Deep learning in water protection of resources, environment, and ecology: achievement and challenges.

The breathtaking economic development put a heavy toll on ecology, especially on water pollution. Efficient water resource management has a long-term influence on the sustainable development of the economy and society. Economic development and ecology preservation are tangled together, and the growth of one is not possible without the other. Deep learning (DL) is ubiquitous in autonomous driving, medical imaging, speech recognition, etc. The spectacular success of deep learning comes from its power of richer representation of data. In view of the bright prospects of DL, this review comprehensively focuses on the development of DL applications in water resources management, water environment protection, and water ecology. First, the concept and modeling steps of DL are briefly introduced, including data preparation, algorithm selection, and model evaluation. Finally, the advantages and disadvantages of commonly used algorithms are analyzed according to their structures and mechanisms, and recommendations on the selection of DL algorithms for different studies, as well as prospects for the application and development of DL in water science are proposed. This review provides references for solving a wider range of water-related problems and brings further insights into the intelligent development of water science.

Machine learning predicts which rivers, streams, and wetlands the Clean Water Act regulates.

We assess which waters the Clean Water Act protects and how Supreme Court and White House rules change this regulation. We train a deep learning model using aerial imagery and geophysical data to predict 150,000 jurisdictional determinations from the Army Corps of Engineers, each deciding regulation for one water resource. Under a 2006 Supreme Court ruling, the Clean Water Act protects two-thirds of US streams and more than half of wetlands; under a 2020 White House rule, it protects less than half of streams and a fourth of wetlands, implying deregulation of 690,000 stream miles, 35 million wetland acres, and 30% of waters around drinking-water sources. Our framework can support permitting, policy design, and use of machine learning in regulatory implementation problems.

Building collaboration and synergy among regional blocs to reduce marine plastic waste pollution: A case for the Gulf of Guinea region.

The Gulf of Guinea (GoG) region in sub-Saharan Africa is a vast and diverse region stretching from Senegal to Angola, covering approximately 6000 km of coastline. It is an important shipping zone transporting oil and gas, as well as goods to and from central and southern Africa. Plastic waste has an impact on the major metropolitan centers of this region, as well as the beaches. The Gulf of Guinea region includes the coastal areas of countries such as Nigeria, Cameroon, Equatorial Guinea, Gabon, Sao Tome and Principe, the Republic of the Congo, the Democratic Republic of the Congo, Ghana, Cote d'Ivoire (Ivory Coast), Benin, and Togo. This study uses qualitative methodology to comparatively interrogate how to use legal instruments to reduce marine plastic pollution in the Gulf of Guinea region, which encompasses these coastal nations.

River water quality management using an integrated multi-objective optimization-simulation approach based on bankruptcy rules.

The aim of this research is to allocate the river's self-purification (acceptance capacity of pollution) fairly between the beneficiaries (pollutant sources) using bankruptcy theory. For this purpose, four bankruptcy rules (CAE, CEL, P, and TAL) were called using the link of the water quality simulation model (QULA2Kw) to an evolutionary optimization algorithm (multi-objective imperialist competition algorithm (MOICA)). The objective functions were reducing polluters' wastewater treatment costs and preventing biochemical oxygen demand (BOD) violations of the standard level along the river. The applicability of the approach is demonstrated by the case study that was carried out on the Dez River in Iran. According to the results, the CEL scenario is the most effective method for the Dez River when taking into account the most optimal state for both objective functions (selecting the best compromise solution from the Pareto front). This is because it has the lowest violation value of the standard level for BOD along the river when compared to other scenarios. Alternatively, when considering Solution 20, which focuses on the maximum cost of treating the polluters while staying within the acceptable level of pollution in the river, the results indicated that the CEA rule emerged as the most favorable option. This is due to its lower treatment cost (156.9 (1000$)) and higher pollution discharge to the river (681.91 g/s).

Applying water quality standards to pollution from diffuse sources.

Water quality standards are instrumental in evaluating the status of water bodies, and in providing protective and restorative endpoints. To date, much of the infrastructure used to implement water quality standards has been directed towards remediating and managing pollution from point source discharges. However, pollution from diffuse sources is the leading cause of water quality impairment, especially by nutrients. Although the effects of nutrient enrichment on streams is well studied, and ecological thresholds identified, those thresholds have not been widely adopted as standards primarily because they are not attainable by point sources. Clearly, a framework for adopting and applying standards to manage pollution from diffuse sources needs to be decoupled from those intended for point sources. This paper argues for a relatively unstructured distributional approach to predict how ecological responses might shift in response to management of diffuse sources. The approach calls for first developing a deterministic model of stressor and response variables, followed by a reformulation as a Bayesian model. In the case here, a structural equation model was developed that linked nutrient enrichment, habitat quality, and chloride and manganese concentrations to an index of macroinvertebrate quality. Results from the Bayesian representation suggest that in landscapes where the drainage network has been highly modified for agricultural production, reduction in total phosphorus alone is expected to have a modest (but non-trivial) effect on macroinvertebrate condition, shifting the distribution of scores up by 1 point. The addition of habitat restoration is likely to shift the distribution upwards by 4 points, an effect size observed in Ohio, USA from other large-scale restoration efforts.

