Dataset for “Climatic and Socioeconomic Drivers of Water Use and Their Spatio-Temporal Patterns for Small and Mid-Sized Cities in the Contiguous United States”

Dave, H. and Cai, X. Dec. 11, 2025. Data from: “Climatic and Socioeconomic Drivers of Water Use and Their Spatio‐Temporal Patterns for Small and Mid‐Sized Cities in the Contiguous United States.” Zenodo. DOI: 10.5281/zenodo.16809300.

This dataset contains all code for calibrating and analyzing machine learning models for “Climatic and socioeconomic drivers of water use and their spatio-temporal patterns for small and mid-sized cities in the Contiguous United States”.

This study explores the drivers of urban water use and their spatial-temporal patterns in 142 small and mid-sized cities across the Contiguous United States (CONUS) by analyzing the data directly collected from these cities and using advanced machine learning techniques. We identify five distinguished clusters across CONUS, each showing unique trends of the impact of drivers on water use. We find that socioeconomic factors significantly influence water use in eastern and southwestern cities, while climatic variables such as precipitation and temperature range dominate in central and northwestern regions. Temporal analysis reveals the impacts of major socioeconomic and climatic disruptions on urban water use in the period 2011–2021, including the COVID lockdown, the rapid growth of data centers, and the drought of 2012. In addition, our analysis suggests that economic growth in small and mid-sized US cities continues to be accompanied by rising water use, contrasting with the opposite trend observed in large cities in prior studies. This implies that as smaller cities develop, their water use may increase above current levels until incomes reach a higher threshold, highlighting the need to improve water use efficiency. This study also presents useful insights for developing effective water demand management strategies in response to climatic variability and socioeconomic growth in small and mid-sized cities.

Zenodo:  Python scripts, model data, simulated water use