- guidelines for stabilising streambanks with riparian … off... · 2018. 4. 17. · 4 riparian...

GUIDELINES FOR STABILISING STREAMBANKSWITH RIPARIAN VEGETATION

TECHNICAL REPORT 99/10September 1999

Bruce Abernethy / Ian D. Rutherfurd

C O O P E R A T I V E R E S E A R C H C E N T R E F O R C A T C H M E N T H Y D R O L O G Y

i

Preface

In Australia, poor management practices havefostered the substantial and ongoing degradation ofriparian lands. Removal, fragmentation andalternation of vegetation cover, combined withchanged flow regimes, has increased the incidence ofriverbank erosion. Decreased riverbank stabilityresults in accelerated changes in channel morphology,lost agricultural production and reduced water quality.Poorly managed riparian zones have also led toincreased movement of sediments, nutrients and othercontaminants from surrounding lands into riversystems.

Over the past six years, the Land and WaterResources Research and Development Corporationhas run a national program of research aimed atimproving the management of riparian landsthroughout Australia. Within this framework, theCooperative Research Centre for CatchmentHydrology has conducted a range of scientificinvestigations to quantify the physical influence ofriparian vegetation over river processes, includingbank erosion and nutrient and sediment movementthrough buffers.

Good progress has been made in this research and weare now in a position to communicate some of theresults to landholders and stream managers. One issuethat is of particular interest to stream managers is howwide vegetated riparian strips need to be alongstreams to perform various functions. In this regard,the Queensland Department of Natural Resourcescontracted the Cooperative Research Centre forCatchment Hydrology to write technical guidelines,now produced in this report. The guidelines providetechniques to help specify the width and compositionof vegetated riparian zones, for bank erosion control.A companion report to this one, published by CSIROLand and Water (Karssies & Prosser, 1999), providesdetail on the design of riparian zones to filtersediment and nutrients from overland flow enteringstreams.

COOPERAT IVE RESEARCH CENTRE FOR CATCHMENT HYDROLOGY

Guidelines forstabilisingstreambanks withriparianvegetation

Bruce Abernethy and Ian D. Rutherfurd

Cooperative Research Centre for Catchment Hydrology

Department of Geography and Environmental Studies

University of Melbourne, Parkville, Victoria, 3163

September, 1999


ii

Abstract

This report provides guidelines for establishingriparian plantations that will stabilise riverbanks,within acceptable limits. Riparian vegetation interactswith a range of geomorphological, geotechnical,hydrological and hydraulic factors to affect the typeand extent of riverbank erosion. The enhanced lateralchannel stability offered by well-vegetated riparianzones can also reduce the need for engineeredstabilisation and heavy maintenance.

The density and type of riparian vegetation coverstrongly influence all aspects of riverbank erosion.Riparian forests are typically composed of overstorey,understorey, groundcover and macrophyte species.Vegetation condition is assessed in terms of the Raineand Gardiner (1995) ‘traffic light’ classification:green, yellow red for riparian vegetation in good,intermediate and poor condition. The mainconsideration when designing riparian revegetationworks for bank stability is continuity of cover.

Different types of vegetation affect differentprocesses, so it is imperative to assess the dominanterosion process correctly so that appropriate speciescan be selected for stability. All of the erosionprocesses that act on banks can be grouped togetherinto three erosion-domains. (1) Subaerial erosion(erosion caused by processes external to the streamsuch as cattle, or rain splash). (2) Scour (removal ofindividual sediment particles or aggregates by flow).(3) Mass-failure (slumps). An underlying philosophyof these guidelines is that the dominant bank erosionprocess changes downstream along a river due tochanges in channel scale.

