Beyond Stationarity: Characterizing Drought in the Context of Changing Baselines
Drought occurs when there is less water available for use by people and ecosystems than expected. Traditional drought assessment methods assume that the expected amount and variability of available water in a given season stays constant (stationary) over long periods of time. However, a changing climate is causing the probability of extreme events to change, a phenomenon known statistically as non-stationarity. Some regions are becoming more humid over time, and some are becoming more arid. As a result, we must fundamentally rethink how to characterize drought while baseline conditions are changing.
This project will develop a novel method that dynamically estimates expected water availability over time based on historical trends and then compares the observed amount to these expectations. The team will begin with a geographic focus on Vermont, expand to the area covered by the Northeast Drought Early Warning System, and eventually expand across the Lower 48 U.S. states.
This work will correct distortions in current drought assessments and in doing so improve drought planning, mitigation, and adaptation efforts throughout the United States.
This research is part of the Fiscal Year 2026 NIDIS Coping with Drought research competition. For more information, please contact Britt Parker (britt.parker@noaa.gov).
Research Snapshot
Tom Parris, ISciences LLC
Lesley-Ann L. Dupigny-Giroux, University Distinguished Professor of Geography & Geosciences, Vermont State Climatologist, University of Vermont
What to expect from this research
- To better assess current droughts, the research team will create a new way to analyze water data. First, they will use statistical methods to identify characteristics that influence drought, such as long-term climate trends, changes in seasonal timing, and the effects of ocean cycles.
- Using identified characteristics, researchers will alculate "dynamic normals" that reflect changing expectations over time. Comparing current conditions to expected values that change over time in response to shifting climate patterns will likely produce more accurate drought monitoring and assessment.
