Authors: Nejat Zeydalinejad
Categories: Earth and Planetary Science, Aquifer, Groundwater level, Groundwater resilience, Groundwater resilience evaluation, Groundwater storage, Groundwater sustainability
Source: MethodsX
Authors: Nejat Zeydalinejad
Currently, groundwater resources are under anthropogenic and natural pressures. Depending on various factors, aquifers have diverse responses toward stresses. Resilience is among the performance criteria determining the ability of an aquifer to withstand unfavorable circumstances and to recover to a steady state after a failure. Hence, making use of this concept in groundwater systems is of cruciality in sustainable groundwater resources management. The main objective of the current research is to provide an overview on the employed procedures and existing difficulties appertaining to the appraisal of groundwater resilience. To this end, 36 studies which have evaluated groundwater resilience were regarded. Results highlighted that not only a very limited literature has employed groundwater resilience, but there also exist multiple knowledge gaps and complications, including resilience dimension, data scarcity and inaccuracies. In terms of groundwater quality and saltwater intrusion much little research exist, i.e. 7 studies among 36 ones. Evaluating groundwater resilience quantitatively, 11 methods were distinguished, each of which with its own advantages and disadvantages which may be engaged depending on factors such as available data and type of aquifer. Finally, more groundwater resilience analyses must be conducted at different spatiotemporal scales in order to preserve aquifers for the future generations.
In summary, followings may be regarded to evaluate groundwater •To begin with, it must be specified which aspect of the aquifer's resilience and toward which unsatisfactory condition is planned to be considered.•Provided that no groundwater modeling is intended to be conducted, the observation data may be analyzed, considering the current state, governing violations, and new steady state.•On condition that groundwater simulation is planned to be conducted, subsequent to model's calibration and verification, the response of the aquifer, as a resilience criterion, is determined.
Specifications tableSubject Earth and Planetary SciencesMore specific subject HydrogeologyName of your Groundwater resilience evaluationName and reference of original All groundwater resilience appraisalsResource Not applicable.
Nowadays, human is putting pressure on groundwater systems, and climate change is exacerbating the circumstances with sharper influences on arid and semi-arid countries, e.g. Iran [94]. Indeed, groundwater level is declining and this causes other environmental issues with no necessarily a linear relationship. For instance, even though the correlation between groundwater level drawdown and land subsidence is at first linear, it becomes exponential after trespassing a specific threshold [8]. Determining this limit is of vitality because it demonstrates the failure threshold of a groundwater system, which should never be exceeded in order for invaluable groundwater resources to be maintained.
Rega and Bonifazi [63] discussed the development of resilience in spatial planning and highlighted the various perceptions embedded therein. Long used in structural engineering and materials science [10], [71], resilience sparked debate in regard to its potential indistinctness and the requirement to clarify its components and varieties [28]. From the beginning, the concept of resilience has spread across dissimilar research, e.g. medicine [27], [69] and economics [5]. Resilience is dissimilarly defined according to the field of study employing it [26] from natural resources researches [82] to coral reef surveys [29]. For instance, Timmerman [81] introduced it as the capacity of societies to address and recuperate from the exterior shocks to their infrastructure in the context of engineering and disasters. The reaction of the water system to external forces, including natural and human effects, is primarily considered as the hydrological resilience. This concept should be taken into account along with the social system [31]. Indeed, it is recommended that the socio-hydrological resilience be used [46]. With regard to climate change, resilience is the capability of the system to absorb turbulence, maintain similar functioning, and modify the external changes [32]. In fact, resilience is either the system's ability to buffer the long-term pressures or the recovery time of the system subsequent to a failure [25], [37], [57]. In other words, resilience is the system's ability to withstand a continuous variable environment, and absorb or adapt to recurring exterior stresses without any major changes in the system's operation [46], [18], [19]. Resilience is also considered as the average period of time during which a system is under a temporary constraint [48], [56]. Loucks [42] regarded resilience of the system as the possibility to recuperate after a failure. Fowler et al. [20] explained resilience as the time during which the system is under unsatisfactory condition in relation to the total time period. Kjeldsen and Rosbjerg [36] quantified resilience by explaining three the reverse of the mean time period a system is under unsatisfactory condition, the maximum time period a system is under unsatisfactory condition, and the length of the 90th fractile of the unsatisfactory time period. The second concept gave the best approximation in their research and several other studies. Kundzewicz and Kindler [38] preferred the second concept as it takes into account all trivial events, and Roach et al. [67] assessed the resilience of a water system using this concept.
To evaluate the sustainability of a groundwater system, operational criteria, including groundwater resilience, may be utilized [78]. Groundwater resilience may be defined as the resilience of a coupled human-groundwater system is the system's ability to increase recharge and decrease base flow in order to maintain equilibrium or reach to a new one [64]. Whether the liquid phase or the physical medium of a groundwater system may be regarded in a groundwater resilience assessment [14]. Quantifying groundwater resilience enhances our knowledge about the system. Most aquifers which were previously characterized by high resilience were recently proved to be under variable to high stress [65]. This familiarity may help preserve the aquifers by introducing appropriate adaptation and management strategies [14]. The resilience of an aquifer to pumping, which may create unstable conditions, can be improved by human intervention, provided that a precise understanding of the system is accessible [64].
Groundwater resilience may be appraised on the subject of anthropic influences, which are of crucially significance, e.g. well-nigh 77% of land subsidence may be attributed to man, among which around 60% is as a result of groundwater extraction [8]. On the other hand, groundwater resilience may be studied on the topic of climate, in which long-term (decadal) climate change, and short-term (less than one year) climatic stresses may be taken into account [44]. Moreover, as distinguishing the impact of climate change and human on groundwater may be of difficulty, in most cases, groundwater resilience may be studied with regard to both factors.
Even though groundwater resilience has been used to evaluate the surface water systems for a long term, its application in groundwater has been limited [60], [78]. Moreover, studies often neglect to describe the main techniques for calculating the groundwater resilience [60] even though evaluating the resilience of a system necessitates a concise methodology with standard variables and parameters [26]. Hence, providing the employed approaches in calculating the resilience of a system, including groundwater, is unquestionably of vitality.
The main purpose of this research is to represent the approaches which have hitherto been applied to evaluate groundwater resilience at different spatiotemporal scales. For this purpose, firstly, the approaches used to assess the groundwater resilience in quantitative terms are discussed and categorized. Secondly, a review on groundwater resilience in the matter of contamination and seawater intrusion are briefly introduced. On the other hand, the potential challenges related to appraising groundwater resilience are presented herein. It is of relevance to note in this regard that the approaches used for calculating the resilience of surface water systems and other ecosystems are not included. In addition, the strategies for improving the resilience of groundwater systems are beyond the scope of the current research.
