Rahmani et al. (2026) Wetlands set the pace of annual runoff in the northern Great Plains

Identification

• Journal: Communications Earth & Environment
• Year: 2026
• Date: 2026-03-11
• Authors: Javad Rahmani, I. F. Creed, Pascal Badiou, Ali A. Ameli
• DOI: 10.1038/s43247-026-03318-0

Research Groups

• Department of Earth, Ocean & Atmospheric Sciences, University of British Columbia, Vancouver, Canada
• Department of Physical & Environmental Sciences, University of Toronto, Toronto, Canada
• Institute for Wetland and Waterfowl Research, Ducks Unlimited Canada, Manitoba, Canada

Short Summary

This study reveals that in North America's Prairie Pothole Region, annual wetland inundation extent, rather than climate drivers, is the dominant factor explaining interannual variability in runoff and high-flow in 69% of 109 studied catchments over 38 years, with most catchments exhibiting threshold-like buffering behavior linked to geographically isolated wetlands.

Objective

• To determine whether annual maximum inundated wetland area (MIWA) explains more interannual variance in runoff ratio (ROR) and high-flow ratio (HFR) than annual and intra-annual climatic drivers in the Prairie Pothole Region (PPR).
• To characterize the functional form of the MIWA–runoff relationship (linear vs. threshold-like) and identify factors controlling its variation across PPR catchments.

Study Configuration

• Spatial Scale: The Prairie Pothole Region (PPR) of North America, spanning approximately 780,000 square kilometres across parts of Canada (Alberta, Saskatchewan, Manitoba) and the United States (Montana, North Dakota, South Dakota, Minnesota, Iowa). The study analyzed 109 non-regulated catchments, with a median catchment area of 4,197 square kilometres.
• Temporal Scale: 38 years (1984–2021) of daily hydroclimate and satellite-based inundation data, aggregated and analyzed on an annual (water year: October–September) basis.

Methodology and Data

• Models used:
  • Partial Spearman correlation analysis to disentangle independent effects of drivers.
  • Power-law model (ROR = a * (MIWA)^b + c) to characterize the functional form of MIWA-runoff relationships.
• Data sources:
  • Streamflow: Daily records from 109 gauged stations (United States Geological Survey, Environment and Climate Change Canada).
  • Climate (Dynamic): Daily precipitation and temperature from ERA5-Land reanalysis; daily potential evapotranspiration (PET) from a global land surface dataset. Snow persistence (SP) from MODIS/Terra 8-Day L3 Version 6 snow cover product (500-metre resolution).
  • Wetland Inundation: Annual Maximum Inundated Wetland Area (MIWA) derived from the Landsat-based Global Surface Water dataset (30-metre resolution, 1984–2021), with lakes and rivers removed.
  • Climate (Static) & Physical: 30-metre USGS digital elevation models (elevation, slope, catchment boundaries, area); river density (Han et al., 2023); Height Above Nearest Drainage (HAND) from MERIT Hydro; MODIS land cover data (500-metre resolution, 2001–2021); soil texture (sand, silt, clay) from SoilGrids 2.0; volumetric soil moisture from ERA5-Land; static water table depth (500-metre resolution) from Janssen et al. (2025).
  • Ancillary Data for Wetland Classification: Lakes and River dataset (Canada); National Hydrography Dataset (NHD) for rivers (U.S.); HydroLAKES database for lakes (U.S.).

Main Results

• In 69% of the 109 catchments, annual maximum inundated wetland area (MIWA) was the dominant driver of interannual variability in runoff ratio (ROR) and high-flow ratio (HFR), explaining more variance than any annual or intra-annual climate index.
• Climate, particularly snow persistence, primarily influences annual runoff by modulating wetland inundation extent (median Spearman correlation ρ = 0.63 in ~49% of fill-spill dominated catchments), rather than directly controlling streamflow.
• The relationship between MIWA and ROR across 75 fill–spill–dominated catchments predominantly exhibits threshold-like buffering behavior (64% of catchments, with power-law exponent b > 1.25), where runoff remains low until a critical inundation threshold is exceeded.
• The strength of this buffering behavior (exponent b) is strongly correlated with the prevalence of Geographically Isolated Wetlands (GIWs) (Spearman ρ = 0.65 with GIWs' maximum inundated extent), indicating that larger GIW areas enhance buffering capacity.

Contributions

• Challenges the prevailing climate-centric view of interannual runoff variability by demonstrating that wetland inundation extent is the dominant proximate driver in the wetland-rich Prairie Pothole Region.
• Provides a generalizable, regional-scale understanding of fill-spill dynamics and their functional forms (linear vs. threshold-like) across a large number of catchments, linking these dynamics to the extent of Geographically Isolated Wetlands.
• Offers a landscape-explicit basis for improving flood forecasting, water quality protection, and adaptive water management by integrating dynamic inundation state into hydrologic models and decision-making.
• Provides a mechanism-based rationale for the legal recognition and conservation of Geographically Isolated Wetlands as critical hydrologic infrastructure due to their demonstrated role in buffering hydrologic responses and mitigating flood risks.

Funding

• Natural Sciences and Engineering Research Council (NSERC) Discovery Grant (RGPIN-2020-04664) awarded to Ali Ameli.
• Environment and Climate Change Canada (EDF-CA-2021i023) awarded to Ali Ameli and Irena Creed.
• Computational support provided by Compute Canada.

Citation

@article{Rahmani2026Wetlands,
  author = {Rahmani, Javad and Creed, I. F. and Badiou, Pascal and Ameli, Ali A.},
  title = {Wetlands set the pace of annual runoff in the northern Great Plains},
  journal = {Communications Earth & Environment},
  year = {2026},
  doi = {10.1038/s43247-026-03318-0},
  url = {https://doi.org/10.1038/s43247-026-03318-0}
}