Explore the latest insights from top science journals in the Muser Press daily roundup (September 1, 2025), featuring impactful research on climate change challenges.
In brief:
Revolutionizing biodiesel: how deep learning is transforming sustainable fuel production
As the world grapples with climate change and dwindling fossil fuel reserves, biodiesel emerges as a promising renewable alternative to conventional diesel. However, the journey toward sustainable biodiesel production faces significant hurdles, particularly in selecting the right feedstocks that don’t compete with food supplies.
A groundbreaking comprehensive review reveals how artificial neural networks (ANNs) and deep learning technologies are revolutionizing this field, offering unprecedented solutions to longstanding challenges.
Traditional biodiesel production relies heavily on edible crops like soybean, palm oil, and rapeseed – creating a problematic “food versus fuel” competition. With fossil fuels still accounting for 88% of global energy consumption, the urgency to develop sustainable alternatives has never been greater. Second-generation biodiesel, derived from non-edible sources such as algae and jatropha, presents an attractive solution but faces obstacles including high production costs and limited commercial viability. This is where deep learning enters the picture, offering a transformative approach to feedstock selection and production optimization.
The research demonstrates remarkable achievements through deep learning applications. ANNs have shown superior predictive accuracy compared to traditional statistical methods, with some models achieving R² values exceeding 90% in predicting crucial biodiesel properties like kinematic viscosity and cetane numbers. These neural networks excel at analyzing complex relationships between feedstock characteristics, production parameters, and environmental factors, enabling rapid assessment of diverse feedstock options without extensive experimental testing.
Particularly impressive are the results from hybrid deep learning models that combine generative and discriminative approaches. For instance, researchers using genetic algorithm-based support vector machines (GA-SVM) successfully optimized biodiesel production from waste cooking oil, while others achieved significant yield improvements by integrating ANNs with response surface methodology (RSM). These advances translate to substantial time and cost savings – critical factors for commercial viability.
The integration of deep learning with Internet of Things (IoT) technology promises to revolutionize biofuel production further. Real-time monitoring and optimization through IoT sensors combined with predictive modeling enable unprecedented control over production processes. This synergy allows manufacturers to adapt quickly to varying feedstock qualities and market demands while maintaining optimal efficiency.
Future applications include developing comprehensive ANN models applicable across diverse engine types and fuel variations, enhancing transferability between different geographical regions and feedstock types. The potential for multi-omics integration and advanced data augmentation techniques will address current limitations in dataset size and model generalization, opening doors to previously unexplored feedstock sources.
This comprehensive review underscores deep learning’s pivotal role in accelerating biodiesel development as a sustainable fuel alternative. By dramatically reducing the time and resources needed for feedstock evaluation and process optimization, ANNs are breaking down barriers that have long hindered biodiesel’s commercial expansion. The technology’s ability to uncover hidden correlations within complex datasets not only advances scientific understanding but also paves the way for more efficient, environmentally friendly fuel production.
As we stand at the intersection of artificial intelligence (AI) and renewable energy, the innovative application of deep learning in biodiesel production represents more than technological advancement – it embodies our commitment to a sustainable future where renewable fuels can effectively compete with and eventually replace fossil fuels.
Journal Reference:
Olugbenga Akande, Jude A. Okolie, Richard Kimera, Chukwuma C. Ogbaga, ‘A comprehensive review on deep learning applications in advancing biodiesel feedstock selection and production processes’, Green Energy and Intelligent Transportation 4, 3: 100260 (2025). DOI: 10.1016/j.geits.2025.100260
Article Source:
Press Release/Material by Beijing Institute of Technology Press Co., Ltd
‘Peak water security’ crisis, Texas A&M researcher warns
As the United States passes a tipping point in water security, new research reveals that millions of Americans now face a growing crisis in accessing clean, affordable water.
The findings, published in PLOS Water and PLOS One, were produced by a multi-university team co-led by Dr. Wendy Jepson, professor of geography and director of Environmental Programs at Texas A&M University.
“Our research shows water insecurity in the U.S. is not just a problem of pipes and infrastructure – it’s a human issue that affects health, daily life and dignity,” Jepson said. “Even in the wealthiest country, millions face challenges getting safe and affordable water, often without anyone realizing it.”
A call for immediate water reform
The research team calls on utility industries, public agencies and policymakers to recognize the scope of the crisis, and reform water management approaches.
“Our goal was to bring water insecurity out of the shadows so decision-makers could build equitable, sustainable water systems for all Americans,” Jepson said.
The team emphasizes that addressing the water crisis requires more than fixing pipes; it demands that policies treat water as a basic human need and that they prioritize the needs of those most affected.
The triple threat behind the water crisis
The studies outline how a “triple threat” of degrading infrastructure, accelerating climate change and sluggish or inadequate policy responses have pushed the U.S. past a critical point of clean and clear access to water – what the researchers call “peak water security.”