This study was conducted by making use of the Google scholar web search engine. In order to include all published research, various search terms were applied, including but not limited to groundwater resilience; aquifer resilience; water resources resilience. A number of 29, 2 and 5 studies were found regarding groundwater resilience in quantitative, qualitative, and saltwater intrusion points of view, respectively. Fig. 1 depicts the number of studies assessing groundwater resilience in different years, considering both groundwater quantity, e.g. groundwater level, recharge and storage, and groundwater quality, e.g. seawater intrusion.Fig. 1Number of the conducted groundwater resilience studies in different years.Fig 1
As portrayed in Fig. 2, different observation data may be required for groundwater resilience appraisals, including climatic, hydrologic, hydrogeologic, socio-hydrologic, and economic data. Groundwater modeling may be included or not. Next, an approach may be selected depending on which aspect of groundwater resilience is planned to be evaluated, e.g. groundwater quantity, quality and seawater intrusion.Fig. 2A schematic scheme as a general approach used to assess the resilience of a groundwater system.Fig 2
As tabulated in Table 1, all in all, 11 approaches have been utilized in different studies in order to assess groundwater resilience quantitatively, which are described in the following sections.Table 1A summary of the studies evaluating groundwater resilience in quantitative point of view. (In ** ai: state variable; amax: the maximum of the water table data; amin: the minimum of the water table data; AV: aquifer volume, ranking; BQ: Groundwater outflow or stream baseflow/quickflow, ranking; C: catchment inflow; D: demand; Dis: discharge regime; Dit: water deficit for region i in time t; WDit: water demand for region i in time t; GwL: groundwater level; GwR: groundwater recharge; GwRe: groundwater resilience; HL: groundwater outflow or half-life of stream flow recession, ranking; moment; L: the average groundwater drought; the base year; the half number of neighboring points; P: pumped volume; Pe: permeability; pi: resilience indicator; R: resilience; RI: overall aquifer resilience index; S: reservoir storage; Slo: slope; Sto: storativity; SYit: safe yield for region i in time t; time; Vi: intrinsic vulnerability/resilience; WB: Groundwater inflow or surface water baseflow, ranking; γ: the threshold of the deviation of ai from its adjacent points).Table 1ApproachAquiferModelEquation or descriptionReferenceThe ability of groundwater to feed river's base flowMarlborough aquifers–RI=BQ+HL+AV+WB[13]The percentage recovery to the total groundwater depletionKathmandu Valley, NepalSWAT, MODFLOWGwRem+1=GwR(m+1)GwLm−GwL(m+1)×100[73]Drought or groundwater droughtPingtung Plain, TaiwanHilbert Huang Transform-based methodDrought risk= Hazard Vulnerability Exposure[41]Pang catchment, EnglandNUTDAYR = 1/L[57]Grafham and Wimbleball, EnglandA simple reservoir behavioral model, i.e.St=St-1-Dt+Ct+PtDrought severity index, considering cumulative monthly shortages, were regarded to discuss groundwater resilience.[89]Conjunctive agricultural use in the aquiferRincon Valley, New MexicoRegression modelingSpecified as the extracted groundwater for irrigation per unit time divided by the total volume of surface water used for irrigation per unit time, groundwater- surface water ratio of application was considered as a criterion for resilience.[21]Fluctuations of groundwater levelAlbemarle-Pamlico peninsula, North CarolinaPIHM-Wetland modelGroundwater resilience was regarded (1) the capability of water table to return to its climatological mean value, and (2) groundwater level declining rate.[99]Upper Guadiana Basin, Spain–After a description of the system, e.g. its flows and functions, the perturbation, which was primarily groundwater pumping for agricultural irrigation and to some extent droughts, was evaluated. Thereafter, a new steady states was described, for which the changes in the hydraulic head of a representative observation well was taken into consideration.[14]Groundwater restoration time after pumpingHaw river basin, USAPIHMThe restoration time of groundwater under transient pumping and different potential climate scenarios was regarded as an index of resilience.[72]Northern Ethiopia–The recovery of groundwater systems after exploitation was examined as a criterion for resilience.[43]The standardized groundwater level or discharge rate indexUdaipur and Aravalli, India–Groundwater resilience index (GRI), which is actually based on the modified SPI, was employed.[9]Bibitarkhoun spring, IranANNs[95]Lali region, IranIHACRES, ANNs, MODFLOW, DrinCGroundwater resilience was calculated (1) The adverse of the mean duration of drought, (2) The inverse of the consecutive number of years with the GRI values less than −1, and (3) The ratio of the number of years with the GRI values greater than −1 to the total number of years.[96]Steady states analysis–Coupled hillslope Boussinesq-vertically lumped vadose zone modelThe state-space location of the threshold between the steady states were determined as an indicator ofresilience.[58]–A simple model of a spatially lumped unconfined aquifer, Lyapunov stability curvesSwitching between wet and dry periods occurs by trespassing from two steady state domains.[60]–A semi-distributed hillslope ecohydrological model, Limit Cycle ContinuationMultiple steady states occur under stochastic daily forcing. The change of climate forcing from monthly to daily time scales decreases the range of parameters with two attractors.[59]Water deficitVerde River Basin, Arizonaagent-based modeling (ABM), MODFLOWR=(no.oftimesDit=0followedbyDit>0)/(no.oftimesDit>0),[1]Isfahan-Borkhar, IranSWAT, MODFLOW, MT3DMS, MOPSOwhere Dit={WDit−SYit,if(WDit>SYit)0,if(WDit≤SYit)[33]Geospatial technologyAfricaGISThe considered layers encompassed groundwater recharge, groundwater storage and permeability.[44]Indo-Gangetic basinGISDespite not being combined, different layers were considered, including groundwater recharge and groundwater storage volumes and trends.[45]Dead Sea basinGIS, J2000The integrated thematic layers included transmissivity, storage, recharge, discharge, lithology, and depth of groundwater level.[6]Susan Karst, IranGIS, RES2DINV, IPI2WinFor evaluating the groundwater resilience two maps were created, (1) an integration of six layers, i.e. distance to springs, elevation difference from springs, slope, lithology, fracture density, and fracture length density, and (2) an interpolation of VES curves after scoring them.[53]Peninsular IndiaGIS, conceptual water balance model, a convex modelAi={ai−n,ai−n+1,…,ai−2,ai−1,ai+1,ai+2,…,ai+n−1,ai+n}fij={1,|ai−aj|≥γ0,|ai−aj|<γ[37]Hotan River Basin, ChinaGIS, Dempster-Shafer theoryγ=0.375(amax−amin)pi=∑j=i−nj=i+nfij[88]Durmitor, MontenegroGISVi=5×Pe+8×Sto+5×Dis+2×Slo[74]Groundwater storageSeveral aquifers in the worldGRACEDetermining the excess of water consumption from the available renewable water in the aquifer, renewable groundwater stress ratio was employed as a criterion for groundwater resilience appraisal.[64], [65]Several aquifers in the worldGRACEResilience equals number of times a satisfactory condition follows an unsatisfactory condition per total number of unsatisfactory conditions[78]Central Valley, California[77]
According to this approach which was used for the aquifers in Marlborough [13], an aquifer may supply a river baseflow consistently, relatively inconsistently, and very inconsistently, for which groundwater resilience is high, moderate, and low, respectively. In addition, the overall aquifer's resilience index (RI) was calculated as RI=BQ+HL+AV+WBwhere, BQ is groundwater outflow or stream baseflow/quickflow, HL is groundwater outflow or half-life of stream flow recession, AV is aquifer volume, and WB is groundwater inflow or surface water baseflow, for all of them ranking was used.
