Multivariate methods and artificial neural networks in the assessment of the response of infaunal assemblages to sediment metal contamination and organic enrichment

Understanding the patterns and processes underlying water quality and pollution risk in West-Africa River using self-organizing maps and multivariate analyses

Rivers are dynamic systems in complex interactions with their surrounding environments. Reliable and fast interpretation of water quality is therefore needed for sustainable river management. Unfortunately, water quality and environmental status interactions have not yet been documented sufficiently in West-Africa. This study explored the spatial-latitudinal and seasonal features of water quality along the Sô River Basin (SRB, West Africa) using self-organizing map (SOM) and principal component analysis. Twenty-two water quality variables were measured in the surface layer at 12 different sampling sites during a twenty-four-month period from July 2016 to June 2018. The results revealed three water quality groups, following an upstream-downstream pollution gradient: (1) upstream and middle reach sites with high dissolved oxygen and Secchi disk depth values, which are more suitable for the aquatic biota; (2) downstream sites with high concentrations of ammonium, biochemical oxygen demand, and heavy metals especially in flood period, reflecting both high organic and heavy metal pollution; and (3) brackish downstream sites characterized by less heavy metal and organic pollutions. No significant variation was observed between seasons. However, the SRB relatively suffered from higher risks of heavy metal contamination and organic pollution in wet seasons. Although hydroclimatic processes affect the water quality, anthropogenic inputs of point and non-point sources were identified and discussed as a more prominent factor contributing to variation in the water quality condition. These results offer insights into the water quality dynamics in river-estuary system as well as potential pollution sources, crucial for defining sanitation, and management measures.

Self-organizing map improves understanding on the hydrochemical processes in aquifer systems

The holistic understanding of hydrochemical features is a crucial task for management and protection of water resources. However, it is challenging for a complex region, where multiple factors can cause hydrochemical changes in studied catchment. We collected 208 groundwater samples from such region in Kumamoto, southern Japan to explicitly characterize these processes by applying machine learning technique. The analyzed groundwater chemistry data like major cations and anions were fed to the self-organizing map (SOM) and the results were compared with classical classification approaches like Stiff diagram, standalone cluster analysis and score plots of principal component analysis (PCA). The SOM with integrated application of clustering divided the data into 11 clusters in this complex region. We confirmed that the results provide much greater details for the associated hydrochemical and contamination processes than the traditional approaches, which show quite good correspondence with the recent high resolution hydrological simulation model and aspects from geochemical modeling. However, the careful application of the SOM is necessary for obtaining accurate results. This study tested different normalization approaches for selecting the best SOM map and found that the topographic error (TE) was more important over the quantization error (QE). For instance, the lower QE obtained from min-max and log normalizations showed problems after clustering the SOM map, since the QE did not confirm the topological preservation. In contrast, the lowest TE obtained from Z-transformation data showed better spatial matching of the clusters with relevant hydrochemical characteristics. The results from this study clearly demonstrated that the SOM is a helpful approach for explicit understanding of the hydrochemical processes on reginal scale that may capably facilitate better groundwater resource management.

The potential ecological risk assessment of soil heavy metals using self-organizing map

Heavy metal pollution control zoning aiming at the health maintenance of watershed soil ecosystem has become an important means of soil environmental protection and governance. Based on the self-organizing map (SOM), this study classifies the data sets of eight heavy metals such as Co, Cd, Zn, Cr, Cu, Pb, Ni, and Tl in 354 samples, calculates the potential ecological risk value of soil heavy metals in combination with the potential Hakansom Risk index (HRI), and uses the geographic information system (GIS) for visualization. In the research results, SOM has divided five soil ecological risk categories. The highest average ecological risk value of 85.95 is found in cluster IV, which is clustered and distributed in urban development areas in the upper reaches of the river. The average ecological risk values of cluster I and cluster V are relatively close at 79.64 and 79.19, respectively. Cluster I and cluster V are distributed in the north of the river in a linear and cluster manner, respectively, and are located on a concave bank with a relatively gentle slope. The average ecological risk of soil pollution in cluster II is 77.59, which is linearly distributed on both banks of the river. The ecological risk of soil pollution in cluster III is the lowest (74.39), mainly scattered in the south of rivers with less human activities. The study further identified the environmental factors that affect the soil ecological risk value in different cluster units and put forward the classified and differentiated management and control strategies for different cluster units. The research shows that SOM can cluster the data sets of heavy metals with high sensitivity and low threshold through competitive learning to effectively provide the distribution information of abnormal soil ecological risk areas. This information is helpful for urban environmental management departments and planning departments to take targeted management and recovery measures to avoid the health risks related to soil heavy metal pollution.

