SEARCH

SEARCH BY CITATION

Keywords:

  • river bank;
  • bank failure;
  • channelization;
  • bank stabilization

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. SITE DESCRIPTION
  5. METHODS
  6. RESULTS
  7. DISCUSSION
  8. CONCLUSIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES

Bank failure is a common fluvial process and can be a pervasive fluvial response to natural and anthropogenic disturbances. Previous research has identified causes and types of bank failure, but the conditions that lead to the cessation of bank failure remain poorly explained. This research examines differences between banks with active bank failure and banks that exhibit evidence of past bank failure that ceased (dormant) throughout three West Tennessee (USA) rivers to provide insight into the processes that cause bank failure to end. Bank characteristics were observed at 68 sites, and data from 55 banks were used to create a logistic regression model. Bank characteristics entered into the model included: vegetative cover, failure location, bar association, bank material, channel width-to-depth (w/d) ratio, and average bank angle. Results of the logistic regression suggest that bank angle best explains (p = 0.31 and odds ratio = 8.2) the difference between banks with active and dormant bank failure. Interestingly, vegetative cover and bank material composition, which have been found to be important in bank stabilization by previous researchers, were not significant predictors of bank stability according to the logistic regression model. These results suggest that in absence of drastic differences in bank material resistance (bedrock vs sediment): (1) spatial patterns of bank failure at the system-scale will be diffuse, (2) bank stability can require a multiple decades, and (3) the potential for vegetation to stabilize banks may be limited in some alluvial systems because of positive feedbacks created by repeated human disturbance. Copyright © 2012 John Wiley & Sons, Ltd.


INTRODUCTION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. SITE DESCRIPTION
  5. METHODS
  6. RESULTS
  7. DISCUSSION
  8. CONCLUSIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES

Bank failure naturally occurs in river systems and is understood to include all processes related to the erosion and mass wasting of river banks (Fonstad and Marcus, 2003). It is well documented that bank failure contributes a significant amount, up to 80%, of sediment to the total sediment load of rivers (Bull and Kirkby, 1997; Simon and Darby, 1999; Sekely et al., 2002; Evans et al., 2006). As a result, bank failure can pose a management dilemma when sediment contributions occur in quantities that threaten water quality or aquatic habitat.

Two main geomorphic processes contribute to the occurrence of bank failure in rivers: (i) fluvial entrainment of bank materials during flows and (ii) mass wasting of weathered bank materials, which often is preceded by subaerial erosion processes that weaken bank materials (Thorne, 1982; Lawler et al., 1999; Yumoto et al., 2006; Harden et al., 2009). Typically, bank failure studies have focused on the efficacy of different storm flows to entrain bank materials and have determined the factors contributing to the mass wasting of banks. Fluvial entrainment of bank materials appears to depend on peak flow intensity (Julian and Torres, 2006; Luppi et al., 2009) and the number of stage increases that occur prior to the peak discharge, which can prime bank material for removal during the peak discharge (Rinaldi et al., 2004).

Bank material composition (texture and layering) has been shown to be an important factor involved in bank failure. Banks with high silt and clay content have been found to be more susceptible to subaerial erosion processes but less susceptible to fluvial entrainment (Couper, 2003). Bank moisture content is an important contributor to subaerial weathering and erosion processes (Casagli et al., 1999; Simon et al., 1999; Couper et al., 2002). Pore water pressure affects matric suction of the bank material; and as a result, changes in pore water pressure cause slaking and weathering of bank faces (Thorne and Osman, 1988). Vegetation has been shown to make banks less failure-prone because root networks help keep bank material dry by increasing hydraulic conductivity (Thorne, 1982; Wynn, 2004). More recently, studies have begun to focus on the processes involved in bank undermining by groundwater seepage (Fox et al., 2007; Wilson et al., 2007; Cancienne et al., 2008).

Most of the existing research concerning bank failure focuses on measuring and explaining processes occurring on banks with current (active) bank failure. Studies that examine the geomorphic characteristics of banks with bank failure that has ceased are much needed. Such information is required to determine the factors involved in the cessation of bank failure and, ultimately, bank stabilization, which could help improve bank stabilization and river restoration efforts.

This study examines the characteristics of stream banks with active and dormant (failure activity has ceased) bank failure in three alluvial streams undergoing geomorphic adjustment to channelization that took place over 40 years ago. Examining banks with both active and dormant bank failure found throughout three different watersheds allows differences in bank properties and the locations of banks with active and dormant bank failure to be discerned, providing insight into the processes significant to explaining spatio-temporal variability of bank failure and achieving bank stability.

