Journal of Environmental Engineering

Phytoplankton plays an essential role in the biogeochemical cycle because it is at the top of the food chain and is a source of oxygen. Eutrophication causes coastal areas to deteriorate as industrialization accelerates, leading to harmful algal blooms (HABs), severely affecting human and ecological health. The frequency and extent of HAB events potentially may increase due to climate change. HAB outbreaks have led to substantial losses for major coastal economies globally, and therefore have emerged as a critical research focus in environmental sciences. However, the lack of an overview of diverse factors influencing HABs complicates the cause identification and the effective countermeasure development for HAB occurrence, thereby impeding the formulation of targeted strategies for prediction and mitigation. Therefore, this review summarizes the influential factors affecting HABs in coastal areas, including water quality factors (nutrients, salinity, stratification, and biological factors) and climatological factors (temperature, pH and ${\mathrm{pCO}}_2$, and irradiance and light). Recent work with several harmful algae species suggested that warmer temperatures combined with nutrient variation, stronger stratification, and ocean acidification may increase the growth of some toxic dinoflagellate species. Although the effects of factors vary for different species and locations, the intensification of anthropogenic activities and climate change likely will increase the frequency, outbreak scale, and severity of most coastal HABs. Because predicting HABs is crucial for understanding the factors and synergy affecting their growth and minimizing losses for decision makers and stakeholders, we reviewed models for predicting HABs, including process-based models, traditional statistical-empirical models, and data-driven machine learning models. Predicting HABs becomes more challenging as the spatial distribution of harmful algae is influenced by future climate patterns. This review paper presents a comprehensive overview of the various factors impacting HABs in coastal areas, serving as a valuable resource for decision makers and researchers to design targeted research and mitigation strategies.

Machine learning (ML) is increasingly implemented to model water infrastructure dynamics. Common ML models are primarily data-driven and require a significant amount of data for robust training. Often, obtaining robust data at higher temporal and spatial resolutions in water systems can be challenging due to cost and time considerations. In such scenarios, integrating the existing scientific knowledge into an ML model as physics-informed ML can be advantageous to enhance predictive capability and generalizability. This study examines the predictive capability and generalizability of physics-informed ML and common ML models for typical unit operation and process (UOP) system dynamics in urban water treatment. The systems studied are (1) a continuous stirred-tank reactor, (2) activated sludge reactor, and a (3) fixed-bed granular adsorption reactor. Applications of physics-informed neural networks (PINNs) are presented. Results demonstrate when the availability of data is limited: (1) common ML models are not necessarily robust for predicting water system dynamics, except when system dynamics exhibit simpler periodic patterns. Common ML models also do not generalize across different loading conditions. (2) In contrast, the developed PINN models yield high predictive capability and generalizability. (3) Benefiting from the embedded prior knowledge, PINNs require significantly reduced data sets for robust predictions. These results suggest hybridizing physics principles and domain knowledge into the ML framework can be critical for robust UOP and water systems modeling.

Abstract

One of the greatest challenges faced when using gravity settling is achieving efficient particle removal and retention rates under high inflow rates. This paper proposes a new method for enhancing sedimentation and protecting the settled particle. This is accomplished by placing one layer of the virtual bed above the real bed to divide the water body into hydraulically different upper and lower regions. The normal structure of the newly-proposed virtual bed is a flat plate with many small perforations. When the water flows along the top surface of the virtual bed, it creates two effects: (1) it isolates and protects water in the area below the virtual bed from being disturbed by the fast and turbulent flows above, which enhances sedimentation of the particles in the lower region and protects the sediment which could be eroded; and (2) the vertical vortex generated by the surficial flows passing over the openings helps the nearby suspended particles to enter the quiescent water region below. To assess the performance of the newly-proposed virtual bed, a rectangular settling tank was used to conduct comparison tests of particle removal with three particle size ranges, four inflow rates, five virtual bed structural designs, and multiple experimental conditions. The results clearly showed that, compared with a traditional settling tank (without the virtual bed), the addition of the proposed settling structure notably enhanced the particle settling rate by 10%–15% for the particles tested and experimental conditions assessed. The virtual bed method easily can be applied to various water treatment devices and facilities to enhance the suspended particle removal efficiencies in treatments of storm runoff, wastewater, and many other kinds of water without the need for chemical additions or energy-intensive processes.

