Application of artificial neural networks to modeling and prediction of ambient ozone concentrations

The deterministic modeling of ambient O3 concentrations is difficult because of the complexity of the atmospheric system in terms of the number of chemical species; the availability of accurate, time-resolved emissions data; and the required rate constants. However, other complex systems have been successfully approximated using artificial neural networks (ANNs). In this paper, ANNs are used to model and predict ambient O3 concentrations based on a limited number of measured hydrocarbon species, NOx compounds, temperature, and radiant energy. In order to examine the utility of these approaches, data from the Coastal Oxidant Assessment for Southeast Texas (COAST) program in Houston, TX, have been used. In this study, 53 hydrocarbon compounds, along with O3, nitrogen oxides, and meteorological data were continuously measured during summer 1993. Steady-state ANN models were developed to examine the ability of these models to predict current O3 concentrations from measured VOC and NOx concentrations. To predict the future concentrations of O3, dynamic models were also explored and were used for extraction of chemical information such as reactivity estimations for the VOC species. The steady-state model produced an approximation of O3 data and demonstrated the functional relationship between O3 and VOC-NOx concentrations. The dynamic models were able to the adequately predict the O3 concentration and behavior of VOC-NOx-O3 system a number of hourly intervals into the future. For 3 hr into the future, O3 concentration could be predicted with a root-mean squared error (RMSE) of 8.21 ppb. Extending the models further in time led to an RMSE of 11.46 ppb for 5-hr-ahead values. This prediction capability could be useful in determining when control actions are needed to maintain measured concentrations within acceptable value ranges.     

Assessment of heavy metal contamination in Hindon River sediments: a chemometric and geochemical approach

The aim of this study was to assess the level of heavy metals (Cd, Cr, Cu, Fe, Mn, Ni, Pb, and Zn) in the surface sediments of the Hindon River, India that receives both treated and untreated municipal and industrial discharges generated in and around Ghaziabad, India. Mean metals concentrations (mg kg⁻¹) were in the range of; Cu: 21.70-280.33, Cd: 0.29-6.29, Fe: 4151.75-17318.75, Zn: 22.22.50-288.29, Ni: 13.90-57.66, Mn: 49.55-516.97, Cr: 17.48-33.70 and Pb: 27.56-313.57 respectively. Chemometric analysis was applied to identify contribution sources by heavy metals while geochemical approaches (enrichment factor and geo-accumulation index) were exploited for the assessment of the enrichment and contamination level of heavy metals in the river sediments. Chemometric analysis suggested anthropic origin of Cu, Cd, Pb, Zn, and Ni while Fe showed lithogenic origin. Mn and Cr was associated and controlled by mixed origin. Geochemical approach confirms the anthropogenic influence of heavy metal pollution in the river sediments. The study suggests that a complementary approach that integrates chemometric analysis, sediment quality criteria, and geochemical investigation should be considered in order to provide a more accurate appraisal of the heavy metal pollution in river sediments. Consequently, it may serve to undertake and design effective strategies and remedial measures to prevent further deterioration of the river ecosystem in future.     

Microbial indicators of heavy metal contamination in urban and rural soils

Urban soils and especially their microbiology have been a neglected area of study. In this paper, we report on microbial properties of urban soils compared to rural soils of similar lithogenic origin in the vicinity of Aberdeen city. Significant differences in basal respiration rates, microbial biomass and ecophysiological parameters were found in urban soils compared to rural soils. Analysis of community level physiological profiles (CLPP) of micro-organisms showed they consumed C sources faster in urban soils to maintain the same level activity as those in rural soils. Cu, Pb, Zn and Ni were the principal elements that had accumulated in urban soils compared with their rural counterparts with Pb being the most significant metal to distinguish urban soils from rural soils. Sequential extraction showed the final residue after extraction was normally the highest proportion except for Pb, for which the hydroxylamine-hydrochloride extractable Pb was the largest part. Acetic acid extractable fraction of Cd, Cu, Ni, Pb and Zn were higher in urban soils and aqua regia extractable fraction were lower suggesting an elevated availability of heavy metals in urban soils. Correlation analyses between different microbial indicators (basal respiration, biomass-C, and sole C source tests) and heavy metal fractions indicated that basal respiration was negatively correlated with soil Cd, Cu, Ni and Zn inputs while soil microbial biomass was only significantly correlated with Pb. However, both exchangeable and iron- and manganese-bound Ni fractions were mostly responsible for shift of the soil microbial community level physiological profiles (sole C source tests). These data suggest soil microbial indicators can be useful indicators of pollutant heavy metal stress on the health of urban soils.     