SITE DESCRIPTION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. SITE DESCRIPTION
  5. METHODS
  6. RESULTS
  7. DISCUSSION
  8. CONCLUSIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES

Human disturbance history

The study streams, Richland, Jeffers, and Dry Creeks are located in the Lower Hatchie River (LHR) Basin of West Tennessee (USA) (Figure 1). The total stream length varies with Jeffers Creek being ~10 km, Richland Creek being ~11 km, and Dry Creek being ~14 km. These tributaries of the LHR are an ideal location for examining bank failure processes because of the high occurrence of failures of many different types throughout each system. The LHR is a low gradient, large, alluvial river that begins in northern Mississippi, flows northward into western Tennessee, and joins the Mississippi River north of Memphis, Tennessee. The main stem of the Hatchie River is unchannelized, but most of its tributary streams in Mississippi and Tennessee have been channelized in some portion in the past (Diehl, 2000). Tributaries were channelized to ameliorate channel aggradation caused by historic human activities in the basin, including the land clearance and intense cultivation in upland areas, which began in the mid-1800s and accelerated hillslope erosion processes, resulting in gully and rill formation in headwater areas. Presently, these upland locations are forested with second growth vegetation and, when cleared, are used for pasture or hay production (National Resources Conservation Service, 1997). Channelization efforts consisted of resectioning (widening and deepening the channel to increase channel capacity) and straightening, which occurred between 1920 and the 1970s (Diehl, 2000). River sections channelized in the past often still retain a mainly trapezoidal cross-sectional shape that is indicative of channelization (Figure 2). Because channelization, the streams have incised as much as 2 m and are currently in a state of increased channel widening and aggradation (Davis, 2007, 2009).

image

Figure 1. The three study streams, Richland, Jeffers, and Dry Creeks located in the Lower Hatchie Basin of west Tennessee, USA

Download figure to PowerPoint

image

Figure 2. Example of a reach in Jeffers Creek with trapezoidal cross-sectional morphology that is strongly suggestive of channelization. Note lack of streamflow. This figure is available in colour online at wileyonlinelibrary.com/journal/rra

Download figure to PowerPoint

Hydrology and geology

The study streams are not gaged and have very little flow year-round, except in the spring when the most precipitation is received (National Resources Conservation Service, 1997). During the summer, autumn, and winter, stream channels are mostly dry (Figure 2). This may be related to the unconsolidated sediment that underlies each watershed being very thick, which is conducive to water table fluctuations, and/or high evapotranspiration rates in the summer, or anthropogenic activities that have resulted in water table lowering, such as crop irrigation. Communications with a landowner whose family has owned and operated a farm in the Richland Creek watershed for over a century revealed that many of the tributaries, including Richland Creek, were once navigable by small steamboats year-round. Such anecdotal evidence supports an anthropogenic cause for streams being seasonally dry given that the change seems to have occurred within the timeframe of European settlement of the area. In addition, the streams have a flashy hydrologic response, presumably in part because of the increased gradient that came with channelization, and also because of increased runoff from land clearance and ditches dug specifically to expediently drain agricultural fields.

The Lower Hatchie Basin lies within the Lower Mississippi Embayment and, as a result, contains a variety of unconsolidated fluvial and coastal deposits. The geology is mainly composed of Holocene fluvial deposits, Pleistocene/Pliocene loess and fluvial deposits, and Tertiary Coastal Plain sediments (Miller, 1974; Luther, 1977; National Resources Conservation Service, 1997). Substrates found within the study watersheds vary from the tributaries' own Holocene alluvium in their floodplains to occasional deposits of Pleistocene loess, which range in thickness, and Tertiary coastal plain sediments found in upland and hilltop areas (National Resources Conservation Service, 1978; National Resources Conservation Service, 1995; National Resources Conservation Service, 1997). However, Pleistocene/Pliocene fluvial deposits, consisting of quartz-rich sand, some silt and clay, are the most commonly occurring substrate in upland areas and hilltops (National Resources Conservation Service, 1978; National Resources Conservation Service, 1995; National Resources Conservation Service, 1997). The lack of bedrock outcroppings and minimal occurrence of cohesive clay exposures in all but a few upland portions of the watersheds translates to geology largely composed of unconsolidated sediments, which contributes to low bank resistance to erosion and high incidence of failure.

METHODS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. SITE DESCRIPTION
  5. METHODS
  6. RESULTS
  7. DISCUSSION
  8. CONCLUSIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES

Bank process observations

Bank processes were observed at a total of 68 locations across the three study streams (two banks observed in 34 reaches, each 100 m long). Sites were chosen by dividing each of the study watersheds into sub-watersheds using ArcGIS (ArcHydro extension) (Release 9.3.1, ESRI, Redlands, CA, USA) and digital elevation models of the watersheds, and this information was used to identify sub-watersheds for sampling.

