Not All Water Consumption Is Equal: A Water Stress Weighted Metric for Sustainable Computing

2025-06-28

AI Summary

Current computational sustainability research largely overlooks the spatiotemporal heterogeneity of water stress, leading to substantial underestimation of the water-related environmental impacts of water-intensive computing activities (e.g., AI training). Method: We propose SCARF—a novel framework that introduces the first spatially and temporally resolved water stress weighting scheme, formalized via the Adjusted Water Impact (AWI) metric. SCARF integrates high-resolution water stress data, geospatial information, and time-series analysis into a dynamic weighted computation model. Contribution/Results: We validate SCARF across three critical domains—LLM inference services, data center operations, and semiconductor manufacturing—demonstrating that spatiotemporal coordination (e.g., load shifting, strategic siting) significantly reduces water impact. SCARF provides the first scalable, actionable, and quantitatively grounded decision-support tool for water-aware green computing.

Technology Category

• Machine Learning: Efficient ML / Green AI
• Humans and AI: Other Foundations of Human Computation & AI
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Systems and Infrastructure for Web, Mobile and WoT: Sustainability and carbon-aware systems for Web, mobile, and WoT
• Responsible Web: Sustainability and climate impact of web technologies

Abstract

Water consumption is an increasingly critical dimension of computing sustainability, especially as AI workloads rapidly scale. However, current water impact assessment often overlooks where and when water stress is more severe. To fill in this gap, we present SCARF, the first general framework that evaluates water impact of computing by factoring in both spatial and temporal variations in water stress. SCARF calculates an Adjusted Water Impact (AWI) metric that considers both consumption volume and local water stress over time. Through three case studies on LLM serving, datacenters, and semiconductor fabrication plants, we show the hidden opportunities for reducing water impact by optimizing location and time choices, paving the way for water-sustainable computing. The code is available at https://github.com/jojacola/SCARF.

Problem

• Assessing water impact of computing with spatial-temporal water stress variations
• Developing a metric to evaluate water consumption considering local stress levels
• Identifying optimization opportunities for water-sustainable computing practices

Innovation

• Spatial-temporal water stress weighted metric
• Adjusted Water Impact (AWI) calculation
• Location-time optimization for sustainability