Multilayer perceptron neural network for flow prediction

Artificial neural networks (ANNs) have proven to be a tool for characterizing, modeling and predicting many of the non-linear hydrological processes such as rainfall-runoff, groundwater evaluation or simulation of water quality. After proper training they are able to generate satisfactory predictive results for many of these processes. In this paper they have been used to predict 1 or 2 days ahead the average and maximum daily flow of a river in a small forest headwaters in northwestern Spain. The inputs used were the flow and climate data (precipitation, temperature, relative humidity, solar radiation and wind speed) as recorded in the basin between 2003 and 2008. Climatic data have been utilized in a disaggregated form by considering each one as an input variable in ANN(1), or in an aggregated form by its use in the calculation of evapotranspiration and using this as input variable in ANN(2). Both ANN(1) and ANN(2), after being trained with the data for the period 2003-2007, have provided a good fit between estimated and observed data, with R(2) values exceeding 0.95. Subsequently, its operation has been verified making use of the data for the year 2008. The correlation coefficients obtained between the data estimated by ANNs and those observed were in all cases superior to 0.85, confirming the capacity of ANNs as a model for predicting average and maximum daily flow 1 or 2 days in advance.     

Accuracy of bottled drinking water label content

The accurate predictions of ground ozone concentrations are required for proper management, control, and making public warning strategies. Due to the difficulties in handling phenomenological models that are based on complex chemical reactions of ozone production, neural network models gained popularity in the last decade. These models also have some limitations due to problems of overfitting, local minima, and tuning of network parameters. In this study, the predictions of daily maximum ozone concentrations are attempted using support vector machines (SVMs). The comparison between the accuracy of SVM and neural network predictions is performed to evaluate their performance. For this, the daily maximum ozone concentration data observed during 2002–2004 at a site in Delhi is utilized. The models are developed using the available meteorological parameters. The results indicated the promising performance of SVM over neural networks in predicting daily maximum ozone concentrations.     

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.     

城市河流重金属元素对降雨径流的动态响应

以海口市美舍河市区河段为研究对象,于2020年雨季分别采集降雨径流以及降雨径流汇入后的河流瞬时水样,测定并分析主要重金属元素的浓度及其变化规律,探讨其对降雨径流的动态响应。研究结果表明：降雨径流对河流可溶态污染元素质量浓度的贡献不突出,对可溶态锌（Zn）、砷（As）、镉（Cd）、硒（Se）具有一定稀释作用,排水口下游 3 m处是可溶态铬(Cr）、锰（Mn）、钴（Co）对降雨径流响应最显著区域;降雨径流对各悬浮态元素质量浓度贡献较大,排水口下游7 m处是绝大多数悬浮态元素对降〖JP〗雨径流响应最显著区域;Mn、Zn、Cr、镍（Ni）、As、Co、Cd 7种元素的总质量浓度对降雨径流的动态响应表现为先上升后下降。其中,排水口下游3 m处是Cr、As总质量浓度对降雨径流响应最显著区段,排水口下游7 m处是Mn、Co、Ni、Zn及Cd总质量浓度对降雨径流响应最显著区段。降雨径流对河流主要污染元素年输入总量估算结果表明,海口市建成区降雨径流向美舍河等受纳水体直接年输入量最大的是Mn和Zn,其次是Cr、As、Ni,Se与Cd的年输入量最小。     

The magnitude of variability produced by methods used to estimate annual stormwater contaminant loads for highly urbanised catchments