Practical Applications

This paper proposes and tested an innovative method for increasing the removal of suspended particles using a virtual bed. A virtual bed is composed of a plate with numerous holes through it suspended above particle-settling facilities. It enhances particle settling by providing a less disturbed region for settling particles, protecting already-settled particles, and encouraging the suspended particles to move toward the real bed. Laboratory tests in a rectangular tank with different particle sizes and flow rates showed that the virtual bed effectively improved particle removal rates. Various possible virtual bed structure designs are proposed and were examined. The best-performing virtual bed was identified, and possible reasons for its performance were discussed based on hydraulics and physics principles. Experiments with different tank configurations and virtual bed position arrangements were conducted, and based on the experimental results, general suggestions for using the virtual bed in practice more effectively are given. In principle, the proposed virtual bed can be applied to any particle-separation facilities.

Abstract

Climate change will reduce the availability of Colorado River Water (CRW) in the Lower Colorado River Basin (LCRB) and hydroelectric power generation at Lakes Powell and Mead. Efforts to stabilize reservoir levels in the presence of interdecadal drought are based primarily on shortages in water deliveries to LCRB states and Mexico per the 2007 Interim Guidelines (2007-IG) and later the LCRB Drought Contingency Plan (2019-DCP). When climate effects are considered, the expected energy generation (lakes Mead and Powell combined) is projected to decrease from 6500 GWH in 2024 to 5400 GWH in 2060, despite a 400–600 GWH annual increase consequent to 2019-DCP measures. 2019-DCP shortages decrease cumulative water deliveries to the LCRB states and Mexico during 2022–2060, increasing evaporative losses from lakes Mead and Powell. The cumulative present value of LCRB water deliveries during 2022–2060 dwarfs that of hydroelectric energy benefits. However, the differences in benefits under the two sets of measures reveal that the economic impact of water and energy are of the same order of magnitude. Under the 2019-DCP rules, energy production and recreation benefits are positive in all but one scenario, while overall water deliveries are reduced in two-thirds of all scenarios. The median difference in present value benefits under 2019-DCP compared with the 2007-IG is about $+\$100\mathrm{M}$, but the full range of outcomes for all climate-related scenarios investigated is approximately $\pm \$1\mathrm{B}$.

Practical Applications

Mexico and the seven US states that take water from the Colorado River system have opened negotiations to reallocate Colorado River water ahead of 2026. The process will be contentious due to competing interests and uncertainties related to the projected flows and associated benefits. Here, the present values of Colorado River system water deliveries, hydroelectric energy generation at Lakes Powell and Mead and associated recreational benefits are projected through 2060. For purposes of comparison, the operations analyzed conform to requirements of either the 2007 Interim Guidelines (2007-IG) or the 2019 Drought Contingency Plan (2019-DCP). This offers a baseline and framework for economic evaluation of benefits resulting from river management alternatives beyond those of the 2007-IG and 2019-DCP. Results indicate that the present worth of water deliveries in the Lower Colorado River Basin (LCRB) during 2022–2060 will clearly exceed that of energy generation at Lakes Powell and Mead. However, differences in revenue show that the economic DCP-impact of water and energy are of the same order of magnitude (and larger than recreational benefits). The analysis and results described provide an avenue for proposing, comparing, and selecting rational river management measures in the presence of unavoidable uncertainties and competing interests.

