Groundwater Level Time Series Research Articles

Machine learning models combined with wavelet transform and phase space reconstruction for groundwater level forecasting

Reliable forecasting of groundwater levels plays an important role in water resource management and the prevention of environmental problems. This study aims to introduce a novel hybrid model combining machine learning with wavelet transform (WT) and phase space reconstruction (PSR) for groundwater level forecasting. The development of the proposed hybrid model consists of two levels. Firstly, a popular data-driven nonlinear artificial neural network (ANN) model was selected as the base model. Thereafter, the WT is used to decompose the groundwater level time series. In addition, the PSR method is further adopted to select the suitable sub-series components resulting from the WT decomposition processes. At the first level, the hybrid model only combines the WT with ANN (WT-ANN). At the second level, a two-step preprocessing process, WT coupled with PSR, is implemented before the ANN, namely WT-PSR-ANN. Meanwhile, three mother wavelets (Daubechies, Symlets, and Dmeyer) are selected to further investigate the effect of wavelet type. Finally, the model's performance is evaluated using visual analysis and statistical metrics (root means square error, correlation coefficient, and the Nash-Sutcliffe efficiency coefficient). The results show that the WT-ANN model outperformed the vanilla ANN model. Furthermore, the WT-PSR-ANN model achieved the best accuracy for all selected mother wavelets. The most suitable mother wavelet is specific to each well, and there is no one-size-fits-all answer. This study highlights the effectiveness of PSR combined with WT as a data preprocessing step when the ANN model is applied for groundwater level forecasting.

Analyzing Groundwater Recharge and Vadose Zone Dynamics by Combining Soil Moisture and Groundwater Level Data with a Numerical Model in Subarctic Conditions

Quantitative estimates on the magnitude and dynamics of vadose and groundwater zones in subarctic conditions are rare. Furthermore, knowledge on the constraints of different data on numerical groundwater recharge estimates (i.e., data worth analyses) is limited. We built a process-based three-dimensional hydrological model to describe and analyze the dynamics and magnitude of groundwater recharge, vadose zone, and related key hydrological components in subarctic conditions. Three-year time series of daily groundwater level and soil moisture (five different depths) data were used to identify plausible model realizations. Thereafter, long-term simulations over 6 years were conducted. The model adequately corresponded with the magnitude and dynamics of the observations, even though the simulations and observations occasionally differed. Simulations with several different plausible parameterizations showed how soil moisture and groundwater level fluctuations, as well as recharge estimates, are sensitive to water retention parameter variations, whereas precipitation and evapotranspiration controlled the recharge magnitude. Recharge showed clear interannual variability (annual mean 0.28–0.60 m). The fraction of annual mean recharge over precipitation was 39%–66% during our simulation period. We showed how groundwater level data can set more constraints on the recharge estimates than soil moisture data, and how both of the data types together constrained the recharge estimates the most. We showed how the vadose zone had modest dynamics and storage properties (typically 2%–8% of the total soil water) in terms of the bulk soil water storage. However, the average amount of the air-filled pore space (0.70 m) and water (0.38 m) in the vadose zone were proportional to annual precipitation and evapotranspiration, respectively. Thus, the vadose zone can form an important boundary between the soil surface and groundwater zone, and its role is not highly sensitive to hydrometeorological fluctuations. Data worth analyses provide a potential opportunity to move toward increasingly reliable model-based groundwater recharge estimates and system understanding.

