Statistical characterization of arsenic contamination in shallow tube wells of western Bangladesh

A regional assessment of the arsenic (As) contamination scenario in shallow tube wells (depth < 150 m) of western Bangladesh is presented. Comparisons are made in light of bulk geological differences (Pleistocene versus Holocene deposits/northwest versus southwest) and As measurement protocols (field kit (FK) versus atomic absorption spectroscopy (AAS)). Our As database comprised the following: (1) the nationwide As survey completed in 1999 by the British Geological Survey in collaboration with the Department of Public Health Engineering (DPHE); and (2) a regional As survey conducted in southwest Bangladesh by the Japan International Cooperation Agency in collaboration with DPHE in 2002. First, we characterize the error structure of the semi‐quantitative FK As measurements using collocated AAS As measurements as reference from a set of 307 wells located in southwest Bangladesh. The depth distribution of As is identified using a very dense cluster of 2963 wells over a 560 km² domain. The probability of the FK method for successful detection of a well sample as unsafe (safe) was found to be 96·9% (34·1%) and 95·2% (80·3%) for the World Health Organization (WHO) and Bangladesh safe limits, respectively. Similarly, the probability of false alarms and false hopes for WHO (Bangladesh) safe limits were found to be 3·1% (4·8%) and 87·5% (19·7%), respectively. The depth at which the highest fraction of wells exceeding a given safe limit occurred could still be inferred correctly by FK measurements. A simple bias adjustment procedure on FK As data did not result in a more accurate characterization of depth distribution of As. This indicated that simple error statistics are inadequate for advancing the utility of FKs; rather, an understanding of the complex and multidimensional error structure is required. Regional anisotropy in the spatial dependence of As for the northwest was found to be stronger than the southwest. The correlation length for As concentration in the east–west direction of northwest Bangladesh (i.e. across major river floodplains) was found to be almost twice (158·80 km) that of the north–south direction (along the major axis of Pleistocene deposits) (78·21 km). For the southwest region, the ratio of east–west to north–south correlation lengths ranged from 1·40 to 1·51. For the northwest region, because it is well known to have the lowest concentrations of As countrywide, knowledge of this anisotropy appears to suggest the need for drilling twice as many remediation deep wells in the proximity of an unsafe shallow well in the north–south direction than in the east–west direction. Findings from this study are potentially useful in setting priority areas for emergency testing, distributing remediation resources equitably and formulating a regional water resources strategy for western Bangladesh. Copyright © 2005 John Wiley & Sons, Ltd.     

Is correlation dimension a reliable proxy for the number of dominant influencing variables for modeling risk of arsenic contamination in groundwater?

The correlation dimension (CD) of a time series provides information on the number of dominant variables present in the evolution of the underlying system dynamics. In this study, we explore, using logistic regression (LR), possible physical connections between the CD and the mathematical modeling of risk of arsenic contamination in groundwater. Our database comprises a large-scale arsenic survey conducted in Bangladesh. Following the recommendation by Hossain and Sivakumar (Stoch Environ Res Risk Assess 20(1–2):66–76, 2006), who reported CD values ranging from 8 to 11 for this database, 11 variables are considered herein as indicators of the aquifer’s geochemical regime with potential influence on the arsenic concentration in groundwater. A total of 2,048 possible combinations of influencing variables are considered as candidate LR risk models to delineate the impact of the number of variables on the prediction accuracy of the model. We find that the uncertainty associated with prediction of wells as safe and unsafe by LR risk model declines systematically as the total number of influencing variables increases from 7 to 11. The sensitivity of the mean predictive performance also increases noticeably for this range. The consistent reduction in predictive uncertainty coupled with the increased sensitivity of the mean predictive behavior within the universal sample space exemplify the ability of CD to function as a proxy for the number of dominant influencing variables. Such a rapid proxy, based on non-linear dynamic concepts, appears to have considerable merit for application in current management strategies on arsenic contamination in developing countries, where both time and resources are very limited.     