Research on water quality improvement of plain irrigation area based on multi-scenario simulation.

Water diversion projects have proven to be effective interventions to improve water quality in irrigation ditches. This study focused on quantifying the water quality improvement by utilizing a hydrodynamic water quality model in Funing County, Yancheng City. The model performed a spatial analysis of pollution concentrations across the study area. Various optimization scenarios were designed based on the diversion project and hydrological structure connectivity. The model was used to simulate changes in nutrient concentrations under different scenarios. The findings of this study were as follows: (1) Rural areas had lower nutrient concentrations and superior hydrological connectivity than urban areas. (2) The effect of water quality improvement correlated positively with increased flow rates introduced by the diversion project. Specifically, when the flow rate increased by 50%, the average reductions were 20% for NH4+, 5.2% for TN, and 5.1% for TP. Furthermore, introduced clean water led to more pronounced improvements in the overall regional water quality. (3) Although increasing the number of ditches improved water pollution concentration, the impact was not significant. (4) Model simulation results showed that 18 to 45% water diversion intensity effectively improved water quality, and the optimal water diversion intensity was 27 to 30%. The optimal water diversion intensities offered valuable insights for managing this region. The study's methods contributed to the promotion of sustainable development in regional water resources and the integrated management of the water environment.

Exploring the contribution of the river chief system on controlling industrial water pollution under quasi-natural experimental conditions.

The innovative environmental system known as the river chief system (RCS) was developed in China to limit water pollution by appointing provincial, prefectural, county, and township officials as "river chiefs" as guardians of every river under their control. Our research project developed a quasi-natural experiment to investigate if the RCS effectively reduces the intensity of industrial water pollution. A multi-period difference-in-difference (DID) method was employed in the experiment with a total of 3808 samples from 272 prefecture-level cities. The samples gathered represented the years from 2007 to 2020. Our results demonstrated that the southern region has far more industrial water contamination than other locations, and the intensity of industrial sewage discharge decreased by 10.25% in pilot cities following the implementation of the RCS compared with what we observed in non-pilot cities. Furthermore, according to the mechanism analysis we used, cities that increased investment in technical advancement and modernized industrial structures may also account for this decreasing trend. Additionally, the results of geographical heterogeneity demonstrate that the RCS's influence on curbing industrial water pollution is greater in cities with a higher intensity of environmental regulation and higher fiscal balance pressure than in locations with low environmental regulation and low fiscal balance pressure. In summary, the Chinese government needs to continue to strengthen the RCS's implementation; this can be done by adjusting to local circumstances as needed and by carrying out the policy accurately and effectively across the country. Our research sheds light on how to prevent water pollution in China and in other countries and supports the positive effects and effectiveness of RCS.

Assessing water quality in the Dong Nai River (Vietnam): implications for sustainable management and pollution control.

Dong Nai River provides essential water resources for millions of people across 11 provinces and cities in Vietnam. However, the different pollution sources such as household, farming, and industrial operations have caused the river water quality to deteriorate over the past decade. To gain a comprehensive understanding of the river's surface water quality, this study employed the water quality index (WQI) across 12 different sampling sites. In total,144 water samples with 11 parameters were analyzed in accordance with the Vietnamese standard 08:2015/MONRE. Results revealed a range of surface water quality, from poor to good according to the VN-WQI (Vietnamese standard), and a medium even bad level in some months according to the NS-WQI (American standard). The study also identified temperature, coliform, and dissolved oxygen (DO) as strong contributors to WQI values (VN_WQI standard). Principal component analysis/factor analysis was used to determine pollution sources, with the results highlighting agricultural and domestic activities as the main contributors to river pollution. In conclusion, this study underscores the importance of effective planning and management of infrastructure zoning and local activities to improve the river's surface water quality and surrounding areas, as well as safeguard the well-being of the millions who depend on it.

Pollution source identification and abatement for water quality sections in Huangshui River basin, China.