Spatio-Temporal Analysis of Heavy Metals in Arid Soils at the Catchment Scale Using Digital Soil Assessment and a Random Forest Model

Predicting the spatio-temporal distribution of absorbable heavy metals in soil is needed to identify the potential contaminant sources and develop appropriate management plans to control these hazardous pollutants. Therefore, our aim was to develop a model to predict soil adsorbable heavy metals in arid regions of Iran from 1986 to 2016. Soil adsorbable heavy metals were measured in 201 samples from locations selected using the Latin hypercube sampling method in 2016. A random forest (RF) model was used to determine the relationship between a suite of geospatial predictors derived from remote sensing and digital elevation model data with georeferenced measurements of soil absorbable heavy metals. The trained RF model from 2016 was used to reconstruct the spatial distribution of soil absorbable heavy metals at three historical timesteps (1986, 1999, and 2010). Results indicated that the RF model was effective at predicting the distribution of heavy metals with coefficients of determination of 0.53, 0.59, 0.41, 0.45, and 0.60 for Fe, Mn, Ni, Pb, and Zn, respectively. The predicted maps showed high spatio-temporal variability; for example, there were substantial increases in Pb (the 1.5–2 mg/kg−1 class) where its distribution increased by ~25% from 1988 to 2016—similar trends were observed for the other heavy metals. This study provides insights into the spatio-temporal trends and the potential causes of soil heavy metal contamination to facilitate appropriate planning and management strategies to prevent, control, and reduce the impact of heavy metal contamination in soils.

Chemical fractionation and bioaccessibility of potentially toxic elements in area of artisanal gold mining in the Amazon

Artisanal mining may have modified the mobility, bioavailability and bioaccessibility of potentially toxic elements (PTEs) in the Serra Pelada gold mine, eastern Amazon, Brazil, which has not yet been studied. The objectives were to perform chemical fractionation of barium (Ba), chromium (Cr), copper (Cu), nickel (Ni), lead (Pb), and zinc (Zn), and to determine the bioaccessibility of these elements in soils and mining wastes collected in agriculture, forest, mining, and urban areas from the influence zone of the Serra Pelada gold mine. Pseudo total concentrations were obtained by acid digestion, chemical fractionation was performed using the Bureau Community of Reference (BCR) sequential extraction, oral bioaccessibility was obtained by the Simple Bioaccessibility Extraction Test (SBET) and lung bioaccessibility was obtained through Gamble's solution. The pseudo total concentrations indicated contamination by Ba, Cu and Ni. The sequential extraction revealed the predominance of all elements in the residual fraction. However, Ba is in high concentrations in the greater mobility forms, ranging from 166.36 to 1379.58 mg kg^-1. Regardless of the area, Cr and Cu are more oral bioaccessible in the intestinal phase, and Zn in the gastric phase. Ba, Cr and Zn are not lung bioaccessible, while Cu, Ni and Pb are bioaccessible via inhalation. The PTEs studied deserve attention not only due to the high pseudo total concentrations found (which indicate potential risk), but also the concentrations in high mobility forms and bioaccessible fractions, especially in the areas of greatest anthropogenic occupation.

Using self-organizing map for coastal water quality classification: Towards a better understanding of patterns and processes

Self-organizing map (SOM) was used to explore the spatial characteristics of water quality in the middle and southern Fujian coastal area. Nineteen water quality variables (temperature, salinity, pH, dissolved oxygen, alkalinity, chemical oxygen demand, nutrients NH₄-N, H₂SiO₃, PO₄^-, NO₂^-, and NO₃^-, heavy metals/metalloid Cu, Zn, As, Cd, Pb, Hg, and Cr⁶⁺, and oil) were measured in the surface, middle, and bottom water layers at 94 different sampling sites. Patterns of water quality variables were visualized by the SOM planes, and similar patterns were observed for those variables that correlated with each other, indicating a common source. pH, COD, As, Hg, Pb, and Cr⁶⁺ likely originated from industries, while nutrients NH₄-N, NO₂^-, NO₃^-, and PO₄^3- were mainly attributed to agriculture and aquaculture. The k-means clustering in the SOM grouped the water quality data into nine clusters, which revealed three representative water types, ranging from low salinity to high salinity with different levels of heavy metal/metalloid pollution and nutrient pollution. Spatial changes in water quality reflected the impacts of natural factors (riverine outflows, tides, and alongshore currents), as well as anthropogenic activities (mariculture, industrial and urban discharges, and agricultural effluents). Principal component analysis (PCA) confirmed the clustering results obtained by SOM, while the latter provides a more detailed classification and additional information about the dominant variables governing the classification processes. The results of this study suggest that SOM is an effective tool for a better understanding of patterns and processes driving water quality.