By assessing the groundwater system in Kathmandu Valley under climate change, Nepal, Shrestha et al. [73] described groundwater resilience as the percentage recovery to the total groundwater depletion at a given time, as the following GwRem+1=GwR(m+1)GwLm−GwL(m+1)×100in which, m, GwRe, GwR, and GwL are the base year, groundwater resilience, groundwater recharge, and groundwater level, respectively.
When groundwater systems are influenced by drought, groundwater recharge at first declines, and then, groundwater level and discharge decrease. These droughts are defined as groundwater droughts, which may occur at dissimilar time spans from months to years [85]. Peters et al. [57] described an approach to compute the groundwater drought time period in Pang catchment, England. They generated ten 1000-year time series of groundwater discharge and recharge using NUTDAY model. Next, they employed daily observational climate data in 37 years considering nearest neighbor resampling approach. The groundwater system was simulated utilizing a linear reservoir model, and drought lengths were extracted from the groundwater recharge and discharge observations by threshold level approach. Resilience (R) was defined as the reverse of the average groundwater drought time (L), as follows [57]:R=1L
Considering groundwater resilience as the reverse of drought vulnerability, Lin et al. [41] computed the spatiotemporal drought risk in Pingtung Plain, Taiwan, as Droughtrisk=Hazard×Vulnerability×Exposure
Drought hazard was described using the standardized precipitation index (SPI) and the dynamic drought intensity (DDI). The latter itself was calculated based on the SPI values, and its formula is as the
DDI=droughtmagnitude×droughtduration
Drought magnitude is the cumulative amount of three-month Standardized Precipitation Index (SPI) values in a drought event, and drought duration is the duration (month) of the drought event. Drought vulnerability was calculated considering Hilbert Huang Transform (HHT) approach. As it was declared, the lower the drought vulnerability is, the stronger the groundwater resilience is expected, and vice versa. After analyzing the water table data using HHT, the influences of various environmental factors, e.g. precipitation and pumping, became more obvious. The proportion of precipitation-affected signal in groundwater was considered as drought vulnerability and quantified from the amplitude of precipitation-affected Intrinsic Mode Functions (IMF) signal to the original average. To end with, drought exposure was considered as the water demand to supply people's livelihoods for a certain population density. In other words, drought exposure is population density, which may be defined as population per area (km^2^).
Watts et al. [89] considered the drought severity index as an indicator of resilience of water systems. Cumulative monthly shortage was applied to determine this index, according to that, a drought begins by beginning a period with negative deficiency, and it ends when a termination criterion is reached, e.g. when flow reaches above the average discharge for at least three months. However, as it was declared, the approach is primarily applied to surface water systems.
Fuchs et al. [21] quantified groundwater resilience in the Rincon Valley, New Mexico, considering the conjunctive use for agricultural irrigation in a constrained groundwater system. Indeed, the water table fluctuation (WTF) approach was regarded to spatiotemporally evaluate the conjunctive use controls regarding the net storage changes. To this end, groundwater- surface water ratio of application (GSRA) was introduced as an indicator of the system's steady state and resilience. GSRA was specified as the extracted groundwater for irrigation per unit time divided by the total volume of surface water used for irrigation per unit time. The term was applied to a common river basin and hydraulically interrelated groundwater system. A condition in which net change in the aquifer's storage is zero was announced as the steady state. The benefit of GSRA is that it can be applied even in the scarcity of some information about the groundwater budget elements, such as evapotranspiration which its calculation accompanies by uncertainties. Moreover, for the considered aquifer, groundwater exploitation together with consumption of available surface water for irrigation are the main variables affecting the aquifer's storage.
The resilience of a wetland in Albemarle-Pamlico peninsula, North Carolina, was studied using Penn State Integrated Hydrological Model for wetland hydrology (PIHM-Wetland) model on daily, seasonal and annual time scales [99]. Groundwater resilience was regarded as one of the most indicators of the wetland, for its evaluations two properties were (1) the capability of water table to return to its climatological mean value, and (2) groundwater level declining rate. For this purpose, the mean groundwater level along with its standard deviation was illustrated to track the variations in hydraulic head of the aquifer. The fluctuations in groundwater level were considered as an index of their resilience, for which its deviation from the mean with one standard deviation of water table together with the climatological mean precipitation were regarded. A possible tipping point, a term in resilience appraisal, was suggested to reach when an apparent groundwater level drawdown is distinguished. Lastly, it was stated that the multi-scale variants of groundwater level under dry climatic situations may be a good gauge of wetland hydrologic resilience to drought.
The approach conducted by de la Hera-Portillo et al. [14] included three steps to evaluate the groundwater resilience in the Upper Guadiana Basin, Spain, which is under climatic and anthropic stresses. First step comprised a description of the system, e.g. its flows and functions. Called as a thermometer, a wetland was announced as the major discharging point of groundwater in the region. The stage before pumping the groundwater system was considered as the reference stage in terms of groundwater resilience. The perturbation was described at the second step, which was primarily groundwater pumping for agricultural irrigation and to some extent droughts. Third step encompassed an explanation of the new steady state, for which the changes in the hydraulic head of a representative observation well was taken into consideration.
The fully-coupled model of PIHM was developed by [72] in order to compute the restoration time of both surface water and groundwater under transient pumping and different potential climate scenarios in Haw river basin, USA. In fact, the restoration time was taken into regard as a criterion for resilience. Results highlighted that surface water recovered more quickly than groundwater to a steady state subsequent to pumping ending. Moreover, in the studied area, groundwater was shown to be more vulnerable to climate change than pumping.
MacDonald et al. [43] monitored the performance of springs and wells over the northern Ethiopia in 18 months. After gathering data from these water supplies, their recovery after groundwater abstraction was examined as a criterion for their resilience. It was found that hand-dug wells are the most unswerving water supply over the locality, such that they recover in less than hours after daily abstractions.