Mineralogical approach in elucidation of contamination mechanism for toxic trace elements in the environment: Special reference to arsenic contamination in groundwater

Mineralogy and mineral-microbe interactions play a significant role in the environmental contamination of toxic trace elements in groundwater and wetland ecosystems. This review has described the importance of use of TEM that is an essential tool to study the environmental mineralogy and mineral-microbe interactions. Biomineralization process that can advance the succession and evolution of biosphere on the earth has also been summarized briefly. This study explained the role of minerals to control the mechanism of adsorption, desorption, mobility and transport of toxic trace elements, especially arsenic (As) in the groundwater. Finally, this review has explored the evolution of As contamination in groundwater of the Quaternary age deposits including the Holocene and Pleistocene with special reference to groundwater of Bangladesh, West Bengal (India), Vietnam and Inner Mongolia (China).     

Contrasting Influence of Geology on E. coli and Arsenic in Aquifers of Bangladesh

Arsenic in groundwater has been a concern in South and Southeast Asia for more than a decade. We explore here the possibility that hydrogeologic factors recently shown to influence the distribution of arsenic might also affect the level of contamination of shallow (<20 m) wells with microbial pathogens. A total of 96 shallow tube wells in two nearby villages of Bangladesh were surveyed during the wet and dry seasons, along with 55 deeper wells in neighboring villages. One of the two villages is located in a particularly sandy environment where recharge is rapid and shallow wells contain little arsenic. Shallow aquifers in the other village are capped with an impermeable clay layer, recharge is an order of magnitude slower, and arsenic levels are high. The fecal indicator E. coli was detected in 43% of shallow wells, compared with 12% of deeper wells. More shallow wells contained E. coli during the wet season (61%) than during the dry season (9%). In the wet season, a higher proportion of shallow wells in the village with low arsenic levels (72%) contained E. coli compared with the village having high arsenic levels (43%). Differences in arsenic and E. coli distributions between the two sites are likely due to the differences in permeability of near‐surface sediments although differences in average well‐depth between the two villages (9 ± 4 vs. 15 ± 3 m) may play a role as well. Hydrogeologic conditions that favor high levels of fecal contamination but low levels of arsenic in shallow groundwater should be taken into account during arsenic mitigation throughout South and Southeast Asia.     

Searching for strange attractor in wastewater flow

 Chaos is a complex and irregular world in contrast with simple and regular natures of linear systems. Scientists and engineers have invoked low-dimensional chaos for understanding the nature of real systems. In this study, the complex behavior of a daily wastewater flow and evidence of deterministic nonlinear dynamics are investigated. The analysis involves both a metric approach of the correlation dimension and a topological technique called the close returns plot. The estimation procedure of delay time and delay time window is reviewed using a new technique called the C–C method for the state space reconstruction. And both parameters are used for estimating the correlation dimension. As a result, the daily wastewater flow shows no evidence of chaotic dynamics, which implies that stochastic models rather than deterministic chaos may be more appropriate for representing an investigated series.     

