Modification of the DRASTIC Framework for Mapping Groundwater Vulnerability Zones

The DRASTIC technique is commonly used to assess groundwater vulnerability. The main disadvantage of the DRASTIC method is the difficulty associated with identifying appropriate ratings and weight assignments for each parameter. To mitigate this issue, ratings and weights can be approximated using different methods appropriate to the conditions of the study area. In this study, different linear (i.e., Wilcoxon test and statistical approaches) and nonlinear (Genetic algorithm [GA]) modifications for calibration of the DRASTIC framework using nitrate (NO₃) concentrations were compared through the preparation of groundwater vulnerability maps of the Meshqin-Shahr plain, Iran. Twenty-two groundwater samples were collected from wells in the study area, and their respective NO₃ concentrations were used to modify the ratings and weights of the DRASTIC parameters. The areas found to have the highest vulnerability were in the eastern, central, and western regions of the plain. Results showed that the modified DRASTIC frameworks performed well, compared to the unmodified DRASTIC. When measured NO₃ concentrations were correlated with the vulnerability indices produced by each method, the unmodified DRASTIC method performed most poorly, and the Wilcoxon–GA–DRASTIC method proved optimal. Compared to the unmodified DRASTIC method with an R² of 0.22, the Wilcoxon–GA–DRASTIC obtained a maximum R² value of 0.78. Modification of DRASTIC parameter ratings was found to be more efficient than the modification of the weights in establishing an accurately calibrated DRASTIC framework. However, modification of parameter ratings and weights together increased the R² value to the highest degree.     

Comparison of machine learning models for predicting fluoride contamination in groundwater

Groundwater is an especially important freshwater source for water supplies in the Maku area of northwest Iran. The groundwater of the area contains high concentrations of fluoride and is, therefore, important in predicting the fluoride contamination of the groundwater for the purpose of planning and management. The present study aims to evaluate the ability of the extreme learning machine (ELM) model to predict the level of fluoride contamination in the groundwater in comparison to multilayer perceptron (MLP) and support vector machine (SVM) models. For this purpose, 143 water samples were collected in a five-year period, 2004–2008. The samples were measured and analyzed for electrical conductivity, pH, major chemical ions and fluoride. To develop the models, the data set—including Na⁺, K⁺, Ca²⁺ and HCO₃ ^? concentrations as the inputs and fluoride concentration as the output—was divided into two subsets; training/validation (80% of data) and testing (20% of data), based on a cross-validation technique. The radial basis-based ELM model resulted in an R ² of 0.921, an NSC of 0.9071, an RMSE of 0.5638 (mg/L) and an MABE of 0.4635 (mg/L) for the testing data. The results showed that the ELM models performed better than MLP and SVM models for prediction of fluoride contamination. It was observed that ELM models learned faster than the other models during model development trials and the SVM models had the highest computation time.     

An optimization on machine learning algorithms for mapping snow avalanche susceptibility

Natural Hazards - Mapping avalanche-prone areas to mitigate damages is important and vital for safety and development planning. New hybrid models are introduced for snow avalanche susceptibility...     

Flux of nutrients and heavy metals from the Melai River sub-catchment into Lake Chini, Pekan, Pahang, Malaysia

This study was carried out to determine the flux of nutrients and heavy metals from the Melai sub-catchment into Lake Chini through the process of erosion. Melai River is one of the seven feeder rivers that contributed to the present water level of Lake Chini. Three properties of soils, such as particle size, organic matter content, and soil hydraulic conductivity and three chemical soil properties, such as available nutrients, dissolved nutrients, and heavy metals, were analyzed and interpreted. Potential soil loss was estimated using the revised universal soil loss equation model. The results show that the soil textures in the study area consist of clay, silty clay, clay loam, and sandy silt loam. The organic matter content ranges from 3.40 to 9.92 %, while the hydraulic conductivity ranges from 5.2 to 25.3 cm/h. Mean values of available P, K, and Mg amount was 8.5 ± 3.7 μg/g, 24.5 ± 3.4 μg/g, and 20.7 ± 18.6 μg/g, respectively. The highest concentration of soluble nutrients was SO ₄ ^?2 (815.8 ± 624.1 μg/g), followed by NO₃ ^?-N (295.5 ± 372.7 μg/g), NH₄ ⁺-N (24.5 ± 22.1 μg/g) and PO₄ ^3? (2.0 ± 0.8 μg/g). The rainfall erosivity value was 1658.7 MJ mm/ha/h/year. The soil erodibility and slope factor ranges from 0.06 to 0.26 ton h/MJ/mm and 7.63 to 18.33, respectively. The rate of soil loss from the Melai sub-catchment in the present condition is very low (0.0028 ton/ha/year) to low (18.93 ton/ha/year), and low level flow of nutrients and heavy metals, indicating that the Melai River was not the contaminant source of sediments, nutrients, and heavy metals to the lake.     