The guidelines stipulate minimum riparian zoneestablishment-widths. Minimum riparian zone widthsare calculated individually for each site on the basisof the present site conditions (observed bankgeometry) and the past erosion history (measured orestimated bank erosion rate). The basic allowanceforthe width of any riparian plantation designed for bankstabilisation should not be less than 5 m measuredonto the floodplain from the bank crest. As banksbecome higher they become less stable. Hence, inaddition to the basic allowance, we recommend thatthe width of riparian strips also include a height

allowancenot less than the height of the bankmeasured vertically from the bank toe to the bankcrest. Time must be allowed for the plants to growbefore they can begin to stabilise the bank, so wherebanks are actively eroding an establishmentallowance should also be included in the final riparianzone width. The establishment allowance isdetermined by multiplying the erosion rate by thetime required for the plantation to mature.

The material presented in this document issummarised and collated into a series of tables thatcan be used to guide and focus the practitioner’sapproach to planning riverbank stability works withvegetation. The report does not account forecological, or sediment and nutrient filtering, or othercriteria that may dictate specific riparian managementregimes beyond those appropriate for bankstabilisation. The width and character of plantationsdesigned for other criteria may be quite different thanplantations designed for bank erosion control. Thus,these guidelines should be considered in the light ofother requirements for riparian zone management.


iii

Acknowledgements

The research, on which these guidelines are based,was funded by the Land and Water ResourcesResearch and Development Corporation and theCooperative Research Centre for CatchmentHydrology. Further funding from the QueenslandDepartment of Natural Resources allowed translationof the original research to its present form.

Special thanks are due to the following individuals.

• Marek Komarzynski, Fiona Cavanagh andKathryn Jerie (Cooperative Research Centre forCatchment Hydrology) for assistance withfieldwork.

• Allan Pedder and Richard Crook (Latrobe Riverlandholders) for access to field sites.

• Ross Scott and Phil Taylor (Lake WellingtonRivers Authority) for loan of field equipment.

• Roy Goswell and Roger Doulis (Department ofCivil Engineering, Monash University) fortechnical assistance with field and laboratoryequipment.

• David Schmiede (Queensland Department ofNatural Resources) for his helpful reviews ofearlier drafts.

• John Amprimo (Queensland Department ofNatural Resources) who initiated the guidelines.

iv


1 Introduction 1

1.1 Purpose of the guidelines 1

1.2 Principles of using vegetation for bank

erosion control 2

1.3 Terms 3

2 Guidelines for determining minimum riparian zone widths 5

2.1 Recommendation 5

2.2 Decision tree 7

2.3 Worked example 8

3 Riparian forest structure 11

3.1 Overstorey 11

3.2 Understorey 11

3.3 Groundcover 11

3.4 Macrophytes 11

4 Riparian vegetation and streambank stability 13

4.1 Mass failure 13

4.2 Fluvial scour 14

4.3 Subaerial preparation 15

5 Assessment of existing riparian condition 17

5.1 Reach assessment 17

5.2 Bank assessment 17

5.3 Vegetation assessment 18

6 Riparian plantation design 19

6.1 Background 19

6.2 Site considerations 20

7 Maintenance regimes 23

8 Alternatives to vegetation 23

9 Bibliography 24

Appendix A: Assessment tables 26

Table 1: Catchment level assessment 26

Table 2: Reach level assessment 27

Table 3: Site level assessment 28

Table 4: Intervention 29

Table 5: Reassessment 30


v


1

1 Introduction

All streams erode but human impacts haveaccelerated the natural and essential processes of riverchannel adjustment, often to unacceptable levels.Riparian and in-channel vegetation can reduce streamerosion rates but it is unrealistic to expectrevegetation to eliminate all erosion. By acceptingsome channel change as inevitable, a managedriparian zone provides space within which river formand processes can be allowed to adjust.

