Dissolved oxygen (DO) is one of the most important indicators of water quality and an essential measure for the aquatic organisms and the local ecosystem. DO concentrations in karst spring flow ...involves complex hydrological processes because of the heterogeneous nature of the karst system. In this study, an interpretable and explainable model that integrates the shapley additive explanations (SHAP) algorithm with the long short‐term memory network model (LSTM) is proposed to evaluate the contributions of karst spring discharge, precipitation, water temperature, and specific conductance to DO concentrations in karst spring flow. The hybrid model can predict the temporal fluctuations of DO levels and provide a robust characterization of DO behaviours. To demonstrate the applicability of the proposed model, we adopted the hydrological and meteorological data from Barton Springs, situated within a highly karstified segment of the Edwards aquifer characterized by extensive fractures and openings. The optimal prediction performance is achieved with a 14‐day time step, which is considered as the effective response time between various hydrological processes and DO concentrations at Barton Springs. It reveals that the influence of karst spring discharge, precipitation, water temperature, and specific conductance in previous 14 days collectively contribute to the current DO concentration in karst spring flow. The SHAP values of input features provide both local and global explanations, demonstrating the magnitude and the direction of each feature's impact on DO levels in karst spring flow. In descending order, the contributions of various hydrological processes to DO are ranked as follows: precipitation, karst spring discharge, temperature, and specific conductance. Precipitation and discharge exhibit positive SHAP values, indicating that increases in these hydrological processes contribute to higher DO levels in the karst flow. Water temperature and specific conductance have negative SHAP values, suggesting that higher water temperature and specific conductance will lead to decreased DO levels in the karst flow.
The long short‐term memory network model (LSTM) layers are integrated with shapley additive explanations (SHAP) to provide an explainable and interpretable deep learning model for dissolved oxygen (DO) concentrations in karst spring flow. The effective response time is determined when the SHAP‐LSTM hybrid model obtains its best performance with time step of 14 days. By assigning SHAP values to all input features, the hybrid model reveals contributions of precipitation, spring discharge, water temperature and specific conductance to DO in karst spring flow.
Karstic spring waters of the coastal area of Bejaia represent important sources of drinking water. The present study covers the identification of the geochemical processes responsible of the ...mineralization of spring waters and their characterization. Forty-two water samples were collected from six important springs within the study area then analyzed according to major ions. In addition to traditional diagrams, Piper, Gibbs and scatter diagrams, multivariate statistical analyses, namely principal component analysis (PCA) and hierarchical cluster analysis (HCA), were applied. Hydrochemical parameters such as temperature (T), pH, and electrical conductivity (EC) of the samples were measured on site, and major ion analysis (Ca
2+
, Mg
2+
, Na
+
, K
+
, HCO
3
−
, SO
4
2−
, and Cl
−
) was conducted in the laboratory. Results showed three different geochemical groups of samples: (1) The first group, characterized by water flowing through fractures of Jurassic limestone (Toudja, Aiguades and Cascade), was lowly mineralized and had Ca–HCO
3
water type. (2) The second group was highly mineralized (samples from Derguina spring), characterized by Ca–Mg–HCO
3
-SO
4
water type, indicating that waters are flowing through Triassic evaporitic formations. (3) The third group, which represents samples from Source Bleue and Minena springs, was moderately mineralized and had Ca–Mg–HCO
3
water type with important concentrations of SO
4
2−
. The water of this group might be the result of an interaction between both limestone and evaporite within abnormal contacts.
•Contents of hydrochemical constituents in the YD1 is much higher than in the YD2.•Dissolution of albite, halite and ion exchange result in high content of Na in YD1.•UCC-normalized REE patterns of ...the springs are of increasing fluctuation trend.•YD1 is a deep circulation spring, while YD2 is a shallow circulation spring.•The Red River fault zone controls the conduction of water and heat in the YD1.
The Xiaguan Hot Spring and the Butterfly Spring near Dali in Yunnan of China were sampled 5 and 3 times, respectively and their hydrogeological characteristics are discussed in this study. The TDS and hydrochemical type of these two springs are 1360 mg/L and HCO3-SO4-Na type, and 246 mg/L and HCO3-Ca-Mg type, respectively. The Butterfly Spring is enriched in Ca2+, Mg2+ and HCO3− due to dissolution of carbonate rocks. In contrast, the Xiaguan Hot Spring is enriched in Na+, K+, HCO3− and SO42− due to dissolution of albite, halite and ion exchange. The concentrations of the rare earth elements (REEs) in the Xiaguan Hot Spring and the Butterfly Spring range from 0.033 μg/L to 0.071 μg/L and 0.013 μg/L to 0.018 μg/L, respectively. The absolute concentrations of the medium rare earth elements (MREEs) and heavy rare earth elements (HREEs) are relatively higher than those of the light rare earth elements (LREEs) in the Xiaguan Hot Spring. The concentrations of the LREEs in the Butterfly Spring are relatively high. The δ2H and δ18O values of the Xiaguan Hot Spring and the Butterfly Spring vary from −102‰ to −107‰ and −11.6‰ to −13.8‰, respectively, showing negative trends due to altitude effect. The data points of δ2H and δ18O of the two spring samples are close to the Global Meteoric Water Line and the Local Meteoric Water Line with a slight 18O shift, indicating that the springs are meteoric in origin with a slight 18O exchange with the surrounding rocks. The geothermal reservoir temperature of the Xiaguan Hot Spring estimated with SiO2 geothermometers ranges from 107.4 °C to 114.6 °C. The Xiaguan Hot Spring occurs in the metamorphic rocks of the Paleoproterozoic Cangshan Group, and is recharged from infiltration of precipitation and snow-melting water. After undergoing a deep circulation, the groundwater is heated by heat flow and rises along the Tangzipu fault to form a hot spring of deep groundwater circulation type. The Butterfly Spring emerges in the Silurian limestone. Groundwater receives recharge from precipitation and snow-melting water in the western mountainous area, flows eastward along the fissures, and emerges on the lower slopes as a spring of shallow groundwater circulation type with ambient temperature.
