Application of the Morris algorithm for sensitivity analysis of the REALM model for the Goulburn irrigation system

The REALM modelling shell is widely used in Australia as a water allocation modelling tool. It has been used to develop the Goulburn System Model (GSM) of the Goulburn, Broken, Loddon and Campaspe Rivers in northeastern Victoria. REALM represents the river and irrigation system as a network of storages and carriers. The model has been optimised to best represent the water harvesting and allocation for use by water management authorities. The model is analysed to assess the sensitivity of a subset of the model outputs, to a subset of the system parameters. The New Morris algorithm uses sampling paths generated in the space of the parameters, to generate points at which the model is run (to generate the model outputs). These model runs are then used to estimate the first and second-order effects of the parameters on the outputs. The results illustrate the mild linkage of the Goulburn and Broken systems, and the Broken system also shows differences between minimum and average outflows. The Goulburn is more sensitive to some of the numerical convergence parameters used in the allocation software, while the Broken is less sensitive to these factors. The numerical convergence factors also lead to important second-order effects.     

Selection of appropriate sampling stations in a lake through mapping

Much valuable information is obtained from water quality measurements and monitoring of lakes around the world. A powerful tool is the use of mapping techniques, as it offers potential use in water quality research. Both remote sensing techniques and traditional water quality monitoring are required to collect data at sampling stations. This study suggests another approach to determine the most appropriate distribution of sampling stations in water reservoirs that will be mapped for water quality parameters. Tests were conducted for the proposed approach for Secchi disc depth (SDD), chlorophyll-a, turbidity and suspended solids parameters in Lake Beysehir, Turkey. Results of analysis are available for a total of 30 sampling stations in August 2006. Ten sampling stations were used to model Lake Beysehir while the others were used for validation of the model. Sampling stations that offered the best representation of the lake for each parameter were determined. Then, the best representative sampling stations for all parameters in the study were determined. Moreover, in order to confirm the accuracy of these re-determined sampling stations, modelling was performed on the results of the analysis of June 2006, and it was found that the values obtained from the re-determined sampling stations were acceptable.     

Calculating of river water quality sampling frequency by the analytic hierarchy process (AHP)

River water quality sampling frequency is an important aspect of the river water quality monitoring network. A suitable sampling frequency for each station as well as for the whole network will provide a measure of the real water quality status for the water quality managers as well as the decision makers. The analytic hierarchy process (AHP) is an effective method for decision analysis and calculation of weighting factors based on multiple criteria to solve complicated problems. This study introduces a new procedure to design river water quality sampling frequency by applying the AHP. We introduce and combine weighting factors of variables with the relative weights of stations to select the sampling frequency for each station, monthly and yearly. The new procedure was applied for Jingmei and Xindian rivers, Taipei, Taiwan. The results showed that sampling frequency should be increased at high weighted stations while decreased at low weighted stations. In addition, a detailed monitoring plan for each station and each month could be scheduled from the output results. Finally, the study showed that the AHP is a suitable method to design a system for sampling frequency as it could combine multiple weights and multiple levels for stations and variables to calculate a final weight for stations, variables, and months.     

Development of an Overall Index of Pollution for Surface Water Based on a General Classification Scheme in Indian Context

Various National and International Agencies involved in water quality assessment and pollution control have defined water quality criteria for different uses of water considering different indicator parameters. Classification schemes for water quality criteria/standards developed by these agencies differ inaddition to terminologies used such as Action level, Guide level etc. in defining the concentration values in these classes. In the present article a general classification scheme viz. Excellent,Acceptable, Slightly Polluted, Polluted and Heavily Polluted water is proposed for surface water quality assessment. The concentration ranges in these classes are defined in Indianscenario considering Indian Standards and CPCB criteria. Standardsby the European Community (EC), WHO etc. and the reported factsabout the pollution effects of important water quality indicatorparameters on the surrounding were also taken into account. Themathematical equations to transform the actual concentrationvalues into pollution indices are formulated and correspondingvalue function curves are plotted. Based on the individual indexvalues, an `Overall Index of Pollution' (OIP) is estimated. Theapplication of OIP is demonstrated at a few sampling stations onriver Yamuna based on observed water quality data. The general classification scheme along with concentrationranges defined in these classes will be of immense use fordetermining the surface water quality status with reference tospecific individual parameter, and the OIP for assessing theoverall water quality status in Indian context.     

