This study investigates the use of machine learning based image classification techniques to detect debris blocking of urban waterways. Using a dataset comprising 1089 labelled CCTV images of a trash screen located in Cardiff, UK and a comprehensive re-sampling approach, we investigate not only the ability of selected machine learning algorithms to correctly identify images, but also to evaluate the uncertainty of these algorithms conditional on the datasets presented to them. For each candidate model, we considered two datasets: an imbalanced dataset and an under-sampled dataset. The results demonstrate that the performance of a simple logistic regression model was broadly comparable to that of more advanced machine learning models such as vision transformers. The best performing models (vision transformers and logistic regression) achieved an accuracy of more than 80%, while the NetRes50 model achieved an accuracy in the low 70%. This is an important result that opens the possibility for implementing these techniques as part of an operational real-time flood warning system utilising already existing cameras.

Ensuring easy access to clean and safe drinking water using low-cost technology is essential to mitigate the rising water scarcity in emerging economies. Commercial large-scale desalination technologies need significant investment, making them unsuitable for off-grid and small-scale applications. However, this operation can be carried out using a low-cost desalination technology based on renewable energy, known as the solar still. In this research work, a modified basin solar still (basin solar still + internal mirrors + 8 kg gravel + black ink (400 ppm per litre)) was developed and experimentally tested in Visakhapatnam (17.68°N, 83.22°E), India, to determine its appropriateness for sustainable seawater desalination. It produced 14% to 23% more desalinated water than a conventional basin solar still. In addition, its thermal efficiency was between 41% and 42%, which was significantly greater than other basin solar stills reported in literature. In addition, high-quality desalinated water was generated at a cost that was around three times less than the drinking water offered at Indian Railways kiosks. Moreover, the ability to mitigate significant CO2 emissions while also addressing water scarcity demonstrated that the modified basin solar still continues to contribute effectively to the United Nations Sustainable Development Goal 6 (Clean Water and Sanitation).

Sustainable Development Goal 6.1 seeks universal access to safe drinking water for all by 2030, yet persistent disparities remain even in high-income countries. Indigenous, remote and small communities are disproportionately affected by poor drinking water quality, but comparable evidence to evaluate performance across communities is very limited due to inconsistent monitoring and reporting. To this end, we constructed a community-level meta-panel dataset of 839 communities (4,137 observations) across 4 Australian jurisdictions (Northern Territory, South Australia, Victoria and Western Australia) and Ontario, Canada, over the period 2018–2022. Drinking water quality was assessed using the Australian Drinking Water Guidelines and Canadian Boil Water Advisories. Logistic regression was employed to estimate the probability of accessing good-quality drinking water, with Indigenous status, remoteness, population size and socio-economic condition as key explanatory variables. Results reveal systematic disparities: Indigenous and very remote communities are statistically significantly less likely to have good-quality drinking water than non-Indigenous and regional communities after controlling for other factors. Our findings indicate that structural inequities – rather than geographic or demographic variation alone – are critical determinants of poor drinking water outcomes in small, Indigenous communities in both Australia and Canada.

Massive open online courses (MOOCs) have emerged as powerful educational tools for disseminating complex scientific knowledge. This study examines the RECONECT MOOC, an innovative online learning platform designed to educate stakeholders about nature-based solutions (NbS) for hydrometeorological risk reduction. Developed by a multidisciplinary team of international experts, the course drew on diverse professional and geographical perspectives, across a range of national and sectoral contexts. Its primary aim to bridge the knowledge gap in climate adaptation strategies across diverse professional sectors. The research investigated the effectiveness of interactive, multimedia-based learning approaches in communicating advanced NbS concepts. Using the Sharable Content Object Reference Model (SCORM) framework, the course integrated multiple learning methods, including reading, listening, inquiry-based learning, discussion forums and practical application exercises. The MOOC was divided into six modules and targeted local authorities, policymakers, academics, postgraduate students and other stakeholders interested in climate resilience. Findings revealed that 90% of participants gained additional NbS knowledge, 69% reported a shift in perception and 62% felt confident applying what they learned. Participants valued the interactive resources and flexible, self-paced format. Overall, the study demonstrates that MOOCs can effectively support scientific knowledge dissemination and strengthen understanding of complex environmental solutions.

