[0198] (1) Changes in precipitation

[0199] Under the RCP8.5 scenario, over a 30-year period, compared to the baseline, the changes in the 20 g / L salinity isolines in the ACCESS1-0 model are basically the same; the 10 g / L salinity isolines exhibit both contraction towards the ocean and expansion towards the inland, with a maximum contraction of 500 m and an expansion of 2000 m; the changes in the 5 g / L salinity isolines are basically the same; the 1 g / L isoline (the brackish water boundary) shifts 100–1000 m towards the ocean; and the 3 g / L isoline shifts 100–500 m towards the ocean.

[0200] like Figure 5 As shown, compared to the ACCESS1-0 model, the 20 g / L isoline in the CCSM4 model is basically the same; the 10 g / L isoline contracts towards the ocean by about 0–200 m, but there is a slight inland expansion in some areas; the 5 g / L isoline contracts towards the ocean by 0–300 m; the 3 g / L isoline shows both contraction towards the ocean and expansion towards the inland, with a maximum contraction of 100 m and expansion of 500 m; the 1 g / L isoline is basically the same. Therefore, it can be seen that increased precipitation leads to a decrease in groundwater salinity, the 1 g / L salinity isoline (the brackish water boundary) shifts towards the ocean, and seawater intrusion weakens.

[0201] (2) Changes in evaporation

[0202] like Figure 6 As shown, with the increase in evaporation, the changes in the salinity isopleths are basically the same as the baseline. Since the groundwater volume decreases with evaporation, the salinity is generally slightly higher than the baseline data. However, because the change in evaporation is very small relative to the baseline, the range of salinity changes is almost imperceptible.

[0203] Although embodiments of the invention have been disclosed for illustrative purposes, those skilled in the art will understand that various substitutions, variations, and modifications are possible without departing from the spirit and scope of the invention and the appended claims. Therefore, the scope of the invention is not limited to the contents disclosed in the embodiments.

Claims

1. A method for predicting the impact of different climate changes on shallow groundwater seawater intrusion, characterized in that: The method includes the following steps: (1) Establish a hydrogeological conceptual model of the study area, determine the boundary range of the study area, and analyze the topography, natural environment and hydrogeological conditions of the study area through literature review and field investigation, so as to generalize the aquifer, groundwater and boundary conditions, and determine the initial conditions of source and sink terms, groundwater flow field and salinity field. (2) A numerical model of groundwater flow and a model of groundwater solute transport in the study area were established using the groundwater model GMS. For the groundwater flow numerical model, the simulated groundwater level and the groundwater level observation well data were compared to ensure that the simulated groundwater level value was basically the same as the observed value and that the simulated groundwater flow field was close to the measured groundwater flow field. For the groundwater solute transport model, the simulated groundwater salinity and the groundwater salinity observation well data were compared to ensure that the simulated groundwater salinity value was basically the same as the observed value and that the simulated groundwater salinity field was close to the measured groundwater salinity field. (3) Using multiple GCM models that have been used in the study area and can reproduce historical temperature and precipitation, the climate data after Cmhyd downscaling is input into the groundwater solute transport model for simulation, the predicted groundwater salinity distribution and dynamic changes are obtained, and the influence of different climate factors on groundwater seawater intrusion is analyzed. The specific steps (3) are as follows: Using the CMIP5 or CMIP6 climate datasets, future changes in climate variables can be predicted using the global climate model GCM based on greenhouse gas emission drivers. The model also considers the medium radiation forcing scenario RCP4.5 and the high radiation forcing scenario RCP8.5 in the representative concentration pathway RCP, and performs bias correction and downscaling using the CMhyd climate model for hydrological simulation. Historically well-performing GCM models within the study area were selected, and bias corrections were performed using different bias correction methods from CMhyd. Each GCM model and each bias correction method were grouped together. The bias-corrected climate data for each group were compared with historical data, and the correlation R and root mean square (RMS) were calculated for each group. The Taylor Skill Score (TSS) is generated, and the TSS of each combination is compared. The precipitation or temperature combination with the highest TSS value in different GCM models is selected as the combination of predicted climate data. Based on the project requirements, the prediction time was selected. After the combination with the highest TSS value was predicted in the Cmhyd climate model, the climate combination with climate conditions and extreme climate phenomena that had never appeared in history was removed. The data of the remaining combinations were input into the groundwater solute transport model to construct groundwater solute transport prediction models under different climate scenarios and analyze the degree of groundwater seawater intrusion under the influence of different climate changes. The formula for calculating the Taylor Skill Score (TSS) is as follows: In the formula: The maximum achievable correlation is set to 1. , These are the correlation of the model and the root mean square after normalization, respectively. in A value close to 1 indicates that the observed value is closer to the simulated value, while a decrease towards 0 indicates a worse model performance; the model's correlation R and normalized root mean square (RMS) are also relevant. It was calculated.

2. The method according to claim 1, characterized in that: The hydrogeological conceptual model in step (1) specifically includes the following steps: 1) Based on the research needs and the hydrogeological conditions of the study area, select complete hydrogeological units to determine the boundary range of the study area; 2) Based on the properties, structure, and lithology of the aquifer, aquifers are generalized into homogeneous or heterogeneous aquifers, isotropic or anisotropic aquifers, impermeable or poorly permeable aquifers. 3) Groundwater is generalized into steady or unsteady flow, unconfined or confined water; 4) Based on the geological structure of aquifers and impermeable layers, the characteristics of boundary flow, and the hydraulic connection between water bodies, the boundary conditions are generalized into three categories: the first category is a given head boundary that determines the head at each moment; the second category is a given flow boundary that determines the flow rate per unit area; and the third category is a mixed boundary that considers both head and flow rate.

