STREAMBANK CRITICAL SHEAR STRESS ESTIMATIONS USING SSURGO AND EMPIRICAL EQUATIONS

  • : Ms Word Format
  • : 70 Pages
  • : ₦3000
  • : 1-5 Chapters
  •  
  • Click to DOWNLOAD Materials

STREAMBANK CRITICAL SHEAR STRESS ESTIMATIONS USING SSURGO AND EMPIRICAL EQUATIONS

Abstract:

 

Sedimentation of waterways is a growing problem.  Sediment is the largest pollutant to the Chesapeake Bay, which can affect a variety of human elements.  In watersheds across the country, eroding banks contribute the primary source of sediment to streams.  To address bank erosion and stream instability, stream restoration is a common practice.  Bank stabilization is a common component of stream restoration projects.  Engineers and researchers utilize bank stability and erosion models to evaluate designs and to enhance resistance to bank erosion.  A commonly cited criticism of stream restoration projects is the focus on the reach scale, yet the reach responds according to its contributing watershed. Few watershed-scale modeling studies exist due to current data limitations.  This study evaluated the use of the SSURGO soil database combined with empirical equations to determine the critical shear stress of streambank soils across the country.  Two phases of research were executed; first, measured soil parameters were compared to estimated soil database parameters. Second, existing soil empirical equations were evaluated for accuracy.  The results show that on both site and national scales, SSURGO does not accurately represent streambank soil properties.  Additionally, the results show that existing empirical equations are not accurate predictors of critical shear stress. The barrier of soil parameterization still exists for large-scale modeling. However, an unexpected result discovered in this research was the power law trend of critical shear stress in cohesive sediments has an exponent of -0.4.  The coefficient of the power law relationship accounts for site-specific conditions and may include factors such as climate, land use, and vegetation. Further work is needed to develop the relationships for the coefficient.

 

 

 

 

 

 

 

 

Table of Contents

 

List of Figures………………………………………………………………………………………………………………….. vi

List of Tables………………………………………………………………………………………………………………….. viii

List of Equations……………………………………………………………………………………………………………….. ix

Acknowledgements…………………………………………………………………………………………………………….. x

Introduction………………………………………………………………………………………………………………………. 1

Hypothesis / Objective / Research Question……………………………………………………………………………… 4

Literature Review………………………………………………………………………………………………………………. 5

Watershed Degradation……………………………………………………………………………………………………. 5

History of Stream Restoration……………………………………………………………………………………………. 7

Types of Bank Erosion…………………………………………………………………………………………………….. 9

Bank Erosion Monitoring……………………………………………………………………………………………….. 11

Bank Erosion Modeling………………………………………………………………………………………………….. 12

Soil Parameters, Testing, and Theory…………………………………………………………………………………. 13

Research Areas and Methodology………………………………………………………………………………………… 16

Introduction…………………………………………………………………………………………………………………. 16

Site Description……………………………………………………………………………………………………………. 16

Fieldwork……………………………………………………………………………………………………………………. 18

Laboratory Analysis………………………………………………………………………………………………………. 20

Bank Erosion Model Selection (CONCEPTS)……………………………………………………………………… 21

CONCEPTS Conceptual Framework…………………………………………………………………………………. 21

Regional Parameterization Framework………………………………………………………………………………. 25

Database Introduction…………………………………………………………………………………………………….. 26

Results & Discussion………………………………………………………………………………………………………… 28

Oliver Run Design & Morphology……………………………………………………………………………………. 28

Field Surveys…………………………………………………………………………………………………………….. 29

Bed Material Sampling………………………………………………………………………………………………… 32

Comparison between Measured and SSURGO estimated Soil Properties…………………………………… 33