Accurately obtaining the pollution sources and their contribution rates is the basis for refining watershed management. Although many source analysis methods have been proposed, a systematic framework for watershed management is still lacking, including the complete process of pollution source identification to control. We proposed a framework for identification and abatement of pollutants and applied in the Huangshui River Basin. A newer contaminant flux variation method based on a one-dimensional river water quality model was used to calculate the contribution of pollutants. The contributions of various factors to the over-standard parameters of water quality sections at different spatial and temporal scales were calculated. Based on the calculation results, corresponding pollution abatement projects were developed, and the effectiveness of the projects was evaluated through scenario simulation. Our results showed that the large scale livestock and poultry farms and sewage treatment plants were the largest sources of total nitrogen (TP) in Xiaoxia bridge section, with contribution rates of 46.02% and 36.74%, respectively. Additionally, the largest contribution sources of ammonia nitrogen (NH3-N) were sewage treatment plants (36.17%) and industrial sewage (26.33%). Three towns that contributed the most to TP were Lejiawan Town (14.4%), Ganhetan Town (7.3%) and Handong Hui Nationality town (6.6%), while NH3-N mainly from the Lejiawan Town (15.9%), Xinghai Road Sub-district (12.4%) and Mafang Sub-district (9.5%). Further analysis found that point sources in these towns were the main contributor to TP and NH3-N. Accordingly, we developed abatement projects for point sources. Scenario simulation indicated that the TP and NH3-N could be significantly improved by closing down and upgrading relevant sewage treatment plants and building facilities for large scale livestock and poultry farms. The framework adopted in this study can accurately identify pollution sources and evaluate the effectiveness of pollution abatement projects, which is conducive to the refined water environment management.

Nonpoint source pollution measures in the Clean Water Act have no detectable impact on decadal trends in nutrient concentrations in U.S. inland waters.

The Clean Water Act (CWA) of 1972 regulates water quality in U.S. inland waters under a system of cooperative federalism in which states are delegated implementation and enforcement authority of CWA provisions by the U.S. Environmental Protection Agency. We leveraged heterogeneity in state implementation of the CWA to evaluate the efficacy of its nonpoint source provisions in reducing nutrient pollution, the leading cause of water quality impairment in U.S. inland waters. We used national survey data to estimate changes in nutrient concentrations over a decade and evaluated the effect of state-level policy implementation. We found no evidence to support an effect of (i) grant spending on nonpoint source pollution remediation, (ii) nutrient criteria development, or (iii) water quality monitoring intensity on 10-year trends in nutrient concentrations. These results suggest that the current federal policy paradigm for improving water quality is not creating desired outcomes.

Quantifying uncertainty of marine water quality forecasts for environmental management using a dynamic multi-factor analysis and multi-resolution ensemble approach.

Unpredictable climate change and human activities pose enormous challenges to assessing the water quality components in the marine environment. Accurately quantifying the uncertainty of water quality forecasts can help decision-makers implement more scientific water pollution management strategies. This work introduces a new method of uncertainty quantification driven by point prediction for solving the engineering problem of water quality forecasting under the influence of complex environmental factors. The constructed multi-factor correlation analysis system can dynamically adjust the combined weight of environmental indicators according to the performance, thereby increasing the interpretability of data fusion. The designed singular spectrum analysis is utilized to reduce the volatility of the original water quality data. The real-time decomposition technique cleverly avoids the problem of data leakage. The multi-resolution-multi-objective optimization ensemble method is adopted to absorb the characteristics of different resolution data, so as to mine deeper potential information. Experimental studies are conducted using 6 actual water quality high-resolution signals with 21,600 sampling points from the Pacific islands and corresponding low-resolution signals with 900 sampling points, including temperature, salinity, turbidity, chlorophyll, dissolved oxygen, and oxygen saturation. The results illustrate that the model is superior to the existing model in quantifying the uncertainty of water quality prediction.

Strategy for cost-effective BMPs of non-point source pollution in the small agricultural watershed of Poyang Lake: A case study of the Zhuxi River.

