Declining Groundwater Levels Research Articles

Rainfall Variability and Rice Production: An Assessment Based on the Variability in Indian Monsoon and Ground Water Level of Punjab, India

In Punjab, 99 per cent of total cultivated area is under irrigation but it is at the cost of declining ground water table. Due to rainfall variability and very high requirement of water in rice crop, which occupy large area in the state during kharif season, has further deteriorated the situation of water resources. Therefore, this study was conducted to identify the relationship between declining groundwater level, variable monsoonal rainfall and rice yield of different agroclimatic zones of Punjab. The historical data of rainfall for different locations of Punjab viz. Amritsar (1998-2017), Ballowal Saunkhri (1998-2017), Bathinda (1998-2017), Ludhiana (1998-2017) and Patiala (1998-2017) was collected from the Department of Climate Change and Agricultural Meteorology, PAU, Ludhiana and Met Centre, IMD Chandigarh. The productivity data of rice was collected from indiastat.com and Statistical Abstract of Punjab whereas the data related to groundwater table (1998-2017) were collected from the Central Ground Water Board, Punjab. It was observed that the ground water table level showed significant decrease after 2005 at all locations and there was a positive correlation between rice yield and ground water table at all the locations except at Amritsar where it was negatively correlated. The deficit rainfall years showed very less impact on rice yield which has increased continuously. However, the relationship between groundwater table and rainfall indicated that deficit monsoonal rainfall led to further decline in groundwater table as it was not recharged.

Modeling the influence of climate on groundwater flow and heat regime in Brandenburg (Germany)

This study investigates the decades-long evolution of groundwater dynamics and thermal field in the North German Basin beneath Brandenburg (NE Germany) by coupling a distributed hydrologic model with a 3D groundwater model. We found that hydraulic gradients, acting as the main driver of the groundwater flow in the studied basin, are not exclusively influenced by present-day topographic gradients. Instead, structural dip and stratification of rock units and the presence of permeability contrasts and anisotropy are important co-players affecting the flow in deep seated saline aquifers at depths >500 m. In contrast, recharge variability and anthropogenic activities contribute to groundwater dynamics in the shallow (<500 m) freshwater Quaternary aquifers. Recharge fluxes, as derived from the hydrologic model and assigned to the parametrized regional groundwater model, reproduce magnitudes of recorded seasonal groundwater level changes. Nonetheless, observed instances of inter-annual fluctuations and a gradual decline of groundwater levels highlight the need to consider damping of the recharge signal and additional sinks, like pumping, in the model, in order to reconcile long-term groundwater level trends. Seasonal changes in near-surface groundwater temperature and the continuous warming due to conductive heat exchange with the atmosphere are locally enhanced by forced advection, especially in areas of high hydraulic gradients. The main factors controlling the depth of temperature disturbance include the magnitude of surface temperature variations, the subsurface permeability field, and the rate of recharge. Our results demonstrate the maximum depth extent and the response times of the groundwater system subjected to non-linear interactions between local geological variability and climate conditions.

Assessing the impact of climate change on a coastal karst aquifer in a semi-arid area

This study investigates the potential future state of a coastal karst aquifer located in the Salento peninsula of southern Italy, examining the impact of climate change. After a historical and future overview of the climate in the study area, an analysis of meteorological indices (SPIs and SPEIs) and groundwater levels (GWLs) during the historical period revealed a strong correlation between them. Then, straightforward linear regression models were applied to establish relationships between SPIs/SPEIs and GWLs for the analysed wells. To estimate future SPIs and SPEIs, an ensemble of regional climate models was used, and estimates of GWLs until the end of the century were obtained by incorporating these projections into the regression models. Two emission pathways, the RCP4.5 and the RCP8.5, were considered in the analysis. The results based on SPIs suggest no significant changes in future GWLs across all wells. However, when considering also temperature through the use of SPEIs, a general decline in groundwater levels is expected, with the decrease being more pronounced under the RCP8.5.