Thorne (1998) was used as a guide for identifying bank failure types and making bank observations. In this study, field observations of bank processes focused on the main types of bank failures extensively described and photographed by Thorne (1998). These include shallow slides, referred to as slumps (which are common in low-cohesion banks and often occur after rotational slips and slab failures), rotational slips (which occur in highly cohesive materials), slab-type block failure, referred to as slab failures (associated with steep bank angles and cohesive materials), cantilever failure (failure of overhanging material, which usually occurs in banks with layers of material that alternate from cohesive to non-cohesive), pop-out failure (which results from bank seepage processes in a cohesive, steep bank), and piping failure (collapse of a portion of a bank because of groundwater flow rates and pressures).

Bank characteristics observed in the field were also developed using guidelines discussed in Thorne (1998). An average bank angle was calculated for each bank by averaging measurements of the upper, middle, and lower bank slopes made with an Abney level. Bank angles used to calculate an average bank angle for each site were taken with the Abney level at 25% of the slope for upper bank measurements, 50% of the slope for middle bank measurements, and 75% of the slope for lower bank measurements. Bank characteristics were examined at each site. Characteristics used to model bank failure status using statistical analyses included: bank material composition, bank failure location, bank-face vegetation characteristics, bank failure status, and bar association. Bank material composition was determined in the field by hand from bulk samples pulled from each part of the bank face—upper, middle, and lower and described using USDA textural classes (silty clay, silty clay loam, silt loam, silt, sandy clay, sandy clay loam, sandy loam, loamy sand, sand, clay loam, and loam). Subsequently, the texture of each bank was classified into one of three groups for statistical analysis based on the dominant texture of each bank. Dominant texture was based on the first term or the single term in some instances (clay, silt, sand, and loam) of the USDA textural classification for soil separates for upper, middle, and lower banks at each site. If a textural term was used as the first term or the only term for two or more sections of the bank face, then it was chosen as the overall textural characterization for the bank. After reviewing the bank textural classifications made in the field, it was determined that the overall texture of the banks fit into one of three groups: sandy, loamy, or composite (sand and/or loam with a clay unit present). These categories represent varying degrees of bank resistance to erosion and mass failure, with sandy being the most failure prone and composite the least. Bank failure location was noted as either inside bend, outside bend, or straight, and was noted to determine whether the propagation of bends through the system as part of increased lateral migration play a role in determining occurrence of bank failures. Bank-face vegetation was characterized as being herbaceous cover, tree cover, or no vegetation present to determine any effects that vegetation may have on stabilizing bank faces.

Bank failure status was rated as active, dormant, or none. Active bank failures showed evidence of recent failure activity, including uprooted, damaged, or dead vegetation, exposed roots, convex or vertical bank faces, fresh (unweathered) bank material exposures, and unworked deposits of failed bank material at the base of banks. Dormant bank failures exhibited evidence that failures had occurred in the past but had since ceased. Such evidence included visible, weathered failure scars with one more of the following: mature trees with bent trucks (the presence of reaction wood) that ultimately straighten, concave/‛relaxed bank slopes, and bank material vegetated with mature trees or saplings. If no evidence could be found to support either active or dormant bank failure, the bank was rated as not having failures present (‘none’) and was excluded from further analyses. The presence of bars and their location relative to banks (opposite bank, attached to bank, or no bar present) was also noted to determine whether bars attached to banks initiate bank stabilization (bank failure dormancy) by protecting the bank from shear stresses experienced during storm flows. Additionally, an average channel w/d ratio was calculated for each reach using cross-sectional measurements of average channel width and average depth (from top of banks) made using a laser level with detector with a manufacturer specified accuracy of 1 cm.

Modeling bank failure status

With exception of average bank angle and channel w/d ratio, most of the data collected in the field yielded categorical data. The classification of bank status as either active or dormant meant that the dependent variable was binary. As a result of both of these factors, logistic regression was chosen to model the effects of different bank characteristics on determining bank failure status.

Logistic regression is commonly used in the social sciences and, in recent years, has become increasingly prevalent in the physical sciences. In geomorphology, logistic regression has been used to model and predict the location of landslides (Eeckhaut Van Den et al., 2006; Chang et al., 2007), to assess shallow earthflow susceptibility (Tolga et al., 2005), and to predict changes in channel pattern (Bledsoe and Watson, 2001). It is primarily used for two purposes: (i) to determine which independent variables (predictors) are important to the occurrence or non-occurrence of the dependent variable and (ii) to create a parsimonious equation to be used to predict the occurrence of the dependent variable. In this study, multiple logistic regression was used to determine which independent variables (average bank angle, channel w/d ratio, bank material composition, bank-face vegetation, failure location, and bar association) are important to determining the likelihood that banks will have active or dormant bank failure (the dependent variable), and in so doing, highlight the factors that contribute to bank stabilization over time.