Springs are biodiversity hotspots and unique habitats that are threatened, especially by water overdraft. Here we review knowledge on ambient-temperature (non-geothermal) freshwater springs that ...achieve sufficient oversaturation for CaCO3 -by physical CO2 degassing and activity of photoautotrophs- to deposit limestone, locally resulting in scenic carbonate structures: Limestone-Precipitating Springs (LPS). The most characteristic organisms in these springs are those that contribute to carbonate precipitation, e.g.: the mosses Palustriella and Eucladium, the crenophilous desmid Oocardium stratum, and cyanobacteria (e.g., Rivularia). These organisms appear to be sensitive to phosphorus pollution. Invertebrate diversity is modest, and highest in pools with an aquatic-terrestrial interface. Internationally, comprehensive legislation for spring protection is still relatively scarce. Where available, it covers all spring types. The situation in Europe is peculiar: the only widespread spring type included in the EU Habitat Directive is LPS, mainly because of landscape aesthetics. To support LPS inventorying and management to meet conservation-legislation requirements we developed a general conceptual model to predict where LPS are more likely to occur. The model is based on the pre-requisites for LPS: an aquifer lithology that enables build-up of high bicarbonate and Ca2+ to sustain CaCO3 oversaturation after spring emergence, combined with intense groundwater percolation especially along structural discontinuities (e.g., fault zones, joints, schistosity), and a proper hydrogeological structure of the discharging area. We validated this model by means of the LPS information system for the Emilia-Romagna Region (northern Italy). The main threats to LPS are water diversion, nutrient enrichment, and lack of awareness by non-specialized persons and administrators. We discuss an emblematic case study to provide management suggestions. The present review is devoted to LPS but the output of intense ecological research in Central Europe during the past decades has clearly shown that effective conservation legislation should be urgently extended to comprise all types of spring habitats.
Display omitted
•Limestone Precipitating Springs (LPS) are ideal to study biocalcification.•Spring-habitat protection is limited globally; in Europe there is a focus on LPS.•We present a conceptual model to predict LPS occurrence to meet EU directives.•Main impacts on LPS are water overdraft and lacking appreciation of their relevance.•LPS should be a flagship to achieve widespread conservation of springs in general.
In this article, a leaf spring insertion to the stator back iron is proposed to reduce vibration and acoustic noise in a switched reluctance machine (SRM). The back iron of the tooth tip is curved ...out and the material made up of carbon spring steel is placed to dampen the vibration due to radial force. A 6/4 stator/rotor pole SRM is designed and its natural vibration mode frequencies are determined through finite-element analyses. The optimal current waveform is applied to the SRM model using the multidisciplinary coupled simulation method to determine the radial force with the harmonic orders. In the final step, the proposed leaf spring architecture is optimized to achieve the most dampening for the effective radial force harmonic orders. The simulation and experimental results confirmed the effectiveness of the leaf spring insertion in reducing the noise and vibration in SRMs.
The study aim was to assess the intrinsic vulnerability of the middle Valseriana (northern Italy) through a methodology that could fully consider the peculiar characteristics of the Pre-Alps region. ...After applying classical methods for karst environments (COP and EPIK), the integrative COPA+K method was developed for the Nossana (dominant drainage system) and Ponte del Costone (dispersive circulation system) spring catchments. The COPA+K approach exploits the potential of the COP method with the addition of two factors: the influence of the main discontinuities on the water system according to their distance from the spring (A factor), and the development of the karst network (K factor). The COP and COPA+K maps were validated based on the characteristic correlation, for the study area, between the values of δ
18
O and the elevation of recharge areas. Compared to COP, the COPA+K results appeared more stringent in the identification of areas characterized by high vulnerability, which passed from 35.6% to 23.6% of the entire study area. COPA+K also made possible to better differentiate the behavior of the two catchments. In terms of high susceptible areas, their percentage difference increased by 12.3%, correctly emphasizing the greater susceptibility of the Nossana drainage system in comparison with the Ponte del Costone dispersive one. For the COPA+K map, in the Nossana catchment, the δ
18
O values indicated a good agreement between the mean elevation of high vulnerable and water recharge areas (± 106 m). For Ponte del Costone, the isotopic data indicated a possible mixing with the waters of the Serio River.