Prediction of daily ground-level ozone concentration maxima over New Delhi

The pollution levels in New Delhi from industrial, residential, and transportation sources are continuously growing. As one of the major pollutants, ground-level ozone is responsible for various adverse effects on both humans and foliage. The present study aims to predict daily ground-level ozone concentration maxima over a site situated in New Delhi through neural networks (NN) and multiple-regression (MR) analysis. Although these methodologies are case and site specific, they are being developed and used widely. Therefore, to test these methodologies for New Delhi where no such study is available for ground-level ozone, six models have been developed based on NNs and MR using the same input data set. The changes in the performance capability of the two methods are sensitive to the selection of input parameters. The results are encouraging, and remarkable improvements in the performance of the models have been observed.     

Correlated Parameters Uncertainty Propagation in a Rainfall-Runoff Model,Considering 2-Copula; Case Study: Karoon III River Basin

Hydrological models are widely used to investigate practical issues of water resources. Parametric uncertainty is considered as one of the most important sources of uncertainty in environmental researches. Generally, it is assumed that the parameters are independent mutually, but correlation within the parameter space is an important factor having the potential to cause uncertainty. The objective and innovation of this study was to address the effects of parameters correlation on a continuous hydrological model uncertainty. HEC-HMS with soil moisture accounting (SMA) infiltration method was used to model daily flows and simulate certainty bounds for Karoon III basin, southwest of IRAN, in two scenarios, independent and correlated parameters using 2-copula. The parameters were represented by probability distributions, and the effect on prediction error were evaluated using Latin hypercube sampling (LHS) on Monte Carlo simulation (MCS). Saturated hydraulic conductivity (K), Clark storage-coefficient (R), and time of concentration (tc) were chosen for investigation, based on observed sensitivity analysis of simulated peak over threshold (POT). One hundred runs were randomly generated from 100 parameter sets captured from LHS of parameters distributions in each sub-basin. Using generated parameter sets, 100 continuous hydrographs were simulated and values of certainty bounds calculated. Results showed that when 2-copula correlated R and tc, with 0.656 Kendall’s Tau and 0.818 Spearman’s Rho coefficients, were propagated, decreasing of outputs’ sharpness was more than when considering K and R (K-R), with 0.166 and 0.262; therefore, incorporation of correlations in the MCS is important, especially when the correlation coefficients exceed 0.65. The model was evaluated at the outlet of the basin using daily stream flow data. Model reliability was better for above-normal flows than normal and below-normal. Reliability increases of simulated flow when considering correlated R-tc was more than K-R because of the correlation values. Incorporation of copula for K-tc not only did not improve the model reliability but also decreased it. Results showed improvement of model reliability, by decreasing predicted error of hydrologic modeling, when dealing with correlated parameters in the system.     

Description and evaluation of a sampling system for monitoring hydrocarbons in sediments

A composite random sampling design was used to estimate the concentrations of hydrocarbons in sediments from two near-shore areas of Scotland (Firth of Clyde and Firth of Forth). The aim of this work was to estimate a mean value for each parameter in these areas, and to determine whether this can be done with more thorough coverage (better representation), better precision and less variance at lower analytical cost through a composite random sampling scheme rather than a simple random sampling scheme, and thereby contribute to the re-design of the UK National Marine Monitoring Programme (NMMP), re-named the UK Clean Seas Environmental Monitoring Programme (CSEMP) in 2006. Samples were collected using a simple random sampling design during 2005. All sediment samples were analysed for their particle size distribution and total organic carbon (TOC). All sediments were analysed for polycyclic aromatic hydrocarbons (PAHs) and n-alkanes. The concentrations of PAHs and n-alkanes in the study areas are described, and sources of PAHs were investigated through the PAH distributions and n-alkane profiles. Individual sediment samples from each area were combined to give a series of composite sub-samples, each comprised of 5 individual sediment samples. These composite samples were re-analysed for the same parameters as the individual samples. Mean total PAH (2- to 6-ring parent and branched) concentrations, based on the individual original sediment samples collected through simple random sampling, were 1858 microg kg(-1) dry weight (SE = 196 microg kg(-1) dry weight, n = 25) and 532.4 microg kg(-1) dry weight (SE = 59 microg kg(-1) dry weight, n = 25) in the Clyde and Forth, respectively. Mean total PAH concentrations of the composite samples were 1745 microg kg(-1) dry weight (SE = 121.0 microg kg(-1) dry weight, n = 5) in the Clyde and 511.6 microg kg(-1) dry weight (SE = 37.4 microg kg(-1) dry weight, n = 5) in the Forth. No significant differences were found between the mean PAH concentrations from the two sampling designs. This study demonstrated that the composite random sampling design gave a mean value with less variance than the simple random sampling design, at significantly reduced analytical effort (and cost).     

