Date of Award
Fall 2026
Abstract
Flooding is one of the most dangerous natural disasters in the United States, posing serious risks to human lives, damaging infrastructure, and necessitating costly repairs. Human modifications to rivers have led to channel incision and disconnection from floodplains, increasing fluvial erosive power and driving channel instability. The incised cobble-bedded Zealand River, located in northern New Hampshire, was recently reconnected to its historic floodplain via a diversion in hopes of restoring form and function. This study combines high-resolution remote sensing data and field observations to assess the stability of the restoration during the first year after project completion. UAS-based LiDAR and Structure from Motion point clouds were used with known grain size distributions at surveyed locations to estimate sediment grain size throughout the study reach of the Zealand River. Predictors derived from the statistical distribution of topographic roughness metrics were leveraged within a linear regression to generate spatially variable estimates of grain sizes. SfM-based models provided the strongest predictive power for grain size estimation (D50 R2=0.854, D84 R2=0.690); however, LiDAR-based models produced very comparable results and demonstrated superior performance when fine-grained material was considered (D50 R2=0.666, D84 R2=0.573). To evaluate the stability of the Zealand River’s restoration, a 2D hydraulic model was developed utilizing high-resolution LiDAR DEMs and surveyed bathymetric cross sections. Reach-scale grain size estimates from the best performing LiDAR-based model were used to develop spatially variable Manning’s roughness values, allowing the model to better constrain overbank flow. Model results for the highest flows seen during the study period, 13.16 cms, indicate that the channel remained relatively stable, which was corroborated with field observations. Simulated flows at the 100-year event indicate an overtopping of the diversion berm, consistent with restoration design to dissipate flood energy. However, localized estimates of shear stress indicate high levels of erosive power, presenting a risk of avulsion. Ultimately, these findings suggest that channel reactivation can represent a potential approach to restoring high-gradient watersheds.
Document Type
Master's Thesis
First Advisor
Anne Lightbody
Second Advisor
Michael Palace
Third Advisor
Landon Gryczkowski
Department or Program
Earth Sciences
Degree Name
Master of Science
Recommended Citation
Omslaer, Ryan, "Using Remote Sensing and Field Observations to Assess the Stability of a Restored Cobble-Bedded River" (2026). Master's Theses and Capstones. 2142.
https://scholars.unh.edu/thesis/2142