In recent years, Poyang Lake has been affected by severe agricultural non-point source (NPS) pollution, a global water pollution problem. The most recognized and effective control measure for agricultural NPS pollution is the strategic selection and placement of best management practices (BMPs) for critical source areas (CSAs). The present study employed the Soil and Water Assessment Tool (SWAT) model to identify CSAs and evaluate the effectiveness of different BMPs in reducing agricultural NPS pollutants in the typical sub-watersheds of the Poyang Lake watershed. The model performed well and satisfactorily in simulating the streamflow and sediment yield at the outlet of the Zhuxi River watershed. The results indicated that urbanization-oriented development strategies and the Grain for Green program (returning the grain plots to forestry) had certain effects on the land-use structure. The proportion of cropland in the study area decreased from 61.45% (2010) to 7.48% (2018) in response to the Grain for Green program, which was mainly converted to forest land (58.7%) and settlements (36.8%). Land-use type changes alter the occurrence of runoff and sediment, which further affect the nitrogen (N) and phosphorus (P) loads since sediment load intensity is a key factor affecting the P load intensity. Vegetation buffer strips (VBSs) proved the most effective BMPs for NPS pollutant reduction, and the cost of 5-m VBSs proved the lowest. The effectiveness of each BMP in reducing N/P load ranked as follows: VBS > grassed river channels (GRC) > 20% fertilizer reduction (FR20) > no-tillage (NT) > 10% fertilizer reduction (FR10). All combined BMPs had higher N and P removal efficiencies than the individual measures. We recommend combining FR20 and VBS-5m or NT and VBS-5m, which could achieve nearly 60% pollutant removal. Depending on the site conditions, the choice between FR20+VBS and NT + VBS is flexible for targeted implementation. Our findings may contribute to the effective implementation of BMPs in the Poyang Lake watershed and provide theoretical support and practical guidance for agricultural authorities to perform and direct agricultural NPS pollution prevention and control.

Drinking Water From Private Wells and Risks to Children.

Drinking water for approximately 23 million US households is obtained from private wells. These wells can become contaminated by pollutant chemicals or pathogenic organisms, leading to significant illness. Although the US Environmental Protection Agency and all states offer guidance for construction, maintenance, and testing of private wells, most states only regulate the construction of new private water wells. With a few exceptions, there is little regulation after construction. Well owners are responsible for their own wells. Children may also drink well water at child care or when traveling. Illness resulting from children's ingestion of contaminated water can be severe. This report reviews relevant aspects of groundwater and wells; describes the common chemical and microbiologic contaminants; gives an algorithm with recommendations for inspection, testing, and remediation for wells providing drinking water for children; and provides references and Internet resources for more information.

Multi-objective optimization of hydrant flushing in a water distribution system using a fast hybrid technique.

As a critical element in preserving the health of urban populations, water distribution systems (WDSs) must be ready to implement emergency plans when catastrophic events such as contamination events occur. A risk-based simulation-optimization framework (EPANET-NSGA-III) combined with a decision support model (GMCR) is proposed in this study to determine optimal locations for contaminant flushing hydrants under an array of potentially hazardous scenarios. Risk-based analysis using Conditional Value-at-Risk (CVaR)-based objectives can address uncertainties regarding the mode of WDS contamination, thereby providing a robust plan to minimize the associated risks at a 95% confidence level. Conflict modeling by GMCR achieved an optimal compromise solution within the Pareto front by identifying a final stable consensus among the decision-makers involved. A novel hybrid contamination event grouping-parallel water quality simulation technique was incorporated into the integrated model to reduce model runtime, the main deterrent in optimization-based methods. The nearly 80% reduction in model runtime made the proposed model a viable solution for online simulation-optimization problems. The framework's capacity to address real-world problems was evaluated for the WDS operating in Lamerd, a city in Fars Province, Iran. Results showed that the proposed framework was capable of highlighting a single flushing strategy, which not only optimally reduced risks associated with contamination events, but provided acceptable coverage against such threats, flushing 35-61.3% of input contamination mass on average, and reducing average time-to-return to normal conditions by 14.4-60.2%, while employing less than half of the initial potential hydrants.

A New Policy of Water Resources and Environmental Regulation in China.

As a developing country, China is facing serious water pollution and scarcity, which indicates the need for integrated water-resource and environmental regulations. Zoning policies have undergone significant advancements to enhance water-resource utilization in China. However, conflicts and overlaps still exist among these policies. To integrate these zoning policies and regulations, the "Three Lines One Permit" (TLOP) water-environment policy was formulated as a new framework, which included the goal for water quality, upper limits on water-resource utilization, and a permit list. This study presents the main achievements of the TLOP as a case-study in Jinan. The territories of Jinan were divided into 158 water-environment control-units (WECUs) and classified into two types of protected zones, three types of pollution-control zones, and ordinary zones. The total maximum pollutant-loads in the 158 WECUs, and 138 townships were calculated. The water-resource-utilization indicator values and ecological demand of key rivers were specified. The permit lists for the water environment at macroscale, mesoscale, and microscale were compiled from four perspectives: spatial constraints, emissions control, risk prevention, and resource utilization. Finally, suggestions were proposed to promote a more scientific and efficient TLOP policy to enhance human-water harmony.

Willingness to Pay to Reduce the Ecological and Health Risks from Water Pollution in Emerging Urban Regions: The Case of the Red River Delta in Vietnam.