Multivariate geostatistical analysis and source identification of heavy metals in the sediment of Poyang Lake in China

Heavy metals in lake sediment have become a great concern because their remobilization has frequently occurred under hydrodynamic disturbance in shallow lakes. In this study, heavy metals (Cr, Cu, Cd, Pb, and Zn) concentrations in the surface and core sediments of the largest freshwater lake in China, Poyang Lake, were investigated. Geostatistical prediction maps of heavy metals distribution in the surface sediment were completed as well as further data mining. Based on the prediction maps, the ranges of Cr, Cu, Cd, Pb, and Zn concentrations in the surface sediments of the entire lake were 96.2-175.2, 38.3-127.6, 0.2-2.3, 22.5-77.4, and 72.3-254.4mg/kg, respectively. A self-organizing map (SOM) was applied to find the inner element relation of heavy metals in the sediment cores. K-means clustering of the self-organizing map was also completed to define the Euclidian distance of heavy metals in the sediment cores. The geoaccumulation index (I_geo) for Poyang Lake indicated a varying degree of heavy metal contamination in the surface sediment, especially for Cu. The heavy metal contamination in the sediment profiles had similar pollution levels as those of surface sediment, except for Cd. Correlation matrix mapping and principal component analysis (PCA) were used to support the idea that Cr, Pb, and Zn may be mainly derived from both lithogenic and human activities, such as atmospheric and river inflow transportation, whereas Cu and Cd may be mainly contributed from anthropogenic sources, such as mining activities and fertilizer application.

Sediment core data reconstruct the management history and usage of a heavily modified urban lake in Berlin, Germany

Urban surface waters face several stressors associated with industry and urban water management. Over much of the past century, the wastewater treatment in Berlin, Germany, relied on inefficient sewage farms, which resulted in severe eutrophication and sediment contamination in the recipient surface waterbodies. A prominent example is Lake Tegel, where a multitude of management measures were applied in the last decades for the purpose of ecosystem restoration. In this study, we analyzed sediment cores of three lakes with X-ray fluorescence spectroscopy: Lake Tegel, Lake Großer Wannsee, which is environmentally similar but has a different management history, and Lake Userin, which serves as a reference located in a nature protection area. Multivariate statistical methods (principal component analysis, k-means clustering, and self-organizing maps) were used to assess the sediment quality and to reconstruct the management history of Lake Tegel. Principal component analysis established two main gradients of sediment composition: heavy metals and lithogenic elements. The impact of the management measures was visualized in the lake sediment composition changing from high abundance of heavy metals and reducing redox conditions to less-impacted sediments in recent layers. The clustering techniques suggested heterogeneity among sites within Lake Tegel that probably reflect urban water management measures. The abundance of heavy metals in recent lake sediments of Lake Tegel is similar to a lake with low urban impact and is lower than in Lake Großer Wannsee suggesting that the management measures were successful in the reduction of heavy metals, which are still a threat for surface waters worldwide.

Identifying trace metal distribution and occurrence in sediments, inundated soils, and non-flooded soils of a reservoir catchment using Self-Organizing Maps, an artificial neural network method

The Lancang-Mekong River is a trans-boundary river which provides a livelihood for over 60 million people in Southeast Asia. Its environmental security is vital to both local and regional inhabitants. Efforts have been undertaken to identify controlling factors of the distribution of trace metals in sediments and soils of the Manwan Reservoir catchment in the Lancang-Mekong River basin. The physicochemical attributes of 63 spatially distributed soil and sediment samples, along with land-use, flooding, topographic, and location characteristics, were analyzed using the Self-Organizing Map (SOM) methodology. The SOM permits the analysis of complex multivariate datasets and gives a visual interpretation that is generally not easy to obtain using traditional statistical methods. Across the catchment, enrichments of trace metals are rare overall, despite the severely enriched cadmium (Cd). The analysis of SOM showed that flooded levels and land-use types were associated with high concentrations of Cd. Sediments and inundated soils covered with shrub and open woodlands in downstream always have a high concentration of Cd. The results demonstrate that SOM is a useful tool that can aid in the interpretation of complex datasets and help identify the environment of enriched metals on a catchment scale.