Assessing automated gap imputation of regional scale groundwater level data sets with typical gap patterns

Large groundwater level (GWL) data sets are often patchy with hydrographs containing continuous gaps and irregular measurement frequencies. However, most statistical time series analyses require regular observations, thus hydrographs with larger gaps are routinely excluded from further analysis despite the loss of coverage and representativity of an initially large data set. Missing values can be filled in with different imputation methods, yet the challenge is to assess the imputation performance of automated methods. Assessment of such methods tends to be carried out on randomly introduced missing values. However, large GWL data sets are commonly dominated by more complex patterns of missing values with longer contiguous gaps. This study presents a new artificial gap introduction approach (TGP- typical gap patterns) that improves our understanding of automated imputation performance by mimicking typical gap patterns found in regional scale groundwater hydrographs. Imputation performance of machine learning algorithm missForest and imputePCA is then compared with commonly applied linear interpolation to prepare a gapless daily GWL data set for the Baltic states (Estonia, Latvia, Lithuania). We observed that imputation performance varies among different gap patterns, and performance for all imputation algorithms declined when infilling previously unseen extremes and hydrographs influenced by groundwater abstraction. Further, missForest algorithm substantially outperformed other methods when infilling contiguous gaps (up to 2.5 years), while linear interpolation performs similarly for short random gaps. The TGP approach can be of use to assess the complexity of missing observation patterns in a data set and its value lies in assessing the performance of gap filling methods in a more realistic way. Thus the approach aids the appropriate selection of imputation methods, a task not limited to groundwater level time series alone. The study further provides insights into region-specific data peculiarities that can assist groundwater analysis and modelling.

Assessing long-term climate change impact on spatiotemporal changes of groundwater level using autoregressive-based and ensemble machine learning models

To evaluate the long-term climate change impacts on groundwater fluctuations of the Ardabil plain, Iran, a groundwater level (GWL) modeling was proposed in this study. Accordingly, the outputs of Global Climate Models (GCMs) under the sixth report of Coupled Model Intercomparison Project (CMIP6) and future scenario of the Shared Socioeconomic Pathway 5-8.5 (SSP5-8.5), were used as climate change forcing to the Machine learning (ML) models. The GCM data were first downscaled and projected for the future via Artificial Neural Networks (ANNs). Based on the results, compared to 2014 (the last year of the base period), the mean annual temperature may increase by 0.8 °C per decade until 2100. On the other hand, the mean precipitation may decrease by about 8% compared to the base period. Then, the centroid wells of clusters were modeled by Feedforward Neural Network (FFNN), examining different input combination sets to simulate both autoregressive and non-autoregressive models. Since each of the ML models can extract different kinds of information from a dataset, after finding the dominant input set via FFNN, GWL time series were modeled via various ML methods. The modeling results indicated that the ensemble of shallow ML models could lead to a 6% more accurate outcome than the individual shallow ML models, and 4% than the deep learning models. Also, the simulation results for future GWLs illustrated that temperature can impact groundwater oscillations directly, whereas precipitation may not have uniform impacts on the GWLs. The uncertainty evolving in the modeling process was quantified and observed to be in acceptable range. Modeling results showed that the main reason for the declining GWL in the Ardabil plain could be primarily linked to the excessive exploitation of the water table, while climate change impact could be also notable.

Predicting Groundwater Levels in Ogallala Aquifer Wells Using Hierarchical Cluster Analysis and Artificial Neural Networks

The Ogallala Aquifer, located in the Central Plains of the United States, is essential for agricultural irrigation and public water supply. Indiscriminate pumping from the aquifer has caused several negative impacts, such as deterioration of water quality and depletion of groundwater levels, which urgently demand better management. This paper applies hierarchical cluster analysis (HCA) and artificial neural networks (ANNs) for predicting annual groundwater levels in 403 wells of the Ogallala Aquifer. First, the methodology employed HCA to cluster homogeneous wells based on the time series of groundwater levels. Then, the study calibrated an ANN model for each cluster (composed of one or more wells) using previous annual values of groundwater levels as input. The HCA results showed a particular pattern in the spatial distribution of the 30 found clusters, revealing that the Ogallala Aquifer holds higher groundwater levels in the western part, which gradually decrease, advancing to the east. The ANN models provided proper predictions even for wells outside of the calibration data set. This investigation concludes that the integration of HCA and ANN enabled single models to accurately forecast annual groundwater levels for sets of wells in the Ogallala Aquifer.

