Nitrogen pollution of groundwater has created problems worldwide. Riparian zones form a connection hub for terrestrial and aquatic ecosystems. As a potential source of ammonium in groundwater, aquitards have an important effect on the environment of riparian zones. The spatial distribution and factors influencing the ammonium content in the riparian zone aquitard of a small watershed were analyzed through three geological boreholes with increasing distances from the river: boreholes A > B > C. The results show that the distribution of ammonium was closely related to the lithology of sediments. Under the influence of the river and floods, the average content of ion exchange form of ammonium of sediments in borehole A (stable sedimentary environment) was 94.31 mg kg⁻¹, accounting for 21.2% of the transferable ammonium. The average proportions of ion exchange form of ammonium in the transferable ammonium of boreholes B and C (unstable sedimentary environment) were 19.1% and 17.4%, respectively. The carbonate and iron-manganese oxide forms of ammonium content of sediments in three boreholes were 0.96–15.28 mg kg⁻¹ and 2.3–54.4 mg kg⁻¹, respectively; this was mainly affected by the pH and Eh of the sedimentary environment. Organic sulfide, the form of transferable ammonium of sediments mainly exists in organic matter. The ammonium content in pore water generally increased with depth and was mainly derived from the mineralization of humic-like organic matter in borehole A. The ammonium in pore water in boreholes B and C mixed with ammonium from the mineralization of organic matter and the desorption of ion exchange form ammonium within sediments. The ammonium content in the pore water (up to 5.34 mg L⁻¹) was much higher than the limit for drinking water of 0.5 mg L⁻¹ in China. Therefore, the aquitard has a high risk of releasing ammonium and poses a certain threat to the quality of groundwater.

The adverse health effects associated with the inhalation and ingestion of naturally occurring radon gas produced during the uranium decay chain mean that there is a need to identify high-risk areas. This study detected radon-prone areas using a geographic information system (GIS)-based probabilistic and machine learning methods, including the frequency ratio (FR) model and a convolutional neural network (CNN). Ten influencing factors, namely elevation, slope, the topographic wetness index (TWI), valley depth, fault density, lithology, and the average soil copper (Cu), calcium oxide (Cao), ferric oxide (Fe₂O₃), and lead (Pb) concentrations, were analyzed. In total, 27 rock samples with high activity concentration index values were divided randomly into training and validation datasets (70:30 ratio) to train the models. Areas were categorized as very high, high, moderate, low, and very low radon areas. According to the models, approximately 40% of the study area was classified as very high or high risk. Finally, the radon potential maps were validated using the area under the receiver operating characteristic curve (AUC) analysis. This showed that the CNN algorithm was superior to the FR method; for the former, AUC values of 0.844 and 0.840 were obtained using the training and validation datasets, respectively. However, both algorithms had high predictive power. Slope, lithology, and TWI were the best predictors of radon-affected areas. These results provide new information regarding the spatial distribution of radon, and could inform the development of new residential areas. Radon screening is important to reduce public exposure to high levels of naturally occurring radiation.

Beach sediments of Kanyakumari at the southernmost India were evaluated for metals and trace elements and to assess their possible impact on coastal ecosystems. Positive correlations (except for Cd and Sr) between them indicated metamorphic lithologies and heavy mineral deposits as possible sources. Significant-extremely high enrichment and very high contamination of Th, Zr, Mo, Ti and U reflected the presence of different heavy minerals. The geo-accumulation index, however, mirrored their variable abundances at different sites. Association of Cd with P suggested the influence of anthropogenic solid waste from fishing industry. It might have caused >41-fold enrichment of Cd and the Fe- Mn-oxides possibly acted as scavengers for 13-fold enrichment of As compared to UCC. Concentrations of Zn and Cr between ERL and ERM in 13% and 93% of the samples, and Ni > ERM in 87% of sediments suggest their bioavailability to seawater with a potential risk for coastal aquifers.