Assessment of the benthic macrofauna in an artificial shell reef zone in Shuangdao Bay, Yellow Sea

The effects of artificial shell reef (ASR) on the benthic macroinvertebrates were studied in Shuangdao Bay, Yellow Sea, China. Results showed that the biomass of macroinvertebrates in the ASR increased with the age of the ASR. Based on self-organizing map (SOM), the macroinvertebrate community of short-term artificial reef (SAR), medium-term artificial reef (MAR) and long-term artificial reef (LAR) emerged as a cluster, which may indicate that the benthic community in the ASR formed after three years. The age of the ASR was the main factor affecting the benthic community. The macroinvertebrates belonged to six phyla, Platyhelminthes, Nemertea, Annelida, Mollusca, Arthropoda and Echinodermata, among which the latter four were the ones that contributed the most for abundance. The biomass of Mollusca increased dramatically with age. The dissimilarity of the species composition of Mollusca was mainly caused by Meretrix meretrix and Protothaca jedoensis. The two species accounted for 15.61%, 28.05% and 75.11% of the macroinvertebrate biomass found in SAR, MAR and LAR, respectively. The ASR might be served as a bivalve stock enhancement tool. We conclude that ASR could assemble macrobenthos effectively and increase the environmental quality of the adjacent area, being a valid option for marine habitat restoration purposes.

Estimating spatio-temporal dynamics of stream total phosphate concentration by soft computing techniques

This study attempts to model the spatio-temporal dynamics of total phosphate (TP) concentrations along a river for effective hydro-environmental management. We propose a systematical modeling scheme (SMS), which is an ingenious modeling process equipped with a dynamic neural network and three refined statistical methods, for reliably predicting the TP concentrations along a river simultaneously. Two different types of artificial neural network (BPNN-static neural network; NARX network-dynamic neural network) are constructed in modeling the dynamic system. The Dahan River in Taiwan is used as a study case, where ten-year seasonal water quality data collected at seven monitoring stations along the river are used for model training and validation. Results demonstrate that the NARX network can suitably capture the important dynamic features and remarkably outperforms the BPNN model, and the SMS can effectively identify key input factors, suitably overcome data scarcity, significantly increase model reliability, satisfactorily estimate site-specific TP concentration at seven monitoring stations simultaneously, and adequately reconstruct seasonal TP data into a monthly scale. The proposed SMS can reliably model the dynamic spatio-temporal water pollution variation in a river system for missing, hazardous or costly data of interest.

Application of chemometric analysis and self Organizing Map-Artificial Neural Network as source receptor modeling for metal speciation in river sediment

Present study deals with the river Ganga water quality and its impact on metal speciation in its sediments. Concentration of physico-chemical parameters was highest in summer season followed by winter and lowest in rainy season. Metal speciation study in river sediments revealed that exchangeable, reducible and oxidizable fractions were dominant in all the studied metals (Cr, Ni, Cu, Zn, Cd, Pb) except Mn and Fe. High pollution load index (1.64-3.89) recommends urgent need of mitigation measures. Self-organizing Map-Artificial Neural Network (SOM-ANN) was applied to the data set for the prediction of major point sources of pollution in the river Ganga.

