Mirzakhani et al. (2025) Dendrohydrological analysis of flood rings in Quercus lyrata Trees: Unveiling Southeast USA flood history

Identification

• Journal: Journal of Hydrology Regional Studies
• Year: 2025
• Date: 2025-11-20
• Authors: Mahsa Mirzakhani, Matthew D. Therrell, Amobichukwu C. Amanambu, Norman L. Jones, Lisa Davis, Scott St. George
• DOI: 10.1016/j.ejrh.2025.102918

Research Groups

• Department of Geography and the Environment, The University of Alabama, Tuscaloosa, AL, United States
• Department of Biological science, The University of Alabama, Tuscaloosa, AL, United States
• Weather & Climate Research at WTW, Minneapolis, MN, United States

Short Summary

This study reconstructs multi-century flood histories in the southeastern United States using anatomical flood rings in Quercus lyrata trees, validating these records against instrumental data. It reveals that flood-ring chronologies extend observed flood variability back to the early 1700s, capturing major 19th-century events, and highlights distinct flood dynamics between large, regulated river systems and smaller, free-flowing ones.

Objective

• To reconstruct multi-century flood histories in the southeastern United States using anatomical flood rings in Quercus lyrata.
• To evaluate the relationship between flood-ring occurrence and instrumental stage-height records.
• To examine how flood-ring signals vary between large regulated rivers and smaller free-flowing systems.
• To build a regional reconstruction of flood variability extending back to the early 1700s.

Study Configuration

• Spatial Scale: Six flood-plain sites across five major river basins in the Southeastern United States: Lower Mississippi River (DNS, BTO), White River (SNA), Pascagoula River (PGO), Altamaha River (MOO), and Suwannee River (SUW). River basin areas range from 25,123 km² to 2,980,000 km².
• Temporal Scale: Tree-ring chronologies spanning 1702–2021. Instrumental stage-height data from approximately 1885–2021.

Methodology and Data

• Models used:
  • Response Index (RI) calculation: Proportion of trees with flood rings at each site per year.
  • Duration-based threshold for flood events: Stage height exceeding a site-specific threshold for at least 10 consecutive days during March–June.
  • Statistical analyses: Spearman's rho correlation, Fisher’s Exact Test, Theil–Sen slope estimator, 31-year sliding window analysis.
• Data sources:
  • Tree-ring samples: 323 increment cores from Quercus lyrata trees.
  • Daily stage height data: U.S. Geological Survey (USGS) and U.S. Army Corps of Engineers (USACE) gauges.
  • Historic Crest data: National Oceanic and Atmospheric Administration (NOAA).

Main Results

• 136 flood ring years were identified across the six tree-ring chronologies spanning 1763–2021.
• Approximately 50% of flood rings during the instrumental period corresponded with recorded daily gauge data, indicating spring flooding events with inundation lasting at least 10 days.
• Significant positive correlations were found between the Response Index (RI) and stage heights at the three Mississippi River sites (SNA, DNS, BTO) for April and May, and at the Altamaha (MOO) and Suwannee (SUW) rivers for March.
• Flood-ring detection accuracy was significantly higher in large river basins (Odds Ratio = 2.15, p = 0.04 for true positives) compared to smaller basins, with significantly lower false negative rates (Odds Ratio = 0.46, p = 0.04).
• The flood-ring record extended observed flood variability back to the early 1700s, capturing major 19th-century events (e.g., 1858, 1865, 1867, 1882 Mississippi River floods) that are absent from instrumental data.
• Large river systems (Mississippi, White) exhibited widespread spring inundation with peak flows in April and May, while smaller rivers (Pascagoula, Altamaha, Suwannee) showed more localized flood patterns and earlier seasonal peaks in March and April.
• Theil–Sen slope analysis indicated a significant positive trend in RI frequency before 1927 (0.0028 per year at SNA, 0.0027 per year at DNS) followed by a significant negative trend at SNA (−0.0013 per year) and a weak decline at DNS after 1927, suggesting reduced flood occurrence following large-scale river engineering.

Contributions

• Provides the first multi-century flood history reconstruction for the southeastern United States using Quercus lyrata flood rings across a diverse range of river basins.
• Validates the reliability of flood rings as indicators of historical flood events, extending instrumental records back to the early 1700s and capturing pre-instrumental major flood events.
• Quantifies and contrasts distinct flood dynamics between large, regulated river systems (widespread, spring-peaking) and smaller, unregulated systems (localized, earlier seasonal peaks).
• Offers critical long-term context for modern flood-risk planning, highlighting the limitations of short observational records and informing the reassessment of flood control measures and adaptation strategies under future climate change.

Funding

• U.S National Science Foundation Geography and Spatial Sciences Program [award#BSC 1359801]

Citation

@article{Mirzakhani2025Dendrohydrological,
  author = {Mirzakhani, Mahsa and Therrell, Matthew D. and Amanambu, Amobichukwu C. and Jones, Norman L. and Davis, Lisa and George, Scott St.},
  title = {Dendrohydrological analysis of flood rings in Quercus lyrata Trees: Unveiling Southeast USA flood history},
  journal = {Journal of Hydrology Regional Studies},
  year = {2025},
  doi = {10.1016/j.ejrh.2025.102918},
  url = {https://doi.org/10.1016/j.ejrh.2025.102918}
}