Short‐Term Memory and Regional Climate Drive City‐Scale Water Demand in the Contiguous US

Gaining insights into current and future urban water demand patterns and their determinants is paramount for water utilities and policymakers to formulate water demand management strategies targeted to high water‐using groups and infrastructure planning strategies. In this paper, we explore the comp...

Full description

Saved in:
Bibliographic Details
Published in:Earth's future Vol. 13; no. 1
Main Authors: Hao, Wenjin, Cominola, Andrea, Castelletti, Andrea
Format: Journal Article
Language:English
Published: Bognor Regis John Wiley & Sons, Inc 01.01.2025
Wiley
Subjects:
Arid regions
Arid zones
Aridity
Artificial intelligence
Causality
Cities
Climate
Climate change
Climatic classifications
Comparative analysis
Data processing
Demand analysis
Drought
Drought conditions
Environmental aspects
Environmental conditions
Feature selection
Forecasting
Households
Machine learning
Measurement
Metropolitan areas
Neural networks
Pressure effects
Regional climates
Regional development
Regional planning
Sociodemographics
Time series
United States
Vapor pressure
Variables
Water
Water demand
Water shortages
Water use
Water utilities
United States
United States > US
ISSN:2328-4277, 2328-4277
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Gaining insights into current and future urban water demand patterns and their determinants is paramount for water utilities and policymakers to formulate water demand management strategies targeted to high water‐using groups and infrastructure planning strategies. In this paper, we explore the complex web of causality between climatic and socio‐demographic determinants, and urban water demand patterns across the Contiguous United States (CONUS). We develop a causal discovery framework based on a Neural Granger Causal (NGC) model, a machine learning approach that identifies nonlinear causal relationships between determinants and water demand, enabling comprehensive water demand determinants discovery and water demand forecasting across the CONUS. We train our convolutional NGC model using large‐scale open water demand data collected with a monthly resolution from 2010 to 2017 for 86 cities across the CONUS and three Köppen climate regions—Arid, Temperate, and Continental—utilizing this globally recognized climate classification system to ensure a robust analysis across varied environmental conditions. We discover that city‐scale urban water demand is primarily driven by short‐term memory effects. Climatic variables, particularly vapor pressure deficit and temperature, also stand out as primary determinants across all regions, and more evidently in Arid regions as they capture aridity and drought conditions. Our model achieves an average R2 ${R}^{2}$ higher than 0.8 for one‐month‐ahead prediction of water demand across various cities, leveraging the Granger causal relationships in different spatial contexts. Finally, the exploration of temporal dynamics among determinants and water demand amplifies the interpretability of the model results. This enhanced interpretability facilitates discovery of urban water demand determinants and generalization of water demand forecasting. Plain Language Summary Water is essential for human activities in urban areas. As urban water demand is changing due to population growth, climate change, and urbanization, we need to plan strategies to meet future water demands, especially as water shortages become increasingly frequent in the Continental US. It is also natural to ask what the main causes driving demand changes are. In this study, we discover the factors that determine how much water people use in 86 cities in the United States through a causality perspective and use established causal relationships to make water demand forecasts. Our proposed causal modeling approach improves our understanding of leading determinants for urban water demand, that is, past water use habits and climatic conditions. Further, it achieves accurate predictions of future water demands. Our results will help city planners and officials make informed decisions to ensure future water security. Key Points A Neural Granger Causal model is developed to discover urban water demand determinants and enhance generalization of demand forecasting Key determinants of urban water demand in the contiguous US are historical demand, vapor pressure deficit, and temperature The Neural Granger Causal model interpretability and its capacity to explore temporal structures aid urban water demand management
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2328-4277
2328-4277
DOI:10.1029/2024EF004415