Nonlinear ensemble prediction of chaotic daily rainfall

The significance of treating rainfall as a chaotic system instead of a stochastic system for a better understanding of the underlying dynamics has been taken up by various studies recently. However, an important limitation of all these approaches is the dependence on a single method for identifying the chaotic nature and the parameters involved. Many of these approaches aim at only analyzing the chaotic nature and not its prediction. In the present study, an attempt is made to identify chaos using various techniques and prediction is also done by generating ensembles in order to quantify the uncertainty involved. Daily rainfall data of three regions with contrasting characteristics (mainly in the spatial area covered), Malaprabha, Mahanadi and All-India for the period 1955–2000 are used for the study. Auto-correlation and mutual information methods are used to determine the delay time for the phase space reconstruction. Optimum embedding dimension is determined using correlation dimension, false nearest neighbour algorithm and also nonlinear prediction methods. The low embedding dimensions obtained from these methods indicate the existence of low dimensional chaos in the three rainfall series. Correlation dimension method is done on the phase randomized and first derivative of the data series to check whether the saturation of the dimension is due to the inherent linear correlation structure or due to low dimensional dynamics. Positive Lyapunov exponents obtained prove the exponential divergence of the trajectories and hence the unpredictability. Surrogate data test is also done to further confirm the nonlinear structure of the rainfall series. A range of plausible parameters is used for generating an ensemble of predictions of rainfall for each year separately for the period 1996–2000 using the data till the preceding year. For analyzing the sensitiveness to initial conditions, predictions are done from two different months in a year viz., from the beginning of January and June. The reasonably good predictions obtained indicate the efficiency of the nonlinear prediction method for predicting the rainfall series. Also, the rank probability skill score and the rank histograms show that the ensembles generated are reliable with a good spread and skill. A comparison of results of the three regions indicates that although they are chaotic in nature, the spatial averaging over a large area can increase the dimension and improve the predictability, thus destroying the chaotic nature.     

Risk characterization for arsenic-impacted water sources, including ground-truthing

One of the most severe issues for water supply in Bangladesh is high arsenic concentrations. Widespread implementation of tubewells was determined to be highly problematic due to arsenic contamination in the groundwater. For substantial areas in Bangladesh, groundwater cannot be utilized for water supply without treatment to reduce arsenic levels. Thus, calculations for the incremental excess lifetime cancer risk are used to demonstrate that indeed, seeing elevated arsenicosis must be expected. By observing the correlations between arsenic and iron and the prevalence of arsenicosis patients, the relationships among the concentrations of arsenic and iron, and the percentage of contaminated wells are significant for determining the prevalence of arsenicosis. The results declare that high arsenic concentrations and high percentages of contaminated wells, as well as the lower iron concentrations, lead to a higher prevalence of arsenicosis patients. It demonstrates the validity of the risk calculation procedure and single arsenic concentrations cannot provide good prediction of arsenicosis risk. When the concentration of arsenic and iron are both high, the ratio of arsenic/iron will decrease, and the arsenicosis rates will lessen, as well. The findings indicate that co-precipitation of iron and arsenic can remove some of the arsenic before the groundwater is consumed.     

Evidence for Elevated Levels of Arsenic in Public Wells of Bangladesh Due To Improper Installation

One of the mainstays of mitigation to reduce the exposure of the rural population of Bangladesh to arsenic (As) from private, mostly <90‐m deep wells over the past 15 years has been the installation of over 300,000 deeper community wells. A comprehensive testing campaign previously conducted across a 180 km² of area of Bangladesh identified 9 out of total of 927 wells >90 m deep that contained >50 µg/L arsenic. We show here that for five of these nine wells, conductivity profiles obtained after spiking the well bore with salt indicate a shallow leak that could explain the high As in the well water. In two of the five leaky wells, the presence of additional screens at the depth of the leak was documented with a downhole camera. The downhole camera did not detect anomalies in the construction of the remaining three leaky wells or in the four wells that did not leak. The four wells that did not leak were all >150‐m deep and located in two villages separated by less than 500 m. Excluding these two villages and a handful of leaky wells, the results indicate an aquifer that is consistently low in As over a sizeable area at depths >90 m. Isolated cases of public wells that are elevated in As that have been reported elsewhere in Bangladesh may therefore reflect improper installation rather than actual contamination of the deep aquifer.     

Comparison of ordinary kriging and artificial neural network for spatial mapping of arsenic contamination of groundwater

In this technical note, we investigate the hypothesis that ‘non-linearity matters in the spatial mapping of complex patterns of groundwater arsenic contamination’. The spatial mapping pertained to data-driven techniques of spatial interpolation based on sampling data at finite locations. Using the well known example of extensive groundwater contamination by arsenic in Bangladesh, we find that the use of a highly non-linear pattern learning technique in the form of an artificial neural network (ANN) can yield more accurate results under the same set of constraints when compared to the ordinary kriging method. One ANN and a variogram model were used to represent the spatial structure of arsenic contamination for the whole country. The probability for successful detection of a well as safe or unsafe was found to be atleast 15% larger than that by kriging under the country-wide scenario. The probability of false hopes, which is a serious issue in public health monitoring was found to be significantly lower (by more than 10%) than that by kriging.     