Use of a general toxicity test to predict heavy metal concentrations in residential soils

Significant clusters of developmental delay and mental retardation (DD/MR) were identified in children born in South Carolina. Although it is difficult to identify one factor that causes DD/MR, environmental insult including exposure of pregnant women to heavy metals can induce DD/MR in their children. Because it is expensive to measure the concentrations of individual metals in large numbers of environmental samples, the general Microtox toxicity test was used to identify highly toxic soil samples. Approximately 100 soil samples were collected from residential areas and analyzed to determine an effective concentration (EC(50)) of soil required to inhibit 50% light emission of the luminescent bacterial test organism (Vibrio fischeri). The EC(50) values were then transformed to relative toxicity units (RTU). A subset of 56 high and low toxicity soil samples was then analyzed by inductively coupled plasma-atomic emission spectrometry (EPA method 6010) for arsenic, lead, and chromium, which are known neurotoxins. The highest measured arsenic concentration was 30 times higher than the South Carolina residential soil limit. Significant correlations were found between the RTU and soil arsenic and chromium concentrations. Microtox also identified some low arsenic and chromium samples as toxic, presumably because additional unidentified toxicants were present in the soil. In general, however, the Microtox test was effective in identifying soils with elevated concentrations of arsenic and chromium, even in residential neighborhoods where limited soil toxicity was expected.     

Response of the lichen Cladonia rei Schaer. to strong heavy metal contamination of the substrate

The phenomenon of mass occurrence of the lichen Cladonia rei in extremely contaminated post-smelting slag dumps was studied in relation to metal accumulation capacity of this lichen. The research was aimed to evaluate the relationships between element contents in the thalli and in the corresponding substrate. The study was conducted in terms of a wide spectrum of Zn, Cd, Pb and As contents. The concentrations of these elements in the lichen thalli and substrate samples were measured. Various regression models were considered to find the best fitted one that greatly reflects the dependencies. Various Cladonia species and the hyperaccumulator Diploschistes muscorum were also included in the study for comparison purposes. Specific non-linear regression models described by a power function reflected relationships between Zn and Cd contents in C. rei thalli and in the host substrate in the most reliable way. The relationship for As was also noted, but none significant model was found. Contrarily, Pb concentrations in the thalli varied independently of the metal levels in the substrate. Nevertheless, the concentrations of all measured heavy metals in C. rei thalli are relatively low considering the frequently enormous substrate contamination. Different Cladonia species demonstrated a generally similar accumulation capacity and could be considered as weak accumulators. The restrained accumulation pattern may be one of the important attributes of C. rei which facilitates its colonisation of extremely contaminated dumps. This finding highlights ecological importance of this species as stable and resistant pioneer in such affected sites.     

Trace organic and heavy metal pollutants in the Mississippi River

A study was conducted to characterize and measure organic and heavy metal pollutants in the Mississippi River. Water samples were collected along the entire length of the river, and were screened for semivolatile organics by capillary GC and for heavy metals by atomic absorption spectrophotometry. Four water samples were further examined for semivolatile organics by capillary GC/MS. Eight heavy metals and more than sixty distinct organic chemicals were identified including alkylbenzenes, various halogenated organics, five herbicides or derivatives, plasticizers, polynuclear aromatic hydrocarbons (PAHs), saturated hydrocarbons, and three miscellaneous organics. All organic compounds were detected at the parts-per-trillion (pptr) level. In spite of the limited nature of the sampling effort, the large number of data derived from this study suggests the need for a more rigorous monitoring of the river for a wide spectrum of chemical pollutants.     

Reflectance properties and physiological responses of Salicornia virginica to heavy metal and petroleum contamination

Wetland ecosystems of California are located in highly populated areas and subject to high levels of contamination. Monitoring of wetlands to assess degrees of pollution damage requires periodic retrieval of information over large areas, which can be effectively accomplished by rapidly evolving remote sensing technologies. The biophysical principles of remote sensing of vegetation under stress need to be understood in order to correctly interpret the information obtained at the scale of canopies. To determine the potential to remotely characterize and monitor pollution, plants of Salicornia virginica, a major component of wetland communities in California, were treated with two metals and two crude oil types to study their sensitivity to pollutants and how this impacted their reflectance characteristics. Several growth and physiological parameters, as well as shoot reflectance were measured and correlated with symptoms and contamination levels. Significant differences between treatments were found in at least some of the measured parameters in all pollutants. Reflectance was sensitive to early stress levels only for cadmium and the lightweight petroleum. Pollutants that differ in their way of action also had different plant reflectance signatures. The high degree of correlation between reflectance features and stress indicators highlights the potential of using remote sensing to assess the type and degree of pollution damage.     