Groundwater resilience of hard rock formations in Udaipur and Aravalli, India, was evaluated in order to improve the sustainability of these resources. Although the concept of SPI had been used to assess drought, the Groundwater Resilience Index (GRI) being based on the modified SPI was used therein. This index expresses the normalized continuous groundwater level variability function. GRI values usually vary between −4 and +4, in which smaller numbers indicate more stressful conditions on the aquifer. Results presented less changes in GRI compared to SPI in the studied area, which indicates less variations in groundwater than precipitation [9].
Zeydalinejad et al. [95] forecasted the resilience of Bibitarkhoun karst spring, southwest Iran, to climate change in the future time period, i.e. 2021–2050, by taking into account the standardized discharge rates and dynamic storage of the aquifer as the GRI values. The former and latter were calculated based on Artificial Neural Networks (ANNs) and spring's loss curves, respectively. Results unveiled that the resilience is probably more reliable if determined by the spring's discharge rates than the dynamic storage volumes of the aquifer. The reason was attributed to the logarithmic nature of the method used to determine the spring's dynamic storage.
Zeydalinejad et al. [96] proposed an approach to calculate the resilience of different water systems in the Lali region, southwest Iran, toward climate change. After calculating the SPI and GRI using the Drought Indices Calculator (DrinC) [80], three approaches were introduced to compute the resilience as a number between 0 and 1, (1) The adverse of the mean duration of when a drought occurs, the GRI values become equal or less than −1. It ends when the values reach to a positive number. (2) The inverse of the consecutive number of years with the GRI values less than −1, and (3) The ratio of the number of years with the GRI values greater than −1 to the total number of years. Results highlighted that the alluvial aquifer and one of the limestone wells in the region had the highest and lowest resilience to climate change, respectively.
A groundwater system encompasses multiple steady states, such that a groundwater steady state may vary and reach to a new one subsequent to a perturbation [58]. The resilience of an attractor, which is a steady state, may be defined as the distance of the state variable from the attractor to a repellor, which is the threshold between steady states [40]. Groundwater resilience was quantified using this definition by Peterson et al. [58], [60] and Peterson and Western [59].
By evaluating multiple steady states in hydrologic models, Peterson et al. [58] declared that they are not considered in most surface water and groundwater physical approaches. Hence, after a non-steady hydrologic perturbation with any magnitude, model may reach to a similar steady state, in such a way that resilience may be incomprehensible. On the topic of this difficulty, a numerical ecohydrological model, i.e. coupled hillslope Boussinesq-vertically lumped vadose zone, was used, in which different groundwater levels occur at steady states for the similar parametric set. Two groundwater steady states were depicted for a coupled groundwater-vadose zone model. It was concluded that the groundwater level steady state may not be dependent on the initial conditions. On the other hand, as it was announced, all basins may not show multiple steady states, and it hinges on the factors controlling the horizontal saturated flow and recharge, including hydrogeology, boundary conditions and basin's shape.
The groundwater resilience of an unconfined aquifer was assessed using an analytical approach by Peterson et al. [60]. A simple lump model along with one- and two-parameter continuation analyses of single ordinary differential equations were regarded. The shape of every steady state attractor basin was displayed utilizing Lyapunov stability curves, considering a quantitative precipitation domain. It was found that switching between wet and dry periods occurs by trespassing from two steady state domains. Plus, temporary steady states were determined using the approach; however, the analytical method may not be employed as coupled non-linear differential equations may be primarily representative of complex groundwater systems.
Peterson and Western [59] proved that inclusion of positive feedbacks in hydrological models lead to compound behavior with multiple steady states. When a large disruption, such as a long-term drought, is simulated, the hydrological situations may change. When the disturbance culminates, the hydrological situation may not return to the state before the disturbance, and it may remain in an alternative state. These multiple steady states happen in the parameterization of each individual model and make the system have a specific resilience to transient disturbances. A semi-distributed ecohydrological model was used to determine whether multiple hydrological attractors exist or not when a basin is exposed to random daily forcing. The model exhibited multiple steady states under average monthly climate forcing. The attractors and the threshold between them were quantitatively determined using the Limit Cycle Continuation (LCC) technique upscaling climate forcing from daily to monthly time steps. In fact, the method was used to determine the horizontal saturated hydraulic conductivity by considering multiple stable states. Results revealed that multiple steady states exist under stochastic daily forcing. Nevertheless, the change of climate forcing from monthly to daily significantly reduces the range of parameters in which two attractors exist, which means fewer basins with multiple attractors exist in comparison with what previously considered.
Sustainable groundwater management was regarded by developing an integrated quantitative-qualitative simulation-optimization model in Isfahan-Borkhar region using SWAT, MODFLOW and MT3DMS model in GMS software environment, Multiple Objective Particle Swarm Optimization (MOPSO) and MATLAB code [33]. The output of the SWAT model, i.e. groundwater recharge rate and its temporal-spatial distribution, was considered as the input of MODFLOW in order to investigate the changes in groundwater storage and flow. The groundwater velocity obtained from the MODFLOW was considered as the input of MT3DMS in order to evaluate the state and transport of Total Dissolved Solids (TDS) concentration. MODFLOW-MT3DMS-MOPSO integrated model was used to solve the groundwater management problem of the aquifer. Sustainability indicators were calculated using the performance criteria of reliability, resilience and vulnerability for each cell using MATLAB software, considering three management scenarios, viz. maintaining the current conditions, 30% increase in performance and 30% decrease in performance. MATLAB software receives the hydraulic head and TDS concentration in each cell, which are the output files of MODFLOW and MT3DMS, respectively, and then the performance indicators are calculated. Finally, the performance indicators obtained for each cell was shown in GMS software. The average of all grid cells was considered as an indicator of the stability of the whole aquifer. Using the equations initially represented by Sandoval-Solis et al. [70], groundwater resilience was quantified. Firstly, deficit (Dit) was discussed, which is zero for a time period (t) when water demand (WDit) in a city (i) is less than safe yield (SYit), as Dit={WDit−SYit,if(WDit>SYit)0,if(WDit≤SYit)
Then, Resilience (R) of the water system was defined as its ability to adjust to ever-changing circumstances, as
R=(no.oftimesDit=0followedbyDit>0)/(no.oftimesDit>0)
Furthermore, the socio-hydrologic groundwater resilience in an aquifer in Verde River Basin, Arizona, was evaluated by Al-Amin et al. [1] using the same equations and an agent-based modeling (ABM) combined with a groundwater model, i.e. MODFLOW.
Different Geographic Information System (GIS)-based layers which have influence on groundwater resilience may be produced, weighted and combined in order for a groundwater resilience map to be created in a spatiotemporal approach.