强震地面运动的混沌特性分析

引入非线性动力学理论和混沌时间序列分析方法考察强震地面运动加速度时程的非线性特征。首先采用功率谱分析法、主成份分析法和Cao方法定性判断地震动加速度时程具有混沌特性,然后应用混沌时间序列分析方法定量计算了30条地震动加速度时程的三个非线性特征参数。计算表明,这些地震动时程的关联维数为2．0～4．0的分数维,Kolmogorov熵K2为大于零的有限正值,最大Lyapunov指数在o～i．0之间。结果说明,强震地面运动具有混沌特性,地震动的高度不规则和复杂性是地震过程强非线性的反映。     

Correlating Bedrock Folds to Higher Rates of Arsenic Detection in Groundwater,Southeast Wisconsin,USA

Arsenic in private drinking water wells is a significant problem across much of eastern Wisconsin, USA. The release mechanism and stratigraphic distribution of sulfide and iron (hydr)oxide sources of arsenic in bedrock aquifers are well understood for northeastern Wisconsin. However, recent geologic mapping has identified numerous small bedrock folds to the south, and the impact of these geologic structures on local groundwater flow and well contamination has been little studied. This paper examines the hydrologic and structural effects of the Beaver Dam anticline, southeast Wisconsin, on arsenic in groundwater in the region. Multivariate logistic regression shows wells near the Beaver Dam anticline are statistically more likely to detect arsenic in groundwater compared to wells farther away. Structural and hydrologic changes related to folding are interpreted to be the cause. Core drilled near the fold axis is heavily fractured, and many fractures are filled with sulfides. Elevated hydraulic conductivity estimates are also recorded near the fold axis, which may reflect a higher concentration of vertical fractures. These structural and hydrologic changes may have led to systematic changes in the distribution and concentration of arsenic-bearing mineral hosts, resulting in the observed detection pattern. For areas with similar underlying geology, this approach may improve prediction of arsenic risk down to the local level.     

Temporal scaling in river flow: can it be chaotic? / L’invariance d’échelle de l’écoulement fluvial peut-elle être chaotique?

Abstract

Abstract Identification of the presence of scaling in the river flow process has been a challenging problem in hydrology. Studies conducted thus far have viewed this problem essentially from a stochastic perspective, because the river flow process has traditionally been assumed to be a result of a very large number of variables. However, recent studies employing nonlinear deterministic and chaotic dynamic concepts have reported that the river flow process could also be the outcome of a deterministic system with only a few dominant variables. In the wake of such reports, a preliminary attempt is made in this study to investigate the type of scaling behaviour in the river flow process (i.e. chaotic or stochastic). The investigation is limited only to temporal scaling. Flow data of three different scales (daily, 5-day and 7-day) observed in each of three rivers in the USA: the Kentucky River in Kentucky, the Merced River in California and the Stillaguamish River in Washington, are analysed. It is assumed that the dynamic behaviour of the river flow process at these individual scales provides clues about the scaling behaviour between these scales. The correlation dimension is used as an indicator to distinguish between chaotic and stochastic behaviours. The results are mixed with regard to the type of flow behaviour at individual scales and, hence, to the type of scaling behaviour, as some data sets show chaotic behaviour while others show stochastic behaviour. They suggest that characterization (chaotic or stochastic) of river flow should be a necessary first step in any scaling study, as it could provide important information on the appropriate approach for data transformation purposes.     