The soaring economic development of export activities of handicrafts centralized in emerging urban regions in Vietnam has accelerated the increase in the occurrence of diseases and threats to ecosystems induced by water pollution. We design a discrete choice experiment to elicit the willingness-to-pay of handicraft enterprises to restore the environment and diminish health risks from polluted wastewater through water quality improvement under different scenarios. Estimates from five latent classes reveal that one-half of entrepreneurs strongly value the provision of wastewater treatment services, and their decisions are mostly driven by preferences to reduce the risk of sickness caused by water pollution. This finding lends support to the argument that self-interested preferences predominate pro-environmental behavior in the readiness to pay for water quality services. While entrepreneurs' preferences attributed to ecological remediation seem to vary according to their educational background, the status-quo group shows low environmental awareness. This divergent behavioral pattern suggests that the design of wastewater management policies requires a mixture of measures that aim at different groups of individuals pursuing economic incentives and the creation of awareness.

Design and Application of an Early Warning and Emergency Response System in the Transboundary Area of the Taihu Lake Basin.

The inter-provincial transboundary area of the Taihu Lake Basin is characterized by a complex river network and reciprocating flow. Frequent environmental pollution events in recent years have become a major safety hazard for the water quality in the Taihu Lake Basin. There are few early warning systems for environmental pollution events in China, the ability to simulate risk is insufficient, and systematic research on technology, development, and application is lacking. Thus, water management requirements are not met in the inter-provincial transboundary area of the Taihu Lake Basin. This paper proposes a cross-border risk management plan for pollution sources in the transboundary areas of the Taihu Lake Basin and an early warning and emergency response system for water pollution events using modern information technology. We used this system to assess and classify 2713 risk sources for nitrogen and phosphorus pollution into 5 categories. We simulated the discharge of a pollutant into a tributary and the early warning and emergency response for the transboundary region. The results indicate that the proposed early warning and emergency response system substantially improved the transboundary water environment and lowered the risk of pollution in the Taihu Lake watershed.

Where to measure water quality ? Application to nitrogen pollution in a catchment in France.

Information on the water quality of rivers can be used to judge the effectiveness of past policies or to guide future environmental policies. Consequently, the location of water quality monitoring stations (WQMSs) plays an important role in river pollution control. In the 2000s, a literature developed on the optimization of WQMS location to identify pollution hot spots, average quality, or to minimize the detection time of a potential source of accidental pollution. This article is part of a new literature aimed at locating WQMSs in order to optimize the economic value of information (EVOI) generated by water quality monitoring networks (WQMNs). The field of study is a catchment in northeastern France where the purpose of quality measurement is to define a policy of reduction of agricultural nitrogen fertilizers in order to reach the standard of 50 mg/l of nitrate at the WQMS. Agro-hydrological and economic models estimate the net benefit of input reduction depending on the location of the WQMS on the basis of different assumptions concerning the ecological damage generated by nitrate. We show that the magnitude of the ecological damage and, consequently, the perception of the contamination generated by nitrate in water, play a decisive role on the optimal location of the WQMS, as well as on the benefit of the economic optimization of locations, compared to traditional optimization. Locating WQMSs in a way that maximizes EVOI will be more attractive for very high or very low levels of damage. However, in this context, linking damage to nitrate concentration or to concentration coupled with riparian population density alone will have little impact.

Evaluation of the costs of agricultural diffuse water pollution abatement in the context of Lithuania's water protection goals and climate change.

This study aimed at evaluating the scale and costs of an environmentally and economically optimal set of Best Management Practices (BMPs) for agricultural pollution abatement in Lithuania in order to reach water protection goals in both inland and marine waters by distributing BMPs optimally in space, while taking climate change impacts into consideration. The assessment of BMPs impact involved the use of the SWAT model by applying two climate change representative concentration pathways (RCP4.5 and RCP8.5) and two time horizons (mid-century and end-century), as well as five BMPs (arable land conversion to grasslands, reduced fertilization, no-till farming, catch-crops, and stubble fields throughout winter). The optimization of the set of BMPs employed a genetic algorithm. The results suggest that the need for BMPs application will increase from 52% of agricultural areas in the historical period up to 65% by the end of century in the RCP8.5 scenario. This means less arable land could actually be used for crop production in the future if water protection targets are met. The high costs for reaching water targets would rise even more, i.e. by 173% for RCP4.5, and by 220% for the RCP8.5 scenario, reaching approximately 200 million euros/year. In such a context, the BMP optimization approach is essential for significant reduction of the costs. Winter cover crops and reduced fertilization show the best effectiveness and cost balance, and will therefore be essential in pursuing water protection targets.