Lanthanoids in the southern Gulf of California (GC) seawater are reported for the first time. Lanthanoids showed differences between peninsular and continental coastline, coastal or marine ecosystems, and dry or rainy season. The chondrite-normalized values showed high variability but followed a same pattern. Light lanthanoids were more enriched than heavy ones. Values of ∑Ln and La/Lu were higher in continental than peninsular coastlines, coastal than adjacent marine ecosystems, and rainy than dry season. Differences were related to the lithology and perturbation degree of the ecosystem watersheds. The chondrite-normalized patterns are typical of geological origin. Slightly negative Ce anomaly was related to the low levels of oxygen in water for the oxidation of Ce (III) to Ce (IV) and its posterior scavenging. Negative δEu anomaly is explained by an influx of fluvial and eolian materials from the upper continental, while a positive Eu anomaly related to hydrothermal vent inputs was non-evidenced.

The present study attempts to decipher the seasonal variations in hydro-geochemistry of groundwater in the Terekhol River Basin, western coastal region, Maharashtra, India. A total of 65 groundwater samples of post-monsoon (POMS) and pre-monsoon (PRMS) seasons were collected and analyzed for major ion composition using standard analytical procedures of APHA. Piper and Gibbs plots is used to elucidate the controlling factors which altering the groundwater composition. Scatter plots of ions indicate that major ions from lithologies exposed in the study area and anthropogenic activities are altering the groundwater chemistry. Statistical analysis includes correlation, factor analysis and cluster analysis used to interpret the hydrochemical data. As compared to the WHO drinking standards, all the groundwater samples are fit for drinking. Irrigation water suitability was ascertained based on SAR, %Na and KR indices. Overall, the groundwater chemistry in study area is reflects changes in natural processes rather than anthropogenic inputs.

The present study illustrates the delineation of the groundwater potential zones in one of the most critical and drought-affected areas under Bundelkhand region of Uttar Pradesh (India). Hydrological evaluations were carried out using GIS tools and remote sensing data which ultimately yielded several thematic maps, such as lineament density, land use/land cover, drainage density, lithology, slope, geomorphology, topographic wetness index (TWI), DEM, and soil. Thematic layers were assigned relative weightages as per their groundwater potential prospects under multi-criteria decision making (MCDM) method through analytical hierarchy process (AHP). To recognize the groundwater potential zone, weighted overlay analysis was also performed. Additionally, for testing of the Dempster-Shafer model, 16 wells in the study area have been selected. Based on the probability of the groundwater occurrence, the belief factor was equated to delineate groundwater potential zones which illustrate five different potential zones. According to the AHP model, the northwest side of the study area is characterized with very high potential zones whereas the northeast and southeast regions constitute medium and low groundwater potential zones respectively. According to the DS model, very high groundwater potential zones constitute 17% and the remaining area falls under low potential. Overall accuracy of the DS model is higher than AHP model.

