UVM Theses and Dissertations
Format:
Online
Author:
McBride, Maeve
Dept./Program:
Civil and Environmental Engineering
Year:
2007
Degree:
PhD
Abstract:
A three part investigation into the effects of riparian reforestation on small streams demonstrated the timing, nature, and processes of morphologic change. First, measurements of two small streams in northeastern Vermont collected in 1966 and 2004 - 2005 documented considerable change in channel width following a period of passive reforestation. Channel widths of several tributaries to Sleepers River were measured in 1966 when the area had more non-forested riparian vegetation than today. A longitudinal survey in 2004 of two of these tributaries, followed by detailed measurements at specific reaches in 2005, provided information on channel size, large woody debris (LWD), and riparian vegetation. Reforested reaches have widened and incised markedly since 1966. Reaches with the oldest forest were widest for a given drainage area, and the non-forested reaches were substantially narrower. A conceptual model was developed that describes a multi-phase process of incision, widening, and recovery following riparian reforestation of non-forested areas.
Second, a fixed-bed hydraulic model of one of the streams was developed to evaluate the impact of forested riparian vegetation on near-bank turbulence during overbank flows. Flume experiments with kinematic similitude and a 1:5 scale represented half a channel and its floodplain, mimicking the size of a non-forested reach. Two types of vegetation were simulated: non-forested, with synthetic grass carpet; and forested, where wooden dowels were added. Three-dimensional velocities were measured with an acoustic Doppler velocimeter. Velocities, turbulent kinetic energy (TKE), and Reynolds shear stress showed significant differences between forested and non-forested runs. Forested runs exhibited a narrow band of high TKE in the near-bank region that was roughly two times greater than in non-forested runs. Hydraulic characteristics of forested runs appear to create an environment with higher erosion potential, thereby indicating a possible driving mechanism for channel widening in reforesting stream reaches.
Third, Light Detection and Ranging (LiDAR) data from Chittenden County were analyzed to develop a method capable of classifying riparian buffers into broad classes according to forest type and age. The geospatial characteristics of the LiDAR data in forested areas were explored using semivariogram analysis, and LiDAR-based metrics were derived in a geographic information system (GIs) to quantify vegetation height and variance. The LiDAR-based metrics were then used in two discriminant analysis procedures that distinguished: 1) forest type as deciduous or coniferous; and 2) forest age in four age classes. With the resulting linear discriminant functions, a GIs-based classification method was developed. The classification method was highly successful at determining forest type but only moderately successful at determining forest age.
Second, a fixed-bed hydraulic model of one of the streams was developed to evaluate the impact of forested riparian vegetation on near-bank turbulence during overbank flows. Flume experiments with kinematic similitude and a 1:5 scale represented half a channel and its floodplain, mimicking the size of a non-forested reach. Two types of vegetation were simulated: non-forested, with synthetic grass carpet; and forested, where wooden dowels were added. Three-dimensional velocities were measured with an acoustic Doppler velocimeter. Velocities, turbulent kinetic energy (TKE), and Reynolds shear stress showed significant differences between forested and non-forested runs. Forested runs exhibited a narrow band of high TKE in the near-bank region that was roughly two times greater than in non-forested runs. Hydraulic characteristics of forested runs appear to create an environment with higher erosion potential, thereby indicating a possible driving mechanism for channel widening in reforesting stream reaches.
Third, Light Detection and Ranging (LiDAR) data from Chittenden County were analyzed to develop a method capable of classifying riparian buffers into broad classes according to forest type and age. The geospatial characteristics of the LiDAR data in forested areas were explored using semivariogram analysis, and LiDAR-based metrics were derived in a geographic information system (GIs) to quantify vegetation height and variance. The LiDAR-based metrics were then used in two discriminant analysis procedures that distinguished: 1) forest type as deciduous or coniferous; and 2) forest age in four age classes. With the resulting linear discriminant functions, a GIs-based classification method was developed. The classification method was highly successful at determining forest type but only moderately successful at determining forest age.