For this research, a six parameter multiple logistic regression equation was used in the form of the following:

  • display math(1)

where x1 is average bank angle, x2 is channel w/d ratio, x3 is bank material composition, x4 is bar association, x5 is failure location, and x6 is bank-face vegetation. Because having a large range of values can make interpreting regression coefficients difficult, the measurements for average bank angle and channel w/d ratio were converted to z-scores and normalized prior to being put into the logistic regression model. Categorical independent variables were coded prior to analyses. Categories for each of the independent variables are summarized in Table 1.

Table 1. Summary of categories for independent and dependent variables used in the logistic regression model
Bank angleW/D ratioVegetationErosion locationBar associationBank materialFailure status
No categoriesNo categoriesNoneAbsentAbsentCompositeDormant
  HerbaceousOutside meanderBehind barLoamyActive
  TreesInside meanderOpposite barSandy 

In cases where a variable had two categories, the number 0 was assigned to one category and the number 1 to the other. For example, bank status had two categories: dormant = 0 and active = 1. In the case of independent variables with three categories (the maximum possible in this study), values 0–2 were assigned, and these were re-coded by spss (Release 19.0.0, IBM SPSS Statistics for Macs, Armonk, NY, USA) to be binary. In such instances, spss treats one category, specified by the user, as a reference value, and codes the other two categories as either 0 or 1, which are then compared with the reference value to determine their relative importance within the group before being added to the model. For example, the variable ‘bank composition’ had three categories: sandy, loamy, and composite (having unconsolidated layers with a cohesive clay unit). It was coded by spss as follows: composite = reference, 0 = loamy, and 1 = sandy. When doing logistic regression, it is advisable to run the model in a variety of ways (entering all independent variables at once or one at a time) to discern which method results in the most robust model overall (Elliott and Woodward, 2007). For this study, the logistic regression was run in spss in two ways. Firstly, by having all independent variables included at the start; and secondly, by adding independent variables one at a time to see if the model improved with each addition. Both methods resulted in the same variables being statistically significant/insignificant, but the first method (all independent variables entered at once) yielded the more robust model overall according to standard model diagnostics (model summary output, Hosmer and Lemeshow Test, and classification table output, all of which are discussed in more detail later). As a result, output from the first model was chosen for reporting.

RESULTS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. SITE DESCRIPTION
  5. METHODS
  6. RESULTS
  7. DISCUSSION
  8. CONCLUSIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES

Bank process observations

As expected, given the disturbance history of the three study streams, bank failure was found to be widespread throughout each of the streams. Of the 68 sites surveyed, 55 exhibited evidence of active or dormant bank failure, and 13 showed no evidence of either active or dormant bank failure (Tables 2-4). There were 36 banks that showed evidence of active bank failure, and 19 that showed evidence of dormant bank failure. Bank failure processes included cantilever, pop-out, slump, slab, and in some cases, there was evidence that soil piping played a role in failures (Figure 3). Slab failure and slumping were the most prevalent types of bank failure modes in all three streams.

Table 2. Description of bank failures in Richland Creek
ReachBankModeLocationStatus
R1aSlumpMid and upperActive
bSlumpMid and upperActive
R2aNoneNoneNone
bSlabMid and upperActive
R3aSlabUpperNone
bCantileverUpperActive
R4aSlumpUpperNone
bSlumpWholeNone
R5aSlumpMid and upperActive
bNoneNoneNone
R6aSlumpMid and upperActive
bSlumpUpperDormant
R7aNoneNoneNone
bSlumpWholeActive
R8aSlumpUpperDormant
bSlumpUpperActive
Table 3. Summary of bank failures in Jeffers Creek
ReachBankModeLocationStatus
J1aSlumpUpperActive
bSlumpUpperActive
J2aSlabUpperDormant
bSlabUpperActive
J3aSlabUpperActive
bSlabUpperActive
J5aSlumpUpperActive
bSlumpUpperActive
B1aSlumpMiddleNone
bSlumpMid/upperNone
B2aPop-outWholeActive
bSlabUpperNone
B3aSlabUpperNone
bSlab; pop-outMid/upperActive
B4aSlumpUpperDormant
bSlabUpperActive
RB1aPop-out/pipingUpperActive
bSlabUpperDormant
RB2aCantilever/pipingUpperNone
bSlabUpperActive
RB3aSlabUpperDormant
bSlumpWholeDormant
RB4aSlumpUpperActive
bSlumpWholeActive
RB5aSlumpUpperDormant
bSlumpUpperDormant
RB6aSlumpUpperActive
bSlabUpperActive
Table 4. Summary of bank failures in Dry Creek
ReachBankModeLocationStatus
D1aSlabUpperDormant
bSlabUpperActive
D2aSlumpUpperDormant
bSlab/pipingUpperActive
D3aSlumpUpperDormant
bSlumpUpperActive
D4aSlabUpperNone
bSlabUpperActive
D5aNoneNoneActive
bSlumpUpperNone
D6aSlumpWholeActive
bSlumpUpperDormant
D7aSlabUpperDormant
bSlabUpperActive
D8aSlabUpperDormant
bSlabUpperDormant
D9aSlabUpperDormant
bSlump/pipingUpperActive
D10aSlumpMid/upperActive
bSlumpUpperActive
D11aSlump/slabMid/upperActive
bSlabUpperDormant
D12aSlump/slabUpperDormant
bSlumpUpperActive
image