基于可视角度下重金属污染物在介质中的迁移规律

介质中污染物迁移的可视化研究有助于厘清污染物及介质间微观作用机制。依托透明土的可视化理念,以Cu(Ⅱ)、Cr(Ⅵ)污染土为研究对象,分析电场及温度场驱动作用下重金属污染物在透明土及实际土体中的迁移分布规律,探讨可视化研究的可靠性并建立重金属污染物的界面迁移机制。结果表明,透明土可用于研究多孔介质中重金属污染物的迁移扩散性,电场及温度场作用下同点位污染物质量分数的波动率均在10%以内;电场作用下污染物的迁移由离子赋存状态所带电荷决定,Cu(Ⅱ)、Cr(Ⅵ)分别以阳离子、酸根阴离子状态存在,迁移方向相反。Cu(Ⅱ)因铜胶体絮凝物生成易产生聚集效应,迁移受阻;温度场作用下污染物的迁移取决于温度梯度,因土颗粒高低温处结合水厚度的差异及水分子间的吸力会促使孔隙流体迁移以达平衡,重金属污染物在对流和分子扩散的协调作用下,均表现为向低温处迁移趋势。在同外力驱动作用下,重金属污染物间的迁移存在差异性,Cr(Ⅵ)的迁移性均强于Cu(Ⅱ)的,化学沉淀物质的产生及重金属阳离子静电吸附作用是影响迁移效率的内在关键。本研究结果可为土壤污染预测及修复研究提供参考。     

Molecular diversity of arbuscular mycorrhizal fungi in relation to soil chemical properties and heavy metal contamination

Abundance and diversity of arbuscular mycorrhizal fungi (AMF) associated with dominant plant species were studied along a transect from highly lead (Pb) and zinc (Zn) polluted to non-polluted soil at the Anguran open pit mine in Iran. Using an established primer set for AMF in the internal transcribed spacer (ITS) region of rDNA, nine different AMF sequence types were distinguished after phylogenetic analyses, showing remarkable differences in their distribution patterns along the transect. With decreasing Pb and Zn concentration, the number of AMF sequence types increased, however one sequence type was only found in the highly contaminated area. Multivariate statistical analysis revealed that further factors than HM soil concentration affect the AMF community at contaminated sites. Specifically, the soils’ calcium carbonate equivalent and available P proved to be of importance, which illustrates that field studies on AMF distribution should also consider important environmental factors and their possible interactions.     

Application of principal component analysis for the estimation of source of heavy metal contamination in surface sediments from the Rybnik Reservoir

The concentrations of metals, loss of ignition and nutrient (N, P) were determined in the bottom sediments of the Rybnik Reservoir (southern Poland). The mean concentrations of the metals in the bottom sediments were: Cd 25.8 microgram/g, Cu 451.7 microgram/g, Zn 1583.4 microgram/g, Ni 71.1 microgram/g, Pb 118.6 microgram/g, Cr 129.8 microgram/g, Fe 38782 microgram/g and Mn 2018.7 microgram/g. The bottom sediments are very heavily loaded with zinc, manganese, copper, nickel, phosphorus and lead (percentage enrichment factor), and cadmium, phosphorus and zinc (index of geoaccumulation). The increase of cadmium, lead, nickel and zinc concentrations was connected with the inflow of the contaminated water of the river Ruda and long-range transport. The contamination of the reservoir with copper and manganese resulted mainly from atmospheric precipitation. The variability of the bottom sediment loading with metals during the investigations was affected in the first place by changes in the concentration of iron, but also those elements whose concentrations in the bottom sediment were elevated compared to the concentrations in shale--cadmium, nickel and lead.     