Determining the groundwater potential recharge zone and karst springs catchment area: Saldoran region,western Iran

Assessing the groundwater recharge potential zone and differentiation of the spring catchment area are extremely important to effective management of groundwater systems and protection of water quality. The study area is located in the Saldoran karstic region, western Iran. It is characterized by a high rate of precipitation and recharge via highly permeable fractured karstic formations. Pire-Ghar, Sarabe-Babaheydar and Baghe-rostam are three major karstic springs which drain the Saldoran anticline. The mean discharge rate and electrical conductivity values for these springs were 3, 1.9 and 0.98 m³/s, and 475, 438 and 347 μS/cm, respectively. Geology, hydrogeology and geographical information system (GIS) methods were used to define the catchment areas of the major karstic springs and to map recharge zones in the Saldoran anticline. Seven major influencing factors on groundwater recharge rates (lithology, slope value and aspect, drainage, precipitation, fracture density and karstic domains) were integrated using GIS. Geology maps and field verification were used to determine the weights of factors. The final map was produced to reveal major zones of recharge potential. More than 80 % of the study area is terrain that has a recharge rate of 55–70 % (average 63 %). Evaluating the water budget of Saldoran Mountain showed that the total volume of karst water emerging from the Saldoran karst springs is equal to the total annual recharge on the anticline. Therefore, based on the geological and hydrogeological investigations, the catchment area of the mentioned karst springs includes the whole Saldoran anticline.     

Analysis of karst spring recession curves,west of Iran

The recession hydrographs of karst springs provide important information about aquifer characteristics such as storage properties and drainage potential, karstification degree, and other hydrological features. The Faresban, Famaasb, Gonbad-e-Kabood, and Gian springs are the important karst springs which drain main karst aquifers in west of Iran. In this study, we evaluate hydrological characteristics of these four karst springs using different analytic equations. Almost, all the recession curves were well fitted by a function that consists of three main exponential terms. The initial portion of the recession curve represents the fast drainage of large fractures and conduits, ending with the slowly decreasing curve, where the drainage of rock matrix and small fissures or baseflow is dominant. The Gian spring has the largest storage volume but the lowest drained water (7%). Using Mangin’s equation, it is apparent that the baseflow supplies over 85% of the total drained water to the Gian spring, while a small quantity of total outflow is made up of quickflow periods. Of the water available to the Famaasb, Faresban, and Gonbad-e-Kabood springs, 85, 80, and 70% are drained during period of baseflow, respectively, whereas the quickflow periods have minor importance in some springs. Based on the Coutagne’s equation results, it can be concluded that the catchment of the Famaasb and Gonbad-e-Kabood springs have a karstified saturation zone with drainage channels that provide a rapid discharge with little storage capacity. At Gian catchment area, the initial discharge is not as rapid as karstification but is fairly homogeneous throughout the aquifer.     

Multivariate statistics and hydrogeochemical modeling for source identification of major elements and heavy metals in the groundwater of Qareh-Ziaeddin plain,NW Iran