In addition to global population growth due to migration from rural areas to urban areas, population density is constantly increasing in certain regions, thereby necessitating the introduction of new settlements in these regions. However, in the selection of settlement areas, no sufficient preliminary examinations have been conducted; consequently, various natural disasters may cause significant life and property losses. Herein, the most suitable settlement areas were determined using GIS (geographic information systems) in Canik District, where the population is continuously increasing. Therefore, this study aimed to incorporate a new perspective into studies on this subject. Within the scope of the study, landslide and flood risks, which are among the most important natural disasters in the region, were primarily evaluated, and high-risk areas were determined. Elevation, slope, aspect, curvature, lithology, topographic humidity index (TWI), and proximity to river parameters were used to produce flood susceptibility maps. A digital elevation model (DEM) of the study area was produced using contours on the 1/25,000 scaled topographic map. The elevation, slope, aspect, curvature, and TWI parameters were produced from the DEM using the relevant analysis routines of ArcGIS software. The raster map of each parameter was divided into 5 subclasses using the natural breaks classification method. In the reclassified raster maps, the most flood-sensitive or flood-prone subclasses were assigned a value of 5, and the least sensitive subclasses were assigned a value of 1. Then, the reclassified maps of the 7 parameters were collected using the “map algebra” function of ArcGIS 10.5 software, and the flood susceptibility index (FSI) map of the study area was obtained. The flood susceptibility map of the study area was obtained by dividing the FSI into 5 subclasses (very low, low, moderate, high, and very high) according to the natural breaks classification method. Thereafter, suitable and unsuitable areas in terms of biocomfort, which affects people’s health, peace, comfort, and psychology and is significant in terms of energy efficiency, were determined. At the last stage of the study, the most suitable settlement areas that were suitable in terms of both biocomfort and low levels of landslide and flood risks were determined. The calculated proportion of such areas to the total study area was only 2.1%. Therefore, because these areas were insufficient for the establishment of new settlements, areas that had low landslide and flood risks but were unsuitable for biocomfort were secondarily determined; the ratio of these areas was calculated as 56.8%. The remaining areas were inconvenient for the establishment of settlements due to the risk of landslides and floods; the ratio of these areas was calculated as 41.1%. This study is exemplary in that the priority for the selection of settlement areas was specified, and this method can be applied for selecting new settlements for each region considering different criteria. Due to the risk of landslides or flooding in the study area, the areas unsuitable for establishing a settlement covered approximately 41.1% of the total study area. The areas that had low flood and landslide risks but were suitable for biocomfort constituted only 2.1% of the study area. In approximately 56.8% of the study area, the risk of landslides or floods was low, and these areas were unsuitable in terms of biocomfort. Therefore, these areas were secondarily preferred as settlement areas. The most suitable areas for settlements constituted only 0.19% of the total study area, and these areas will not be able to meet the increasing demand for settlement area. Therefore, it is recommended to select areas that do not have the risk of landslides and floods but are unsuitable for biocomfort. This study reveals that grading should be performed in the selection of settlement areas. When choosing a settlement area in any region, possible natural disasters in the region should be identified first, and these disasters should be ordered in terms of their threat potential. Moreover, biocomfort areas suitable for settlements should be considered. In the next stages of settlement area selection, the criteria that affect the peace and comfort of people, such as distance to pollution sources, distance to noise sources, and proximity to natural areas, should also be evaluated. Thus, a priority order should be created for the selection of settlement areas using various other criteria.

Floods are among the most devastating environmental hazards that directly and indirectly affect people’s lives and activities. In many countries, sustainable environmental management requires the assessment of floods and the likely flood-prone areas to avoid potential hazards. In this study, the performance and capabilities of seven machine learning algorithms (MLAs) for flood susceptibility mapping were tested, evaluated, and compared. These MLAs, including support vector machine (SVM), random forest (RF), multivariate adaptive regression spline (MARS), boosted regression tree (BRT), functional data analysis (FDA), general linear model (GLM), and multivariate discriminant analysis (MDA), were tested for the area between Safaga and Ras Gharib cities, Red Sea, Egypt. A geospatial database was developed with eleven flood-related factors, namely altitude, slope aspect, lithology, land use/land cover (LULC), slope length (LS), topographic wetness index (TWI), slope angle, profile curvature, plan curvature, stream power index (SPI), and hydrolithology units. In addition, 420 actual flooded areas were recorded from the study area to create a flood inventory map. The inventory data were randomly divided into training group with 70% and validation group with 30%. The flood-related factors were tested with a multicollinearity test, the variance inflation factor (VIF) was less than 2.135, the tolerance (TOL) was more than 0.468, and their importance was evaluated with a partial least squares (PLS) method. The results show that RF performed the best with the highest AUC (area under curve) value of 0.813, followed by GLM with 0.802, MARS with 0.801, BRT with 0.777, MDA with 0.768%, FDA with 0.763, and SVM with 0.733. The results of this study and the flood susceptibility maps could be useful for environmental mitigation, future development activities in the area, and flood control areas.