Framework for comprehensive assessment of ecological water conveyance based on long-term evolution forecast of riparian forests in arid floodplains: synergistic consideration of flood inundation and groundwater

Ecological water conveyance (EWC) is a crucial measure to address ecological degradation in arid regions. To scientifically formulate ecological water conveyance pattern (WCP), and provide a basis for ecological flow management of reservoirs, this study proposes a framework to assess the effect of EWC, and applied it to the Irtysh River basin as a case study. Following the framework, we first developed a calculation method to assess the evolution of the reproduction and growth of riparian forests caused by hydrological variations. Additionally, simulations of surface water and groundwater conditions under different scenarios of water resources development were conducted based on hydrodynamic and groundwater models. The study further predicted the long-term evolution of riparian forest reproduction and growth based on the simulation results. The results indicated that when EWC was not carried out (reservoirs operating according to the designed rules), after 30 years, floodplain flood (FF) inundation area decreased by 47%, duration shortened by 73%, and groundwater levels in 80% of the region declined by 1.5–2 m, causing severe and moderate impacts on the reproduction and growth of riparian forests, respectively. Alternatively, adopting EWC (reservoirs operating according to ecological rules) led to a 24% decrease in FF inundation area, a 63% reduction in its duration, and a groundwater level decline of 0.5–1 m in 87% of the region, and these changes caused severe impacts on riparian forest reproduction, while growth remained unaffected. These findings indicate that the current EWC effectively safeguards the growth of riparian forests but fall short of mitigating the impacts of water conservancy project construction and operation on the reproduction of riparian forests. Given the current trajectory, riparian forests are anticipated to undergo a gradual decline. Consequently, a sustainable EWC should ensure that the water volume discharged into the region exceeds 53% of the natural inflow and urgently requires measures to extend the duration of FF inundation. This study can provide new insights for the early evaluation, comparison and optimization of various WCPs before their implementation.

A numerical modelling approach to investigate the fate of brine reject of farm scale desalination plants on groundwater aquifers in arid environments

Farm-scale desalination units are gaining popularity for agricultural irrigation in arid countries, such as Kuwait to meet freshwater demands. However, less attention has been given to the management of environmentally hazardous brine reject water they produce. In this study we investigated the fate of brine water produced by the inland desalination units on the underlying aquifers using numerical modelling and field investigations. The methodology involved developing groundwater flow and solute transport models using Flex VMF-SEAWAT to simulate the movement of reject brine. The field investigations included collecting 150 water samples and conducting pumping tests on newly drilled wells. This numerical simulation considered advection, dispersion, and adsorption processes with variable groundwater density following rigorous validation and calibration of the developed numerical models. The results show that the RO reject brine will significantly increase groundwater salinity, exceeding 10,000 mg/L when accounting for advection, dispersion, and adsorption processes. The sustainable yield of the aquifer, with a salinity of <10,000 mg/L, averages 500 Mm3 but is expected to be depleted within 16 years with the current extraction rate. The resulting hydraulic properties are favourable with K about 100 m/d, T > 1000 m2/day, and Sy just >0.1. The adopted values for dispersivity and adsorption coefficients for chloride and sulphate salts in the aquifer were 10 m and 1 × 10−7 [mg/L]−1 respectively. Chemical and numerical analyses indicate a mixing ratio between the reject brine and groundwater in the study area of approximately 10 %. Uncontrolled groundwater extraction, combined with the surface disposal of RO reject brine, has led to a significant decline in groundwater levels and an increase in the salinity. The adsorption ratio of simulated brine plume was 13 %. The authors recommend to dispose the RO reject water in a safe location or transfer it to the nearest wastewater treatment plant for proper treatment and reuse.