Oliver Run Soil Sampling and Analysis…………………………………………………………………………… 33

Database Comparison of Properties………………………………………………………………………………… 35

Evaluation of Empirical Equations……………………………………………………………………………………. 39

Improvements to Soil Parameterization………………………………………………………………………………. 45

Conclusions…………………………………………………………………………………………………………………….. 48

Bibliography……………………………………………………………………………………………………………………. 50

Appendix A: Research Timeline………………………………………………………………………………………….. 57

Appendix B: Program of Study (MS)……………………………………………………………………………………. 58

Introduction

Sediment is one of the primary pollutants of the Chesapeake Bay [Hassett et al., 2005].  Waterway sedimentation can lead to a variety of ecological and infrastructure problems from reduced biodiversity such as fish and macroinvertebrate populations, to reservoir sedimentation, which can affect power generation and navigation. Current bank erosion modeling efforts feature limited feedbacks between surface water and groundwater.  More complex coupled models exist; however, they require significant input data.  The deficiency of necessary data for complex models is often a justification for the use of simpler models.  Both the complexity and data requirements of bank erosion models continue to increase. The time and cost to collect the data needed to parameterize these types of models create a formidable obstacle. In this study, the parameterization and application of a physically based bank erosion model for predicting erosion at a stream restoration site using readily available data are tested.

Different entities implement stream restoration projects across the United States to restore ecological function and return morphological stability. A significant component of many, if not most, of these projects is bank stabilization [Hassett et al., 2005]. Channel banks can destabilize over engineering time scales for many reasons, but most commonly due to channel modifications and urbanization. As bank degrade and destabilize, channels can widen, leading to scour and erosion predominantly during high flow events. In extreme cases, a new channel may be formed. Increased peak flows and flashier storm flows, resulting from urbanization and land use changes, can further exacerbate channel degradation Physical modification of stream channels as part of stream restoration efforts can affect, and be affected, by all of these issues.      The design of stream restoration projects requires a multi-disciplinary approach, which may include processes from hydrology, hydraulics, ecology, and geomorphology. A morphologically stable stream exhibits no net erosion or deposition of sediment over time.  A stable channel configuration, or geometry, is dependent on the hydrologic regime as well as the type and volume of sediment transported. The hydrologic regime is a function of climate, topography, land use, and groundwater interactions. The accurate prediction of the hydrologic regime on a watershed scale requires a model. For sediment modeling, the type of sediment determines the governing equations.  Channel bed and streambank soil properties can be determined through soil sampling. However, soil properties are highly heterogeneous and can vary by orders of magnitudes along a bank profile [Sutarto et al., 2014].

The addition and removal of instream structures modifies how a channel adjusts its longterm morphology. Grade control structures affect streams for decades or even longer.  One example is the legacy effects of millponds throughout the North East.  The creation of millponds reduces the slope of a short section of river which then leads to sedimentation [Schenk and Hupp, 2009].  The lack of sediment in the water below the pond can result in localized scour as the river attempts to regain its equilibrium sediment concentration. This phenomenon is seen below the Hoover Dam. The removal of dams results in changes to sediment transport since a significant amount of legacy sediment is now available to erode and be transported through the system [Schenk and Hupp, 2009]. The effect of stream restoration structures on a longer, multi-decadal, time scale is not well understood; as some structures perform as designed while others have failed [Thompson, 2002].

Bank erosion models provide a method of predicting erosion following bank stabilization.  Models exist in a wide range of complexity, from simplistic to highly complex. A persistent problem area of erosion modeling is the difference between cohesive and non-cohesive soil.  Non-cohesive soils erode as a function of sediment size, whereas cohesive soils are dependent on additional factors.  Further complications in modeling result from the mixing of cohesive and non-cohesive fractions.  Additionally, soil properties exhibit spatial heterogeneity; as soils can vary considerably within just a small area [Hanson and Simon, 2001]. Advances in soil theory have enabled the development of 3-D erosion models, such as Virtual StreamLab (VSL3D), yet these models require parameters that are seldom readily available [Khosronejad et al., 2014].  Typically, researchers and engineers apply these models on the reach scale. Model setup requires extensive soil testing to determine the necessary properties. Due to the time and financial requirements for testing soil properties, models are seldom seen as a viable option for consulting engineers.