The present study aims to evaluate the possible source of major and some minor elements and heavy metals in the groundwater of Qareh-Ziaeddin plain, NW Iran with respect to chemical elements, saturation index, and multivariate statistics including correlation coefficient, cluster analysis, and factor analysis. Groundwater samples were collected in Jun 2016 and measured with respect to EC, pH, major and some minor elements and heavy metals including Fe, Mn, Zn, Cr, Pb, Cd, Al, and As. Among all the measured parameters, some of the samples exceed the World Health Organization (WHO) guideline value for EC, Na, Mg, HCO₃, SO₄, Cl, NO₃, F, As, Zn, and Pb. The results of correlation analysis show that weathering and dissolution of minerals especially evaporites and silicates, water-rock interaction, and cation exchange are dominant occurred processes in the groundwater of the study area. Also, denitrification process is occurred in the groundwater system. Cluster analysis categorizes the samples into three distinct groups which are different based on their EC and dependent variables, e.g., Na, Ca, Cl, SO₄ and pH, Pb, Cd, and As. It can be found that volcanic, evaporite, and clay formations have the least impact on the chemistry of the cluster 1 samples while clay and evaporite formations have the highest impact on the cluster 3 and also calcareous formations on cluster 2. Factor analysis shows that five factors, with total variance of 83%, are effective in the release of heavy metals and groundwater chemistry which are mostly geogenic.     

Vertical distribution and source identification of polycyclic aromatic hydrocarbons in anoxic sediment cores of Chini Lake, Malaysia: Perylene as indicator of land plant-derived hydrocarbons

Four anoxic sediment cores were collected from Chini Lake, Malaysia in order to investigate the variability of polycyclic aromatic hydrocarbon (PAH) and perylene concentrations. The study also determined significant differences of perylene concentrations in different sediment layers. Total PAH concentrations ranged from 248 to 8098 ng g⁻¹ in the samples. Diagnostic PAH ratios such as methylphenanthrenes/phenanthrene (MP/P), phenanthrene/anthracene (P/A) and fluoranthene/(fluoranthene + pyrene) (Fl/(Fl + Py) revealed a dominance of pyrogenic influences and partial petrogenic inputs to the top sediment layers. Perylene concentrations were high in the top layers (<12 cm) and increased with increasing depth. There is a significant positive correlation (r = 0.705, p = 0.01) between perylene concentrations and TOC. Analysis of variance (ANOVA) and LSD revealed significant differences (p < 0.05) in TOC-normalized perylene concentrations between the upper (<12 cm) and bottom layers (>12 cm). The average perylene concentrations accounted for 26–50% (0–12 cm) and 50–77% (12–36 cm) of pentacyclic-aromatic hydrocarbon isomers (PAI) present whereas it made up 10–34% (0–12 cm) and 46–66% (12–36 cm) of the total PAH. The average pyrene concentrations decreased with increasing depth and accounted for 62% (0–3 cm), 20–23% (3–12 cm) and 3–1.4% (12–36 cm) of perylene present. The results of hierarchical cluster analysis based on these ratios suggested different input sources for the top and bottom layers. It is concluded that the activity of termites on woody plants produced perylene which is supplied to the lake by run-off from the heavy and frequent rains in this Asian tropical climate. In addition, there was also in situ formation of perylene in the bottom layers due to diagenetic processes.     

Geochemistry of subsurface Late Quaternary ironstones in Rajshahi and Bogra Districts,Bangladesh: implications for genetic and depositional conditions

The present study deals with the geochemistry of Late Quaternary ironstones in the subsurface in Rajshahi and Bogra districts, Bangladesh with the lithological study of the boreholes sediments. Major lithofacies of the studied boreholes are clay, silty clay, sandy clay, fine to coarse grained sand, gravels and sands with (fragmentary) ironstones. The ironstones contain major oxides, Fe₂O₃* (* total Fe) (avg. 66.6 wt%), SiO₂ (avg. 15.3 wt%), Al₂O₃ (avg. 4.0 wt%), MnO (avg. 7.7 wt%), and CaO (avg. 3.4 wt%). These geochemical data imply that the higher percentage of Fe₂O₃* along with Al₂O₃ and MnO indicate the ironstone as goethite and siderite, which is also validated by XRD data. A comparatively higher percentage of SiO₂ indicates the presence of relative amounts of clastic quartz and manganese-rich silicate or clay in these rocks. These ironstones also have significant amounts of MnO (avg. 7.7 wt%) suggesting their depositional environments under oxygenated condition. Chemical data of these ironstones suggest that the source rock suffered deep chemical weathering and iron was mostly carried in association with the clay fraction and organic matter. Iron concretion was mostly formed by bacterial build up in swamps and marshes, and was subsequently embedded in clayey mud. Within the coastal environments, the water table fluctuates and goethite and siderite with mud and quartz became dry and compacted to form ironstone.