The surface water is a significant feature in the hydrological system and is a vital compound for life growth. Assessment of trace elements in the water bodies is essential since it poses huge threats to aquatic organisms and humans if present in high concentrations. This study was carried out to assess the seasonal changes in the dissolved trace elements concentration in Bhavani river, which is one of the major rivers of Tamil Nadu, southern India and also to assess the human health risk due to its consumption. A total of 46 surface water samples were collected along the river during pre-monsoon and post-monsoon of 2018 and were analyzed for various trace elements such as Zn, Cu, Fe, Ni, and Pb. The variation in trace element concentration is observed spatially, where higher concentration is found in samples from agricultural and urban areas than the samples from the undisturbed natural-mountain terrains. The results highlighted that the concentrations of trace elements differ temporally where the concentration is greater during the monsoon due to increased discharge of sewage and agricultural run off to the river. Multivariate statistical analysis indicates stronger relationship between trace elements and other physio-chemical parameters hinting that natural and anthropogenic sources alters the riverine chemistry. Thus, the rainfall–runoff characteristics along with lithology, topography, and landuse of the basin plays a dominant role in the seasonal variation of dissolved trace elements. The water quality index value shows “good/excellent” during pre-monsoon and “marginal/fair” during monsoon season and the Heavy Metal Pollution Index values were also low during both the seasons. The river water samples which defy these indices were found to be either from urban or agricultural lands. The oral and dermal ingestion health risk to adults was assessed, which indicates that the risks posed to humans by consumption of water were minimal. The trace metal concentration of the river was then compared with the other rivers of world and India, where it shows that Zn, Cu, and Ni concentration was higher in Bhavani than in most of the rivers. Thus, the study highlighted that the urban settlements and agricultural lands have a considerable influence on river quality thereby triggering the increase in trace element concentrations. Therefore, the study necessitates on the continuous monitoring of river along with adoption of stringent discharge protocols.

Understanding the spatial distribution of soil salinity is required to conserve land against degradation and desertification. Against this background, this study is the first attempt to predict soil salinity in the Jaghin basin, in southern Iran, by applying and comparing the performance of four deep learning (DL) models (deep convolutional neural networks—DCNNs, dense connected deep neural networks—DenseDNNs, recurrent neural networks-long short-term memory—RNN-LSTM and recurrent neural networks-gated recurrent unit—RNN-GRU) and six shallow machine learning (ML) models (bagged classification and regression tree—BCART, cforest, cubist, quantile regression with LASSO penalty—QR-LASSO, ridge regression—RR and support vectore machine—SVM). To do this, 49 environmental landsat8-derived variables including digital elevation model (DEM)-extracted covariates, soil-salinity indices, and other variables (e.g., soil order, lithology, land use) were mapped spatially. For assessing the relationships between soil salinity (EC) and factors controlling EC, we collected 319 surficial (0–5 cm depth) soil samples for measuring soil salinity on the basis of electrical conductivity (EC). We then selected the most important features (covariates) controlling soil salinity by applying a MARS model. The performance of the DL and shallow ML models for generating soil salinity spatial maps (SSSMs) was assessed using a Taylor diagram and the Nash Sutcliff coefficient (NSE). Among all 10 predictive models, DL models with NSE ≥ 0.9 (DCNNs was the most accurate model with NSE = 0.96) were selected as the four best models, and performed better than the six shallow ML models with NSE ≤ 0.83 (QR-LASSO was the weakest predictive model with NSE = 0.50). Based on DCNNs-, the values of the EC ranged between 0.67 and 14.73 dS/m, whereas for QR-LASSO the corresponding EC values were 0.37 to 19.6 dS/m. Overall, DL models performed better than shallow ML models for production of the SSSMs and therefore we recommend applying DL models for prediction purposes in environmental sciences.