玉米叶片光谱弱差信息的重金属污染定性分析

不同重金属污染下的玉米叶片光谱响应存在细微差异。设计不同浓度的铜（Cu）和铅（Pb）污染实验,并测量不同浓度铜离子和铅离子胁迫下玉米叶片的光谱反射率、Cu含量、Pb含量及对应叶绿素含量。针对光谱弱差信息甄别的难点,结合光谱微分处理、光谱角、正切函数增强和波谱分段检测等构建一种甄别玉米叶片Cu、Pb胁迫的分段微分光谱角正切（segmentation derivative spectral angle tangent,SDSAT）模型。甄别玉米叶片不同生长程度的玉米叶片在Cu、Pb胁迫下的光谱响应差异,并通过区分玉米冠层Cu、Pb胁迫光谱的差异以验证该模型有效可行。实验结果表明,SDSAT模型甄别Cu、Pb胁迫玉米光谱主要差异在“紫光”“红边”和“近峰B”波段区间,这些差异也是鉴别Cu、Pb污染的定性分析标志,并可以根据模型测度结果分析玉米的污染程度。     

Modelling of dissolved oxygen content using artificial neural networks: Danube River, North Serbia, case study

The aims of this study are to create an artificial neural network (ANN) model using non-specific water quality parameters and to examine the accuracy of three different ANN architectures: General Regression Neural Network (GRNN), Backpropagation Neural Network (BPNN) and Recurrent Neural Network (RNN), for prediction of dissolved oxygen (DO) concentration in the Danube River. The neural network model has been developed using measured data collected from the Bezdan monitoring station on the Danube River. The input variables used for the ANN model are water flow, temperature, pH and electrical conductivity. The model was trained and validated using available data from 2004 to 2008 and tested using the data from 2009. The order of performance for the created architectures based on their comparison with the test data is RNN > GRNN > BPNN. The ANN results are compared with multiple linear regression (MLR) model using multiple statistical indicators. The comparison of the RNN model with the MLR model indicates that the RNN model performs much better, since all predictions of the RNN model for the test data were within the error of less than ±10 %. In case of the MLR, only 55 % of predictions were within the error of less than ±10 %. The developed RNN model can be used as a tool for the prediction of DO in river waters.     

Periphyton as monitors for heavy metal pollution in the Calcasieu River estuary

The levels of copper, lead, chromium, zinc, cadmium, arsenic and silver were determined in periphyton specimens obtained with a diatometer collector. Stations selected were along three important bayous of the Calcasieu River system. Distributions of some metals in the organisms were similar to those found in sediment from the same locations, while other metals appeared to be similar to water concentrations. Concentration ratios of periphyton over sediment greatly exceeded one for the metals chromium, zinc, cadmium, arsenic and silver. The concentrations of heavy metals in the periphyton appeared to yield more information about pollutants than either water or sediment samples collected at the periphyton stations.     

Climate change impacts on the leaching of a heavy metal contamination in a small lowland catchment

The Keersop catchment (43km(2)) in the south of The Netherlands has been contaminated by the emissions of four zinc ore smelters. The objective of this study was to assess the effects of future projected climate change on the hydrology and the leaching of heavy metals (i.e. Cd and Zn) in the catchment. The numerical, quasi-2D, unsaturated zone Soil Water Atmosphere Plant model was used with 100-year simulated daily time series of precipitation and potential evapotranspiration. The time series are representative of stationary climates for the periods 1961-1990 ("baseline") and 2071-2100 ("future"). The time series of future climate were obtained by downscaling the results of eight regional climate model (RCM) experiments, driven by the SRES A2 emissions scenario, using change factors for a series of climate statistics and applying them to stochastic weather generator models. The time series are characterized by increased precipitation in winter, less precipitation in summer, and higher air temperatures (between 2°C and 5°C) throughout the year. Future climate scenarios project higher evapotranspiration rates, more irrigation, less drainage, lower discharge rates and lower groundwater levels, due to increased evapotranspiration and a slowing down of the groundwater system. As a result, lower concentrations of Cd and Zn in surface water are projected. The reduced leaching of heavy metals, due to drying of the catchment, showed a positive impact on a limited aspect of surface water quality.     

The effects of heavy metal contamination on the soil arthropod community of a shooting range

Soils in clay pigeon shooting ranges can be seriously contaminated by heavy metals. The pellets contained in ammunition are composed of Pb, Sb, Ni, Zn, Mn and Cu. The total concentrations of these metals in soils, and the effects of their increasing levels on the arthropod community were investigated at seven sampling sites in a clay pigeon shooting range and compared with two controls. Research revealed that the spatial distribution of Pb and Sb contamination in the shot-fall area was strongly correlated with the flight path of the pellets. Ordination obtained through Redundance Analysis showed that Collembola, Protura and Diplura were positively correlated with major detected contaminants (Pb, Sb), while Symphyla showed a negative correlation with these pollutants. Determination of the soluble lead fraction in soil, and of its bioaccumulation in the saprophagous Armadillidium sordidum (Isopoda) and the predator Ocypus olens (Coleoptera), showed that a significant portion of metallic Pb from spent pellets is bioavailable in the soil and can be bioaccumulated by edaphic organisms, entering the soil trophic network, but without biomagnification.     

