Data centers are Texas's fastest-growing electricity and water claimants, with interconnection requests exceeding 400 GW against an 86 GW peak. We develop a four-region system dynamics model that endogenizes compute demand, siting, and cooling choice around a within-region allocation cascade with irrigated agriculture as residual claimant. The model is anchored to Texas Water Development Board and ERCOT data and reproducing four independent observations without fitting. Results showed that local siting restraint never relieves nexus pressure. It displaces the buildout into water-scarce agricultural regions in a closed state, or exports the industry once demand turns footloose. Under a drought of record, irrigation absorbs nearly all shortage, accumulating 6.3 million acre-feet of foregone Panhandle supplies, while an 80% irrigation-protection reservation changes nothing. A dry-cooling mandate cuts direct water 78% at a 1.6% indirect penalty, and chip efficiency moves 2050 outcomes more than any policy lever. Governance must target allocation and siting jointly, not volumes alone.
Agri-food chains, especially fruit, produce about 10% of global GHG emissions, with more than half arising after harvest. Yet current methods, such as life cycle assessment (LCA), lack the spatial, temporal and physiological detail needed to resolve stage-specific drivers of environmental impact. We aim to overcome these limitations by developing a spatio-temporal, physics-based life cycle model that quantifies climate change (CC) and fossil resource use (RU,f) impacts for individual fruits and shipments across a transcontinental maritime supply chain. The proposed framework is demonstrated using a case study of the citrus supply chain connecting production in South Africa to consumption in the Netherlands. We show that variability among fruits and between shipments substantially alters CC and RU,f impact outcomes, demonstrating the need for dynamic rather than static supply-chain assessment. Energy use, maritime transport and food loss dominate impacts, with the farm-to-packhouse and overseas shipping stages contributing up to 0.5 kg CO₂-eq kg⁻¹ and 2 MJ kg⁻¹ of fruit consumed, depending on local conditions and cold-chain management. Stage-specific interventions, such as reducing field heat prior to forced-air cooling or increasing shipping temperatures from −1 °C to 1–3 °C under validated phytosanitary conditions, can meaningfully lower impacts. This dynamic framework offers a more realistic basis for identifying and achieving feasible decarbonization in global fruit supply systems.
The development of Secondary Materials Markets (SMMs) is a promising approach to diversify and strengthen critical materials supply chains. SMMs involve the recovery of valuable materials from discarded products and reintroduce valuable materials into the economy through cooperative business partnerships. This research addresses the need for analytical tools, which can enable SMM stakeholders to evaluate potential outcomes under various operating conditions and make informed decisions regarding SMM participation and development. A novel agent based model, integrated with existing techno-economic analysis and lifecycle assessment metrics, is developed to quantitatively evaluate the feasibility and impacts of SMMs for enabling secondary critical material feedstocks. Multiple interdependent factors are considered simultaneously, including techno-economic developments, demand uncertainty, inconsistent used product management scenarios, and changing stakeholder business strategies and interactions. The model’s analytical capabilities are demonstrated using a hypothetical-realistic photovoltaic (PV) waste management case study. Under the baseline tested conditions, the model estimated recovery of approximately 72 % of regional PV waste providing secondary materials to meet the demands of co-located manufacturing firms. For example, approximately 45 % of the aluminum demand was met via recovered materials leading to 23 million USD (approximate) in savings for the partner firm buying recovered aluminum under baseline tested conditions. In this manner, this research lays the foundation for analytical tool development to support decisions related to critical material recovery through SMMs.
Grazing intensity optimization is a natural climate solution that contributes to carbon sequestration and consequent climate mitigation. However, it may threaten food security by affecting ruminant production. Here, we evaluate the trade-off between carbon sequestration and ruminant production from grazing intensity optimization in global grasslands and investigate global supply chain drivers. Results show that optimizing grazing intensity could unleash 29.7 ± 2.5 billion tons of carbon sequestration potential (CSP), roughly equivalent to three years of global carbon emissions. However, it would reduce grassland-based ruminant production by 20%. Ruminant production loss is pronounced in low- and lower-middle income countries where local ability of managing grazing intensity is limited. International trade contributes 25% of the untapped CSP, with half from exports in sub-Saharan Africa, indicating the solution of local trade-off through global joint effort. These findings inform demand-side interventions and international cooperation to optimize grazing intensity in grasslands for carbon sequestration and food security.