Figure 3. Examples of bank failure found in the three study streams: (a) slab failure, (b) slumping, and (c) pop-out failure. Slab failure and slumping were the two most common failure types documented. This figure is available in colour online at wileyonlinelibrary.com/journal/rra

Download figure to PowerPoint

Logistic regression model of bank failure status

The variables vegetative cover—trees, failure location—inside meander, failure location—outside meander, bar association—behind bar, bar association—opposite bar, and bank material—loamy all displayed inverse relationships (negative β coefficients) with active bank failure status, whereas the variables vegetation—herbaceous cover, bank material—sandy, w/d ratio, and bank angle displayed positive relationships with active bank failure status. However, only two variables, failure location—inside meander (Figure 4) and bank angle (Figure 5), were found to be statistically significant predictors (p = 0.037 and p = 0.031, respectively) of whether a bank had active or dormant bank failure (Table 5). The odds ratio for bank angle indicates that with each increase in bank angle degree, there is an 8.2 times increase in the likelihood that a bank will have active bank failure. The odds ratio for failure location—inside meander bend indicates that the likelihood of the occurrence of active bank failure decreases by 0.052 times if the location is inside a meander bend.

image

Figure 4. Predicted probabilities of active bank failure as a function of erosion location. X-axis values represent locations, with 0 = inside meander bend, 1 = outside meander bend, and 2 = straight section

Download figure to PowerPoint

image

Figure 5. Predicted probabilities of active bank failure as a function of bank angle z-scores

Download figure to PowerPoint

Table 5. Results of logistic regression model.
 βSEWalddfSig.Exp(β) – odds ratio
  1. X-axis values are normalized z-scores of bank angles measured in degrees. Bank angle measurements were normalized and converted to z-scores prior to logistic regression to reduce the range of values and make the interpretation of β coefficients easier.

Vegetation  3.49220.175 
Herbaceous cover1.1881.4970.62910.4283.280
Trees−1.9171.2702.27910.1310.147
Failure location  4.80720.090 
Inside Meander−2.9611.3744.64510.0310.052
Outside Meander−1.1671.2640.85210.3560.311
Bar association  0.18020.914 
Behind bar−20.4849453.0570.00010.9980.000
Opposite bar−20.0119453.0570.00010.9980.000
Bank material  0.10020.951 
Loamy−0.4821.6430.08610.7690.618
Sandy0.0051.0490.00010.9961.005
W/D ratio0.6840.5241.69910.1921.981
Bank angle2.1021.0074.35510.0378.183
Constant23.0519453.0570.00010.99810255653115.124

In terms of model performance, the model classified or predicted 79% of dormant bank failures correctly and 92% of active bank failures correctly (Table 6). The other measure of model efficacy used in this study was also within an acceptable range. The Hosmer–Lemeshow test, which is an inferential goodness-of-fit test, had a chi-square of 8.29, 7 degrees of freedom, and an insignificant p value (p = 0.31), suggesting that the model fit the data well, that is, the null hypothesis that the model fit the data was not rejectable (Peng et al., 2002).

Table 6. Classification table for logistic regression model. The cut value was 0.500
Classification table
ObservedPredicted
Failure statusPercentage correct
DormantActive
Failure statusDormant15478.9
Active33391.7
Overall percentage  87.3

DISCUSSION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. SITE DESCRIPTION
  5. METHODS
  6. RESULTS
  7. DISCUSSION
  8. CONCLUSIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES

Bank process observations

The two most common types of bank failure that occurred in the three study streams were slumps and slab failures. These mass failure modes are common in unconsolidated, highly erodible bank material, which is present throughout the study streams. System-wide bank failure is a common geomorphic adjustment process that occurs after channelization (Simon, 1992; Simon and Hupp, 1992). In these study streams, only 13 of the 68 banks surveyed did not show some evidence of either active or dormant bank failure, 36 sites exhibited active failure, and 19 exhibited dormant failure. It is somewhat remarkable that, after at least 40 years of geomorphic adjustment, many of the banks of the three study streams have yet to reach a stable state. This somewhat prolonged state of geomorphic adjustment is probably due to two factors. First, the unconsolidated material that dominates the three study watersheds provides very little resistance to either mass wasting or fluvial erosion processes. Second, the study streams are tributaries with smaller watershed areas, meaning that there is less spatial variability of bank material; and thus, bank resistance does not vary significantly from one location to another, creating a broad distribution of failures throughout each system.