Managing the riparian zone to improve riverbankstability involves planting native (particularlyindigenous) trees, shrubs, grasses and macrophytes inthe stream, on the banks and on the stream margins.Stabilisation schemes that incorporate riparian plantsare ecologically superior to hard engineeredalternatives and generally provide satisfactoryimprovements to bank stability. Vegetation interactswith a range of geomorphological, geotechnical,hydrological and hydraulic factors to affect the typeand extent of riverbank erosion (Abernethy andRutherfurd, 1996; 1998a; 1998b). Riparian vegetationreduces flow velocities, directly reinforcesriverbanks, intercepts and slows surface runoff, andlimits access to the bank by stock. The enhancedlateral channel stability offered by well-vegetatedriparian zones can also reduce the need for engineeredstabilisation and heavy maintenance.

This document provides guidance for designingvegetated riparian zones intended for streambankerosion control. The guidelines are divided into anumber of sections. As part of our introductorycomments, Section 1 defines the purpose, scope andunderlying principles of the guidelines along withsome definitions of specific terms used throughoutthe document. Section 2 outlines our proposedtechnique for determining the width of the riparianzone to stabilise the riverbank under variousconditions. Sections 3-6 provide background materialthat explains the interaction between riparianvegetation and bank erosion processes, how to assessthe present condition of the riparian zone andconsiderations for plantation design. Maintenanceregimes of riparian plantations and alternatives tostabilisation schemes that rely completely onvegetation are briefly touched on in Sections 7-8.

All of the material presented in this document issummarised and collated into a series of tables thatmay be used to guide and focus the practitioner’s

approach to planning riverbank stability works withvegetation. The five tables in Appendix A arepresented in a top-down fashion that cover issues atvarious scales, from whole of catchment to regionalto on-site, to provide context for design and on-siteworks.

1.1 Purpose of the guidelines

These guidelines will provide assistance in planningand implementing riparian revegetation worksspecifically designed to retard bank erosion rates tomore natural levels. Our intention is to provide somegeneral rules for deciding on the structure and widthof vegetated riparian zones for bank erosion control.The width and character of plantations designed forhabitat retention (Cummins, 1993), or for interceptingnutrients and sediment in runoff (Karssies & Prosser,1999), may be quite different than plantationsdesigned for bank erosion control. Thus, theseguidelines should be considered in the light of otherrequirements for riparian zone management.

Readers should be aware of the following points.a) These guidelines concentrate on the role of

vegetation in riverbank erosion withoutconsidering other forms of erosion, such asgullies or channel incision.

b) The guidelines are intended for use in regionswhere water is extracted from streams forirrigation. While we recognise that these streamsare typically large, meandering rivers, flowingthrough alluvial sediments, we also includeguidance on revegetating smaller tributaries.

c) Bank erosion is a natural process. Given enoughtime, even fully vegetated, natural streams erodeback and forth over their floodplains. Indeed,there is a growing body of evidence that somechannel erosion is essential to the long-termhealth of stream ecosystems. A healthy, vegetatedriparian zone should not be expected to provideabsolute stability to a stream.

d) In degraded streams, there are many situations inwhich vegetation alone will not stabilise streams.One of the essential tasks for managers is torecognise such situations.

e) Finally, these guidelines do not discuss weeds invegetated riparian zones. In Queensland,particularly, weeds are a critical managementproblem but, in some instances, they can alsocontribute to bank stability.


2

1.2 Principles of using vegetation for bank erosion control

The underlying philosophy of these guidelines is thatthe dominant bank erosion process changesdownstream along a river due to changes in channelscale. Different types of vegetation affect differentprocesses, so it is imperative to assess the dominanterosion process correctly so that appropriate speciescan be selected for stability.

Principle 1. Bank erosion is the result of a number ofdifferent processes that all act to destabilise riverchannels.

All of the erosion processes that act on banks can begrouped together into three erosion-domains. (1)Subaerial erosion (erosion caused by processesexternal to the stream such as cattle, or rain splash).(2) Scour (removal of individual sediment particles oraggregates by flow). (3) Mass-failure (slumps).

Principle 2. Each of the three erosion-domains actthroughout catchments but at any given location oneof the groups dominates over the others.

The dominant process varies down a stream system:subaerial processes tend to be more important onsmaller streams, the effects of scour are mostpronounced in mid-catchment reaches while mass-failure is increasingly important as bank heightincreases in lowland settings.