Spatiotemporal variation of water balance components in Mashhad catchment, Iran: Investigating the impact of changes in climatic data and land use

Abstract The research aims to investigate the spatiotemporal changes of water balance components and distinguish the relative impacts of climatic data and land-use on groundwater level in northeastern Iran. This investigation employs the WetSpass-M model to estimate water balance and the Mann-Kendall test alongside Sen's slope estimator to evaluate trend. The study also assesses mean annual water balance components, considering diverse combinations of land-use and soil. The findings offer a hydrological insight, revealing that 14% of precipitation results in runoff, 29% of that recharging the aquifer; the remaining portion is lost through evapotranspiration. The trends in precipitation and simulated water components are not significant but a significant downward trend in groundwater is observed beyond a specific point in time. Based on this outcome, as well as the analysis of land-use changes, it was speculated that human activities in this fast-developing region might be implicated in the decline in groundwater levels. Analysis of water balance components in various soil and land-use combinations indicates that evapotranspiration exhibits greater variability within the land-cover class, while recharge is more influenced by soil texture. These findings enhance our understanding for identifying potential sites for artificial recharge and determining sustainable groundwater withdrawals based on spatiotemporal recharge patterns.

Monitoring the Impacts of Artificial Recharge Structures on Water Table at Ambedkar Nagar, Uttar Pradesh, India

To study the groundwater recharge through rainfall and artificial recharge structures in selected dry well in different locations of UP was experimented. The data was recorded (1998-2017) to execute the artificial recharge structure at the appropriate locations with best geological condition to enhance the recharge rate at least cost for control of declining groundwater level. Study reveals that the stochastic auto regressive lime series model is an effective tool for management of ground water resource at pre and post monsoon. The variation of pre and post monsoon ground water level is maximum as the physical soil characteristics including hydraulic conductivity may enhance the recharge rate at least cost.

Changes in water surface area of the Middle Atlas-Morocco lakes: A response to climate and human effects

The lakes represent crucial surface water resources and an integral part of wetlands. The most concerning aspect of the degradation of these areas is the complete drying up of the lakes. In the Mediterranean region, successive changes in land use practices in the context of climate change have strongly influenced wetland areas. In this study, we used Landsat TM, OLI, and OLI-2 satellite images to monitor the water surface area in two representative lakes (Aoua and Ifrah) of the Tabular Middle Atlas and to map land use across the entire study area. To extract information related to lakes and land use, we employed the Support Vector Machine machine learning algorithm, widely used in remote sensing studies. However, we identified drought periods from precipitation data using the Standardized Precipitation Index (SPI) recommended by the World Meteorological Organization (WMO). The results obtained from the processing of Landsat satellite images indicate a significant reduction in the surface area of the lakes, with periods of drying for Aoua lake, endangering their fragile ecosystems and biodiversity. The critical situation of the two lakes is attributed to a combination of natural and anthropogenic factors. The analysis of climatic data shows a significant climate change from the 1980s, with long periods of drought. In parallel, the study area has undergone remarkable modifications in land use patterns, mainly characterized by a significant extension of irrigated agricultural surfaces to the detriment of grazing and rainfed lands. In three decades, the area of irrigated crops has increased from approximately 1300 hectares in 1985 to 7070 hectares in 2022, representing an increase of 542%. The findings presented in this study reveal the extent of lake degradation in the TMA and reflect the alarming decline in groundwater levels. This situation indicates the necessity of formulating a strategy to protect water resources and wetlands in the Middle Atlas.

Observations of Decadal-Scale Brine Geochemical Change at the Bonneville Salt Flats

Over the past century, the Bonneville Salt Flats, which lies on the western edge of the Great Salt Lake watershed, has experienced changing environmental conditions and a unique history of land use, including resource extraction and recreation. The perennial halite salt crust has decreased in thickness since at least 1960. An experimental restoration project to return mined solutes began in 1997, but it has not resulted in anticipated salt crust growth. Here, primary observations of the Bonneville Salt Flats surface and subsurface brine chemistry and water levels collected from 2013 to 2023 are reported. Spatial and temporal patterns in chemistry, focused on density and water stable isotopes, are evaluated and compared with observations across seven periods of research spanning from 1925 to 2023. Declining salinity in the areas to the east of extraction ditches and south of Interstate 80 were observed. Brine extracted for potash production decreased in salinity as extraction rates increased. Between the years 1964 and 1997, the salinity of the shallow aquifer brine located beneath and to the east of the crust experienced a decrease. However, following this period, the salinity stabilized and subsequently increased. Salinity recovery was concurrent with declines in brine extraction and the salt restoration project, with the largest decrease in brine extraction being concurrent with the largest recovery in salinity. The specific impact of the restoration project on the brine salinity increase remains unclear. To the west, the shallow aquifer in the area between the Silver Island Mountains and the salt crust has increased in salinity. This increase is accompanied by a decline in groundwater levels, which enables the underground movement of solutes from east to west, following a salinity gradient away from the saline pan. Over the past 25 years, the alluvial-fan aquifer along the Silver Island Mountains has markedly declined, leading to increasingly more saline and isotopically heavier basinal waters to be extracted for industrial use. This change is concurrent with the onset of the salt restoration project, which relies on alluvial-fan aquifer waters. This compilation of changes in groundwater chemistry provides an important resource for stakeholders working to understand and manage this dynamic and ephemeral evaporite system. It also offers an example of decadal-scale change in a highly managed Great Salt Lake watershed saline system.