Stormwater contaminant loading estimates using event mean concentration (EMC), rainfall/runoff relationship calculations and computer modelling (Model of Urban Stormwater Infrastructure Conceptualisation—MUSIC) demonstrated high variability in common methods of water quality assessment. Predictions of metal, nutrient and total suspended solid loadings for three highly urbanised catchments in Sydney estuary, Australia, varied greatly within and amongst methods tested. EMC and rainfall/runoff relationship calculations produced similar estimates (within 1 SD) in a statistically significant number of trials; however, considerable variability within estimates (～50 and ～25 % relative standard deviation, respectively) questions the reliability of these methods. Likewise, upper and lower default inputs in a commonly used loading model (MUSIC) produced an extensive range of loading estimates (3.8–8.3 times above and 2.6–4.1 times below typical default inputs, respectively). Default and calibrated MUSIC simulations produced loading estimates that agreed with EMC and rainfall/runoff calculations in some trials (4–10 from 18); however, they were not frequent enough to statistically infer that these methods produced the same results. Great variance within and amongst mean annual loads estimated by common methods of water quality assessment has important ramifications for water quality managers requiring accurate estimates of the quantities and nature of contaminants requiring treatment.     

The role and distribution of total suspended solids in the macrotidal coastal waters of Korea

Spatial and temporal distribution patterns of total suspended solids (TSS) in the shallow and macrotidal regions of the Korean peninsula indicated there were significant changes in TSS concentrations. These were seasonally influenced by the wind, river input and tidal cycle. There were high TSS values at estuarine and river mouth stations and during low tide due to the re-suspension of bottom sediment by strong wind action during winter months, in addition the land input through rivers and strong tidal current during ebbing. Monthly mean values of TSS significantly correlated with wind speed and nitrate concentration (p < 0.01). This indicated that the resuspension of surface sediment was a more important source of TSS than the river input, and that nitrate was introduced into the water column during the resuspension process. TSS were seven times higher at low tide than in high tide. Light penetration was significantly inhibited by TSS; as >98% of incident light was absorbed within 2 m and zero photosynthetically active radiation (PAR) under 2 m in the estuarine stations during winter. Removal of heavy metals and nutrients by TSS in the water column was evident. Over 80% of the initial concentration of nutrients was removed within 10 min under various concentrations of TSS and also TSS contained significantly higher concentration of heavy metals than surface sediment. The concentration levels of nutrients and chemical oxygen demand in the west coast were comparable with the East and South Sea, even the major rivers in the Korean peninsula flow into the West Sea with major pollutant loadings into the coastal areas. High concentration of TSS is likely to contribute to the removal process of these pollutants, resulting in relatively lower levels of nutrients and organic materials in these coastal waters.     

Total metal levels in crayfish Astacus leptodactylus (Eschscholtz, 1823), and surface sediments in Lake Terkos, Turkey

The aim of this study was to determine the total metal accumulation (aluminium, copper, manganese, lead, cadmium and iron) in different organs and eggs of Astacus leptodactylus (Eschscholtz, 1823) and sediments total metal contents (aluminium, copper, manganese, lead, cadmium, iron, zinc, chromium, nickel) in Lake Terkos. Water and sediment samples were collected from two stations at two different depths (1 and 2 m) of Lake Terkos in May 2008. Crayfish samples were collected by trammel net at the same region. Primary hydrographic conditions, such as temperature (13.6–19.4°C), salinity (0.27–0.34‰), dissolved oxygen (7.04–12.30 mg l^???1) and pH (7.42–8.51), were recorded for each sampling point. Moreover, the total organic carbon (1.65–5.44%) and the total calcium carbonate contents (19.44–41.16%) of sediment samples were determined. According to the Turkish Food Codex (J Zool 26:283–288, 2002), the maximum allowable Pb and Cd levels in crayfish are 0.5 mg/kg wet weight. Accordingly, the Pb and Cd levels determined in A. leptodactylus samples are below this limit. However, when compared with the acceptable metal limits defined by WHO, Australian National Health and Medical Research Council and Ministry of Agriculture in United Kingdom (UK), it is clear that the Cu level is at the limit and the Cd results exceed the limit. When the metal contents in sediment samples from Lake Terkos are examined, it is seen that the Al, Fe, Mn, Ni and Cu contents are lower while Zn, Cr, Cd and Pb contents are higher than the crustal average values. The high values draw attention to the land-based domestic and industrial inputs. Lake Terkos sediments have high enrichment factors (EF) of Zn, Cr, Cd and Pb metals which corroborate this result. The low EFs of Fe, Ni and Cu are due to the natural (terrigeneous) inputs. Additionally, there is no Al, Fe, Ni and Cu metal enrichment in these lake sediments because of the low contamination factor (CF) values. However, it is moderately contaminated by Zn, Cr and Pb, and heavily contaminated by Cd.     