Logistic regression model of bank failure status

Previous researchers have identified several different factors and processes that are significant to bank failure at individual bank locations, including the presence or absence of vegetation, bank material moisture status, bank material texture, mass wasting processes, and subaerial erosion processes. This study attempted to use many of these known bank failure factors in a logistic regression model in order to discern differences between banks with active failure and those with dormant failure (i.e. changes in bank failure over time) and to help identify processes that help explain why bank failure can occur simultaneously at multiple locations throughout a river system (i.e. spatial variability).

The results of the logistic regression model suggest that two factors are most significant to the cessation of bank failure and the distribution of active bank failure throughout the study systems. They include the location of failure, specifically inside meander bends, and bank angle. Both variables were statistically significant, but each behaved differently in the model.

The β coefficient for failure ‘location —inside meander’ was negative, and this indicates an inverse relationship between banks located inside bends and the occurrence of active bank failure. At first glance, such a relationship makes sense given that flow is often deflected to outside bend locations by meandering thalwegs, resulting in bank retreat in banks on the outside bend and not the inside of bends. However, bank failure ‘location—outside meander’ also had a negative β coefficient (although not statistically significant), suggesting an inverse relationship for it as well. The odds ratio for both ‘inside bend’ and ‘outside bend’ locations was relatively small (0.052 and 0.311, respectively), meaning the decrease in active bank failure likelihood was very small in both cases, and as a result, it is difficult to accept that there is any real effect provided by the bank failure location variable.

Bank angle was the other statistically significant factor, with p = 0.031 and an odds ratio of 8.2. The large odds ratio combined with the statistically significant p value (<0.05), means that it is more plausible that bank angle is a determining factor in the progression of bank failure from an active to a dormant state. Bank angle is probably the most significant variable in the model because of the dominance of unconsolidated bank material found in the three study streams. Because banks are mostly unconsolidated, once bank failure begins, whatever the initial mechanism, it does not conclude until a stable bank angle has been reached. Bank failures likely occur throughout each system because there are few locations with lithological constraints, such as bedrock or cohesive sediments, to limit the spatial extent of bank failure.

The results of the bank observations and logistic regression model are interesting in that they highlight the longevity and diffuse nature of channelization impacts, particularly in settings with unconsolidated sediments. They also confirm the work of others (Simon, 1992, 1994), in that they suggest that bank failure persists in incised streams until bank angles and presumably bank heights are reduced. The results of this study are also interesting for what they did not reveal. Bank material composition (texture) was not an important factor in the logistic regression model and neither was vegetative cover. There are several possibilities for why bank material was not a significant predictor of bank failure status. It could be that the groupings used to describe overall texture for each bank were too simplistic; and therefore, the model did not detect differences based on texture. However, the textural groupings used in the logistical regression model represented a range of mass wasting/fluvial erosion resistance based on different levels of cohesiveness, with ‘sandy’ representing low resistance, ‘loamy’ representing medium resistance, and ‘composite/clay unit present’ representing the most resistance. Therefore, it seems most plausible that the relative homogeneity of bank texture (i.e. mostly unconsolidated sediments) makes bank texture less of an issue in the study streams. The fact that vegetation was not a significant predictor of bank failure status is contradictory to previous studies (Simon and Hupp, 1992) conducted in larger, channelized streams where vegetation was found to help banks stabilize. There are several possible explanations for this contradictory result. It is possible that the flashy hydrology that exists in the three study streams as a result of the channelization creates flows that are not conducive to the establishment of vegetation, limiting the role vegetation can play in bank stabilization. Another possible explanation is that there has not been enough time for fluvial geomorphic adjustment to occur, meaning that channel incision and widening is still active and habitually creating banks poised to fail. These results suggest that in highly disturbed streams, such as those that have experienced channelization, the ameliorative or protective role that vegetation plays in stabilizing banks in some river systems can be superseded by the severity of the impacts of channelization.

Study limitations

There were two main limitations to this study. Firstly, the data used in the logistic regression model heavily relied on accurate determination of bank characteristics and bank failure status. Every effort was made to be diligent and consistent in making bank observations, and protocols outlined in Thorne (1998) were followed as closely as possible. Second, a large number of banks were observed in the field for this study but, in retrospect, it would have been beneficial to the model to have more samples of dormant bank failure, because the majority of the banks surveyed happened to be ones with active bank failure. According to the model diagnostics, the model did a better job of predicting banks with active bank failure than with dormant failure, and this was probably related to the smaller sample size of banks with dormant bank failure.