Assessing Farmers’ Awareness of Climate Change Impact: A Case of the Bundelkhand Region, India

The study was conducted in 2023-24 in the Bundelkhand region, India and examined the awareness of climate variability and its impacts on farmers. Analytical and descriptive research design was used in the present study. The present study was carried out in two districts namely, Jalaun and Datia of Bundelkhand. Four blocks were selected randomly from these districts, and a total of 320 respondents were selected randomly. The results revealed that most of the respondents were well informed about changes in crop cycles (81.87%), new pests and diseases (63.12%) and natural hazards like earthquakes and cyclones (69.06%). There are still considerable gaps in their knowledge of topics like biodiversity loss and groundwater depletion. Factor analysis revealed four principal components influencing awareness; natural hazards (.717), soil productivity (.681), pest and disease emergence (.775) excess water in surface (.818), and decline in groundwater level (.738). Correlation and regression analysis indicate that education, landholding, socio economic status, information-seeking behaviour, economic motivation and scientific orientations are significantly and positively associated with climate change awareness. The findings suggest that focused awareness efforts need to be made that utilize media and extension advisory services to improve farmers’ awareness and ability to adapt.

Understanding the vulnerability of surface–groundwater interactions to climate change: insights from a Bavarian Forest headwater catchment

Headwaters play a crucial role in maintaining forest biodiversity by providing unique habitats and are important for the regulation of water temperature and oxygen levels for downstream river networks. Approximately 90% of the total length of streams globally originate from headwaters and these systems are discussed to be especially vulnerable to impacts of climate change. This study uses an integrated hydrological model (HydroGeoSphere) in combination with 23 downscaled ensemble members from representative concentration pathways (RCPs) 2.6, 4.5 and 8.5 to examine how climate change affects water availability in a headwater catchment under baseflow conditions. The simulations consistently predict increasing water deficits in summer and autumn for both the near (2021–2050) and far future (2071–2099). Annual mean water deficits were estimated to be 4 to 7 times higher than historical levels. This is mainly due to a projected reduction in precipitation inputs of up to – 22%, while AET rates remain similar to those observed during the historical reference period (1992–2018). The declining groundwater storage reserves within the catchment are expected to result in a significant decline in surface water availability during summer and autumn, with a reduction in mean annual stream discharge by up to 34% compared to the reference period. Due to declining groundwater levels, upstream reaches are predicted to become intermittent in summer leading to a reduction of the total stream flow length by up to 200 m. Findings from this study will enhance our understanding of future water availability in headwater systems and may aid in the development of effective management strategies for mitigating local impacts of climate change and preserving these vulnerable ecosystems.