With the continuous application of lithium-ion batteries (LIBs) in electric vehicles and energy storage, the number of retired batteries has surged, making their recycling an urgent environmental and economic challenge that must be addressed. This work explores the application of machine learning (ML) and life cycle assessment (LCA) in the recycling of retired LIBs. This review summarizes relevant studies and analyzes the current use of ML, focusing on its roles in predicting recycling potential, assessing health status, forecasting remaining lifespan, classifying battery materials, and optimizing recycling processes. LCA focuses on quantifying environmental impacts, energy conservation, and economic benefits throughout the recycling process. It seeks to identify critical stages within the recycling process to facilitate process improvement and increase efficiency. Based on this, a new strategy for collaborative integration of ML and LCA is proposed, alongside the establishment of a decision support system. This system aims to provide a robust scientific basis and empirical technical data to support both corporate strategic planning and the formulation of governmental regulatory frameworks. The integration of ML and LCA accelerates efficient, green, and safe recycling of spent LIBs, offering a novel strategy for industrial sustainability with significant theoretical and practical value.
The climate benefits of bio-plastics can be undermined by the environmentally intensive production of their synthesis catalysts, an upstream burden that is often overlooked. Within the electrochemical oxidation of 5-hydroxymethylfurfural (HMF) to 2,5-furandicarboxylic acid (FDCA), a key monomer for bioplastics poly(ethylene 2,5-furandicarboxylate), PEF, this burden would offset the advantages over petroleum-based polyethylene terephthalate (PET). Here, we designed a synergistic Ni–Sc catalyst and, more critically, introduced a mineral upcycling strategy to break this trade-off. A scandium-doped nickel metal–organic framework (NiSc-MOF) promotes sequential alcohol/aldehyde oxidation, achieving 97.4% FDCA yield at 1.44 V. By directly converting laterite ore into NiSc-MOF, we bypass the energy- and emission-intensive production of conventional high-purity salts, enabling stable 200 h operation with 92.9% Faradaic efficiency. Life-cycle assessment reveals that this approach reduces the climate impact and resource depletion of FDCA synthesis by 14% and 24%, respectively. When extended these benefits to the full PEF production system, it achieves a 33% reduction in greenhouse-gas emissions and a 49% decrease in fossil-resource consumption compared to PET, promoting a more resource-efficient and sustainable pathway for next-generation bioplastics.
Food waste in university cafeterias has become a challenge for campus management. However, methods such as manual weighing and vision-based assessment still suffer from low efficiency or limited quantitative capability, making food waste estimation difficult. To address this gap, we introduce the University Canteen Food Waste Dataset (UCFW-D), integrating images and weight records of cafeteria leftovers and comprising 109,978 samples from 20 universities in China. We propose a segmentation-stacking framework that integrates an enhanced Efficient Multi-scale Attention and Deformable Convolution-Enhanced DeepLabv3+ Network (EMD-DeepLabv3+), improving mean Intersection over Union (mIoU) and mean Pixel Accuracy (mPA) by 2.7 and 3.3 percentage points to 74.5% and 86.3%, respectively. The stacking model learns a mapping from segmented area features to food waste weight, demonstrating predictive performance. Finally, we develop a Food Waste Intelligent Segmentation and Carbon Accounting System (FWA) for weight estimation and carbon accounting. This supports food waste assessment in university catering systems.
Sorted post-household mixed plastic packaging waste requires additional treatment steps to ensure that its pyrolysis products satisfy hydrotreatment specifications. Cost and contaminant minimisation should be considered when designing a feedstock upgrading process. A Superstructure–Decision-Making (SS–DM) framework was developed to address this challenge by evaluating more than 3,000 sorting and washing sequences from a multi-level perspective. The results reveal that optimal performance is achieved with (i) a 4-step NIR-sorting cascade including NIR-cleaner(s) (high % polyolefin (PO) recovery and purity, low cost), and (ii) either cold washing with density separation or dry washing. Increasing the PET/MPO/film content while decreasing the PO content in the input decreases %PO purity, %PO recovery, and contaminant removal efficiency. In varying decision-making priorities, balancing cost and quality allows fulfilment of all constraints. Beyond decision support, SS–DM functions as a fast-track tool that generates high-resolution outputs across all units, serving as valuable inputs for pyrolysis modelling.
Solar power production will generate tens of millions of tons of end-of-life solar modules by 2050, with glass comprising ∼75% of their mass. This study investigates recycling waste ground solar glass as a supplementary cementitious material (SCM), addressing both PV waste management and the shrinking supply of traditional pozzolans. Solar glasses were predominantly amorphous but had a higher Mg and a lower Fe content than traditional soda-lime glasses. Reactivity of ground solar glasses was governed primarily by fineness and surface pore volume, with the finer solar glass exhibiting the highest dissolution and ASTM C1897 reactivity. Ground solar glass in cement pastes also refined pore structure, consumed portlandite through pozzolanic reaction, and enabled long-term strength recovery. Moreover, trace element leaching remained below regulatory limits, verifying its viability as a safe SCM. Repurposing end-of-life solar glass thereby alleviates waste management burdens and increases SCM supply to construction industry, while fostering sustainable industrial symbiosis.