Using stable lead isotopes to trace heavy metal contamination sources in sediments of Xiangjiang and Lishui Rivers in China

Lead isotopes and heavy metal concentrations were measured in two sediment cores sampled in estuaries of Xiangjiang and Lishui Rivers in Hunan province, China. The presence of anthropogenic contribution was observed in both sediments, especially in Xiangjiang sediment. In the Xiangjiang sediment, the lower ²⁰⁶Pb/²⁰⁷Pb and higher ²⁰⁸Pb/²⁰⁶Pb ratio, than natural Pb isotope signature (1.198 and 2.075 for ²⁰⁶Pb/²⁰⁷Pb and ²⁰⁸Pb/²⁰⁶Pb, respectively), indicated a significant input of non-indigenous Pb with low ²⁰⁶Pb/²⁰⁷Pb and high ²⁰⁸Pb/²⁰⁶Pb. The corresponding concentrations of heavy metals (As, Cd, Zn, Mn and Pb) were much higher than natural values, suggesting the contaminations of heavy metals from extensive ore-mining activities in the region.     

人工神经网络和专家系统在污水生物处理系统中的应用

对近年来国内外污水生物处理系统中人工神经网络和专家系统的应用进行了简要的回顾。分析了废水生物处理工艺难于控制的原因及人工神经网络和专家系统的结构和特点。结果表明．国外智能控制发展迅速,并且应用领域遍及污水生物处理的各个方面,国内尚处于起步阶段。简要探讨了废水生物处理智能控制今后应深入研究的问题及方向。     

Multivariate statistical study of heavy metal enrichment in sediments of the Pearl River Estuary

The concentrations and chemical partitioning of heavy metals in the sediment cores of the Pearl River Estuary were studied. Based on Pearson correlation coefficients and principal component analysis results, Al was selected as the concentration normalizer for Pb, while Fe was used as the normalizing element for Co, Cu, Ni and Zn. In each profile, sections with metal concentrations exceeding the upper 95% prediction interval of the linear regression model were regarded as metal enrichment layers. The heavy metal accumulation mainly occurred at sites in the western shallow water areas and east channel, which reflected the hydraulic conditions and influence from riparian anthropogenic activities. Heavy metals in the enrichment sections were evaluated by a sequential extraction method for possible chemical forms in sediments. Since the residual, Fe/Mn oxides and organic/sulfide fractions were dominant geochemical phases in the enriched sections, the bioavailability of heavy metals in sediments was generally low. The 206Pb/207Pb ratios in the metal-enriched sediment sections also revealed the influence of anthropogenic sources. The spatial distribution of cumulative heavy metals in the sediments suggested that the Zn and Cu mainly originated from point sources, while the Pb probably came from non-point sources in the estuary.     

某矿区土壤和地下水重金属污染调查与评价

为了解湘南某矿区土壤和地下水重金属污染状况,对该矿区东河流域附近重金属污染源进行了调查,同时,对地下水和土壤样品进行了采样分析,结果表明：（1）该矿区东河流域附近的主要污染源有18个,其中有色金属选厂、尾矿库、采矿场和冶炼厂是排放重金属较多的污染源;（2）20个采样点中土壤重金属Pb、Cd、Zn、As和Hg大部分超过国家土壤环境质量标准（GB15618-1995）,综合污染指数P综〉1,该矿区主要的重金属污染元素为Cd、As和Hg,且土壤中Cd、Zn和As的含量两两之间存在着极显著的正线性相关关系;（3）重金属元素在土壤中的纵向迁移不明显,该矿区附近20个采样点的地下水并未受到污染,综合污染指数P综〈1。20个采样点地下水Pb、Cd、Zn、As、Hg浓度均能达到地下水质量标准（GB／T14848．9）中的Ⅲ类标准。     

Prediction of ambient PM10 and toxic metals using artificial neural networks

In this study, an artificial neural network is employed to predict the concentration of ambient respirable particulate matter (PM10) and toxic metals observed in the city of Jaipur, India. A feed-forward network with a back-propagation learning algorithm is used to train the neural network the behavior of the data patterns. The meteorological variables of wind speed, wind direction, relative humidity, temperature, and time are taken as input to the network. The results indicate that the network is able to predict concentrations of PM10 and toxic metals quite accurately.