Abstract

Nitrogen removal from municipal wastewater through partial nitritation-denitrification (nitritation) is challenging to accomplish in a membrane-aerated biofilm reactor (MABR) due to the reactor configuration, which potentially interferes with nitrite-oxidizing bacteria inhibition. This study investigated the impact of intermittent aeration on the development and sustenance of nitritation in a lab-scale MABR for the treatment of municipal wastewater. The study was accomplished in four phases (Phases I–IV) using a combination of continuous and intermittent aeration modes with aerated and nonaerated cycles of 10 min (5 on/5 off), 20 min (10 on/10 off), and 25 min (10 on/15 off), respectively, and a constant hydraulic retention time of 2.5 h. Biofilm development and stabilization were completed using a continuous aeration condition (Phase I). Nitrite accumulation rate, nitrate production rate, and ammonium nitrogen removal efficiency achieved in Phases II–IV were, 35%, 12%, and 99%; 76%, 3.4%, and 98%; and 94%, 1%, and 98%, respectively. Intermittent aeration significantly improved total inorganic nitrogen removal efficiency by $\sim 20\%$. Between the initiation of intermittent aeration and termination of the study, ammonia-oxidizing bacteria activities within the reactor increased by $>150\%$ from 4.53 to $12.6\mathrm{mgN}/\mathrm{h}·\mathrm{g}$ volatile suspended solids (VSS). In contrast, nitrite-oxidizing bacteria activities declined by $>60\%$ from 1.17 to $0.46\mathrm{mgN}/\mathrm{h}·\mathrm{gVSS}$. The consistent lagging of nitrate production rate behind nitrite accumulation rate, increase in ammonia-oxidizing bacteria activities, and decline in nitrite-oxidizing bacteria activities over the operation period indicates the establishment of nitritation. This study demonstrates that using intermittent aeration, nitritation can be developed and sustained in MABR under mainstream conditions.

Practical Applications

Biological nitrogen removal from municipal wastewater is typically through the conventional activated sludge system using the traditional nitrification and denitrification pathway. However, three major issues abound with this pathway: high aeration energy, external organic carbon requirement, and the potential release of nitrous oxide. An alternative to the aforementioned process is the MABR. MABR is a low-energy technology with potential for pollutants removal through nitritation. Nitritation compares favorably to simultaneous nitrification-denitrification in terms of energy efficiency, sustainability, and operational costs. However, nitrite-oxidizing bacteria activities suppression is required for the successful establishment of nitritation. Studies have suggested that nitrite-oxidizing bacteria activity suppression can be accomplished in biological systems when operated under specific conditions, but although some of these conditions are inherent in some systems, they are absent in mainstream wastewater and MABR. Therefore, establishing nitritation in MABR for the treatment of conventionally collected sewage is challenging. This study recommends using continuous aeration to develop the biofilm and establish stable performance within the MABR, and subsequently switching the aeration mode to predetermined intermittent aeration cycles. This will reduce the dissolved oxygen concentration, suppress nitrite-oxidizing bacteria activities, and stabilize suppressed nitrite-oxidizing bacteria activities within the MABR biofilm.

Increasing urbanization has put significant pressure on water and wastewater management. To reduce dependency on freshwater, treated sewage can be reutilized for various purposes such as irrigation, horticulture, washing, flushing, and so on. A wastewater treatment plant (WWTP) is a unit that improves the polluted raw wastewater into an environmentally friendly water (usually called treated sewage) through various treatment steps and electromechanical equipment. Nanobubble (NB) technology in the wastewater treatment industry is one of the promising concepts that uses gas-infused NBs to treat the wastewater for reducing the energy consumption and improving the treated water quality. In the current work, the efficiency of NB technology has been studied by upgrading an existing WWTP with NB injection for improving the treated effluent quality. The NB unit has been installed at the filter feed tank to treat the incoming sewage. Improvement in the wastewater treatment plant performance has been identified by analyzing the commonly used water quality parameters, i.e., pH, turbidity, total dissolved solids (TDS), conductivity, biochemical oxygen demand (BOD), chemical oxygen demand (COD), total suspended solids (TSS), total Kjeldahl nitrogen (TKN), ${\mathrm{NH}}_3$, total phosphorus (TP), and fecal coliform at inlet and outlet of the WWTP. The results are then compared with the baseline WWTP working without NB, and it was observed that the physiochemical parameters were reduced to acceptable levels after treatment with NB infusion.

