Determination of dominant biogeochemical processes in a contaminated aquifer-wetland system using multivariate statistical analysis.
ABSTRACT Determining the processes governing aqueous biogeochemistry in a wetland hydrologically linked to an underlying contaminated aquifer is challenging due to the complex exchange between the systems and their distinct responses to changes in precipitation, recharge, and biological activities. To evaluate temporal and spatial processes in the wetland-aquifer system, water samples were collected using cm-scale multi-chambered passive diffusion samplers (peepers) to span the wetland-aquifer interface over a period of 3 yr. Samples were analyzed for major cations and anions, methane, and a suite of organic acids resulting in a large dataset of over 8000 points, which was evaluated using multivariate statistics. Principal component analysis (PCA) was chosen with the purpose of exploring the sources of variation in the dataset to expose related variables and provide insight into the biogeochemical processes that control the water chemistry of the system. Factor scores computed from PCA were mapped by date and depth. Patterns observed suggest that (i) fermentation is the process controlling the greatest variability in the dataset and it peaks in May; (ii) iron and sulfate reduction were the dominant terminal electron-accepting processes in the system and were associated with fermentation but had more complex seasonal variability than fermentation; (iii) methanogenesis was also important and associated with bacterial utilization of minerals as a source of electron acceptors (e.g., barite BaSO(4)); and (iv) seasonal hydrological patterns (wet and dry periods) control the availability of electron acceptors through the reoxidation of reduced iron-sulfur species enhancing iron and sulfate reduction.
- SourceAvailable from: Nawaf Abu-Khalaf[Show abstract] [Hide abstract]
ABSTRACT: The multivariate data analysis is used to analyse groundwater samples from 13 different wells along a period of 13 years. The results show that the most significant pollutants for the groundwater resources in the region are Cl and Na which are come mainly from the Wadi Zomer stream as the main source. HCO3, show relatively low concentration and restricted to those wells which have low abstraction rates and, slightly effected by the pollutants, and received a good replenishment of. Ward’s method was used for cluster analysis. It managed to classify the wells into three groups, according their geochemical and locations characteristics. Some wells were clustered near each other, since they share the same effects of the surrounded environment. The geological formations and the layers that water taped from play also a role in the water quality distribution. The results emphasize that there is a pollution-dilution process that the groundwater undergoes between a freshwater end-member from the upper Cenomanian Turonian replenished water that prevailed more HCO3 content, and the polluted water end member that contains high Na/Cl.
- [Show abstract] [Hide abstract]
ABSTRACT: PurposeSedimentary aquifers are prone to anthropogenic disturbance. Measures aimed at mitigation or adaptation require sound information on the reactivity of soil/sediments towards the infiltrating water, as this determines the chemical quality of the groundwater and receiving surface waters. Here, we address the issues of relevant sediment properties, adequate analytical methods, borehole location selection, detail of stratification, and required sample size, to develop a protocol for efficient characterization of subsurface reactivity on a regional scale. Materials and methodsThe sequence of geological formations in the Dutch part of the North Sea Basin is documented in the form of systematic descriptions of some 450,000 borings. The basic data are stored in a database that also includes a limited amount of geochemical data collected for specific research projects. Based on the borehole descriptions, a Digital Geological Model of the Netherlands (DGM) has recently been completed. We combined the results of a statistical analysis of the existing geochemical data with theoretical and practical considerations, to assess the degree of variability of subsurface reactivity, the relevance of different DGM-based stratifications, and the efficiency and possible redundancy of analytical parameters. Results and discussionWe present two protocols for the quantitative characterization of the reactive properties of the soil and subsurface sediments of the Netherlands, down to a depth of about 30m below surface level. As numerous strategies are already available for soil surveying, the facies-based protocol for boring and sampling is aimed at subsoil sediments. Stratification is a combination of regional, lithological, and lithostratigraphical classifications. Selection of borehole locations and sampling depths is first based on the a priori information. Given the results of the first round, additional boring, sampling and analysis are performed when necessary. The analytical protocol also applies to soil surveys. It deploys limited means to obtain the most relevant information on subsurface reactivity in view of the priority environmental issues identified. ConclusionsWith the progress of technologies for aquifer architecture characterization and routine chemical analysis, assessment of subsurface reactivity on a regional scale has now become feasible. Lithological stratification is essential, but regional and lithostratigraphical variability cannot be ignored. With adequate stratification, a sample size of 45 per stratum was found sufficient in most instances. The key analytes chosen appear to be statistically independent; hence, a further reduction in analytical techniques would result in serious loss of information. KeywordsAnalytical protocol–Aquifer sediments–Characterization–Geochemistry–Reactivity–Sampling protocolJournal of Soils and Sediments 02/2010; 11(2):336-351. DOI:10.1007/s11368-010-0313-4 · 2.11 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: The fuzzy cluster and different multivariate statistical techniques were used to evaluate landscape water bodies (LWB) of Wenzhou using data sets of 7 different sites. The fuzzy cluster and hierarchical cluster analysis grouped seven sampling sites into three clusters i.e. highly pollution level (HPL), medium pollution level (MPL) and less pollution level (LPL) sites based on the similarity of water quality characteristics. The factor analysis showed that factor1 included BOD, COD, TP and NH4-N, whereas factor1 included Chla. The order of general pollution in LWB was Chasan north River, Chasan south River, Wenshiyuan River, Maanchi Park, Zhongsan Park, Jiusan Park and Xiusan Park. The results of fuzzy cluster and hierarchical cluster analysis are in good agreement with the discriminant analysis. The fuzzy cluster and multivariate statistical techniques are useful tools of data mining for evaluation and classification of landscape water bodies and may be applicable to analysis and assessment of other surface water.