CONCLUSIONS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. SITE DESCRIPTION
  5. METHODS
  6. RESULTS
  7. DISCUSSION
  8. CONCLUSIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES

Bank failure in alluvial streams with unconsolidated bank materials can be a spatially diffuse and long enduring response to channelization. Without lithological constraints, bank failure is not limited spatially or temporally; and as a result, bank stability requires a long time to be achieved. When bank stability is achieved, it occurs primarily as a result of reduced bank angle. Settings such as the study tributaries pose a difficult situation for watershed and natural resource managers in that bank stabilization efforts would need to be carried out throughout whole watersheds to limit the impact of bank failure on sediment loads, but such efforts are usually too costly to be implemented.

ACKNOWLEDGEMENTS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. SITE DESCRIPTION
  5. METHODS
  6. RESULTS
  7. DISCUSSION
  8. CONCLUSIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES

This research was supported by a grant from the National Science Foundation (NSF-BC-GSS-0402503) and a contract from the Tennessee Department of Conservation. The authors thank an anonymous reviewer whose comments greatly improved the quality of this paper and the Cartographic Research Laboratory at the University of Alabama for their assistance with the site location figure.

REFERENCES

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. SITE DESCRIPTION
  5. METHODS
  6. RESULTS
  7. DISCUSSION
  8. CONCLUSIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  • Bledsoe BP, Watson CC. 2001. Logistic analysis of channel pattern thresholds: meandering, braiding, and incising. Geomorphology 38: 281-300.
  • Bull LJ, Kirkby MJ. 1997. Gully processes and modeling. Progress in Physical Geography 21(3): 354374. DOI: 10.1177/030913339702100302.
  • Cancienne R, Fox GA, Simon A. 2008. Influence of seepage undercutting on the root reinforcement of streambanks. Earth Surface Processes and Landforms 33(11): 17691786.
  • Casagli N, Rinaldi M, Gargini A, Curini A. 1999. Pore water pressure and streambank stability: results from a monitoring site on the Sieve river, Italy. Earth Surface Processes and Landforms 24: 10951114. DOI: 0.1002/(SICI)10969837(199911)24:12<1095::AID-ESP37>3.0.CO;2-F.
  • Chang KT, Chiang SH, Hsu ML. 2007. Modeling typhoon- and earthquake-induced landslides in a mountainous watershed using logistic regression. Geomorphology 89(3-4): 335-347. DOI: 10.1016/j.geomorph.2006.12.011.
  • Couper P. 2003. Effects of silt-clay content on the susceptibility of river banks to the subaerial erosion. Geomorphology 56: 95108. DOI: 10.1016/SO169-555X(03)00048.5.
  • Couper P, Stott T, Maddock I. 2002. Insights into river bank erosion processes derived from analysis of negative erosion-pin recordings: observations from three recent UK studies. Earth Surface Processes and Landforms 27: 5979. DOI: 10.1002/esp.285.
  • Davis L. 2007. Spatial patterns of geomorphic adjustment in channelized, tributary streams. Physical Geography 28(4): 301310. DOI: 10.2747/0272-3646.28.4.301.
  • Davis L. 2009. Sediment entrainment potential in modified alluvial streams: implications for re-mobilization of stored in-channel sediment. Physical Geography 30(3): 249- 268. DOI: 10.2747/0272-3646.30.3.249.
  • Diehl TH. 2000. Shoals and valley plugs in the Hatchie River Watershed. Water-Resources Investigations Report 00-4279. U.S. Geological Survey.
  • Eeckhaut Van Den M, Vanwalleghem T, Poesen J, Govers G, Verstraeten G, Vandekerckhove L. 2006. Prediction of landslide susceptibility using rare events logistic regression: a case-study in the Flemish Ardennes (Belgium). Geomorphology 76 (3-4): 392-410. DOI: 10.1016/j.geomorph.2005.12.003.
  • Elliott AC, Woodward WA. 2007. Statistical Analysis Quick Reference Guidebook. Sage Publications: Thousand Oaks, California.
  • Evans DJ, Gibson CE, Rossell RS. 2006. Sediment loads and sources in heavily modified Irish catchments: A move towards informed management strategies. Geomorphology 79(1–2): 93113.
  • Fonstad M, Marcus AW. 2003. Self-organized criticality in riverbank systems. Annals of the Association of Geographers 93(2): 281-296. DOI: 10.1111/1467-8306.9302002.