This triple threat disproportionately impacts low-income households and historically marginalized communities, which face higher rates of water contamination, shutoffs and exclusion from infrastructure improvements.
“We know water insecurity exists in the U.S.,” said Dr. Amber Pearson, co-author and associate professor at Michigan State University. “But we’ve lacked the right tools to measure it.”
A new tool to track America’s water crisis
To measure and track the crisis, the researchers introduced a new tool: the Household Water Insecurity Experiences (HWISE), a survey-based measurement originally created for lower-income countries but scaled to the U.S. context.
The tool uses data from more than 1,000 households in over 15 at-risk communities across 2,770 Americans. Using the tool, the researchers are evaluating how well it predicts real-world outcomes and metrics like reliance bottled water and stress related to water access.
While the study is ongoing, the researchers believe the tool will have major implications for targeted infrastructural investments, integrated public health efforts and strategies aimed at closing the water equity gap.
“This scale will help us understand the real, everyday struggles families face and guide more fair policies and investments,” Pearson said.
Journal Reference:
Jepson W, Wutich A, Pearson AL, Beresford M, Brewis A, Cooperman A, et al., ‘Beyond peak water security: Household-scale experiential metrics can offer new perspectives on contemporary water challenges in the United States’, PLOS Water 4 (8): e0000413 (2025). DOI: 10.1371/journal.pwat.0000413
Pearson AL, Jepson W, Brewis A, Osborne-Gowey J, Wutich A, Beresford M, et al., ‘A protocol for the development of a validated scale of household water insecurity in the United States: HWISE-USA’, PLoS One 20 (8): e0330087 (2025). DOI: 10.1371/journal.pone.0330087
Article Source:
Press Release/Material by Zaid Elayyan | Texas A&M University (TAMU)
AI model maps building emissions to support fairer climate policies
An open-source artificial intelligence model to accurately map the carbon emissions of buildings across multiple cities could become a powerful new tool to help policymakers plan targeted and equitable decarbonisation strategies.
The model, developed by researchers at the College of Design and Engineering (CDE) at the National University of Singapore (NUS), offers city planners a detailed picture of how building carbon emissions are distributed and what drives them, with a view to helping authorities design smarter, fairer strategies to cut emissions.
The model is the result of research led by Assistant Professor Filip Biljecki from the Department of Architecture at CDE. The team’s findings were published in the journal Nature Sustainability.
“Our model estimates operational carbon emissions of individual buildings at the scale of entire cities,” said Department of Architecture PhD student Winston Yap, Lead Author of the study.
“Unlike previous approaches that rely on proprietary data, our open approach is designed to be transferable across cities, including those with different data availability conditions.”
Applied to data mapping over half a million buildings in five cities – Singapore, Melbourne, New York City (Manhattan), Seattle and Washington DC – the researchers say their model explained up to 78 per cent of the variation in emissions. The results revealed significant differences in how emissions are distributed within cities and identified key factors that influence building energy use, including urban form, planning history, and income levels.
“Building emissions are not just about size or density, they’re deeply shaped by the unique context of each city, from its planning legacy to climate and economic conditions,” said Asst Prof Biljecki. “By using only open data, we’ve built a flexible framework that cities around the world can use to better understand their carbon footprint and plan more effective responses.”
One of the key insights from the study is the complex relationship between building density and carbon emissions. While taller buildings tend to be more energy-efficient per unit area due to economies of scale, dense urban cores may also experience higher cooling demands due to urban heat island effects. Suburban areas, typically associated with detached low-rise buildings, were found to be significant contributors to total emissions, sometimes rivalling those of city centres.
The research also uncovered stark inequalities. In most cities studied, wealthier neighbourhoods were found to have disproportionately high per capita emissions. In Manhattan, for example, more than half of total building emissions were attributed to just a handful of large buildings.
“Uniform carbon pricing or blanket regulations risk placing an unfair burden on lower-income communities that may already be struggling with older, less efficient infrastructure,” said Asst Prof Biljecki. “Our findings highlight the need for place-based strategies that take both emissions intensity and socioeconomic vulnerability into account.”
The framework integrates diverse data sources including satellite imagery, street view photos, population maps, road networks, and local climate data using graph neural networks, a form of deep learning that captures spatial relationships between urban elements.
By making their approach entirely open, the researchers say they want to support global efforts to reduce emissions from the built environment and to help cities meet their climate targets.
“This work demonstrates the potential of open science and AI to accelerate urban sustainability,” said Asst Prof Biljecki. “It’s not just about understanding where emissions come from, but also ensuring that climate action is both effective and fair.”
Journal Reference:
Yap, W., Wu, A.N., Miller, C. et al., ‘Revealing building operating carbon dynamics for multiple cities’, Nature Sustainability (2025). DOI: 10.1038/s41893-025-01615-8
Article Source:
Press Release/Material by National University of Singapore (NUS) | College of Design and Engineering
Featured image credit: Gerd Altmann | Pixabay