Bioretention systems use plants and soil media to treat pollutants in stormwater. Biochar can be an effective amendment to bioretention soil media for removing pollutants due to its adsorption properties. This study evaluates biochar as an amendment to bioretention soil media for the removal of contaminants in stormwater. Columns with varying ratios of biochar to bioretention soil media (0%, 25%, and 50% biochar) were tested to evaluate the removal efficiencies of copper, zinc, phosphorus, nitrogen, and microplastics from stormwater. Five trials were conducted with stormwater collected from parking lot runoff. Results show that the addition of biochar in bioretention systems improves the removal efficiency of zinc (average of 79% in columns with 50% biochar compared to 54% in columns without biochar) and copper (average of 73% in columns with 50% biochar compared to 25% in columns without biochar). Stormwater effluent from the biochar-amended columns also had significantly lower turbidity (21 NTU) than that of the control columns (58 NTU). Consistent leaching of nitrogen and phosphorus species was evident in the control as well as the biochar-amended columns, with average removal efficiencies of $-\mathrm{1,114}\%$ for total nitrogen and $-703\%$ for total phosphorus. Microplastics were effectively removed by all columns, regardless of biochar amendment, with an average removal of 86%. This study indicates that biochar has the potential to reduce metal concentrations in stormwater when used as an amendment to bioretention systems but has a limited impact on nutrients and microplastics.

Abstract

To investigate the effect of gas activation on the low-temperature ammonia removal performance of activated carbon during nitric oxide (NO) removal, nitrogen (${\mathrm{N}}_2$), carbon dioxide (${\mathrm{CO}}_2$), and air were used as activated gases. Orthogonal experiments were designed to examine the activation conditions of each factor. The influence of various activation conditions on the physicochemical properties of activated carbon has been investigated. The interaction between various factors and their effect on the ${\mathrm{NH}}_3$ removal performance of NO were thoroughly investigated. Results show smooth pore walls and large average pore size after ${\mathrm{N}}_2$ activation. Moreover, some nitrogen-containing functional groups are introduced to the surface and help improve the NO conversion rate. ${\mathrm{CO}}_2$ activation can readily disrupt pore structures, causing surface pores to become disordered and some lignin and other functional groups to decompose; thus, the NO conversion rate by the activated carbon following ${\mathrm{CO}}_2$ activation is low. The strong oxidation of air activation causes the pore wall of activated carbon to collapse, the formation of new pores on the surface, and the introduction of oxygen-containing functional groups, so that the denitration rate increases with increasing activation temperature. The kind of activated gas has the most significant effect on the NO removal performance of activated carbon, while the activation temperature has minimal effect on the NO conversion rate of activated carbon. ${\mathrm{N}}_2$ activation at 225°C for 60 min is the best physical activation condition.

Practical Applications

Herein, we focused on three factors: activation gas; activation temperature; and activation time, which influence the nitric oxide conversion rate of activated carbon. Further, the effect of these factors and their interactions on nitric oxide removal performance were studied. Relevant research findings can be utilized as a reference for the physical activation of activated carbon and optimization of the low-temperature ${\mathrm{NH}}_3$ removal performance of activated carbon during nitric oxide removal.

Abstract

The accumulation of fat, oil, and grease (FOG) deposits in sanitary sewer systems is a global issue responsible for billions of dollars in annual maintenance costs, environmental impact, and public health concerns. Laboratory and data-driven studies have explored the physical and chemical characteristics of FOG deposits, formation kinetics, and factors influencing their accumulation in sewer networks, such as pipe sags, pipe age, and pipe material. However, a scalable modeling platform capable of incorporating FOG kinetics and the key variables driving accumulation is not currently available. This study introduces a new platform for modeling the formation and attachment of FOG deposits in sewer systems, the FOG Sewer Waste Management System (FOG-SWMS). FOG-SWMS was utilized to quantify the impact of pipe sags, pipe age and roughness, and pipe material on the accumulation of FOG deposits in sewer lines. Results indicate that under typical sewer conditions, pipe sags and pipe age may increase the accumulation of FOG deposit mass by as much as approximately 23% and 150%, respectively. Surface $\mathrm{pH}\ge 8$, as a proxy for cementitious pipe materials, also resulted in more than a 172% increase in FOG deposit mass accumulation. These results confirm data from previous laboratory studies and machine learning algorithms, and also provide a mechanistic explanation of formation processes. Two full-scale sewer network case studies, Study Area 1 (SA1) and Study Area 2 (SA2), were analyzed using FOG-SWMS, to evaluate the predictive capability of the model to identify FOG deposit accumulation zones. FOG-SWMS successfully predicted approximately 85% and 73% of known accumulation zones in SA1 and SA2, respectively. Significant changes in the spatial and temporal distribution of FOG deposits in response to new commercial and residential development also were demonstrated with the model.