Comparison of Three Imputation Methods for Groundwater Level Timeseries

This study compares three imputation methods applied to the field observations of hydraulic head in subsurface hydrology. Hydrogeological studies that analyze the timeseries of groundwater elevations often face issues with missing data that may mislead both the interpretation of the relevant processes and the accuracy of the analyses. The imputation methods adopted for this comparative study are relatively simple to be implemented and thus are easily applicable to large datasets. They are: (i) the spline interpolation, (ii) the autoregressive linear model, and (iii) the patched kriging. The average of their results is also analyzed. By artificially generating gaps in timeseries, the results of the various imputation methods are tested. The spline interpolation is shown to be the poorest performing one. The patched kriging method usually proves to be the best option, exploiting the spatial correlations of the groundwater elevations, even though spurious trends due to the the activation of neighboring sensors at times affect their reconstructions. The autoregressive linear model proves to be a reasonable choice; however, it lacks hydrogeological controls. The ensemble average of all methods is a reasonable compromise. Additionally, by interpolating a large dataset of 53 timeseries observing the variabilities of statistical measures, the study finds that the specific choice of the imputation method only marginally affects the overarching statistics.

Datasets of Groundwater Level and Surface Water Budget in a Central Mediterranean Site (21 June 2017–1 October 2022)

This note makes available five years of data gathered in a measurement site equipped with a micrometeorological station and two monitoring wells. Series of data of hydrological and atmospheric variables make it possible to estimate the flux of water across the atmosphere-land interface and to calculate the water budget, which are crucial topics in climate and environmental sciences. The water-table measures began during 2017, one of the driest years of the whole instrumental period of climate history for the Central Mediterranean. Data from the micrometeorological station have been used to construct two more datasets of daily and monthly totals of different terms of the surface water budget, from which the net infiltration has been estimated. An apparent decreasing trend characterizes both the data time series of groundwater level and estimated infiltration in the considered period.

Calculation model and application of ecological water demand of watercourses in low-carbon sponge urban areas based on DPSA algorithm

Abstract To improve the prediction ability of river ecological water requirement in sponge low-carbon urban areas, a prediction model of it based on big data analysis was proposed. A statistical analysis model of river ecological water requirement in sponge and low-carbon urban areas was established. Differential Particle Swarm Optimization Algorithm (DPSA) was used to calculate the characteristic quantity of river ecological water requirement in sponge and low-carbon urban areas. The detection model of ecosystem hydrothermal balance and lognormal distribution time series of groundwater level were constructed for the prediction of river ecological water requirement in sponge and low-carbon urban areas. Combined with the integrated control method of river ecological environment water demand forecasting and information exchange, a decentralized control model of river ecological water demand in low-carbon sponge urban areas was established. Through applicability analysis and model measurement analysis, combined with comprehensive index measurement, the prediction model of river ecological water requirement in low-carbon sponge urban areas was optimized. The empirical analysis results show that this method has good adaptability and high prediction accuracy for the prediction of river ecological water requirement in sponge low-carbon urban areas and improves the reliability of the prediction of river ecological water requirement and the matching level of urban ecological environment.

Groundwater Variability in a Semi-Arid River Basin, Central India

The Betwa River basin, a semi-arid catchment that has been classified as a major hotspot of groundwater depletion in Central India. The rainfall and streamflow intermittency have affected agricultural practices due to the variability of groundwater availability for irrigation. This study evaluates the spatial and temporal variations of groundwater level (GWL) in the last 25 years (1993–2018) in the catchment. We applied a nonparametric Seasonal Trend decomposition based on the Loess (STL) method to decompose the GWL time series into the seasonal, trend, and remainder components. We observed that the GWL in the northeastern regions of the basin has depleted about 3–5 mbgl in the last two decades. During the same period, the basin has experienced a reduction in the rainfall magnitude (2.07 mm/yr). We observed that the overexploitation of groundwater for irrigation and rainfall variability have greatly impacted the GWL condition in the study area. Further, if the groundwater extraction continues at present rates, the Betwa River basin may experience severe depletion in the future.