A probabilistic water quality index for river water quality assessment: a case study

Available water quality indices have some limitations such as incorporating a limited number of water quality variables and providing deterministic outputs. This paper presents a hybrid probabilistic water quality index by utilizing fuzzy inference systems (FIS), Bayesian networks (BNs), and probabilistic neural networks (PNNs). The outputs of two traditional water quality indices, namely the indices proposed by the National Sanitation Foundation and the Canadian Council of Ministers of the Environment, are selected as inputs of the FIS. The FIS is trained based on the opinions of several water quality experts. Then the trained FIS is used in a Monte Carlo analysis to provide the required input-output data for training both the BN and PNN. The trained BN and PNN can be used for probabilistic water quality assessment using water quality monitoring data. The efficiency and applicability of the proposed methodology is evaluated using water quality data obtained from water quality monitoring system of the Jajrood River in Iran.     

Pesticide concentrations in air and precipitation in the Netherlands

Atmospheric deposition may be an important source of persistent organic compounds (POP) and pesticides for the Dutch coastal and inland waters. Current estimates of the atmospheric input have been made using atmospheric dispersion models. The uncertainty is however large. A project was defined with the aim to assess the input on the basis of measurements. For a period of two years (1999-2001) a monitoring network was operated. At eighteen stations, located across the whole country, air and precipitation samples were taken on a weekly and monthly basis. In these samples the concentrations of pesticides, PCB's and PAH's were determined. Up to 50 different pesticides were observed in precipitation and air. The concentration of 17 of these in precipitation exceeded the maximum permissible level for surface water and 22 exceeded the standard for drinking water of 100 ng l(-1). The input from the atmosphere to Dutch inland waters appeared to be as large as the input of pesticides by other sources such as spray drift. Model calculations were also carried out to identify the sources of these compounds. The occurrence of atrazine could be related to emissions outside the Netherlands.     

Characteristics of Stormwater Quality from Two Urban Watersheds in Singapore

This paper presents the results of a study on the quality of stormwater runoff from two adjacent urban watersheds in Singapore. The study involved continuous recording of rainfall and streamflow within the basins, and systematic sampling of storm runoff during rainfall events. The water samples collected were analysed for its physical and chemical constituents. The ranges and mean concentrations of common quality parameters have been established, and the flushing effects of selected parameters were studied. Correlations of event mean concentrations and pollution loads of total suspended solids and chemical oxygen demand to both antecedent dry weather period and rainfall characteristics were also carried out.     

Probabilistic risk assessment and risk mapping of metals in tropical estuarine sediments using the 1 quarisk model

Total sediment concentrations of Cd, Cu, Pb, Zn, As, and Hg obtained from the Ankobra, Sakumo II, and Volta estuaries in Ghana were used to generate contaminant probability density distributions and species sensitivity distributions in AQUARISK. Results of the tier 1 assessment showed Cu, Cd, Zn, and Pb were not of concern in the Ankobra as their measured values and the 99th percentile of the fitted distributions were lower than the SQG low-trigger values. Mercury (Hg) and As were however, identified to be of concern in this estuary. In the Sakumo II estuary, Cu, Cd, Pb, and Hg have been identified to be of concern because their concentrations are higher than the SQG low-trigger values. Hg has been identified as the only metal of concern in the Volta estuary. The total proportion of species likely to be affected by the combined concentration of Cd, Cu, and Zn measured from Ankobra, Sakumo II, and Volta were 14%, 16%, and 12%, respectively, according to the Bur III distributional analysis of the ecotoxicology data. The measured median sediment concentrations of As and Hg in the Ankobra estuary greatly exceeded the median sediment concentration targets to achieve a 5% or less exceedence of the SQG low value. Similarly, in the Sakumo II estuary, the measured median sediment concentrations of Cu, Pb, and Hg greatly exceeded the median sediment concentration targets to achieve a 5% or less exceedence of the SQG low. For the Volta estuary however, other metals except Hg fall below the target values.     