Modeling of SO<Subscript>2</Subscript> distribution in Istanbul using artificial neural networks

Urban air pollution is a growing problem in developing countries. Some compounds especially sulphur dioxide (SO₂) is considered as typical indicators of the urban air quality. Air pollution modeling and prediction have great importance in preventing the occurrence of air pollution episodes and provide sufficient time to take the necessary precautions. Recently, various stochastic image-processing algorithms such as Artificial Neural Network (ANN) are applied to environmental engineering. ANN structure employs input, hidden and output layers. Due to the complexity of the problem, as the number of input–output parameters differs, ANN model settings such as the number of neurons of these layers changes. The ability of ANN models to learn, particularly capability of handling large amounts (or sets) of data simultaneously as well as their fast response time, are invariably the characteristics desired for predictive and forecasting purposes. In this paper, ANN models have been used to predict air pollutant parameter in meteorological considerations. We have especially focused on modeling of SO₂ distribution and predicting its future concentration in Istanbul, Turkey. We have obtained data sets including meteorological variables and SO₂ concentrations from Istanbul-Florya meteorological station and Istanbul-Yenibosna air pollution station. We have preferred three-layer perceptron type of ANN which consists of 10, 22 and 1 neurons for input, hidden and output layers, respectively. All considered parameters are measured as daily mean. The input parameters are: SO₂ concentration, pressure, temperature, humidity, wind direction, wind speed, strength of sunshine, sunshine, cloudy, rainfall and output parameter is the future prediction of SO₂. To evaluate the performance of ANN model, our results are compared to classical nonlinear regression methods. The over all system finds an optimum correlation between input–output variables. Here, the correlation parameter, r is 0.999 and 0.528 for training and test data. Thus in our model, the trend of SO₂ is well estimated and seasonal effects are well represented. As a result, we conclude that ANN is one of the compromising methods in estimation of environmental complex air pollution problems.     

Monte Carlo assessment of uncertainty in the simulated hydrological response to land use change

We quantified the uncertainty in hydrological response for a set of land use change scenarios by varying plant parameters within realistic uncertainty bounds in a Monte Carlo analysis. The results show that simulated hydrological fluxes significantly change after the introduction of outwintering suckler cow management, despite the presence of a significant amount of output uncertainty due to uncertainty in the plant parameterisation. The key to a proper uncertainty assessment was to consider the uncertainty in the difference between the scenarios instead of the absolute uncertainty of each single scenario. Additionally, a sensitivity analysis showed that changing soil properties in response to land use change does not result in significantly different results in the scenario analysis.     

Characterization of the Leachate in an Urban Landfill by Physicochemical Analysis and Solid Phase Microextraction-GC/MS