Principle 3. The influence of vegetation over each ofthe processes remains relatively constant throughoutcatchments.

Suitably selected and placed vegetation will reducethe rate of the bank erosion processes. The overstorey

has a greater influence over the processes of mass-failure than do understorey, groundcover ormacrophyte species. However, shrubs and grasses onthe bank face, and macrophytes at the bank toe, areequally important for controlling sub-aerial erosionand scour.

Principle 4. Know your erosion processes. The key toplanting vegetation to control bank erosion is to knowwhat the dominant erosion process is, and to knowhow each of the vegetation types will affect thatprocess.

For example, lowland riverbanks generally retreat bycyclical combination of fluvial scour of the bank toefollowed by mass failure under gravity, followed thenby removal of the failed material by further scour. Allcomponents of the cycle are affected to some degreeby bank material loosening or weakening due tosubaerial weathering processes. The subaerialmechanisms of weakening and weathering aregenerally controlled by climatic conditions, whilstfluvial processes are associated with channel flowhydraulics. Mass failure is usually triggered when acritical stability condition is exceeded, either byreduction of the internal strength of the bank (oftendue to subaerial processes) or a change in profilegeometry (typically the result of scour). The rate atwhich material is transported away from a particularsite ultimately controls the rate of bank retreat overtime (Little et al., 1982; Thorne, 1982; Chang, 1988;Alonso and Combs, 1990). Revegetation to stabilise ariverbank undergoing this kind of erosion must ensurethat appropriate plants are established in appropriateplaces on the bank.

Bank crest

Bank toe

Bank face

Bank height

Channel width

Bank height

Riparian zonewidth

Figure 1:Description of features used in the guidelines


3

1.3 Terms

Riparian zone width (Figure 1) is the minimumwidth of riparian forest required for ongoing bankstability. It is measured from the bank crest awayfrom the channel. The minimum width of the riparianzone varies with bank height and bank erosion rate(Section 2).

Bank height (Figure 1) is really a measure of near-bank channel depth measured vertically between thebank toe and bank crest. This will often be the localmaximum depth of the channel and is likely to varyfrom bend to bend.

Bank crest (Figure 1) is the junction of the channelwith the floodplain. In practice, this point can bedifficult to identify. An alternative definition is thelevel reached by the flood that comes on averageevery one to two years, or the point above which youconsider the river to be in flood.

Bank toe (Figure 1) is the junction of the bank withthe bed of the channel. The bank toe is usuallymarked by a break in slope but often the transitionfrom bank to bed is gradual and the toe is difficult todetermine.

Stability is defined as a ‘natural’ rate of bank erosion.This does not imply absolute stability, but the rate ofchange that may have existed before Europeansettlement. Thus, an unstable bank is a bank that iseroding at a faster rate than natural. You may be ableto determine what the natural rate is by comparingyour target bank with a nearby reach in near-naturalcondition.

Erosion rate is the rate at which the bank face moves(metres/year). This will be an average rate over atleast 20 years so that the bank will have beensubjected to some major floods. The erosion rateshould be determined by existing evidence (aerialphotos, distance from fence lines, etc.) but can beestimated as 1.6% of channel width per year (seeSection 6.2).

Channel width (Figure 1) is measured from crest tocrest.

Plantation maturity is defined as the time it takesthe vegetation to have a major influence on theerosion rate (e.g. to a closed canopy). Plantations inthe wet-tropics can mature within ten years but thiscould take up to 50 years in drier areas(see Section 6.2).


