Riparian vegetation covers only about 0.4% of the land surface whereas rivers and streams cover about 0.3 to 0.56% but riparian vegetation still provides substantial ecosystem services including many that affect hydrodynamic processes. The ability to estimate the flow resistance caused by vegetation and the consequences for hydrodynamic processes has long been a vexing problem for engineers. Flow resistance is strongly influenced by local hydraulic conditions (water velocity and depth) and vegetation structures, and in turn, vegetation affects mass and momentum exchange processes. However, riparian vegetation conditions change continuously due to external stressors including climate change (e.g., extended drought in New Mexico) and engineered structures (e.g., dams and levees), which in turn have impacts on river morphology.
In my dissertation, first I investigated novel techniques for characterizing riparian vegetation through field-based and remote sensing techniques and modeling hydraulic roughness due to riparian vegetation in a two-dimensional hydrodynamic model (Chapter 2). Next, I studied the influence of riparian vegetation on channel/floodplain connectivity in terms of mass and momentum transport (Chapter 3). Finally, I explored the use of machine learning techniques to characterize spatiotemporal variations in riparian vegetation and river morphology in response to external hydrodynamic drivers of change (Chapter 4). The overarching goal of this research in general was to advance understanding of the dynamics of river systems in relation to riparian vegetation.
In chapter 2, spatial variations in Manning’s roughness were observed based on vegetation species and discharge that effects the hydraulic parameters in presence of riparian vegetation when comparing the user assigned and iteratively computed hydraulic roughness approaches. Thus, the method proposed here is beneficial for describing the hydraulic conditions for the diverse vegetation in terms of density and species. The analysis from chapter 3 showed that the mass and momentum transfer is sensitive to vegetation but insensitive to changes in vegetation density for real river or irregular channels and is mostly driven by a pattern of channel geometry (i.e., meandering). The results from chapter 4 showed the increasing trends of vegetation cover and reduced channel width along the Rio Grande even having long term drought. The availability of long-term datasets and machine learning algorithms in open-source Google Earth Engine cloud storage and computing platform results demonstrated the potential for large scale spatiotemporal analysis of riparian vegetation and river morphology for similar rivers like the Rio Grande.
This dissertation advanced the understanding of the dynamics of the river system in relation to riparian vegetation. The improved knowledge of the river system provides beneficial information for river management in terms of ecosystem services provided by the river system, water delivery, and flood conveyance purposes.
Two-dimensional hydrodynamic model, riparian vegetation, hydraulic roughness, LiDAR, Google Earth Engine, Random Forest
Level of Degree
First Committee Member (Chair)
Dr. Mark Stone
Second Committee Member
Dr. Julie Coonrod
Third Committee Member
Dr. Liping Yang
Fourth Committee Member
Dr. Ryan Morrison
Chaulagain, Smriti. "An Investigation of Emerging Technologies to Advance the Understanding of Dynamics Between the Floodplain and Main Channel due to Riparian Vegetation." (2022). https://digitalrepository.unm.edu/ce_etds/279