The parameterization of physically based erosion models requires sufficiently detailed soil and bank material data.  A variety of methods exist to either measure or predict soil strength parameters.  Early work focused on developing empirical equations to predict parameters [Smerdon and Beasley, 1961]. Recent work has focused on determining the physical soil strength processes to create physically based equations [Sang et al., 2015].  Parameters required in bank erosion models include, but are not limited to, critical shear stress, erodibility, water content, bulk density, cohesion, and particle size. A major current knowledge gap is the lack of a method to determine soil parameters over a large area without extensive soil testing [Langendoen, 2000].        Many researchers argue that a watershed scale is the most efficient scale for restoration [Wohl et al., 2005; Palmer et al., 2014; Ogston et al., 2015].  The measurement of erosion in a small watershed led to the conclusion that significant variation in erosion exists at individual sites.  The total sediment discharge  of a watershed is affected by the timing and magnitude of discharge events [Palmer et al., 2014]. However, instrumenting at a watershed scale is typically not economically or logistically feasible for many cases. The alternative approach is erosion modeling at the watershed scale.  Working on a watershed scale results in its own unique difficulties.  When applied to stream restoration, the correct link must exist between stream restoration and sediment budgeting. Stream restoration projects typically reduce erosion, which, in turn, reduces the sediment supply. In order to design an effective project, there must be an accurate understanding of the available sediment supply [Smith et al., 2011]. When comparing erosion across watersheds, soil erosion rates measured at one scale are often not representative of sediment yield at another spatial scale [de Vente and Poesen, 2005].  Looking forward, the modeling of erosion at a watershed scale is a research area that would have impacts on watershed planning, land use and restoration projects.

The watershed scale is becoming more important for decisions and prioritizing restoration projects. Bank erosion can be a major contributor to the overall sediment load of a stream, and physically based erosion models have the advantage of being able to determine spatial patterns of erosion.  However, the more complex the model, the more elaborate the data input requirements. Further development of soil parameterization methods would reduce the difficulty in setting up the complex erosion models.  In reality, heterogeneous soil properties lead to hotspots of erosion, which are an important characteristic that should be captured in a model, yet this requires detailed spatial soil data. Since traditional soil testing is often an expensive and time-consuming option, alternative options of soil parameterization will be explored. This research is noteworthy since it creates a method of soil parameterization for a bank erosion model based on currently available datasets.  The results and shortcomings identified will help guide further parameterization work, which will ultimately facilitate the complex modeling of erosion.

Hypothesis / Objective / Research Question

 

Hypothesis: 

Sufficient publicly available soil characterization data exists for the prediction of bank erosion, particularly in the design of stream channel modifications with present bank erosion models.

 

Objective:

The objective of this study was to determine the availability, feasibility, and reliability of soil data to parameterize a selected streambank erosion model that could potentially be coupled with a hydrologic model. Data examined included existing field data, SSURGO data, and local data from a restored stream channel. Changes in the constructed restoration site over a decade was used as a case study to illustrate the geomorphic changes of a stream channel for the case in which the groundwater-surface water regimes and channel responses were largely unknown; hard points were used in the original design to accommodate those uncertainties.

 

Research Questions:

To satisfy the objective, several research questions were addressed. First, what physically-based models exist to model bank processes? What are their limitations and input parameters? Next, the restoration site was studied. How has the Oliver Run site changed over the last 10 years? What effect has the hard points had on channel development? How does this compare to an earlier study’s (Niezgoda, 2004)  geomorphic model (FLUVIAL-12) predicted changes? Since different models are used in academia versus industry, there could be interesting findings. After the site investigation of Oliver Run, the SSURGO dataset will be utilized and there will be several questions related to it. Can simple soil characterization or Spatial Soil Data (SSURGO) be used to parameterize a physically based bank erosion model (CONCEPTS)? For this question, there will be a study site investigation as well as a regional study utilizing SSURGO data. One question remains which ties together all of the previous questions. Based on the results specified above, for watershed-scale modeling of erosional processes in streams, do we have bank material data that is sufficient to support realistic output from bank erosion models?

STREAMBANK CRITICAL SHEAR STRESS ESTIMATIONS USING SSURGO AND EMPIRICAL EQUATIONS

Sharing is caring!

Leave a Reply