Loss of beaver dams decreases floodplain connectivity in Colorado headwater streams
Data files
May 28, 2026 version files 152.44 MB
-
BankLines.zip
83.56 KB
-
CalibrationWSEL.zip
9.29 KB
-
FluxCalculations.py
16.22 KB
-
ModelOutputFiles.zip
119.92 MB
-
ObstructionLocations.zip
5.15 KB
-
PebbleCounts.zip
3.64 KB
-
README.md
4.57 KB
-
TerrainSurfaces.zip
15.25 MB
-
VelocityAndDepthRasters.zip
17.15 MB
Abstract
As ecosystem engineers, beavers (Castor canadensis) modify river corridor form through dam building. When beavers are removed from a system, their unmaintained dams wash out, altering the stream’s hydrologic regime. We quantified the change in floodplain connectivity caused by the loss of beaver dams at three headwater tributary sites in the Kawuneeche Valley, Rocky Mountain National Park, Colorado, USA. From field observations, we developed two-dimensional steady-state hydraulic models to compare metrics of floodplain connectivity under historical (beaver-active) and present (no beaver activity) scenarios. The historical scenarios featured modeled beaver dams matching conditions in the year 1990 when beavers were present in the landscape. We simulated three low-to-moderate recurrence interval flood discharges to assess floodplain connectivity metrics, including the volume of water on the floodplain, the fraction of flow moving through the floodplain, the volumetric flux into the floodplain, and mean site and floodplain residence times. We found that the loss of beaver dams decreases floodplain connectivity across all connectivity metrics (up to a 96.5% loss in connectivity) except mean floodplain residence time, which increased in the absence of dams. Channel velocities also increased following the loss of beaver dams. We conclude that, in headwater streams, beaver dams play an important and quantifiable role in facilitating floodplain connectivity, and floodplain disconnection from the loss of dams has major implications for other ecological and geomorphic floodplain processes.
Dataset DOI: 10.5061/dryad.2z34tmpzd
Description of the data and file structure
This dataset supports the development and analysis of two-dimensional hydraulic models used to compare metrics of floodplain connectivity between scenarios with and without beaver dams on several headwater tributaries in Colorado. It contains various field data (CalibrationWSEL.zip; PebbleCounts.zip), model inputs (BankLines.zip; ObstructionsLocations.zip; TerrainSurfaces.zip), analysis tools (FluxCalculations.py), and results (VelocityAndDepthRasters.zip; ModelOutputs.zip).
Files and variables
File: CalibrationWSEL.zip
Description: Collection of comma-delimited files containing water surface elevation (WSEL) coordinate data associated with measured calibration and validation discharges at Onahu Creek, Lower Baker Creek, and Upper Baker Creek. The site and measured discharge are specified in the .csv name (e.g., Q2.40 = discharge of 2.40 cms). Each row contains a single WSEL point defined by a Name, Easting, Northing, and Elevation. The coordinate system is Colorado State Plane North (m).
File: BankLines.zip
Description: Shapefiles delineating the left and right banks at Onahu Creek, Lower Baker Creek, and Upper Baker Creek. Bank lines determine the boundary between the modeled channel and floodplain roughness values.
File: ObstructionLocations.zip
Description: Shapefiles with locations and approximate orientations of modeled historical beaver dams at Onahu Creek, Lower Baker Creek, and Upper Baker Creek.
File: PebbleCounts.zip
Description: Comma-delimited files containing raw pebble count data measured across multiple transects at Onahu Creek, Lower Baker Creek, and Upper Baker Creek. Transects are specified by a column heading T_# (e.g., T_3 is Transect 3). Each column represents one transect and contains 100 pebble count measurements.
File: FluxCalculations.py
Description: Python script that calculates mass and momentum fluxes across specified bank boundaries using SRH-2D model outputs.
File: TerrainSurfaces.zip
Description: Raster images containing final terrain surface information at Onahu Creek, Lower Baker Creek, and Upper Baker Creek.
File: VelocityAndDepthRasters.zip
Description: Raster images containing velocity and water depth results for each site and modeling scenario.
File: ModelOutputFiles.zip
Description: SRH-2D model output files. This folder contains two types of output files: Extensible Model Data Format (XMDF) files for each site and scenario run (e.g., LowerBaker2yrXMDF.h5, LowerBaker2yrDamsXMDF.h5) and mesh files (e.g., LowerBakerMeshes.h5) for each site. To read and extract datasets from .h5 files, we used the h5py Python package (see associated Python script).
- XMDF files contain results including water surface elevation (Water_Elev_m), water depth (Water_Depth_m), depth-averaged velocity (Vel_Mag_m_p_s and Velocity_m_p_s), Froude number (Froude), and bed shear stress (B_Stress_n_p_m2) at each mesh node. Relevant datasets have the following file paths:
- Datasets/Velocity_m_p_s/Values
- Datasets/Water_Depth_m/Values
- Mesh files contain various datasets related to the model meshes. Specifically, these files contain mesh node coordinates (NodeLocs) which correspond to the XMDF results. Coordinates are in Colorado State Plane North (m). The dataset containing mesh node coordinates has the following file path:
- 2DMeshModule/[InsertMeshName] Mesh/Nodes/NodeLocs
Code/software
FluxCalculations.py is a Python script. We used Python 3.13.2 with the following packages: pandas (2.2.3), numpy (2.2.2), math, h5py (3.12.1), and matplotlib (3.10.0).
Access information
Data was derived from the following sources:
- We used 2-ft digital elevation models (DEMs) available through the Colorado Water Conservation Board's LiDAR Download Portal as part of the final terrain surfaces: https://coloradohazardmapping.com/lidarDownload.
- FluxCalculations.py was modified from an original script by Chaulagain et al. (2025).
- Chaulagain, S., Stone, M.C., Morrison, R.R., and Byrne, C.F., 2025, Mass and Momentum Flux Prediction at the Channel‐Floodplain Interface Associated With Riparian Vegetation: River Research and Applications, p. rra. 4436, doi:10.1002rra. .4436.