Extreme weather events, combined with human-induced factors, such as expanding impervious surfaces and inadequate drainage infrastructure, are driving escalating urban flood risks worldwide. In this study, we present a novel spatiotemporal Long Short-Term Memory (LSTM)-based surrogate of the U.S. Environmental Protection Agency (EPA)’s Storm Water Management Model (SWMM) to predict maximum water depth and inflow at the asset level within urban drainage networks. The high-resolution SWMM model, encompassing the full network of conduits and manholes, was first calibrated and validated using U.S. Geological Survey (USGS) observations. The LSTM surrogate was then trained on data from 5,000 rainfall events across seven Annual Recurrence Intervals (ARIs) ranging from 1 to 100 years. The SWMM-LSTM surrogate model consistently achieves high predictive performance for both water depth and inflow, highlighting its robustness across diverse storm scenarios and ARI conditions. Hyperparameter optimization via grid search revealed task-specific configurations: larger hidden layers with moderate dropout improved water depth predictions, while deeper network architectures with minimal dropout optimized inflow forecasts. By providing rapid, computationally efficient predictions without compromising accuracy, the SWMM-LSTM surrogate offers a practical tool for real-time flood risk assessment, scenario evaluation and actionable decision-making in complex urban drainage systems.

A stakeholder structured engagement process at the Sustainable Water Infrastructure Management (SWIM) conference and workshop was held in December 2024. The participants identified critical current and future issues facing the water sector that are synthesized in this paper. In particular, they highlighted issues of water systems’ vulnerability and lack of resilience to hazards and stressors; inequities associated with water scarcity; and water quality problems – all affected by natural or man-made influences. The Smart One Water (S1W) vision was the baseline for the SWIM 2024 conference. This paper expands the S1W vision with a synthesis of the conference discussions about S1W-related fundamental concepts, practices and implementation barriers. It includes initial recommendations – based on a digital, data-focused, stakeholder-driven approach – with expert representatives of the public and private water supply sectors, academia, government and policymakers tasked to generate real-world adaptable ideas and practical solutions. Specifically, S1W envisions a future where water management and governance silos are eliminated to provide the necessary collaboration to enable efficient, resilient, affordable and equitable water access capable of adapting to a changing environment. This would be a future where communities govern collaboratively through integrated decision-making on policy, management and funding of natural and engineered water systems at the river basin scale.

Environmental modeling is a powerful tool for facilitating discussions among stakeholders involved in decision-making for environmental planning. This article explores how models can be used to structure stakeholder discussions by visualizing the opportunity space – the range of interventions that are deemed to be technically and physically feasible – and how it overlaps (or does not) with the decision space – the set of interventions considered acceptable or desirable by stakeholders. Using the case study of the Aa of Weerijs catchment in the Netherlands and Belgium, we demonstrate how different model contributions (sensitivity analysis, impact assessment and scenario evaluations at catchment and local levels) can aid these discussions by providing structured insights into potential interventions, both within the opportunity and decision spaces. The findings highlight how models can bridge gaps between stakeholders and technical experts by facilitating more effective discussions and better-informed decision-making processes in environmental planning.

Future water demand modeling is of crucial importance for stakeholders, particularly in the era of rapidly changing climate and socioeconomic conditions. The modeling results can be applied to develop effective adaptation strategies that ensure equitable and sustainable allocation of water to various economic sectors, including institutional, commercial, industrial (ICI), residential and agricultural. However, a comprehensive review of existing future water demand modeling methods that consider both climatic and socioeconomic factors as well as the major economic sectors is currently lacking. This review article contributes to fill this knowledge gap while introducing a more streamlined and comprehensive methodological approach for conducting literature reviews in the environmental sciences domain. At the core of this method is a new framework designed to support research questions formulation and literature search strategies named STAR (Systems, Trouble/Treatment, Alternative, Response). In addition, it presents a data-requirement-based metric as well as a new nomenclature for classification of surveyed methods and approaches to guide the selection process of future water demand modeling methods. Furthermore, it proposes a hybrid modeling approach made up of three components (computational intelligence, dynamic systems and probabilistic scenarios) in the form of a theoretical workflow for future water demand modeling. The proposed workflow ensures broad applicability, making it adaptable not only to water demand management but also to a wide range of challenges across the environmental sciences.