Damage distribution during the 1994 Northridge, California, earthquake relative to generalized categories of surficial geology

The spatial distributions of severely damaged buildings (red-tagged) and of breaks in the water distribution system following the 1994 Northridge, California, earthquake (M_L = 6·4) are investigated relative to the local characteristics of surficial geology. The pipe breaks are used as an indicator of nonlinear soil response, and the red-tagged buildings as indicator of severe shaking. The surficial geology is described by several generalized categories based on age, textural character and thickness of the near surface layer. Two regions are studied: the San Fernando Valley and Los Angeles-Santa Monica. The analysis shows that there is no simple correlation between damage patterns and surficial geology. Single family wood-frame buildings were damaged less when built on fine silt and clay (0–3 m thick) from the late Holocene.     

Nonlinear determinism in river flow: prediction as a possible indicator

Whether or not river flow exhibits nonlinear determinism remains an unresolved question. While studies on the use of nonlinear deterministic methods for modeling and prediction of river flow series are on the rise and the outcomes are encouraging, suspicions and criticisms of such studies continue to exist as well. An important reason for this situation is that the correlation dimension method, used as a nonlinear determinism identification tool in most of those studies, may possess certain limitations when applied to real river flow series, which are always finite and often short and also contaminated with noise (e.g. measurement error). In view of this, the present study addresses the issue of nonlinear determinism in river flow series using prediction as a possible indicator. This is done by (1) reviewing studies that have employed nonlinear deterministic methods (coupling phase‐space reconstruction and local approximation techniques) for river flow predictions and (2) identifying nonlinear determinism (or linear stochasticity) based on the level of prediction accuracy in general, and on the prediction accuracy against the phase‐space reconstruction parameters in particular (termed as the ‘inverse approach’). The results not only provide possible indications to the presence of nonlinear determinism in the river flow series studied, but also support, both qualitatively and quantitatively, the low correlation dimensions reported for such. Therefore, nonlinear deterministic methods are a viable complement to linear stochastic ones for studying river flow dynamics, if sufficient caution is exercised in their applications and in interpreting the outcomes. Copyright © 2006 John Wiley & Sons, Ltd.     

Nonlinear and multifractal approaches of the geomagnetic field

Recent nonlinear dynamics techniques have been developed to analyse chaotic time series data. We first summarize the procedure which gives an appropriate reconstruction of the unknown dynamics from scalar measurements in a pseudophase space. It permits, firstly, the representation of the trajectories of the dynamical system—they define an attractor when the system is dissipative—by preserving its topological properties. We then present the invariant measures and ergodic quantities such as the multifractal spectrum and Lyapunov exponents which can be estimated on the reconstructed attractor. The multifractal analysis provides us with a characterization of the scaling energy of the process whereas the Lyapunov exponent gives another statistical measure of the stability of the dynamics. The estimation of these quantities was tested on synthetic data. The nonlinear and multifractal analyses were finally applied to the hourly mean values of the magnetic field recorded at the Eskdalemuir (ESK) observatory over 79 years (692,520 data measurements for each component). The estimations of a 5-dimensional pseudo-phase space and a positive Lyapunov exponent confirm the possibility of low-dimensional deterministic chaos in the magnetic field observations at ESK observatory. The correlation between the solar activity (the Wolf number), the unstable nature of the magnetic field, and the singularity spectrum points out the forcing of the solar cycles on the dynamics of the magnetic field at ESK observatory.     

Analysis of seismic dynamical systems

A procedure is presented for the analysis of complex stationary time series for which the Fourier power spectra reveals broadband noise or broadened pulses. We first determine the Hurst exponent from which we may know whether the time series under study is mainly random or if the data points present correlations. If the data are correlated, a chaotic analysis will reveal whether they may be interpreted as a low dimensional nonlinear system (defined by a low correlation dimension and a finite and positive Kolmogorov entropy and largest positive Lyapunov exponent) or as a stochastic process. We have studied three kind of temporal series: inter-event time series of infrasonic pulses recorded at Stromboli volcano, and, S-coda waves and microseisms, that have been recorded at the eastern Pyrenees. Results show that microseisms and Coda waves can be modeled as a low dimensional deterministic system, Correlation dimensions 2.3, 3.2, respectively. At the contrary infrasonic has resulted stochastic. This chaotic character can be attributed to the medium properties. Coda waves with scattering through a fractal distribution of scatters or to multiple reflection inside resonators (for example sedimentary basins) and microseisms as a propagation of wave guide of variable cross section which have the same temporal characteristics as a nonlinear forced oscillator.     