Different GIS-based layers, e.g. groundwater recharge, groundwater storage and permeability, were produced and combined in order to investigate the groundwater resilience in Africa [44]. In regard to groundwater storage, effective porosity and saturated thickness of the aquifer were used. Besides, transmissivity and thickness of the aquifer were considered to obtain permeability. Since transmissivity and effective porosity data were not available in most parts of Africa, the indices of these parameters were employed. Instead of transmissivity, yield or aquifer productivity in wells with appropriate spatial distribution was used, and the characteristics of flow and storage of different rocks represented the effective porosity. The saturation thickness of the aquifer was estimated using boreholes and available reports. To access the groundwater recharge data, the global WaterGap model [16] was employed. This study showed that groundwater has high resilience toward climate change over the African continent. It was also suggested that increasing the access to groundwater resources through manual pumps might be very successful. However, the strategies related to the availability of extensive volume of groundwater with high sustainability were stated to accompany with issues. Besides, as it was asserted, large groundwater reserves are located in the main sedimentary basins that can accommodate wells with high water yield. However, not only are they far from population centers, but their groundwater depth is also deep.
The groundwater resources in the Indo-Gangetic basin, including the countries of India, Pakistan, Bangladesh and Nepal, were investigated by MacDonald et al. [45] using a GIS-based approach. Considering the uncertainty of precipitation in the future and the possibility of continuous increase in groundwater extraction, evaluating the resilience of groundwater resources was of vitality for the region, as it was articulated. Factors such as freshwater storage volume, permeability of aquifer and long-term recharge were considered. In order to investigate the resilience of groundwater systems, two approaches were 1- identifying the volume and distribution of available freshwater resources as an estimate of the capacity of the aquifer in buffering the changes in recharge or discharge, and 2- determining the changes in groundwater storage in the alluvial aquifer system and the relationship of these changes to the existing groundwater recharge and withdrawal as an indicator of the impacts of future climate and pumping pressures. In terms of the second approach, the changes in groundwater storage were calculated according to the annual deviations of the groundwater level using available maps, databases and groundwater level monitoring points. The volume of the available fresh groundwater resources was estimated by using the specific yield and thickness of the alluvium, and thence, groundwater was divided into different types based on its salinity. Next, the annual change of groundwater level was combined with specific yield maps to estimate the annual changes of groundwater storage. Results indicated that groundwater storage, groundwater recharge, water salinity, and average flow rate of the rivers in the studied area were 7000 km^3^, 200 km^3^, 1000 mg/l, and 1250 km^3^/year, respectively. It was avowed that the high volume of groundwater storage has a buffer capacity on short-term changes in groundwater extraction and climate change influences; however, even a small decrease in groundwater level can affect the ecosystem, river flow and shallow wells. It was also mentioned that even though most springs in the region were not resilient to climate change, a small flow still remained during droughts. A decadal age of groundwater was portrayed in Nepal, indicating the resilience of the aquifer to the annual changes in precipitation. Also, though declining the reliance on springs, the extensive use of shallow groundwater resources by digging hand wells in flood plain sediments in the last five to ten years was announced as a factor in the resilience of communities to climate change. To end with, being typically isolated from the current climate, the deeper groundwater resources in the Bengal basin indicated a relatively high resilience to climate change and pumping.
A combination of different layers in GIS was used to determine the resilience of groundwater resources to climate change in the east of the Dead Sea basin [6]. The weight assigned to transmissivity, storage, recharge, discharge, lithology, and depth of groundwater level was 5, 4, 3, 3, 2, and 1, respectively. J2000 model was used to estimate the groundwater recharge. In the basin, precipitation would decrease by 21%, and temperature would increase by 2–3 °C, leading to changes in groundwater level, storage and recharge in the future time period, i.e. 2020–2080. The outputs of the model indicated a 30% reduction in the amount of groundwater recharge in 2050. To conclude, considering the future climate change and groundwater modeling, significant effects on groundwater resilience was proclaimed to occur in the studied area.
The resilience term was employed to Susan karst, southwest Iran, using geophysical and GIS-based approaches [53]. Vertical electrical soundings (VESs) were interpreted, scored, and interpolated to produce a groundwater resilience map. In regard to the GIS approach, six criterial maps were generated, reclassified, weighted, and overlaid, i.e. distance to springs, elevation difference from springs, slope, lithology, fracture density, and fracture length density. Due to disparate surface and subsurface characteristics of the considered karst, geospatial and geophysical resilience maps disclosed dissimilar resilience for Asmari and Ilam-Sarvak karst formations. All in all, as it was concluded, the Ilam-Sarvak formation with conduit flow and low storage has relatively low resilience, and the Asmari formation with diffuse flow and large storage portrays high resilience.
Stevanović et al. [74] used a GIS-based approach to calculate the vulnerability/resilience of karst aquifers in Durmitor area of Montenegro. An equation for computing the vulnerability/resilience of the considered groundwater systems was proposed, as Vi=5×Pe+8×Sto+5×Dis+2×Slowhere, Vi, Pe, Sto, Dis, and Slo are intrinsic vulnerability/resilience, permeability, storativity, discharge regime, and slope, respectively, among which storativity was declared as the most significant element.
The same equations were used to compute groundwater resilience over peninsular India [37] and Hotan River Basin, China [88]. The considered method simulated groundwater resilience over the region auspiciously. Water table was employed as the state variable (ai), and while the spatial changes were computed using GIS, the temporal changes in groundwater resilience at monthly time scale computed using the resilience indicator (pi). As it was detailed, groundwater resilience can be reflected by the deviance between the state variable and its adjacent values, i.e. ai-n, ai-n+1, …, ai-2, ai-1, ai+1, ai+2,…, ai+n-1, ai+n, in which n equals the half number of neighboring points, such that 2n=8 and n = 4. The larger the deviation is, the groundwater resilience becomes higher, indicating a lower autocorrelation in the time series of the state variable. Changing from 0 to 8, the resilience indicator (pi) was defined as Ai={ai−n,ai−n+1,…,ai−2,ai−1,ai+1,ai+2,…,ai+n−1,ai+n}fij={1,|ai−aj|≥γ0,|ai−aj|<γγ=0.375(amax−amin)pi=∑j=i−nj=i+nfijin which, i is moment, γ is the threshold of the deviation of ai from its adjacent points, and amax and amin represent the maximum and minimum of the water table data, respectively.
Groundwater resilience primarily depends on the storage capacity of groundwater systems [7], [30], [34], [44], [76]. Hence, groundwater storage may be used to appraise the groundwater resilience of an aquifer, as the research carried out by Richey [64], Richey et al. [65], Thomas et al. [78], and Thomas [77].