Practical Applications

Fat, oil, and grease deposits are solid masses that form in sanitary sewer lines and create a variety of issues for municipalities. They are primarily the result of discharged food waste residuals into sewers, which reacts with typical wastewater components to form a solid that can block sewer lines and cause overflow of wastewater. These impacts are common, and equate to an incredible amount of time and money spent on sewer repairs and maintenance. This study introduces a new computer-based model that allows researchers to visualize the formation of FOG deposits in sewer line segments. The model is not limited to specialized researchers; it also is intended to be used by engineers and pretreatment coordinators who may be concerned with the accumulation of FOG deposits in sewer line engineering structures. The main goals of this study were (1) to provide the sewer line workforce with a tool for predicting the accumulation of FOG deposits in sewer networks, especially when there are changes in commercial establishments; (2) to serve as a platform for integrating modern sewer inspection data and sensor deployment studies that are becoming more popular; and (3) to incorporate future advancements in machine learning and AI algorithms.

Raw industrial effluent was collected from a textile dyeing industry in which 2-methylpyridine (2Mp) was used as a solvent in the dyeing and rinsing bath of cotton fabric. The wastewater was characterized and found to contain a higher 2Mp concentration than the discharge limit set by USEPA ($1\mathrm{mg}/\mathrm{L}$). High-performance liquid chromatography (HPLC) was utilized for the detection and quantification of 2Mp in the effluent. Because adsorption is a tertiary level of treatment, a pretreatment using clariflocculation (CF) was imparted to the raw effluent in order to reduce the high organic load, suspended solids, and other unknown interfering species. The CF-pretreated effluent then was subjected to both batch and column sorption using modified graphene oxide nanomaterial crosslinked with chitosan and coated onto sand (GCS) in order to evaluate its 2Mp removal efficiency. This study is a novel attempt at the removal of 2Mp from textile dyeing effluent, which was not addressed in previous studies. Carboxyl-rich graphene oxide nanomaterial was prepared using the method of prior cooling and thermal control. It was found that the 2Mp adsorption capacity of GCS was reduced by about 27.8% in the case of industrial samples compared with synthetic 2Mp samples. The suggested treatment reduced the 2Mp concentration in the industrial effluent to below the safe discharge limit of $1\mathrm{mg}/\mathrm{L}$. The effect of dissolved co-occurring ions on removal of 2Mp by sorption was studied, and it was that whereas the potassium and sulphate ions were most disruptive to the sorption of 2Mp, calcium ion aided the sorption process via complexation.

This study explored the recovery of phosphorus from human urine via electrochemical precipitation, employing magnesium alloy tailings as electrodes. The observed pH change during the process was primarily attributed to ammonium precipitation. However, the hydroxide produced by the cathode could potentially have a negative effect. There was a strong correlation between the amount of precipitate produced and pH, which can serve as a useful parameter for monitoring the effectiveness of the treatment. However, electrode corrosion significantly influenced the purity of the resultant struvite product. To counteract this, a device was designed to separate struvite from the electrode corrosion products using fluid flow, which led to an increase in purity. The pilot-scale application of magnesium electrodes proved to be an effective method for phosphorus recovery, achieving a high removal rate through adjustments to the residence time and current density. However, the limited source of wastewater resulted in a struvite production value that was considerably below the maintenance costs. Hence, the economic feasibility of this process necessitates further enhancement.