Potential Impact of Climate Change on Residential Energy Consumption in Dhaka City

This study aimed to assess the impacts of climate change on residential energy consumption in Dhaka city of Bangladesh. The monthly electricity consumption data for the period 2011–2014 and long-term climate variables namely monthly rainfall and temperature records (1961–2010) were used in the study. An ensemble of six global circulation models (GCMs) of coupled model intercomparison project phase 5 (CMIP5) namely, BCCCSM1-1, CanESM2, MIROC5, MIROC-ESM, MIROC-ESM-CHEM, and NorESM1-M under four representative concentration pathway (RCP) scenarios were used to project future changes in rainfall and temperature. The regression models describing the relationship between historical energy consumption and climate variables were developed to project future changes in energy consumptions. The results revealed that daily energy consumption in Dhaka city increases in the range of 6.46–11.97 and 2.37–6.25 MkWh at 95% level of confidence for every increase of temperature by 1 °C and daily average rainfall by 1 mm, respectively. This study concluded that daily total residential energy demand and peak demand in Dhaka city can increase up to 5.9–15.6 and 5.1–16.7%, respectively, by the end of this century under different climate change scenarios.     

Artificial neural network analysis for reliability prediction of regional runoff utilization

Many factors in the reliability analysis of planning the regional rainwater utilization tank capacity need to be considered. Based on the historical daily rainfall data, the following four analyzing procedures will be conducted: the regional daily rainfall frequency, the amount of runoff, the water continuity, and the reliability. Thereafter, the suggested designed storage capacity can be obtained according to the conditions with the demand and supply reliability. By using the output data, two different types of artificial neural network models are used to build up small area rainfall–runoff supply systems for the simulation of reliability and the prediction model. They are also used for the testing of stability and learning speed assessment. Based on the result of this research, the radial basis function neural network (RBFNN) model, using the Gaussian function that has a similar trend as the nature as basic function, has better stability than using the back-propagation neural network (BPNN) model. Despite the fact that RBFNN was more reliable than BPNN, it still made a conservative estimate for the actual monitoring data. The error rate of RBFNN was still higher than the correction of BPNN 4-3-1-1. This should have significant benefit in the future application of the instantaneous prediction or the development of related intelligent instantaneous control equipment.     

Distribution and bioaccumulation of organochlorine pesticides (OCPs) in food web of Nansi Lake,China

The concentration of 12 organochlorine pesticides (OCPs) were measured in water, sediment, aquatic plant, and animal (shrimp and fish) of Nansi Lake by gas chromatography equipped with an electron capture detector. The total OCPs concentrations were 65.31–100.31 ng L^?1 in water, 2.9–6.91 ng g^?1 dry weight (dw) in sediments, 1.29–6.42 ng g^?1 dw in aquatic plants and 7.57–17.22 ng g^?1 dw in animals. The OCPs composition profiles showed that heptachlor compounds was also the predominant OCPs contaminants in addition to hexachlorocyclohexanes (HCHs) and dichlorodiphenyltrichloroethanes (DDTs) in Nansi Lake. According to the source of HCHs and DDTs in sediment samples, there was no new input and the HCHs pollution mainly came from the use of Lindane in Nansi Lake. Bioaccumulation of OCPs in aquatic biota indicated that DDTs and heptachlor compounds had a strong accumulation, followed by HCHs and drins. The accumulation abilities of fish for OCPs were higher than those of plants and shrimps. The OCPs biota-sediment accumulation factor values of Channa argus was the highest in fish samples, followed by Carassius auratus, and Cyprinus caspio. Risk assessment of sediment showed that heptachlor epoxide had a higher occurrence possibility of adverse ecological effects to benthic species. Based on the calculation of acceptable daily intake and hazard ratio, HCHs in fish and shrimps from Nansi Lake had a lifetime cancer risk of greater than one per million. The risk assessment of water, sediment, and fish indicated the water environment of Nansi Lake is at a safe level at present.     