Emphasizing on the difficulties with the present procedures to assess the condition of large-scale aquifers, Richey [64] and Richey et al. [65] investigated the stress and resilience of the world's largest aquifer systems using Renewable Groundwater Stress (RGS) ratio, being defined as the ratio of groundwater consumption to its availability. The amount of groundwater withdrawal was estimated using the Gravity Recovery and Climate Experiment (GRACE) satellite observations, which provide spatial vertical data on changes in snow, soil moisture, surface water and groundwater storage. While the availability of groundwater was previously defined as the volume of aquifer recharge by some authors, e.g. [15], [55], [66], [86], they considered it as the total storage. By considering the groundwater storage, the buffer capacity of an aquifer can be judged. For instance, an aquifer system with low storage has little capacity to cover drought and pumping compared to an aquifer with high groundwater storage. Thence, the Total Groundwater Stress (TGS) ratio was determined in order to quantify the effect of groundwater withdrawal obtained from the satellite on the resilience and lifespan of the studied aquifers as a function of the total aquifer storage. TGS indicates the number of years in which a certain volume, e.g. 90%, of an aquifer's total reserve is drained. In fact, the aquifers being under high stress conditions in terms of the RGS and TGS ratios were considered to lack resilience due to high discharge with limited buffer capacity. It was declared that when the equilibrium cannot be sustained, at least in the considered time interval, groundwater storage would decrease based on Alley and Leake [4], which was considered as the turning point of trespassing the resilience limit of the aquifer. Results indicated that the highest RGS values were recognized for the aquifers encompassing all types of anthropogenic biomes, especially those in which agriculture and irrigation are dominant. Additionally, it was indicated that insufficient knowledge about the total volume of groundwater reserves still existed worldwide, causing difficulties in groundwater resilience studies.
The stability of groundwater in the main aquifers of the world was studied by Thomas et al. [78] with the criteria of resilience, reliability, vulnerability and their combination as a stability index using the groundwater storage derived from GRACE satellites. They considered resilience as the number of times a satisfactory condition occurs following an unsatisfactory condition per total number of unsatisfactory conditions. After determining the GRACE-Groundwater Drought Index (GGDI) and normalizing it, positive and negative values were considered as favorable and unfavorable conditions, respectively. Performance criteria were normalized by considering the maximum and minimum values in order to draw comparisons. The results of quantifying the resilience of the world's major aquifers showed that in dry areas, i.e. in the case of the North-Western Sahara aquifer system and the Arabian aquifer system, the resilience is low. These two aquifers were previously considered as non-renewable [47], on which excessive stress is applying [22], [23], [65]. Also, results presented that the aquifers with higher resilience, such as the West Siberian basin, the Russian platform basin and the Congo basin, are in the regions with high precipitation and northern latitudes. As it was stated, these aquifers could also be called resilient by considering other criteria, including the relationship between recharge and precipitation. It is of relevance to note that the same definition for resilience was used by Thomas [77] for evaluating the adaptive management of groundwater in the Central Valley, California, considering the Sustainable Groundwater Management Act (SGMA) program.
In terms of groundwater contamination, scarce research has hitherto been undertaken, i.e. two studies to the best knowledge of the author (Table 2). Lapworth et al. [39] researched the residence time of shallow groundwater with a depth of less than 50 m in West Africa using environmental isotopes in order to determine groundwater resilience against climate change. Shallow groundwater is quickly recharged after precipitation, and little evidence of evaporation before recharge exists, as it was affirmed. What is more, in dry regions with precipitation less than 400 mm/year, groundwater recharge was estimated to be up to 20 mm/year. According to different climatic zones, the average residence time of most groundwater systems was pronounced to be about 32 to 64 years. Besides, the considered aquifers exhibited similar hydraulic diffusivity and significant groundwater storage. In fact, the regarded water resources were mentioned to have high resilience against short-term changes in precipitation and recharge. As it was stated, young shallow groundwater systems with 1- to 2-decade age have the most sensitivity to climate variability. In addition, it was concluded that the aquifers with more than 100-year age usually do not receive significant modern recharge, and even though they are not directly sensitive to climate variability, they are vulnerable to groundwater discharge or mining. Besides, as it was stated, intermediate-aged aquifers have the most resilience toward high variability of precipitation. On the whole, it was declared that any future changes in precipitation and recharge due to climate change cannot possibly lead to a failure in improved groundwater supplies on a continental scale.Table 2A summary of the studies evaluating resilience in groundwater quality point of view.Table 2MethodAquiferVariablesDescriptionReferenceGroundwater residence timeWest AfricaEnvironmental isotopes, i.e. CFCs, SF6 and ^3^HGroundwater systems with different resilience were classified depending on the age of groundwater and its vulnerability to climate change and variability.[39]Hydrogeochemical and groundwater isotopic propertiesNdop plain, CameroonMajor ions, TDS, б^18^O and бDThe consistency of groundwater recharge and flow patterns was considered as a criterion for groundwater resilience evaluations to climatic variability in dry and wet seasons.[90]
The other research evaluated the hydrogeochemical and groundwater isotopic properties of the Ndop plain, Cameroon [90]. A key objective was to determine the groundwater resilience to seasonal climatic change. To this end, other than major ions and TDS, б^18^O and бD were regarded, based on which a consistent groundwater recharge and groundwater flow pattern was discerned in both dry and wet seasons. According to this finding, it was concluded that the considered groundwater system has resilience to current seasonal climatic variations.
Though being researched infrequently, evaluating the groundwater resilience of the aquifers for which seawater intrusion is plausible is of high vitality because not only may water table have fluctuations but water quality may also be deteriorated. Therefore, other than groundwater level, the saltwater/fresh water interface may be taken into account.
Katic and Grafton [34] assessed the resilience of a coastal aquifer, i.e. Guarani groundwater system, South America, by considering groundwater pumping – in terms of its rate and depth – in reference to economic payoffs. Accordingly, a dynamic model was developed by considering a perturbation method which was coded in Mathematica, capable of solving derivatives of the functions. To evaluate groundwater resilience, seawater intrusion was regarded as the irrevocable disastrous incidence occurring when groundwater level declines below a specific level, which can be called as the failure threshold. As it was discerned, given that the threshold limit is not certain, a higher economic return may be achieved by controlling both the rate and depth of pumping instead of just controlling its rate.
The resilience of three Mediterranean aquifers to seawater intrusion in the case of over-pumping was appraised by Mazi et al. [49] utilizing a generalized analytical sharp interface model in reference to normalized intrusion performance curves. The normalized position of the sharp interface toe was considered as the criterion for the appraisal of groundwater resilience. Two critical limits of seawater intrusion under groundwater pressures were identified, each of which can be reached, (1) a limit of well intrusion, at which intruded seawater spreads to crucial localities of groundwater pumping, and (2) a tipping point of complete saltwater intrusion up to the prevailing groundwater divide of a coastal aquifer. Results demonstrated different conditions, from seriously to less threatened, for each aquifer in relation to the extent of seawater intrusion.