Differentiating the Spatiotemporal Distribution of Natural and Anthropogenic Processes on River Water-Quality Variation Using a Self-Organizing Map With Factor Analysis

To elucidate the historical improvement and advanced measure of river water quality in the Taipei metropolitan area, this study applied the self-organizing map (SOM) technique with factor analysis (FA) to differentiate the spatiotemporal distribution of natural and anthropogenic processes on river water-quality variation spanning two decades. The SOM clustered river water quality into five groups: very low pollution, low pollution, moderate pollution, high pollution, and very high pollution. FA was then used to extract four latent factors that dominated water quality from 1991 to 2011 including three anthropogenic process factors (organic, industrial, and copper pollution) and one natural process factor [suspended solids (SS) pollution]. The SOM revealed that the water quality improved substantially over time. However, the downstream river water quality was still classified as high pollution because of an increase in anthropogenic activity. FA showed the spatiotemporal pattern of each factor score decreasing over time, but the organic pollution factor downstream of the Tamsui River, as well as the SS factor scores in the upstream major tributary (the Dahan Stream), remained within the high pollution level. Therefore, we suggest that public sewage-treatment plants should be upgraded from their current secondary biological processing to advanced treatment processing. The conservation of water and soil must also be reinforced to decrease the SS loading of the Dahan Stream from natural erosion processes in the future.

Occurrence, distribution and risk assessment of polychlorinated biphenyls and polybrominated diphenyl ethers in nine water sources

Water quality of water sources is a critical issue for human health in South China, which experiences rapid economic development and is the most densely populated region in China. In this study, the pollution of organohalogen compounds in nine important water sources, South China was investigated. Twenty six organohalogen compounds including seventeen polychlorinated biphenyls (PCBs) and nine polybrominated diphenyl ethers (PBDEs) were detected using gas chromatograph analysis. The concentrations of total PCBs ranged from 0.93 to 13.07ngL(-1), with an average value of 7.06ngL(-1). The total concentrations of nine PBDE congeners were found in range not detected (nd) to 7.87ngL(-1) with an average value of 2.59ngL(-1). Compositions of PCBs and PBDEs indicated the historical use of Aroclors 1248, 1254 and 1260, and commercial PBDEs may be the main source of organohalogen compounds in water sources in South China. The nine water sources could be classified into three clusters by self-organizing map neural network. Low halogenated PCBs and PBDEs showed similar distribution in the nine water sources. Cancer risks of PCBs and PBDEs via water consumption were all below 10(-6), indicating the water quality in the nine water sources, South China was safe for human drinking.

Advancing analysis of spatio-temporal variations of soil nutrients in the water level fluctuation zone of China's Three Gorges Reservoir using self-organizing map

The ~350 km² water level fluctuation zone (WLFZ) in the Three Gorges Reservoir (TGR) of China, situated at the intersection of terrestrial and aquatic ecosystems, experiences a great hydrological change with prolonged winter inundation. Soil samples were collected in 12 sites pre- (September 2008) and post submergence (June 2009) in the WLFZ and analyzed for soil nutrients. Self-organizing map (SOM) and statistical analysis including multi-way ANOVA, paired-T test, and stepwise least squares multiple regression were employed to determine the spatio-temporal variations of soil nutrients in relation to submergence, and their correlations with soil physical characteristics. Results showed significant spatial variability in nutrients along ~600 km long shoreline of the TGR before and after submergence. There were higher contents of organic matter, total nitrogen (TN), and nitrate (NO3-) in the lower reach and total phosphorus (TP) in the upper reach that were primarily due to the spatial variations in soil particle size composition and anthropogenic activities. Submergence enhanced soil available potassium (K), while significantly decreased soil N, possibly due to the alterations of soil particle size composition and increase in soil pH. In addition, SOM analysis determined important roles of soil pH value, bulk density, soil particle size (i.e., silt and sand) and nutrients (TP, TK, and AK) on the spatial and temporal variations in soil quality. Our results suggest that urban sewage and agricultural runoffs are primary pollutants that affect soil nutrients in the WLFZ of TGR.

Metal pollution status in Zhelin Bay surface sediments inferred from a sequential extraction technique, South China Sea

Surface sediments collected from Zhelin Bay, the largest mariculture base of eastern Guangdong Province, were analyzed for total metal concentrations and chemical speciation. The results demonstrated that the average total concentration (mg/kg) ranges were 36.7-65.8 (Pb), 53.8-98.8 (Cr), 39.0-87.1 (Ni), 50.9-144.5 (Cu), and 175.0-251.2 (Zn), which were clearly higher with respect to their corresponding benchmark values. The predominant speciation of Pb was reducible and comprised a residual fraction, whereas a major portion (57.6-95.4%) of Cr, Ni, Cu, and Zn was strongly associated with the residual fractions. Taking as a whole, surface sediments of Zhelin Bay had a 21% probability of toxicity based on the mean effects range-median quotient.