Soils and sediments of the Thua Thien-Hue Province (central Vietnam): recognizing trace element sources and the likely influence of natural events

Trace element concentrations have been measured in soil and sediment samples taken from the Thua Thien Hue (TT-H) Province and the Tam Giang-Cau Hai (TG-CH) Lagoon in 2002 and 2004. Results show that the lagoon is only slightly contaminated by elements such as Ag, Cd, Cr, Cu, Ni, Pb and Zn, whereas As, probably naturally enriched in this Asian region, reaches values above the lowest internationally accepted sediment quality guidelines. Concentrations in soils are ascribable to natural sources and distributions in the TG-CH Lagoon are mainly influenced by river inputs, with Ag and Cd undergoing estuarine desorption processes. However, concentration-depth profiles of most trace elements in sediments show a recent increasing trend that might be linked to the economic development of the area. The comparison of porosity and (210)Pb depth profiles (confirmed also by Cd and As) from repeated samplings of the same core locations in the TG-CH Lagoon seems to account for a loss of 5-10 cm of surficial sediment that took place during the time interval between the two samplings. It is hypothesized that extreme meteorological events (e.g. a major typhoon that hit the area in November 2003) could be responsible for such sediment displacement.     

The distribution and composition of hydrocarbons in sediments from the Fladen Ground, North Sea, an area of oil production

The distribution and composition of hydrocarbons in sediment from the Fladen Ground oilfield in the northern North Sea have been investigated. The total PAH concentrations (2- to 6-ring parent and alkylated PAHs, including the 16 US EPA PAHs) in sediments were relatively low (<100 microg kg(-1) dry weight). The PAH, the Forties crude and diesel oil equivalent concentrations were generally higher in sediment of fine grain size and higher organic carbon concentration. PAH distributions and concentration ratios indicated a predominantly pyrolytic input, being dominated by the heavier, more persistent, 5- and 6-ring compounds, and with a high proportion of parent PAHs. The n-alkane profiles of a number of the sediments contained small, high boiling point, UCMs, indicative of weathered oil arising from a limited petrogenic input. The geochemical biomarker profiles of the sediments that contained UCMs showed a small bisnorhopane peak and a high proportion of norhopane relative to hopane, indicating that there was contamination from both Middle Eastern and North Sea oils. Therefore contamination was not directly as a result of oil exploration activity in the area. The most likely source of petrogenic contamination was from general shipping activity.     

Lake Ontario sediments in monitoring pollution

An investigation was carried out to select locations for long-term monitoring of inputs of contaminants into Lake Ontario using fine-grained bottom sediments as an historical record of pollution. The sediment sampling program was designed to determine sediment heterogeneity in the western, central and eastern depositional basins of the lake. Surficial sediments and sediment cores were collected in each basin to obtain information on horizontal distribution and concentration profiles of major and trace elements in the sediments (Si, Al, Ca, Mg, Fe, Na, K, Ti, Mn, P, As, Co, Cu, Cr, Ni, Pb and Zn). The results of the investigation indicated that fine-grained sediments in three Lake Ontario depositional basins are homogeneous to a high degree, and that only a few sediment cores need to be collected within each basin for the long-term monitoring of inputs of contaminants to the lake.     

Impacts of climate-induced changes on the distribution of pesticides residues in water and sediment of Lake Naivasha, Kenya

This study reports evidence of increased chlorpyrifos contamination in sediment and water in Lake Naivasha following its intensive application in the horticultural farms in the catchment area. Analytical results show that levels of chlorpyrifos residues were influenced by climate-induced rainfall pattern with higher levels reported during period of heavy precipitation with significant decrease during low rainfall. On average, the levels ranged between 14.8 and 32.8 ng g^?1 in sediment during rainy season compared to a range of 8.5–16.6 ng g^?1 in the dry season. Additionally, the mean concentration of chlorpyrifos in water ranged between 8.61 and 22.4 μg L^?1 during rainy season and below detection limit (bdl) ?13.6 μg L^?1 in dry season as quantified by enzyme-linked immunosorbent assay. Meanwhile, independent t test analysis indicated that there was significant difference in concentration at p?≤?0.05 between the seasons with respect to sediment and water samples. This demonstrated that climate-induced variations had considerable influence on contamination. While diazinon and carbofuran were equally applied intensively, their levels were below the detection limit in the all the samples analyzed. ELISA results were validated by the capillary-HPLC photodiode-array detector instrument analysis, and statistical comparison showed no significant difference between them. It was evident that chlorpyrifos residues determination in water and sediment by ELISA can be a useful strategy in environmental management and monitoring program, and a complimentary analytical tool to high performance liquid chromatography. Levels of chlorpyrifos detected in sediment and water were found to exceed recommended criteria for protection of aquatic life and preservation of water quality and may be hazardous if not regularly monitored.     

Storm runoff quality and pollutant loading from commercial, residential, and industrial catchments in the tropic

Information on the pollution level and the influence of hydrologic regime on the stormwater pollutant loading in tropical urban areas are still scarce. More local data are still required because rainfall and runoff generation processes in tropical environment are very different from the temperate regions. This study investigated the extent of urban runoff pollution in residential, commercial, and industrial catchments in the south of Peninsular Malaysia. Stormwater samples and flow rate data were collected from 51 storm events. Samples were analyzed for total suspended solids, 5-day biochemical oxygen demand, chemical oxygen demand, oil and grease (O&G), nitrate nitrogen (NO₃-N), nitrite nitrogen, ammonia nitrogen, soluble reactive phosphorus, total phosphorus (TP), and zinc (Zn). It was found that the event mean concentrations (EMCs) of pollutants varied greatly between storm characteristics and land uses. The results revealed that site EMCs for residential catchment were lower than the published data but higher for the commercial and industrial catchments. All rainfall variables were negatively correlated with EMCs of most pollutants except for antecedent dry days (ADD). This study reinforced the earlier findings on the importance of ADD for causing greater EMC values with exceptions for O&G, NO₃-N, TP, and Zn. In contrast, the pollutant loadings are influenced primarily by rainfall depth, mean intensity, and max 5-min intensity in all the three catchments. Overall, ADD is an important variable in multiple linear regression models for predicting the EMC values in the tropical urban catchments.     

Soil Water Content Estimated by Support Vector Machine for the Assessment of Shallow Landslides Triggering: the Role of Antecedent Meteorological Conditions

Soil water content is a key parameter for representing water dynamics in soils. Its prediction is fundamental for different practical applications, such as identifying shallow landslides triggering. Support vector machine (SVM) is a machine learning technique, which can be used to predict the temporal trend of a quantity since training from past data. SVM was applied to a test slope of Oltrepò Pavese (northern Italy), where meteorological parameters coupled with soil water content at different depths (0.2, 0.4, 0.6, 1.0, 1.2, 1.4 m) were measured. Two SVM models were developed for water content assessment: (i) model 1, considering rainfall amount, air temperature, air humidity, net solar radiation, and wind speed; (ii) model 2, considering the same predictors of model 1 together with antecedent condition parameters (cumulated rainfall of 7, 30, and 60 days; mean air temperature of 7, 30, and 60 days). SVM model 2 showed significantly higher satisfactory results than model 1, for both training and test phases and for all the considered soil levels. SVM models trends were implemented in a methodology of slope safety factor assessment. For a real event occurred in the tested slope, the triggering time was correctly predicted using data estimated by SVM model based on antecedent meteorological conditions. This confirms the necessity of including these predictors for building a SVM technique able to estimate correctly soil moisture dynamics in time. The results of this paper show a promising potential application of the SVM methodologies for modeling soil moisture required in slope stability analysis.