The resilience of small-island freshwater lenses toward climatic variables and groundwater exploitation was appraised in Bonriki Island, Kiribati [61]. It was stated that because enough detailed data existed for the studied area, a groundwater resilience appraisal was achievable through a three-dimensional variable-density model, i.e. SEAWAT Version 4 which integrates MODFLOW and MT3DMS. Moreover, groundwater recharge was estimated using the water balance model WATBAL. Field observations along with the developed model could answer several questions, e.g. freshwater storage and the fluxes of submarine groundwater discharge could not be estimated using the detailed field data but they were quantified utilizing the model. Results unveiled the high dependency of the freshwater lens on both pumping and precipitation. It was also announced that the total volume of groundwater discharge, the pumping rate and the timing of discharge are of cruciality in the freshwater lens contraction. Besides, a threshold of 2500 m^3^/d recharge input was considered as the time when the lens thinning occurs. Last of all, in relation to the aquifer's recovery after groundwater pumping, it was pointed out that the island's lens was still contracting even after 27.5 years of pumping, which is accompanied with deteriorating the groundwater quality.
A numerical approach, i.e. MODFLOW, was employed to evaluate the socio-hydrological resilience of the Valley of Santo Domingo– an aquifer prone to saltwater intrusion– toward human-related and climate change scenarios. In truth, after the aquifer was simulated by the model through SWOT and PEST analyses, its response to various potential scenarios was achieved. Groundwater types in regard to saltwater intrusion were evaluated using a Piper trilinear diagram. According to the model's output, it was concluded that provided that the participation of local groundwater users and managers occurs, groundwater quality and quantity is secured. Moreover, different internal and external aspects in human resilience were tabularized [91].
Groundwater resilience was quasi-quantitatively appraised in a coastal aquifer, namely Manglaralto in Santa Elena, Ecuador, by configuring a matrix considering political, economic, social, environmental and cultural aspects by Herrera-Franco et al. [26]. At first, different scores were assigned to dissimilar sub-indicators; then, the resilience value (R) was calculated as a weighted sum of the considered axes, i.e. political or economic axis (P), environmental axis (E), social axis (S) and cultural axis (C). For the purpose, the following equation was R(%)=∑i=1nSPiN+∑i=1nSEiN+∑i=1nSSiN+∑i=1nSCiN4×100,where, N is the addition of the maximum values of every sub-indicator of each axis, 4 is the number of axes being regarded, and n is the number of sub-indicators for each axis. The system was suggested to have very high resilience, high resilience, resilience, and low resilience for the calculated values of 75–100, 50–75, 25–50, and 0–25, respectively. Results unraveled a high resilience in the aquifer because of water harvesting and other practical standards leading to sustainable utilization, management and preservation of the system.
The resilience of a groundwater system may be studied in regard to one disturbance; however, when the ability of the aquifer in obliteration of that unfavorable situation is studied, another new disturbance may be applied to the system. In truth, other than environmental factors, socio-hydrological [31], [91], livelihood [62], economic [34], and even a combination of political, economic, environmental and cultural features [26] may be considered in assessing the groundwater resilience of a system.
In the contemporary era, human and climate change are simultaneously affecting aquifers, such that separating their influence is a difficult task even though a few studies exist which have evaluated aquifers without pumping wells [54], [95], [96], [97]. As an illustration, the Central Valley aquifer was stated to have moderate resilience, which was related to the ineffective recovery from the 2011 drought [78]; however, that moderate resilience was deteriorated due to excessive groundwater extraction and rapid groundwater depletion during drought [17]. The spatial changes of sustainability indicators are attributed to the complex interactions between the climate and groundwater use [77].
In studying groundwater resilience, magnitude, duration and scale of an unsatisfactory condition must be taken into account, as de la Hera-Portillo et al. [14] apprised. The ability of the system to withstand perturbation is also of significance; hence, the hydrodynamic characteristics of the aquifer are of importance. Plus, the initial steady state of a system may vary after a disturbance, and a new one may be reached. As a result, multiple steady states must be regarded [59].
Provided that an aquifer be under a disturbance, its different aspects may show dissimilar resilience, e.g. groundwater level may reach to a new steady state after a stress period while it may take long time for groundwater quality to be improved [14].
Due to inducing stresses on groundwater reserves, different, even some new, challenges may be prevailing on the aquifers, making groundwater resilience calculation more problematic ever-increasingly. For descriptive purposes, the number of groundwater monitoring points may be decreasing, as climate change and human influences are declining the groundwater level. Ergo, the observation wells might be deepened in order that collecting data to be achievable.
Existing enough and accurate data is one of the drawbacks in evaluating groundwater resilience. This problem is more obvious in the developing countries. Climatic data are required for evaluating the groundwater-related climate change studies. Not only are observation data necessary, but also simulated climate data are requisite for forecasting the groundwater response to climate change in the future. These observation data may not exist in some localities. Otherwise, global pre-downscaled climate data sets may be used in such a case [98]. Even if adequate climatic data might be accessible, the climate system is non-linear, and hence, its projection is difficult, especially in regard to precipitation in comparison with temperature [54]. Forecasting a drought period is also difficult, especially in long time spans in the future [100]. Most other environmental phenomena are non-linear in nature, e.g. seawater intrusion [49], and because of this non-linearity not only is determining the required relationships difficult but it also changes temporally.
In respect to hydrogeological observations, hydraulic head is the most important unknown variable in groundwater modeling. However, observation wells may not be sufficient and properly distributed in spatial perspective. Hence, the representative hydrograph of an aquifer may not represent the real characteristics of the groundwater system. The depth of observation wells may not be archetypal of the aquifer, and multilevel groundwater monitoring systems may not exist for assessing the vertical groundwater flow.
Groundwater monitoring points are primarily located in alluvial aquifers, such that less observation wells exist in karst and hard-rock aquifers; ergo, studying these groundwater systems is even more challenging. Even if observation wells occur, modeling the responses of these aquifers and calculating their resilience is difficult due to their complex hydrogeological characteristics [95], [96], [97]. Further, other than a karst spring's discharge rate, other variables must also be monitored to evaluate the karst aquifers, viz. electrical conductivity (EC) and temperature.
The hydrodynamic properties of an aquifer, i.e. transmissivity and specific yield, may not also exist or their precision may not be high enough. Indeed, these characteristics, which are of highly significance for physical (numerical) groundwater modeling [92], may be estimated during the calibration of the model, especially using a trial and error approach. Hence, non-numerical approaches, viz. meta-heuristic ones, may be applied. However, one non-numerical approach may not be able to simulate the aquifer's response considering a specific input; in this case, various approaches might be used, and the best model can be selected based on the statistical criteria [93]. All in all, numerical and non-numerical approaches have their own limitations, as detailed by Zeydalinejad [92], which are among the difficulties with evaluating the groundwater resilience. Even though it was previously stated by Intergovernmental Panel on Climate Change (IPCC) that groundwater resources would decline in most dry subtropical localities, including sub-Saharan Africa, Cuthbert et al. [12] challenged this consensus by accentuating that augmented spatial coverage of long-span groundwater monitoring is required in order to contend the difficulty with the model verification. In addition to the uncertainties with groundwater modeling, the uncertainties occur with the general circulation models, emission scenarios, and downscaling approaches.
