Hydrotwin: an AI-based physics and optimization framework for improved water resource resilience in systems experiencing adverse events
This white paper focuses on water for energy.
As we move ahead into an uncertain future for water resource demand and supply, a massive challenge exists to optimize water systems to be resilient to a range of press/pulse extreme event states and conditions. Optimal decision-making needs include investments for siting of water and energy resources to achieve resource adequacy (the ability of an electricity system to meet demand for power under a wide range of conditions), such as reservoir operations that can prepare for potential future droughts or floods, and flow routing through or around fire-affected pipeline and electrical grid networks. These optimal sitings and routing would aim to improve water and energy resource and systems resiliency to adverse press/pulse events.
In the near term, AI data centers are springing up all around the country and will require vast, sustainable, and stable volumes of water to cool the computing centers that are enabling the AI revolution. Hydrotwin provides an automated tool for on-the-fly representation of the physical system of environmental water flows, water management infrastructure, and water demand from agriculture, industry, and cities. Hydrotwin can determine the holistic implications of how water systems can optimally route flows to supply water for energy system uses under critical shifts in water resources. Where Hydrotwin shines is in its ability to rapidly respond to adverse events that have become common in today's world (e.g., flooding, wildfires, cold and heat outbreaks). Once a system's physical water delivery system requires an update (e.g., routing of water towards a wildfire-affected region or away from flooding areas) then Hydrotwin will rapidly produce thousands optimal candidate strategies to address the adverse event that can be individually considered by decision makers.
The core measures of success center around assessing the extent to which Hydrotwin can inform water-energy security-related threats and opportunities. The overarching goal of Hydrotwin would be to achieve complete revolution of water resource management by enabling fast, accurate water cycle flux simulations to answer fundamental science questions and offer solutions previously unattainable without coordinated partnerships spanning multiple sectors.
Citation Formats
TY - DATA
AB - This white paper focuses on water for energy.
As we move ahead into an uncertain future for water resource demand and supply, a massive challenge exists to optimize water systems to be resilient to a range of press/pulse extreme event states and conditions. Optimal decision-making needs include investments for siting of water and energy resources to achieve resource adequacy (the ability of an electricity system to meet demand for power under a wide range of conditions), such as reservoir operations that can prepare for potential future droughts or floods, and flow routing through or around fire-affected pipeline and electrical grid networks. These optimal sitings and routing would aim to improve water and energy resource and systems resiliency to adverse press/pulse events.
In the near term, AI data centers are springing up all around the country and will require vast, sustainable, and stable volumes of water to cool the computing centers that are enabling the AI revolution. Hydrotwin provides an automated tool for on-the-fly representation of the physical system of environmental water flows, water management infrastructure, and water demand from agriculture, industry, and cities. Hydrotwin can determine the holistic implications of how water systems can optimally route flows to supply water for energy system uses under critical shifts in water resources. Where Hydrotwin shines is in its ability to rapidly respond to adverse events that have become common in today's world (e.g., flooding, wildfires, cold and heat outbreaks). Once a system's physical water delivery system requires an update (e.g., routing of water towards a wildfire-affected region or away from flooding areas) then Hydrotwin will rapidly produce thousands optimal candidate strategies to address the adverse event that can be individually considered by decision makers.
The core measures of success center around assessing the extent to which Hydrotwin can inform water-energy security-related threats and opportunities. The overarching goal of Hydrotwin would be to achieve complete revolution of water resource management by enabling fast, accurate water cycle flux simulations to answer fundamental science questions and offer solutions previously unattainable without coordinated partnerships spanning multiple sectors.