  • Fox GA, Wilson GV, Simon A, Langendoen E, Akay O, Fuchs JW. 2007. Measuring streambank erosion due to ground water seepage: correlation to bank pore water pressure, precipitation, and stream stage. Earth Surface Processes and Landforms 32(10): 15581573. DOI: 10.1002/esp.1490.
  • Harden CP, Foster W, Morris C, Chartrand K, Henry E. 2009. Rates and processes of streambank erosion in tributaries of the Little River, Tennessee. Physical Geography 30(1): 116. DOI: 10.2747/0272-3646.30.1.1.
  • Julian JP, Torres R. 2006. Hydraulic erosion of cohesive riverbanks. Geomorphology 76: 193206. DOI: 10.1016/j.geomorph.2005.11.003.
  • Lawler DM, Grove JR, Copperthwaite JS, Leeks GJL. 1999. Downstream change in river bank erosion rates in the Swale-Ouse system, northern England. Hydrological Processes 13: 977992. DOI: 10.1002/(SICI)1099-1085(199905)13:7<977::AID-YP785>3.0.CO;2-5.
  • Luppi L, Rinaldi M, Teruggi LB, Darby SE, Nardi L. 2009. Monitoring and numerical modelling of riverbank erosion processes: a case study along the Cecina River (central Italy). Earth Surface Processes and Landforms 34: 530546. DOI: 10.1002/esp.1754.
  • Luther ET. 1977. Our Restless Earth: The Geologic Regions of Tennessee. The University of Tennessee Press: Knoxville, USA.
  • Miller RA. 1974. The Geologic History of Tennessee. State of Tennessee Department of Conservation, Division of Geology: Nashville, USA.
  • National Resources Conservation Service. 1978. Soil Survey of Madison County, Tennessee. U.S. Department of Agriculture.
  • National Resources Conservation Service. 1995. Soil Survey of Haywood County, Tennessee. U.S. Department of Agriculture.
  • National Resources Conservation Service. 1997. Soil Survey of Hardeman County, Tennessee. U.S. Department of Agriculture.
  • Peng CJ, Lee KL, Ingersoll GM. 2002. The Journal of Educational Research 96(1): 3-14. DOI: 10.1080/00220670209598786.
  • Rinaldi M, Casagli N, Dapporto S, Gargini A. 2004. Monitoring and modelling of pore water pressure changes and riverbank stability during flow events. Earth Surface Processes and Landforms 29: 237254. DOI: 10.1002/esp.1042.
  • Sekely AC, Mulla DJ, Bauer DW. 2002. Streambank slumping and its contribution to the phosphorus and suspended sediment loads of the Blue Earth River, Minnesota. Journal of Soil and Water Conservation 57(5): 243250.
  • Simon A. 1992. Energy, time, and channel evolution in catastrophically disturbed fluvial systems. Geomorphology 5(3–5): 345372.
  • Simon A. 1994. Gradation processes and channel evolution in modified West Tennessee streams: process, response, and form. U.S. Geological Survey Professional Paper 1470.
  • Simon A, Darby SE. 1999. The nature and significance of incised river channels. In Incised River Channels: Processes, Forms, Engineering and Management, Darby SE, Simon A (eds). Wiley: Chichester, UK.
  • Simon A, Hupp C. 1992. Geomorphic and vegetative recovery processes along modified stream channels of West Tennessee. U.S. Geological Survey Open-File Report, vol. 91-502.
  • Simon A, Curini A, Darby S, Lagendoen EJ. 1999. Streambank mechanics and the role of bank and near-bank processes in incised channels. In Incised River Channels, Darby SE, Simon A (eds). John Wiley & Sons, Ltd: Chichester, UK.
  • Thorne CR. 1982. Processes and mechanisms of river bank erosion. In Gravel-bed Rivers, Hey RD, Bathurst JD, Thorne CR (eds). John Wiley & Sons, Ltd: Chichester, UK.
  • Thorne CR. 1998. Stream Reconnaissance Handbook. Wiley: Chichester, UK.
  • Thorne CR, Osman AM. 1988. The influence of bank stability on regime geometry of natural channels. In River Regime, White WR (ed). John Wiley & Sons, Ltd: Chichester, UK.
  • Tolga C, Nefeslioglu HA, Gokceoglu C, Sonmez H, Duman TY. 2005. Susceptibility assessments of shallow earthflows triggered by heavy rainfall at three catchments by logistic regression analyses. Geomorphology 72(1-4): 205-271. DOI: 10.1016/j.geomorph.2005.05.011.
  • Wilson GV, Periketi RK, Fox GA, Dabney SM, Shields FD, Cullum RF. 2007. Soil properties controlling seepage erosion contributions to streambank failure. Earth Surface Processes and Landforms 32(3): 447459. DOI: 10.1002/esp.1405.
  • Wynn TM. 2004. The effects of vegetation on streambank erosion. PhD thesis, Virginia Polytechnic Institute and State University, Blacksburg, VA.
  • Yumoto M, Ogata T, Matsuoka N, Matsumoto E. 2006. Riverbank freeze-thaw erosion along a small mountain stream, Nikko Volcanic area, Central Japan. Permafrost and Periglacial Processes 17: 325339. DOI: 10.1002/ppp.569.