The aim of this study is to evaluate extensively the characterization and identification of major pollutant parameters by paying attention to the organic chemical pollution for unregulated dumping site leachate in Eskişehir/Turkey. The study that is first and only one research has been very important data related with before new sanitary landfill site in Eskişehir city. For this purpose, in this study leachate samples were collected in-situ at monthly interval for a period of 8 months. Firstly, thirty three physicochemical parameters were monitored. Secondly, SPME technique was used for identification of organic pollutants. Meteorological data were also recorded for the same sampling period to correlate meteorological data and physicochemical parameters. Mean values are used in the correlation analysis. Correlation is shown only for the relationship between air temperature and NO₃ ⁻. No correlation has been found between rain and leachate quality parameters since the amount of rain was very low during the sampling period. However, analysis results were generally decreased in winter season when each parameter and each sampling point are examined separately. According to correlation between every parameter, especially solid content and dissolved oxygen concentration of leachate is affecting to other parameters. Also, sodium and potassium are changing proportionally with same parameters (suspended solids, fixed solids, dissolved oxygen) and high correlation between chloride and heavy metal concentration is showing. The results were statistically evaluated by use of SPSS 10.0 program. Second part of the study, the leachate was extracted by Solid Phase Microextraction (SPME) technique and then analyzed. Of the methodologies tested in this study, the best one selected was based on 100 μ m polydimethylsiloxane coated fiber (PDMS), headspace with heating (Δ HS) sampling mode and an extraction time of 15 min. at a temperature of 50 ^o C. Thirty three organic compounds in leachate were identified by GC/MS.     

Sensitivity analysis of TOPSIS method in water quality assessment: I. Sensitivity to the parameter weights

Sensitivity analysis is becoming increasingly widespread in many fields of engineering and sciences and has become a necessary step to verify the feasibility and reliability of a model or a method. The sensitivity of the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method in water quality assessment mainly includes sensitivity to the parameter weights and sensitivity to the index input data. In the present study, the sensitivity of TOPSIS to the parameter weights was discussed in detail. The present study assumed the original parameter weights to be equal to each other, and then each weight was changed separately to see how the assessment results would be affected. Fourteen schemes were designed to investigate the sensitivity to the variation of each weight. The variation ranges that keep the assessment results unchangeable were also derived theoretically. The results show that the final assessment results will change when the weights increase or decrease by ±20 to ±50 %. The feedback of different samples to the variation of a given weight is different, and the feedback of a given sample to the variation of different weights is also different. The final assessment results can keep relatively stable when a given weight is disturbed as long as the initial variation ratios meet one of the eight derived requirements.     

Use of neural networks for monitoring surface water quality changes in a neotropical urban stream

This paper reports the using of neural networks for water quality analysis in a tropical urban stream before (2002) and after sewerage building and the completion of point-source control-based sanitation program (2003). Mathematical modeling divided water quality data in two categories: (a) input of some in situ water quality variables (temperature, pH, O2 concentration, O2 saturation and electrical conductivity) and (b) water chemical composition (N-NO2(-); N-NO3(-); N-NH4(+) Total-N; P-PO4(3-); K+; Ca2+; Mg+2; Cu2+; Zn2+ and Fe+3) as the output from tested models. Stream water data come from fortnightly sampling in five points along the Ipanema stream (Southeast Brazil, Minas Gerais state) plus two points downstream and upstream Ipanema discharge into Doce River. Once the best models are consistent with variables behavior we suggest that neural networking shows potential as a methodology to enhance guidelines for urban streams restoration, conservation and management.     

Ecotoxicity assessment of aged petroleum sludge using a suite of effects-based end points in earthworm Eisenia fetida

Laboratory investigations were conducted to test the toxicity of aged petroleum sludge collected from Shengli Oil Field, the second largest oilfield in China, to earthworm Eisenia fetida. Various end points were measured in the earthworms, including mortality, growth, cocoon output, juvenile production, and avoidance behavioral response, to determine their comparative sensitivity for assessing harmful effects of petroleum-hydrocarbon-contaminated soil. The results showed that all these assays responded in a concentration-dependent manner, and two chronic end points, juvenile production and cocoon output, as well as avoidance behavioral response appeared to be sensitive end points for detecting toxicity of petroleum-hydrocarbon-contaminated soil. Comparatively, juvenile production exhibited similar sensitivity to avoidance behavior response, both of which were more sensitive than cocoon output, while mortality and adult growth were proposed as the least sensitive parameters. It was suggested that large amounts of petroleum sludge deposited in Shengli Oil Field may pose a potential threat to the local ecosystem, and the utility of multiple effects-based end points in earthworm E. fetida is useful to facilitate ecological risk assessments in hydrocarbon-contaminated sites.     