Assessing climate change impact on French groundwater resources using a spatially distributed hydrogeological model

Today, large-scale modelling tools are needed for anticipating the expected impacts of climate change on hydrosystems and for planning mitigation measures. We developed a spatially-distributed model for the MARTHE hydrogeological modelling computer code as a preliminary tool for France, and carried out a first evaluation over a 10-year period, comparing observed and modelled time series of groundwater levels and river flows. Then, prospective hydrological simulations were undertaken using five regionalized climate simulations. The evaluation results show contrasted performance depending upon the regions and variables considered, but they provide encouraging perspectives. Future hydrological projections anticipate increased recharge by +15% on average over France, accompanied by increases more wetter/dryer groundwater extreme events in the north/south by the 2070-2099 period compared to the 1976-2005 reference period. Based on these results, the MARTHE model can be considered as a promising template for the construction of a nation-wide hydrological model.

Climate impact assessment to the groundwater levels based on long time-series analysis in a paddy field area (Piedmont region, NW Italy): preliminary results

The analysis of the time-series of groundwater level are extremely important to observe the behaviours of groundwater over time and to identify any critical situations. The studied area is an agricultural district characterised by paddy fields, located in the eastern part of Piedmont, on the border with Lombardy. In this area long time-series of groundwater level, starting from the 1960s, have been collected in 16 wells. Water table data have a good completeness (in the majority of the cases >90%). Firstly, the groundwater hydrodynamic behaviour, based on water table levels, was investigated to highlight the response of groundwater to the recharge. A basic statistical analysis was performed (mean, median, standard deviation, maximum, minima), and then trends of water table levels were evaluated in order to better observe the long-term behaviour of groundwater. These analyses allowed to observe a groundwater hydrodynamic behaviour characterised by a repeating annual pattern (minimum in February/March and maximum in August/September) in correspondence to the period of irrigation. Moreover, trend analysis highlighted the presence of both wells with a decreasing water table (with maximum lowering of 4.3 m in 60 years) and wells with an increasing water table (with maximum rises of 2.8 m in 35 years). Furthermore, in most cases, it can be observed that all three trends analysed agree on being positive or negative. Future insights will be the comparison of these long time-series with the meteorological data, and the investigation of other factors (e.g. anthropic withdrawal, variations of cultivation practices and irrigation, geology of the subsoil) to better understand the causes of the water table fluctuations and trends.

Mapping the groundwater memory across Ireland: A step towards a groundwater drought susceptibility assessment

The occurrence of groundwater drought is closely linked to the meteorological input but also to the surface and subsurface properties, which function as lagging filter for the input signal.Knowledge about the potential occurrence of groundwater droughts is relevant for existing and future groundwater users, particularly in regions where climate change is expected to extend or increase the number of dry periods. This study proposes a method to quantify the groundwater memory as basis for characterising the intrinsic susceptibility of groundwater to drought.The memory of groundwater was estimated using a sliding window autocorrelation function applied on 114 groundwater level time series. A Random Forest regressor then modelled the groundwater memory across Ireland, using national digital maps as input files.The key variables explaining groundwater memory are: the relative and absolute surface topography and a thick overburden (>10 m). Accordingly, the lowest memory appears in elevated areas with overburden thicknesses of <10 m, and vice versa. Areas of low, moderate and high groundwater memory relate to high, moderate and low groundwater drought susceptibility. The uncertainty of the results is lowest in areas of low memory and highest in areas of high memory, presumably related to the distribution of the observations.The results are considered relevant in the context of water resources planning across sectors (agriculture, industry, domestic), particularly in the context of climate change adaptation.

Alleviating drought and water scarcity in the Mediterranean region through managed aquifer recharge

Drought and water scarcity can significantly impair the sustainable development of groundwater resources, a scenario commonly found in aquifers in the Mediterranean region. Water management measures to address these drivers of groundwater depletion are highly relevant, especially considering the increasing severity of droughts under climate change. This study evaluates the potential of managed aquifer recharge (MAR) to offset the adverse effects of drought and water scarcity on groundwater storage. Los Arenales aquifer (central Spain), which was unsustainably exploited for irrigation in the second half of the twentieth century, is employed as a case study. Two neighbouring zones within this aquifer are contrasted, namely, Los Arenales (LA) and Medina del Campo (MC). The primary difference between them in terms of water resources management is the wide-scale implementation of MAR systems in LA since the early 2000s. Several groundwater statistical methods are used. Groundwater-level trend analysis and average piezometric levels show in LA a faster recovery of aquifer storage and less susceptibility to drought compared to MC. On the other hand, standardised precipitation indexes and standardised groundwater level indexes of detrended groundwater-level time series, which do not include the effects of MAR, show that LA can be more negatively affected by drought and groundwater abstraction. The sharper recovery of piezometric levels in LA when considering MAR, and bigger drought impacts observed when the effects of this measure are removed, demonstrate that MAR can effectively alleviate the impacts of water scarcity and drought, providing an adaptation solution to climate change worldwide.

Exploring river–aquifer interactions and hydrological system response using baseflow separation, impulse response modeling, and time series analysis in three temperate lowland catchments

Abstract. Lowland rivers and shallow aquifers are closely coupled, and their interactions are crucial for maintaining healthy stream ecological functions. To explore river–aquifer interactions and the lowland hydrological system in three Belgian catchments, we apply a combined approach of baseflow separation, impulse response modeling, and time series analysis over a 30-year study period at the catchment scale. Baseflow from hydrograph separation shows that the three catchments are groundwater-dominated systems. The recursive digital filter methods generate a smoother baseflow time series than the graphical methods. Impulse response modeling is applied using a two-step procedure. The first step of groundwater level response modeling shows that groundwater level in shallow aquifers reacts fast to the system input, with most of the wells reaching their peak response during the first day. There is an overall trend of faster response time and higher response magnitude in the wet (October–March) than the dry (April–September) periods. The second step of groundwater inflow response modeling shows that the system response is also fast and that simulated groundwater inflow can capture some variations but not the peaks of the separated baseflow time series. The time series analysis indicates that groundwater discharge to rivers is likely following groundwater level time series characteristics, with a strong trend and seasonal strengths, in contrast to the streamflow, which exhibits a weak trend and seasonality. The impulse response modeling approach from the groundwater flow perspective can be an alternative method to estimate the groundwater inflow to rivers, as it considers the physical connection between river and aquifer to a certain extent. Further research is recommended to improve the simulation, such as giving more weight to wells close to the river and adding more drainage dynamics to the model input.

WITHDRAWN: Mapping the groundwater memory across Ireland: a step towards a groundwater drought susceptibility assessment

• National assessment of the memory of groundwater across the Republic of Ireland. • The machine learning model is able to simulate the observed memory. • The study provides a first known high resolution map of the groundwater memory. • Groundwater memory informs on the susceptibility of groundwater to drought. • Lowest drought susceptibility appears in low-elevated areas with thick overburden, and vice versa. The occurrence of groundwater drought is closely linked to the meteorological input but also to the surface and subsurface properties, which function as lagging filter for the input signal. Knowledge about the potential occurrence of groundwater droughts is relevant for existing and future groundwater users, particularly in regions where climate change is expected to extend or increase the number of dry periods. This study proposes a method to quantify the groundwater memory as basis for characterising the intrinsic susceptibility of groundwater to drought. The memory of groundwater was estimated using a sliding window autocorrelation function applied on 114 groundwater level time series. A Random Forest regressor then modelled the groundwater memory across Ireland, using national digital maps as input files. The key variables explaining groundwater memory are: the relative and absolute surface topography and a thick overburden (>10 metres). Accordingly, the lowest memory appears in elevated areas with overburden thicknesses of less than 10 m, and vice versa. Areas of low, moderate and high groundwater memory relate to high, moderate and low groundwater drought susceptibility. The uncertainty of the results is lowest in areas of low memory and highest in areas of high memory, presumably related to the distribution of the observations. The results are considered relevant in the context of water resources planning across sectors (agriculture, industry, domestic), particularly in the context of climate change adaptation.

Recognizing the relationship between the groundwater level and hydrological time-series: a case study of the Ardabil Plain

Abstract Two approaches were used in the Ardabil Plain to determine the relationship between hydrological time-series (rainfall and runoff) and the groundwater level (GWL). The wavelet-entropy measure (WEM) and the wavelet transform coherence (WTC) were used as two approaches of the wavelet transform (WT). The WEM was considered as a criterion for the degree of fluctuation of the time-series and the WTC represents a common time–frequency space. In the calculation of the WEM, the monthly rainfall, runoff, and the GWL time-series were divided into three different time periods and decomposed into several frequent time-series. Wavelet energies were then calculated for each sub-series. The WEM results show a decrease in rainfall, runoff, and the GWL. The reduction in the WEM represents the reduction in the natural variation of the time-series. The reduction of entropy for runoff, rainfall, and the GWL time-series was about 1.58, 1.36, and 29%, respectively. It can be concluded that the reduction in the fluctuation of the hydrological time-series has relatively little effect on the oscillation patterns of the GWL signal. In this regard, it could be concluded that human activities such as pumping water from wells play a main role in reducing the GWL in the Ardabil Plain. The WTC results show that runoff has the highest coherence (0.9–1) among hydrological variables with the GWL time-series in the frequency bands of 4–8 and 8–16 months.

Data-driven decision management of urban underground infrastructure through groundwater-level time-series cluster analysis: the case of Milan (Italy)

The significant increase in urbanization has resulted in greater use of the subsurface in urban planning and, therefore, increased interaction between groundwater and underground infrastructure. Numerical models are the primary tool adopted to manage the resulting problems; however, their construction is time- and cost-consuming. Groundwater-level time-series analysis can be a complementary method, as this data-driven approach does not require an extensive understanding of the geological and boundary conditions, even if providing insights into the hydrogeologic behaviour. Thus, a data-driven approach was adopted to analyse groundwater time-series of the shallow aquifer, occupied by several underground structures, beneath Milan city (Northern Italy). Statistical (Mann-Kendall and Sen’s slope estimator, autocorrelation and cross-correlation, hierarchical cluster analysis) and geospatial techniques were used to detect the potential variables influencing the groundwater levels of 95 monitoring wells, covering the period 2005–2019. A general rising trend of the water table was identified, with local hydrogeologic differences in the western and southernmost areas. Based on time-series analysis results, four management areas have been identified. These areas could act as future geographic units with specific groundwater management strategies. In particular, subsurface public car parks can be classified with respect to groundwater flooding as (1) not submerged, (2) possibly critical, or (3) submerged at different groundwater conditions. According to these outcomes, targeted guidelines for constructing new car parks have been elaborated for each management area. The methodology proved to be efficient in improving the urban conceptual model and helping stakeholders design the planned underground development, considering groundwater aspects.

A Method for Evaluating Coastal Underground Barrier Wall Using Groundwater Tidal Response.

Tidal response methods are usually used to estimate the hydraulic parameters of coastal aquifers. In this study an analytical model for aquifer tidal response was used. An existing analytical solution for tidal response of groundwater levels was extended to evaluate a subsurface barrier wall to prevent saltwater penetration in a coastal aquifer. A field feasibility study was conducted at the Komesu Dam, Japan. Groundwater levels were observed at pairs of sites on the seaward and reservoir sides of the wall. Groundwater-level time series data collected from a reservoir-side site near a horizontal hollow pipe penetrating the wall contained a visible sinusoidal tidal component, whereas data from another reservoir-side site did not. Analysis of these observations on the groundwater tidal response derived hydraulic parameters of the barrier wall between the paired observation sites. Although the parameters derived by the used simple formulas seem only approximate or apparent, the difference of the results for the two pairs indicated that the extended tidal response method can be useful for evaluation of the barrier function of the wall.