GROUNDWATER LEVEL PREDICTION USING DEEP RECURRENT NEURAL NETWORKS AND UNCERTAINTY ASSESSMENT

Abstract. Groundwater is one of the most important sources of regional water supply for humans. In recent years, several factors have contributed to a significant decline in groundwater levels (GWL) in certain regions. As a result of climate change, such as temperature increase, rainfall decrease, and changes in relative humidity, it is necessary to investigate and model the effects of these factors on GWL. Although a number of researches have been conducted on GWL modeling with machine learning (ML) and deep learning (DL) algorithms, only a limited number of studies have reported model uncertainty. In this paper, GWL modeling of some piezometric wells has been conducted by considering the effects of the meteorological parameters with Long-Short Term Memory (LSTM) and Gated Recurrent Unit (GRU) algorithms. The models were trained on one piezometric well data and predictions were executed on six other wells. To perform an uncertainty assessment, the models were run 10 times and their means were calculated. Subsequently, their standard deviations were considered to evaluate the outcomes. In addition, the prediction power of the models was validated using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Normalized Root Mean Square Error (NRMSE), and R-Squared (R2). Finally, for all the six wells that did not participate in the training phase, the prediction functions of the trained models were run 10 times and their accuracy was assessed. The results indicate that LSTM (R2=95.6895, RMSE=0.4744 m, NRMSE=0.0558, MAE=0.3383 m) had a better performance compared to that of GRU (R2=95.2433, RMSE=0.4984 m, NRMSE=0.0586, MAE=0.3658 m) on the GWL modeling.

Arizona Groundwater Explorer: interactive maps for evaluating the historical and current groundwater conditions in wells in Arizona, USA

Groundwater is an important water source in Arizona, accounting for about 41% of water use in this mostly arid-to-semiarid state in the southwestern United States, and the availability of groundwater resources in the state is a concern. To provide accessible information from depth-to-groundwater data, a series of web-based interactive maps were developed, called the Arizona Groundwater Explorer (AGEx). Scripts were written to harmonize and synthesize groundwater datasets from the two largest publicly available sources, subset these data to address different groundwater availability questions, and display the results in online, interactive maps. The combined dataset contained 1,820,122 depth-to-groundwater measurements from 1891 through 2022 from 41,918 wells in Arizona. Data views are provided for 20 topics, including recent (2020 or later) depth to groundwater (4,569 wells), historical (pre-1950) depth to groundwater (4,287 wells), wells with long-term (≥50 years) records (1,183 wells), wells with recent groundwater level decline (277 wells), wells with recent groundwater level rise (120 wells), and linear trends in groundwater levels over ten 10-year periods (number of wells ranging from 341 in 1978–1987 to 1,208 in 2003–2012), among others. With ongoing drought in the region resulting in declining surface-water supplies in Arizona, groundwater may play an even larger role in satisfying water needs in the state. The AGEx series of maps provides a nonspecialist audience with an improved understanding of historical, current, and changes in groundwater levels in Arizona.

Kuwait household water demand in 2050: Spatial microsimulation and impact appraisal

AbstractHousehold water demand has increased dramatically in Kuwait over the last few decades, due to rapid population growth and changing lifestyles. Avoiding a water deficit through a supply‐side approach has been the default strategy in Kuwait, yet this approach is unsustainable, associated with declining groundwater levels, and reliance on desalination that results in major carbon emission and environmental impact and that takes a large and growing share of oil revenues. In this study, we forecast household water demand in Kuwait to 2050 under a Business‐As‐Usual (BAU) scenario and evaluate the economic and environmental impacts. A spatial microsimulation, constrained by the national population projection of the Kuwait Institute of Scientific Research (KISR), was developed to overcome data limitations in forecasting household demand. Results show a 45% increase in water demand by 2050, to 664.1 million cubic metres (MCM), relative to the 2019 base year. Annual production costs increase from 1.39 billion USD in 2019 to 1.99 billion USD by 2050, whilst carbon emissions increase from 10.85 to 15.54 million tonnes/year. These results should alert policymakers to the potential impacts of the growing water demand and provide further support for water conservation action to reduce demand.