4


5

2 Guidelines for determining minimum riparian zone widths

The following procedure can be applied to shortreaches of uniform stream. The higher and steeper thebank the more prone it is to erosion. As the bankbecomes steeper and higher, the zone of influence ofthe erosion processes extends deeper into the bank(Sections 4.1 and 6.1). Slump blocks become largerand deeper and the portions of the bank profilevulnerable to scour and subaerial processes increase.So, as a rule the higher and steeper the bank profile,the wider the protective zone that should be replanted.Hence, applying the following technique to banksections on the outside or the inside of meander bendsor at inflection points will result in design criteria forriparian zones of varying widths. In thosecircumstances where the control of sediment/nutrientingress to the stream is also desired, a grassed filterzone in addition to the riparian zone widthsrecommended here is also desirable (see Karssies andProsser, 1999).

Note that vegetation condition is assessed in terms ofthe Raine and Gardiner (1995) ‘traffic l ight’classification (see also Section 5). Other vegetationclassifications, such as that developed by Luke Pen inWestern Australia (Pen, 1994), could also be usefullyadopted for this purpose. Staying with the Raine andGardiner traffic light system, though, we define:

• ‘green’ vegetation to be a contiguous stand ofmature riparian forest established at least therecommended minimum width for its location.

• ‘Yellow’ vegetation is a riparian forest that is notcontiguous (compared to natural remnants or otherlocal expert advice), is immature (perhaps a recent

plantation), or does not meet the recommendedwidth requirement.

• ‘Red’ vegetation lacks one or more structuralelements (Section 3), is in a generally degradedcondition and performs no stabilising role(Table 1).

For the purposes of bank stability, interspeciesvariation within structural groups, (see Section 3) arerelatively small and unimportant. All deep-rootedspecies (overstorey) are more capable of reinforcingriverbanks against mass failure than shallow rootedgroundcover. Trees should be planted aroundpotential slump-block failure planes (Sections 4 and6). However, understorey and groundcover speciesprovide mid- and upper-bank sections with greaterprotection from scour. Lower bank sections that tendto remain wet throughout the year are best protectedby macrophyte species where they can be established.All structural groups interact with and modifysubaerial processes. Trees may dry the bank throughtranspiration and help resist the onset of piping, whilesmaller plants might reduce the affects of rainsplashor rill erosion.

Final design considerations (e.g. planting density andspecies makeup) tend to be controlled by ecologicalconstraints. As a result, our only advice is to aim for anaturally diverse and dense vegetation community.

2.1 Recommendation

The minimum width of any riparian plantationdesigned for bank stabilisation should not be less than5 m measured onto the floodplain from the bank crest.Wherever possible, plants established on the bankface should be integrated with the basic allowance.Also, macrophytes should be incorporated in thedesign where appropriate.

Table 1: Summary of vegetation condition assessment.(� = attribute present, � = attribute absent, ? = attribute either present or absent)

Vegetation condition

Attribute Green Yellow Red

Native � ? ?Mature / established � ? ?Structural elements � ? �Contiguous � ? �Appropriate width � � �


6

As banks become higher they become less stable.Hence, in addition to the (5 m) basic allowance,riparian strips must also include a width componentthat compensates for bank height. The heightallowanceshould not be less than the height of thebank measured vertically from the bank toe to thebank crest (Figure 1).

Time must be allowed for the plants to grow beforethey can begin to stabilise the bank, so where banksare actively eroding an establishment allowanceshould also be included in the final riparian zonewidth. The establishment allowance will permit thecomponents of the riparian zone planted for basic andheight allowance to mature as the present bank-lineerodes towards them. The establishment allowanceisdetermined by multiplying the erosion rate (seeSection 1.3) by the time required for the local riparianforest to mature (Section 6.2).

The above recommendation assumes certain thingsabout bank material, bank hydrology and bankgeometry. One assumption is that the riverbed isstable. Without bed stabilisation, it may not bepossible to achieve bank stability with vegetation

alone in excessively incised streams with high, steepbanks and large failure blocks. Local conditions maydictate that the ultimate design for bank stability mayrequire other treatments at the site. If, for example,rock toe-protection was used to prevent scour of thebank toe then the establishment allowance could berelaxed or avoided.

The last point is an extremely important caveat to allour recommendations. