The science of the total environment

Artificial neural networks (ANNs) constitute a class of flexible nonlinear models designed to mimic biological neural systems. In this entry, we introduce ANN using familiar econometric terminology and provide an overview of ANN modeling approach and its implementation methods. † Correspondence: Chung-Ming Kuan, Institute of Economics, Academia Sinica, 128 Academia Road, Sec. 2, Taipei 115, Taiwan; ckuan@econ.sinica.edu.tw. †† I would like to express my sincere gratitude to the editor, Professor Steven Durlauf, for his patience and constructive comments on early drafts of this entry. I also thank Shih-Hsun Hsu and Yu-Lieh Huang for very helpful suggestions. The remaining errors are all mine.

Artificial neural networks

Economics and Information Theory

There is an increased global demand for activated carbon (AC) in application of water treatment and purification. Water pollutants that have exhibited a greater removal efficiency by AC included but not limited to heavy metals, pharmaceuticals, pesticides, natural organic matter, disinfection by-products, and microplastics. Granular activated carbon (GAC) is mostly used in aqueous solutions and adsorption columns for water treatment. Commercial AC is not only costly, but also obtained from non-renewable sources. This has prompted the search for alternative renewable materials for AC production. Biomass wastes present a great potential of such materials because of their availability and carbonaceous nature. This in turn can reduce on the adverse environmental effects caused by poor disposal of these wastes. The challenges associated with biomass waste based GAC are their low strength and attrition resistance which make them easily disintegrate under aqueous phase. This paper provides a comprehensive review on recent advances in production of biomass waste based GAC for water treatment and highlights future research directions. Production parameters such as granulation conditions, use of binders, carbonization, activation methods, and their effect on textural properties are discussed. Factors influencing the adsorption capacities of the derived GACs, adsorption models, adsorption mechanisms, and their regeneration potentials are reviewed. The literature reveals that biomass waste materials can produce GAC for use in water treatment with possibilities of being regenerated. Nonetheless, there is a need to explore 1) the effect of preparation pathways on the adsorptive properties of biomass derived GAC, 2) sustainable production of biomass derived GAC based on life cycle assessment and techno-economic analysis, and 3) adsorption mechanisms of GAC for removal of contaminants of emerging concerns such as microplastics and unregulated disinfection by-products.

Synthesis and application of Granular activated carbon from biomass waste materials for water treatment: A review

Sugarcane is classed among those plants having a high water requirement and yet it is drought tolerant. The duty of water for cane is fixed at a minimum of 200 acres (≈ 80 ha). It grows well in areas receiving solar energy from 18 to 36 MJ m−2. Rough estimates show that 80% of water loss is associated with solar energy, 14% with wind and 6% with temperature and humidity. The literature is replete with water requirement studies and values vary from ∼ 1000 to c 3000 mm (Table 6.1) depending on the agro-ecological conditions and crop cycle (12–36 months). The amounts of water required to produce 1 g cane, DM and sugar are 50–60, 135–150 and 1000–2000 g, respectively. The peak water use of 0.75–0.85 cm day−1 is observed during the grand growth phase. In Hawaii, Chang et al. (1965) recorded daily levels of evapotranspiration (ET) of 0.3–0.5 cm in winter and 0.7–0.9 cm in summer. A linear relation exists between water use and DM production, and between ET and yield/sugar (Shih 1985).

Irrigation and Drainage

This article explores the suitability of a long short-term memory recurrent neural network (LSTM-RNN) and artificial intelligence (AI) method for low-flow time series forecasting. The long short-term memory works on the sequential framework which considers all of the predecessor data. This forecasting method used daily discharged data collected from the Basantapur gauging station located on the Mahanadi River basin, India. Different metrics [root-mean-square error (RMSE), Nash–Sutcliffe efficiency (ENS), correlation coefficient (R) and mean absolute error] were selected to assess the performance of the model. Additionally, recurrent neural network (RNN) model is also used to compare the adaptability of LSTM-RNN over RNN and naive method. The results conclude that the LSTM-RNN model (R = 0.943, ENS = 0.878, RMSE = 0.487) outperformed RNN model (R = 0.935, ENS = 0.843, RMSE = 0.516) and naive method (R = 0.866, ENS = 0.704, RMSE = 0.793). The finding of this research concludes that LSTM-RNN can be used as new reliable AI technique for low-flow forecasting.

Long short term memory (LSTM) recurrent neural network for low flow hydrological time series forecasting

Impact of urbanization on groundwater recharge and urban water balance for the city of Hyderabad, India

Hyderabad is one of the fastest growing mega cities in India and it is facing many economic, social and environmental problems due to rapid urban growth. For the better planning of resources and to provide basic amenities to its residents, it is necessary to have sufficient knowledge about its urban growth activities. Also, it is necessary to monitor the changes in land use over time and to detect growth activities in different parts of the city. To accomplish these tasks with greater accuracy and easiest way, remote sensing and geographic information system (GIS) tools proved to be very advantageous. This study makes an attempt towards the mapping of land use classes for different time periods and analysis of apparent changes in land use using the Landsat Thematic Mapper (TM) and Enhanced Thematic Mapper (ETM) data for the urban agglomeration of Hyderabad, India. In this study, three different time periods viz. 1989–2000, 2000–2005 and 2005–2011 are chosen for the analysis. The results have shown that high-density urban area had grown during 1989–2011 by encroaching into other land use classes. The urban growth has also affected water resources both, qualitatively and quantitatively in the region. The transformation of other land use types into urban area dynamically continued in the North-East and Southern parts of the city. In the North-East direction, the urban growth was mostly due to growth in industrial and residential area and in Southern part, mostly due to residential growth.

Analysis of urban growth using Landsat TM/ETM data and GIS—a case study of Hyderabad, India

Dissolved oxygen mass balance has been computed for different reaches of River Kali in western Uttar Pradesh (India) to obtain the reaeration coefficient (K 2 ). A total of 270 field data sets have been collected during the period from March 1999 to February 2000. Eleven most popular predictive equations, used for reaeration prediction and utilizing mean stream velocity, bed slope, flow depth, friction velocity and Froude number, have been tested for their applicability in the River Kali using data generated during field survey. The K 2 values computed from these predictive equations have been compared with the K 2 values observed from dissolved oxygen balance measurements in the field. The performance of predictive equations have been evaluated using error estimation, namely standard error (SE), normal mean error (NME), mean multiplicative error (MME) and correlation statistics. The equations developed by Smoot and by Cadwallader and McDonnell showed comparatively better results. Moreover, a refined predictive equation has been developed using a least-squares algorithm for the River Kali that minimizes error estimates and improves correlation between observed and computed reaeration coefficients.

Refinement of predictive reaeration equations for a typical Indian river

An application of remote sensing and geographic information system (GIS) has become one of the essential techniques in the field of hydrogeological science that helps in assessing, monitoring, conserving, and managing groundwater resources. In the present study, modified DRASTIC model has been used to access the groundwater vulnerable zone of Kharun Basin. The modification has been done by replacing hydraulic conductivity parameter to land use. The aim of this study is to determine susceptible zone for groundwater pollution by integrating hydrogeological layers in GIS environment. The final index map shows that the resulting index value varies from 86 to 191; further, the zones are divided into five classes. The results showed that out of the total area of 4191 km2, an area of about 8, 197, 2730, 1229, and 27 km2 lies in very low, low, moderate, high and very high vulnerable zones, respectively. Sensitivity analysis is also performed in this study, and it reveals that depth of water table, land use, and topography parameters cause large variations in vulnerability index as compared to other parameters. Moreover, land use and topography were found to be more effective parameters in assessing groundwater vulnerability than assumed by original model. Model validation shows a good correlation between nitrate sample and resulting DRASTI-LU index. This study produces a very valuable tool for policy makers as it reveals a very comprehensive indication of vulnerability to ground water contamination. The knowledge about high vulnerable zone is important for local authorities to manage and monitor groundwater resources.

Ramakar Jha

Papers

Long short term memory (LSTM) recurrent neural network for low flow hydrological time series forecasting

Impact of urbanization on groundwater recharge and urban water balance for the city of Hyderabad, India

Analysis of urban growth using Landsat TM/ETM data and GIS—a case study of Hyderabad, India

Refinement of predictive reaeration equations for a typical Indian river

Assessment of groundwater vulnerability using modified DRASTIC model in Kharun Basin, Chhattisgarh, India