Reconnaissance to characterisation of land subsidence due to groundwater overdraft and oil extraction in and around Mehsana City, Gujarat, India by long-term hybrid differential interferometric SAR technique

We present a framework for studying reconnaissance to characterisation of land subsidence in the slowly subsiding city of Mehsana, Gujarat, India, due to groundwater overdraft and oil extraction. Considering interferometric synthetic aperture radar (InSAR) data availability constraints, we first conducted a long-term conventional Differential InSAR (DInSAR) investigation followed by time-series advanced DInSAR analyses by Persistent Scatterer InSAR (PSInSAR). We identified land subsidence in different parts of the study area, with subsidence rates ranging from ≤ -1 cm/yr to ≥ -3 cm/yr. During the entire observation period (2004–2019), the core urban area has been consistently subsiding with differential subsidence rates of −1.7 to −2.1 cm/yr during 2004–2006, −1.6 to −1.7 cm/yr during 2007–2011 and −2.6 to −2.9 cm/yr during 2016–2018 in conventional DInSAR and −2.5 to −3.2 cm/yr in advanced DInSAR. The subsidence measurements by collateral DInSAR observations in multifrequency and/or multi-imaging modes during 2016–2017 agree well in terms of the spatial extent of subsiding areas and the rates of subsidence. We characterised land subsidence in four type areas due to the diverse causes based on the temporal behaviour of land subsidence: (a) densely-populated city centre, (b) dairy plants and other industries in urban/peri-urban areas, (c) emerging urban clusters in peri-urban/rural areas, and (d) oilfields with agri-horticulture farms in the rural areas. We found a consistently increasing land subsidence trend in ‘a’, reducing land subsidence in ‘b’, newly emerging and rapidly growing land subsidence in ‘c’, and spatially-extensive subtle land subsidence in ‘d’. From the temporal nature of groundwater level (GWL) decline and land subsidence, we inferred that the land subsidence has been occurring primarily due to groundwater overdraft associated with urban growth, industry and agri-horticultural farms and secondarily due to oil extraction in the southeastern part of the study area. A comparison of annual rainfall difference from the long-term average vs GWL change vs land subsidence shows that the rainfall deficit during 2–3 consecutive years and groundwater overdraft are primarily related to land subsidence in the study area.

Monitoring land subsidence induced by tectonic activity and groundwater extraction in the eastern Gediz River Basin (Türkiye) using Sentinel-1 observations

Alaşehir-Sarıgöl sub-basin (ASSB) in Türkiye faces severe water stress mainly due to intensive agricultural irrigation. This has led to declining groundwater levels, aquifer compaction, and subsequent land subsidence. This study aims to evaluate the roles of tectonic activity and groundwater withdrawal in land-subsidence and investigate additional factors like faults and soft soil thickness. The P-SBAS algorithm was applied, using 98 and 123 Sentinel-1 SAR images in ascending and descending orbits, respectively, from 2016 to 2020. Independent Component Analysis separated long-term displacements from seasonal variations in the InSAR time series data. InSAR analysis showed displacement rates up to −6.40 cm/year. Results reveal a direct correlation between InSAR displacement and soft soil thickness, highlighting aquitard layer compaction due to groundwater withdrawal and piezometric head depletion as the primary causes of land subsidence. It was not possible to disentangle the multi-year displacements with different rates, i.e., to isolate the displacements caused by tectonic activity from displacements due to the soft soil compaction induced by groundwater withdrawal. Nonetheless, the analysis identified two spatiotemporal displacement trends: one linked to long-term, linear plastic compaction of the aquitard due to groundwater withdrawal and the other involving long-term displacements with seasonal rebounds caused by yearly water level fluctuations.

“Ground Water Development Techniques under Water Cup” Bidal Village Dist. Satara

Abstract: Paani Foundation is a non-profit, non-governmental organization actively engaged in drought prevention and watershed management in the state of Maharashtra, India. Founded by Indian actor Amir Khan and his wife Kiran Rao, the foundation is led by CEO Satyajit Bhatkal. The state of Maharashtra has a history of recurring droughts and severe water scarcity over the past few years, including the significant drought in 2013. Approximately one-third of the state falls within the semi-arid climatic zone. Due to the impact of climate change, rainfall patterns have become increasingly erratic. The state's irrigation coverage is only 16% of the total cultivated area, significantly lower than the national average of 42%. As a result, there has been excessive use of bore wells, leading to a rapid decline in groundwater levels. In 2015, the state government officially declared that 60% of its villages were facing a "drought-like condition," meaning their crop yields were less than 50% of the normal yield in the state. This condition currently affects 23,811 out of the state's 39,453 villages, resulting in a significant drop in agricultural output for the year, as stated by officials.