Resilience is not dimensionless, and hence, making a comparison of groundwater resilience between different localities is difficult [57]. Therefore, this concept should be studied more meticulously in order to be applied to groundwater systems in terms of sustainability. Moreover, after a land subsidence, which is irreversible, employing groundwater resilience may be meaningless, as the system may experience a permanent failure [24]. Not only may it take long time to reach to an equilibrium in groundwater systems even for small aquifers such as the Bonriki Island [61], but reaching to a new one may be impossible altogether [24]. In these conditions, assessing the resilience of the aquifer may be inconsequential. As well, it is unrealistic to consider complete drainage for an aquifer because water quality, its availability, and soil properties limit the amount of groundwater which can be extracted [2].
In the current study, the categorized methods for assessing the groundwater resilience are both in quantitative and qualitative viewpoints. On condition that groundwater resilience be merely regarded quantitatively, the applied techniques become even fewer. One difficulty with quantitatively evaluating the groundwater resilience is determining the failure threshold of an aquifer. Irreversible and uncertain events are of vitality in the groundwater resilience science; however, a very limited literature exists accounting both factors, e.g. [11], [34], [35], [51], [52], [83], [84]. Different variables have been considered to assess the groundwater resilience for an aquifer, e.g. groundwater level, discharge rate and storage, among which estimation of the latter includes much more difficulties and uncertainties. Due to the fact that in most cases it is not clear how much groundwater exists, it is difficult to quantify the resilience of the aquifer [65]. In addition, an aquifer that reduces the base flow discharged to the rivers may not be considered resilient in an integrated groundwater-surface water system [65].
Most of the studies of groundwater on a global scale have used remote sensing methods, which have obtained valuable information regarding the changes of groundwater storage on a large scale [77]. However, studies based on remote sensing do not provide sufficient resolution to detect local or even regional scale changes in groundwater [3]. As a result, the use of remote sensing data for sustainable groundwater management is ambiguous [77].
Resilience depends on the scale of study [50], [79], [99], such that a long-term large scale study may not show the response of the system to a short-term small scale disturbance, such as a flash flood, and this process may be more obvious in a karst aquifer than an alluvial aquifer. In the case of the GRACE satellites, the accuracy of the estimate enhances with the increase of the scale of the studied region [87].
One approach for estimating groundwater storage is making use of GRACE satellites, which investigate monthly gravity anomalies related to monthly changes in continental water storage [68]. One of the limitations of the GRACE observations is that it observes changes in groundwater storage in one dimension, while the main aquifer processes occur in three dimensions [3]. Additionally, the effect of soil moisture in extracting the groundwater storage indicator signal obtained from the GRACE satellites is ambiguous [75].
Different portions of an aquifer may have dissimilar groundwater potential depending on hydrogeological characteristics of the system, and therefore, the groundwater resilience differs throughout its domain, like the study conducted by Shrestha et al. [73]. In this case, geospatial technique may be used to portray groundwater resilience over the aquifer area. Otherwise, other approaches may be employed quantifying the groundwater resilience of the whole system.
As declared by Chinnasamy et al. [9], considering different lengths of time series may result in dissimilar groundwater resilience values. In addition, various normalizing approaches may lead to dissimilar GRI values. Considering only one variable at a time, the standardized groundwater level approach may not be employed auspiciously for regions with low seasonal precipitation and short time series.
As groundwater resources are currently under both human and climate induced pressures, appraisal of groundwater resilience may be considered as an emergency need. All the studies which have hitherto evaluated groundwater resilience were overviewed. While little research has appraised groundwater resilience quantitatively, as for groundwater quality there exist even fewer studies.
Quantitative groundwater resilience evaluations contain several approaches, including the ability of groundwater to feed river's base flow, the percentage recovery to the total groundwater depletion, appraisal of groundwater drought, fluctuations of groundwater levels, groundwater restoration time after pumping, application of standardized groundwater level or spring's discharge rate, analysis of steady states, estimation of water deficit as a criterion for groundwater resilience computation, spatiotemporal calculation of groundwater resilience using geospatial technology, estimation of groundwater storage, and the time when a specific percentage of aquifer storage is consumed. The employed approaches have been used for different aquifers from large-scale (e.g. [44]) to small-scale (e.g. [96]) ones. Various groundwater modeling approaches have been employed, including Regression modeling, analytical approach, SWAT, MODFLOW, ANNs, GIS and GRACE (Table 1). In some studies, an appraisal of the observation data has been considered without groundwater modeling (e.g. [14]). In these approaches, the groundwater resilience has been computed as a number (e.g. [96]) or classified, e.g. as high, intermediate or low (e.g. [44]). In each category, different variable/ or variables have been considered as a way to represent groundwater resilience, encompassing drought, groundwater level, spring's discharge rate, groundwater storage, groundwater recovery time, river's base flow, water deficit, and hydrodynamic properties of the aquifer (Table 1).
All the procedures used to quantitatively evaluate the groundwater resilience of an aquifer have their own benefits and drawbacks, among which evaluating the groundwater storage may be considered as one of the best approaches. In fact, by assessing the groundwater storage and groundwater budget of an aquifer, the managerial strategies may be introduced in order to maintain its invaluable groundwater reserves. However, the groundwater storage of an aquifer system may not be quantifiable as it requires enough and accurate hydrogeological data. Therefore, appraisals on groundwater level, whether considering the standardized groundwater level or not, might be proposed as it only needs the hydraulic heads in observation wells.
In relation to groundwater resilience in terms of its quality, even a very limited literature has been published. They consider groundwater residence time [39] and the consistency of groundwater hydrogeochemistry [90] using hydrogeochemical and environmental isotopes. Moreover, in reference to seawater intrusion, groundwater resilience in coastal aquifers have been assessed in few studies, for which other than groundwater level, saltwater intrusion which also deteriorates groundwater quality, has been taken into account.
Different challenges occur in evaluating groundwater resilience, among which data scarcity is the most significant difficulty, which is specifically of more cruciality in developing countries, like Iran. Besides, very limited research together with a lack of a specified way for the appraisal of groundwater are other drawbacks in groundwater resilience studies. Anyway, to study groundwater resilience for an aquifer, each of the approaches or maybe a combination of which may be employed depending on factors such as the approach which is going to be undertaken and the available data. Finally, as resilience has been employed in groundwater very recently, novel approaches may also be taken into consideration.
Nejat Zeydalinejad: Conceptualization, Methodology, Writing- Original draft preparation, Revising the manuscript, Visualization, Investigation.
Not applicable.
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.