Numerical modeling is used to understand the regional scale flow dynamics of the fault‐hosted orogenic geothermal system at the Grimsel Mountain Pass in the Swiss Alps. The model is calibrated ...against observations from thermal springs discharging in a tunnel some 250 m underneath Grimsel Pass to derive estimates for the bulk permeability of the fault. Simulations confirm that without the fault as a hydraulic conductor the thermal springs would not exist. Regional topography alone drives meteoric water in a single pass through the fault plane where it penetrates to depths exceeding 10 km and acquires temperatures in excess of 250°C. Thermal constraints from the thermal springs at Grimsel Pass suggest bulk fault permeabilities in the range of 2e−15 m2–4.8e−15 m2. Reported residence times of >30,000 and 7 years for the deep geothermal and shallow groundwater components in the thermal spring water, respectively, suggest fault permeabilities of around 2.5e−15 m2. We show that the long residence time of the deep geothermal water is likely a consequence of low recharge rates during the last glaciation event in the Swiss Alps, which started some 30,000 years ago. Deep groundwater discharging at Grimsel Pass today thus infiltrated the Grimsel fault prior to the last glaciation event. The range of permeabilities estimated from observational constraints is fully consistent with a subcritical single‐pass flow system in the fault plane.
Plain Language Summary
Observations from warm springs discharging through a fault at Grimsel Pass (2,164 m.a.s.l.) in the Swiss Alps are used to constrain a numerical model of the deep water circulation feeding the springs. The springs are known to have been active for at least 3.3 million years and to be due to ascent of meteoric water that penetrated to depths exceeding 10 km, where it acquired temperatures above 250°C. Simulations show that the circulation along the fault connects a meteoric recharge zone at high altitude to the west with a sub‐vertical permeable discharge zone at Grimsel Pass. A key unknown is the permeability of the fault. Temperature, discharge rate and chemical composition of the spring water depend on flow conditions at depth and can be used to estimate the fault's bulk permeability. Our study shows that the range of fault permeabilities can be narrowed down to roughly half an order of magnitude: 1e−15 m2–5e−15 m2. This permeability range is consistent with a currently stable, single‐pass flow pattern. Long water‐residence times inferred from the isotopic composition of the spring water suggest low recharge rates during the last glaciation and the dominance of a multi‐pass flow pattern during that time.
Key Points
The topography in the Grimsel region drives meteoric water to depths exceeding 10 km, causing discharge of thermal water at Grimsel Pass
Based on thermal and chemical constraints from the spring water, the bulk fault permeability is in the range of 2e−15 m2–4.8e−15 m2
Recent periods of glaciation changed the pattern of flow in the fault plane and are likely the reason for fluid residence times >30,000 years
•Geochemical and isotopic composition of two low temperature thermal springs in south Konkan region, South India were studied.•These thermal waters are of Na-HCO3 type peripheral waters.•Geochemical ...classification indicate dissolution of silicate minerals controls chemical composition of this thermal water system.•Stable isotope composition indicate meteoric origin of thermal waters without isotope exchange by interaction with host rocks.•Results suggest that these thermal springs are most likely part of fault-controlled shallow hydrothermal system.
The west coast of India is one of the important geothermal provinces that hosts many thermal springs of varying discharge characteristics. A total of twenty thermal springs have been reported in this region falling into two different suites of geological formations. Two low temperature geothermal springs located in the south Konkan region were studied with the objectives (1) to understand the geochemical evolution, (2) to identify the recharge source and (3) to propose a reservoir temperature. The hydrogeological and geochemical investigations were carried out on two geothermal springs, viz. Bandaru and Irde, both located in the Dakshina Kannada district of the Karnataka state, India. In this region, low-temperature geothermal activity has been reported in the form of the evolution of warm water from thermal spring systems with a notable presence of faults/lineaments in the geological formation. These faults enable the upward flow of geothermally heated water. The hydrochemistry of water samples was analyzed to determine the dominant hydrogeochemical processes in the region. These thermal springs are categorised as Na-HCO3 type facies. The thermal water is of a moderately alkaline nature like other geothermal springs of the west coast of India. Based on the results for major and trace element concentrations it was found that the hydrochemical process was mainly controlled by the dissolution of silicate minerals with a marginal influence from anthropogenic activity. Evaluation of δD and δ18O stable isotope data suggested that the thermal springs and adjacent surface and groundwater were of meteoric origin. Reservoir temperature was estimated with Na-K-Ca, K2/Mg, quartz and chalcedony geothermometers. The results suggest that the reservoir under consideration is a low enthalpy reservoir with temperatures ranging between 55 and 86 °C.
PDF