An overview of groundwater response to a changing climate in the Murray-Darling Basin, Australia: potential implications for the basin system and opportunities for management

The Murray-Darling Basin (MDB) is a highly allocated and regulated, mostly semiarid basin in south-eastern Australia, where groundwater is a significant water resource. Future climate predictions for the MDB include an expansion of arid and semiarid climate zones to replace temperate areas. The impacts of climate change are already evident in declining groundwater levels and changes in the connection status between rivers and groundwater, and modelling has predicted a further reduction in future groundwater recharge and ongoing declines in groundwater levels. This is predicted to further reduce river baseflow and negatively impact groundwater-dependent ecosystems (GDEs), and these system responses to a changing climate and extreme events are complex and not always well understood. This report provides an overview of the current state of knowledge of groundwater response to a changing climate for the MDB, and outlines challenges and opportunities for future groundwater research and management. Opportunities for the region include improving data systems and acquisition through automation and novel data sources, and growing capability in integrated, risk-based modelling. Quantification of the groundwater/surface-water connection response to declining groundwater levels, and assessing GDE water requirements and thresholds, would enable identification of vulnerable systems and inform the development of metrics for adaptive management, improving the ability to respond to climate extremes. There is potential to adapt policy to support active management of groundwater where required, including conjunctive use and water banking. Improving knowledge sharing and water literacy, including understanding community values of groundwater and GDEs, would support future decision-making.

Artificial intelligence models to evaluate the impact of climate change on groundwater resources

This study develops three different artificial intelligence (AI) models in order to investigate the effects of climate change on groundwater resources using historical records of precipitation, temperature and groundwater levels together with regional climate projections. In particular, the Non-linear Autoregressive Neural Network (NARX), the Long-Short Term Memory Neural Network (LSTM) and the Convolutional Neural Network (CNN) were compared. Considering an aquifer located in northern Italy as a case study, the neural networks were trained to replicate observed groundwater levels by taking as input precipitation and temperature records, and in the case of the NARX also antecedent groundwater levels, on a monthly scale. The trained networks were used to infer groundwater levels until the end of the century based on precipitation and temperature projections provided by an ensemble of 13 Regional Climate Models (RCMs) from the EURO-CORDEX initiative. Two emission pathways were considered: the RCP4.5 and RCP8.5. All the AI models show good performance metrics during the training phase, but NARXs perform poorly compared to the other models during validation and testing. For the future, the NARX and LSTM models predict a decline in groundwater levels, especially for the RCP8.5 scenario, while slight changes are expected using the CNN. As NARXs are not deep learning techniques and CNNs may not be able to extrapolate values outside the training range, LSTMs appear to be better suited for climate change impact evaluations.

Determination and analysis of groundwater ecological level in Changling County

In order to determine and analyze the groundwater ecological level in Changling County, a total of Landsat 8 series remote sensing images from 2017 to 2021, buried depth data of 51 groundwater observation Wells and 47 phreatic salinity detection data in 2021 were collected. Based on the field investigation and laboratory experiment, the spatial and temporal changes of groundwater level and vegetation coverage in the study area from 2013 to 2017 were determined by using remote sensing technology (the Environment for Visualizing Images) and geographic information system. Through the analysis of the vegetation coverage ratio and the relationship between the groundwater depth and diving salinity, we determine the appropriate growth of vegetation in the study area of diving level and diving salinity water. The results showed that from 2017 to 2021 in the average groundwater level and vegetation cover degree decline year by year, the study area northwest shallow depth of groundwater, vegetation cover degree better, however, in the central and eastern regions, the groundwater depth is larger and the vegetation coverage degree is poor. When the depth of local groundwater is 3.3–4.5 m and the phreatic salinity is less than 0.85 g/L, the ecological water level and phreatic salinity are suitable for vegetation growth in the study area. When the depth of local water is less than 3 m, soil salinization occurs in the surface soil near the lake marsh. When the depth of local water was greater than 8.4 m, the phenomenon of soil desertification appeared in some areas of central China.