3. The method according to claim 2, characterized in that: The numerical model for groundwater flow in step (2) is as follows: In the formula: , , : Permeability coefficients (m / d) in the x, y, and z directions, respectively; Water storage rate (1 / m); Water head (m); Source and sink terms of the aquifer (1 / d); :time.

4. The method according to claim 3, characterized in that: The steps for constructing the groundwater flow numerical model include the following: 1) Time Discreteness and Spatial Discreteness Based on the established study area boundary, the study area boundary is drawn in GMS. The size and number of model cells are determined according to the size of the study area. The number of model layers is determined according to the hydrogeological conditions and aquifer thickness of the study area. The elevations of the top and bottom plates of each aquifer after model layering and the initial groundwater head are collected, imported into GMS and interpolated. The stress period and simulation start and end times are determined. 2) Determine the hydrogeological parameters of the model Based on the hydrogeological survey data and pumping test data of the study area, and combined with the topography, geological map and hydrogeological map of the study area, the study area is divided into hydrogeological parameter zones. Determine the initial hydrogeological parameters, including the horizontal / vertical hydraulic conductivity, water storage rate, specific yield, and porosity of the weakly permeable layer and aquifer; 3) Identify source and sink items (1) Based on the hydrogeological map, combined with the precipitation infiltration recharge coefficient of different topographic and geomorphic units, and taking into account the surface soil quality, underlying surface conditions and surface water depth, precipitation infiltration and agricultural irrigation are divided into zones; precipitation and agricultural irrigation infiltration recharge in the study area are treated as areal recharge and imported into the groundwater flow numerical model through Recharge Package. (2) Based on the data, determine the starting and ending elevations of the riverbed, the water level and the hydraulic conductivity of the river section in the study area. The infiltration of the river in the study area is treated as a surface recharge and imported into the groundwater flow numerical model through River Package. (3) The groundwater extraction volume and the distribution of extraction wells are determined based on data, and appropriate well groups are merged and imported into the groundwater flow numerical model; (4) Evaporation as a factor in shallow groundwater loss under given climatic conditions Evaporation determination: Based on the data, the evaporation rate, groundwater evaporation depth and surface evaporation elevation of the study area were determined, and the data were imported into the groundwater flow numerical model using EVT Package; 4) Model identification and validation By using national or organizational groundwater monitoring networks or individual field monitoring, the location of observation points is determined and water level data is collected. This data is then imported into the GMS model, the Observation Point function is activated, observation well data is input, and the groundwater flow numerical model is run. Hydrogeological parameters and source-sink values ​​are continuously changed and repeated until the difference between the calculated groundwater level and the observed groundwater level is controlled within 10% of the annual groundwater level fluctuation. After running the groundwater flow numerical model, each observation point will display a colored band, with the middle point as the observed value, the upper end as the observed value plus the range value, and the lower end as the observed value minus the range value. The applicability of the model is judged by observing and calculating the degree of fit between the simulated value and the observed value, and the absolute error of the groundwater level value during the simulation of the monitoring point should be less than the actual water level fluctuation of the corresponding simulation period of the monitoring point.

5. The method according to claim 4, characterized in that: The groundwater solute transport model in step (2) is calculated using the variable density groundwater flow equation and the solute transport equation, as follows: The equation for the flow of variable-density groundwater is as follows: In the formula: Density of groundwater [ML] -3 ]; Density of fresh water at standard temperature [ML] -3 ]; Dynamic viscosity of groundwater [ML] -1 T -1 ]; Dynamic viscosity of fresh water [ML] -1 T -1 ]; Hydraulic conduction tensor [LT] -1 ]; Vertical coordinate [L]; Freshwater head [L]; Specific water storage ratio [L] -1 ]; Time [T]; Porosity [-]; Salt concentration [ML] -3 ]; : has density Source and sink items [T] -1 ] ; The solute transport equation is as follows: In the formula: Bulk density [ML] -3 ]; Distribution coefficient of type k [L] 3 M -1 ]; Concentration of species k [ML] -3 ]; Hydrodynamic dispersion coefficient tensor [L] -2 T -1 Q: Flow rate [LT] -1 ]; Source or sink concentration of species k [ML] -3 ] .

6. The method according to claim 1, characterized in that: The steps for constructing the groundwater solute transport model are as follows: Based on the groundwater flow numerical model, MT3DMS and SEAWAT modules were added; the Advance package, Dispersion package and Source / sink mixing package in the MT3DMS module were added, and the VDF package in the SEAWAT module was added. The physicochemical properties of salinity TDS, the solute dispersion and molecular dispersion coefficient were adjusted so that the simulated groundwater salinity values ​​are basically close to the measured values. The measured solute data were imported into the GMS model using the 2D Scatter Data module of GMS and interpolated to obtain the initial concentration. The time and spatial discretization were the same as those of the groundwater flow numerical model. The source and sink terms were the same as those of the groundwater flow numerical model, but salinity source and sink terms were added to the river and ocean source and sink terms to obtain the river and ocean salinity. After running the model, the parameters were continuously adjusted to make the simulated groundwater salinity values ​​close to the measured values, and the simulated groundwater salinity contour values ​​tended to be consistent with the measured values. Using multiple GCM models, climate data downscaled by Cmhyd was input into a groundwater solute transport model for simulation, resulting in predicted groundwater salinity distribution and dynamic changes. The impact of different climate factors on groundwater seawater intrusion was analyzed.

7. The application of the method as described in any one of claims 1 to 6 in predicting the impacts of different climate changes on shallow groundwater seawater intrusion.

Images