Recommended Sampling Intervals for Arsenic in Private Wells

Geogenic arsenic in drinking water is a worldwide problem. For private well owners, testing (e.g., private or government laboratory) is the main method to determine arsenic concentration. However, the temporal variability of arsenic concentrations is not well characterized and it is not clear how often private wells should be tested. To answer this question, three datasets, two new and one publicly available, with temporal arsenic data were utilized: 6370 private wells from New Jersey tested at least twice since 2002, 2174 wells from the USGS NAWQA database, and 391 private wells sampled 14 years apart from Bangladesh. Two arsenic drinking water standards are used for the analysis: 10 µg/L, the WHO guideline and EPA standard or maximum contaminant level (MCL) and 5 µg/L, the New Jersey MCL. A rate of change was determined for each well and these rates were used to predict the temporal change in arsenic for a range of initial arsenic concentrations below an MCL. For each MCL and initial concentration, the probability of exceeding an MCL over time was predicted. Results show that to limit a person to below a 5% chance of drinking water above an MCL, wells that are ½ an MCL and above should be tested every year and wells below ½ an MCL should be tested every 5 years. These results indicate that one test result below an MCL is inadequate to ensure long-term compliance. Future recommendations should account for temporal variability when creating drinking water standards and guidance for private well owners.     

Interactions and connectivity between runoff generation processes of different spatial scales

Monitoring runoff generation processes in the field is a prerequisite for developing conceptual hydrological models and theories. At the same time, our perception of hydrological processes strongly depends on the spatial and temporal scale of observation. Therefore, the aim of this study is to investigate interactions between runoff generation processes of different spatial scales (plot scale, hillslope scale, and headwater scale). Different runoff generation processes of three hillslopes with similar topography, geology and soil properties, but differences in vegetation cover (grassland, coniferous forest, and mixed forest) within a small v‐shaped headwater were measured: water table dynamics in wells with high spatial and temporal resolution, subsurface flow (SSF) of three 10 m wide trenches at the bottom of the hillslopes subdivided into two trench sections each, overland flow at the plot scale, and catchment runoff. Bachmair et al. ( 2012 ) found a high spatial variability of water table dynamics at the plot scale. In this study, we investigate the representativity of SSF observations at the plot scale versus the hillslope scale and vice versa, and the linkage between hillslope dynamics (SSF and overland flow) and streamflow. Distinct differences in total SSF within each 10 m wide trench confirm the high spatial variability of the water table dynamics. The representativity of plot scale observations for hillslope scale SSF strongly depends on whether or not wells capture spatially variable flowpaths. At the grassland hillslope, subsurface flowpaths are not captured by our relatively densely spaced wells (3 m), despite a similar trench flow response to the coniferous forest hillslope. Regarding the linkage between hillslope dynamics and catchment runoff, we found an intermediate to high correlation between streamflow and hillslope hydrological dynamics (trench flow and overland flow), which highlights the importance of hillslope processes in this small watershed. Although the total contribution of SSF to total event catchment runoff is rather small, the contribution during peak flow is moderate to substantial. Additionally, there is process synchronicity between spatially discontiguous measurement points across scales, potentially indicating subsurface flowpath connectivity. Our findings stress the need for (i) a combination of observations at different spatial scales, and (ii) a consideration of the high spatial variability of SSF at the plot and hillslope scale when designing monitoring networks and assessing hydrological connectivity. Copyright © 2013 John Wiley & Sons, Ltd.