A review of techniques for parameter sensitivity analysis of environmental models

Mathematical models are utilized to approximate various highly complex engineering, physical, environmental, social, and economic phenomena. Model parameters exerting the most influence on model results are identified through a sensitivity analysis. A comprehensive review is presented of more than a dozen sensitivity analysis methods. This review is intended for those not intimately familiar with statistics or the techniques utilized for sensitivity analysis of computer models. The most fundamental of sensitivity techniques utilizes partial differentiation whereas the simplest approach requires varying parameter values one-at-a-time. Correlation analysis is used to determine relationships between independent and dependent variables. Regression analysis provides the most comprehensive sensitivity measure and is commonly utilized to build response surfaces that approximate complex models.     

Treatment of Urban Areas within a Regional Transport Model of Sulphur and Nitrogen Oxides

The air pollution transport model UGEM (The University of Greenwich Evaluation Model) has been developed to evaluate medium-range transport and deposition of sulphur and oxidised nitrogen from all types of sources of emissions in the UK and to estimate their average annual deposition and concentrations across the UK. The model has been tested for its predictions against the available measurements.This study was focused on a possibility of applying the UGEM model to the assessment of air quality on a local scale. One parameter in the model is crucial, the local deposition fraction. The effect of this parameter on quality of the model predictions has been studied for different scales of UGEM output, such as the whole territory of the UK, a rural region and an urban area.The results of the study show that the magnitude of the local deposition fraction should be different for each grid square to reach the best agreement of predictions of concentrations with measurements. Applying a local value of the parameter to each grid square will improve the model predictions of the concentrations in urban areas in particular and will not affect the quality of model predictions of the wet deposition.     

Confronting Management Challenges in Highly Uncertain Natural Resource Systems: a Robustness–Vulnerability Trade-off Approach

This paper presents a framework for the study of policy implementation in highly uncertain natural resource systems in which uncertainty cannot be characterized by probability distributions. We apply the framework to parametric uncertainty in the traditional Gordon–Schaefer model of a fishery to illustrate how performance can be sacrificed (traded-off) for reduced sensitivity and hence increased robustness, with respect to model parameter uncertainty. With sufficient data, our robustness–vulnerability analysis provides tools to discuss policy options. When less data are available, it can be used to inform the early stages of a learning process. Several key insights emerge from this analysis: (1) the classic optimal control policy can be very sensitive to parametric uncertainty, (2) even mild robustness properties are difficult to achieve for the simple Gordon–Schaefer model, and (3) achieving increased robustness with respect to some parameters (e.g., biological parameters) necessarily results in increased sensitivity (decreased robustness) with respect to other parameters (e.g., economic parameters). We thus illustrate fundamental robustness–vulnerability trade-offs and the limits to robust natural resource management. Finally, we use the framework to explore the effects of infrequent sampling and delays on policy performance.     

Application of multivariate statistical techniques in the assessment of water quality in the Southwest New Territories and Kowloon, Hong Kong

The application of different multivariate statistical techniques for the interpretation of a complex data matrix obtained during 2000?C2007 from the watercourses in the Southwest New Territories and Kowloon, Hong Kong was presented in this study. The data set consisted of the analytical results of 23 parameters measured monthly at 16 different sampling sites. Hierarchical cluster analysis grouped the 12 months into two periods and the 16 sampling sites into three groups based on similarity in water quality characteristics. Discriminant analysis (DA) provided better results both temporally and spatially. DA also offered an important data reduction as it only used four parameters for temporal analysis, affording 84.2% correct assignations, and eight parameters for spatial analysis, affording 96.1% correct assignations. Principal component analysis/factor analysis identified four latent factors standing for organic pollution, industrial pollution, nonpoint pollution, and fecal pollution, respectively. KN1, KN4, KN5, and KN7 were greatly affected by organic pollution, industrial pollution, and nonpoint pollution. The main pollution sources of TN1 and TN2 were organic pollution and nonpoint pollution, respectively. Industrial pollution had high effect on TN3, TN4, TN5, and TN6.     