Influence of low-frequency variability on groundwater level trends

Estimating groundwater level evolution is a major issue in the context of climate change. Groundwater is a key resource and can even account in some countries for more than half of the water supply. Groundwater trend estimates are often used for describing this evolution. However, the estimated trend obviously strongly depends on available time series length, which may be caused by the existence of long-term variability of groundwater resources. In this paper, using a groundwater level database in Metropolitan France as an example, we address this issue by exploring how much trend estimates are sensitive to low-frequency variability of groundwater levels. Database consists of relatively undisturbed groundwater level time series regarding anthropogenic influence (water abstraction by either continuous or periodic pumping). Frequent changes in trend direction and magnitude are detected according to time series length, which can eventually lead to contradictory interpretations of the groundwater resource evolution, as presented in first part of this article. To assess whether low-frequency variability – known to originate from climate variability – can induce such modifications of trends, we explored in a second step the multi-time scale variability of groundwater levels using a methodology based on discrete wavelet transform. Most of the time series displaying changing trends depending on time series length corresponded to aquifers with high-amplitude low-frequency variability of groundwater levels. Two predominant low-frequency components were detected: multi-annual (∼7 years) and decadal (∼17 years). We finally examined how much those two low-frequency components may affect trend estimates on the longer time period available. For this purpose, we individually removed each of both components from the original times series by discrete wavelet filtering and re-estimated trends in the filtered groundwater level time series. The results showed that the groundwater level trends were highly sensitive to the presence of any of these low-frequency components, which may then strongly influence the estimated trends either by exaggerating or mitigating them. These results emphasize that i) attributing the estimated trends only to climate change would be hazardous given the large influence of low-frequency variability on groundwater level trends, ii) estimation of trends in hydrological projections resulting from General Circulation Models (GCM) outputs in which low-frequency variability is not well represented would be subject to strong uncertainty, iii) a potential change in the amplitude of internal climate variability – e.g. increasing or decreasing low-frequency variability – in the next decades may lead to substantial changes in groundwater level trends.

How groundwater time series and aquifer property data explain heterogeneity in the Permo-Triassic sandstone aquifers of the Eden Valley, Cumbria, UK

A novel investigation of the impact of meteorological and geological heterogeneity within the Permo-Triassic Sandstone aquifers of the River Eden catchment, Cumbria (UK), is described. Quantifying the impact of heterogeneity on the water cycle is increasingly important to sustainably manage water resources and minimise flood risk. Traditional investigations on heterogeneity at the catchment scale require a considerable amount of data, and this has led to the analysis of available time series to interpret the impact of heterogeneity. The current research integrated groundwater-level and meteorological time series in conjunction with aquifer property data at 11 borehole locations to quantify the impact of heterogeneity and inform the hydrogeological conceptual understanding. The study visually categorised and used seasonal trend decomposition by LOESS (STL) on 11 groundwater and meteorological time series. Decomposition components of the different time series were compared using variance ratios. Though the Eden catchment exhibits highly heterogeneous rainfall distribution, comparative analysis at borehole locations showed that (1) meteorological drivers at borehole locations are broadly homogeneous and (2) the meteorological drivers are not sufficient to generate the variation observed in the groundwater-level time series. Three distinct hydrogeological regimes were identified and shown to coincide with heterogeneous features in the southern Brockram facies, which is the northern silicified region of the Penrith Sandstone and the St Bees Sandstone. The use of STL analysis in combination with detailed aquifer property data is a low-impact insightful investigative tool that helps guide the development of hydrogeological conceptual models.