AU - Bennett, Katrina
A2 - Schwenk, Jon
A3 - Brelsford, Christa
A4 - Garcia, Manuel
DB - Energy-Water Resilience
DP - Open EI | National Laboratory of the Rockies
DO -
KW - water resource planning
KW - AI and machine learning enabled data fusion
KW - water supply
KW - water management
KW - water system optimization
KW - stakeholder engagement
KW - wildfires
KW - cold and heat outbreaks
KW - droughts
KW - floods
KW - Hydrotwin
KW - AI
KW - ML
KW - artificial intelligence
KW - machine learning
KW - optimization
KW - adverse events
KW - water systems
KW - optimal siting
LA - English
DA - 2026/01/15
PY - 2026
PB - LANL
T1 - Hydrotwin: an AI-based physics and optimization framework for improved water resource resilience in systems experiencing adverse events
UR - https://ewr.openei.org/submissions/12
ER -
Bennett, Katrina, et al. Hydrotwin: an AI-based physics and optimization framework for improved water resource resilience in systems experiencing adverse events. LANL, 15 January, 2026, Energy-Water Resilience. https://ewr.openei.org/submissions/12.
Bennett, K., Schwenk, J., Brelsford, C., & Garcia, M. (2026). Hydrotwin: an AI-based physics and optimization framework for improved water resource resilience in systems experiencing adverse events. [Data set]. Energy-Water Resilience. LANL. https://ewr.openei.org/submissions/12
Bennett, Katrina, Jon Schwenk, Christa Brelsford, and Manuel Garcia. Hydrotwin: an AI-based physics and optimization framework for improved water resource resilience in systems experiencing adverse events. LANL, January, 15, 2026. Distributed by Energy-Water Resilience. https://ewr.openei.org/submissions/12
@misc{EWR_Dataset_12,
title = {Hydrotwin: an AI-based physics and optimization framework for improved water resource resilience in systems experiencing adverse events},
author = {Bennett, Katrina and Schwenk, Jon and Brelsford, Christa and Garcia, Manuel},
abstractNote = {This white paper focuses on water for energy.
As we move ahead into an uncertain future for water resource demand and supply, a massive challenge exists to optimize water systems to be resilient to a range of press/pulse extreme event states and conditions. Optimal decision-making needs include investments for siting of water and energy resources to achieve resource adequacy (the ability of an electricity system to meet demand for power under a wide range of conditions), such as reservoir operations that can prepare for potential future droughts or floods, and flow routing through or around fire-affected pipeline and electrical grid networks. These optimal sitings and routing would aim to improve water and energy resource and systems resiliency to adverse press/pulse events.
In the near term, AI data centers are springing up all around the country and will require vast, sustainable, and stable volumes of water to cool the computing centers that are enabling the AI revolution. Hydrotwin provides an automated tool for on-the-fly representation of the physical system of environmental water flows, water management infrastructure, and water demand from agriculture, industry, and cities. Hydrotwin can determine the holistic implications of how water systems can optimally route flows to supply water for energy system uses under critical shifts in water resources. Where Hydrotwin shines is in its ability to rapidly respond to adverse events that have become common in today's world (e.g., flooding, wildfires, cold and heat outbreaks). Once a system's physical water delivery system requires an update (e.g., routing of water towards a wildfire-affected region or away from flooding areas) then Hydrotwin will rapidly produce thousands optimal candidate strategies to address the adverse event that can be individually considered by decision makers.
The core measures of success center around assessing the extent to which Hydrotwin can inform water-energy security-related threats and opportunities. The overarching goal of Hydrotwin would be to achieve complete revolution of water resource management by enabling fast, accurate water cycle flux simulations to answer fundamental science questions and offer solutions previously unattainable without coordinated partnerships spanning multiple sectors.
},
url = {https://ewr.openei.org/submissions/12},
year = {2026},
howpublished = {Energy-Water Resilience, LANL, https://ewr.openei.org/submissions/12},
note = {Accessed: 2026-11-15}
}
Details
Data from Jan 15, 2026
Last updated Jan 16, 2026
Submitted Jan 15, 2026
Contact
Katrina Bennett
Authors
Keywords
water resource planning, AI and machine learning enabled data fusion, water supply, water management, water system optimization, stakeholder engagement, wildfires, cold and heat outbreaks, droughts, floods, Hydrotwin, AI, ML, artificial intelligence, machine learning, optimization, adverse events, water systems, optimal sitingDOE Project Details
Project Name White Papers on Ideas to Advance Energy-Water Resilience
Project Lead
Project Number WP-012