Stochastic approaches for time series forecasting of boron: a case study of Western Turkey

In the present study, a seasonal and non-seasonal prediction of boron concentrations time series data for the period of 1996–2004 from Büyük Menderes river in western Turkey are addressed by means of linear stochastic models. The methodology presented here is to develop adequate linear stochastic models known as autoregressive integrated moving average (ARIMA) and multiplicative seasonal autoregressive integrated moving average (SARIMA) to predict boron content in the Büyük Menderes catchment. Initially, the Box–Whisker plots and Kendall’s tau test are used to identify the trends during the study period. The measurements locations do not show significant overall trend in boron concentrations, though marginal increasing and decreasing trends are observed for certain periods at some locations. ARIMA modeling approach involves the following three steps: model identification, parameter estimation, and diagnostic checking. In the model identification step, considering the autocorrelation function (ACF) and partial autocorrelation function (PACF) results of boron data series, different ARIMA models are identified. The model gives the minimum Akaike information criterion (AIC) is selected as the best-fit model. The parameter estimation step indicates that the estimated model parameters are significantly different from zero. The diagnostic check step is applied to the residuals of the selected ARIMA models and the results indicate that the residuals are independent, normally distributed, and homoscadastic. For the model validation purposes, the predicted results using the best ARIMA models are compared to the observed data. The predicted data show reasonably good agreement with the actual data. The comparison of the mean and variance of 3-year (2002–2004) observed data vs predicted data from the selected best models show that the boron model from ARIMA modeling approaches could be used in a safe manner since the predicted values from these models preserve the basic statistics of observed data in terms of mean. The ARIMA modeling approach is recommended for predicting boron concentration series of a river.     

Use of PFU protozoan community structural and functional characteristics in assessment of water quality in a large, highly polluted freshwater lake in China

Structural and functional parameters of protozoan communities colonizing on PFU (polyurethane foam unit) artificial substrate were assessed as indicators of water quality in the Chaohu Lake, a large, shallow and highly polluted freshwater lake in China. Protozoan communities were sampled 1, 3, 6, 9 and 14 days after exposure of PFU artificial substrate in the lake during October 2003. Four study stations with the different water quality gradient changes along the lake were distinguishable in terms of differences in the community's structural (species richness, individual abundance, etc.) and functional parameters (protozoan colonization rates on PFU). The concentrations of TP, TN, COD and BOD as the main chemical indicators of pollution at the four sampling sites were also obtained each year during 2002-2003 for comparison with biological parameters. The results showed that the species richness and PFU colonization rate decreased as pollution intensity increased and that the Margalef diversity index values calculated at four sampling sites also related to water quality. The three functional parameters based on the PFU colonization process, that is, S(eq), G and T90%, were strongly related to the pollution status of the water. The number of protozoan species colonizing on PFU after exposure of 1 to 3 days was found to give a clear comparative indication of the water quality at the four sampling stations. The research provides further evidence that the protozoan community may be utilized effectively in the assessment of water quality and that the PFU method furnishes rapid, cost-effective and reliable information that may be useful for measuring responses to pollution stress in aquatic ecosystems.     

An Integrated Multicriteria Decision Analysis System for Reducing Air Emissions from Mining Process

The selection of a best alternative method to minimize air pollution and energy consumption for mine sites is a critical task because it encompasses evaluation of different techniques. The aim of this paper is to select most suitable technology for mining system which helps in reducing air pollution and carbon footprints by implementing the multicriteria decision analysis (MCDA) method. The existing methods or frameworks in the mining sector, which have been used in the past to select the sustainable solution, are lacking aid of MCDA, and there is a need to contribute more in this field with a promising decision system. The MCDA method is applied as a probabilistic integrated approach for a mine site in Canada. The analysis involves processing inputs to the Preference Ranking Organization Method for Enrichment Evaluation (PROMETHEE) method which assists in identifying the alternatives, defining the criteria, and thus outranking of the final choice. Moreover, criteria weighting has been determined using analytical hierarchical process (AHP) method. Three categories: reduction of dust/fugitive emission control strategies, reduction in fuel consumption to minimize carbon footprint, and cyanide destruction methods are selected. The probability distributions of criteria weights and output flows are defined by performing uncertainty analysis using the Monte Carlo simulation (MCS). The sensitivity analysis is conducted using Spearman’s rank correlation method and walking criteria weights. The results indicate that the integrated framework provides a reliable way of selecting air pollution control solutions and help in quantifying the impact of different criteria for the selected alternatives.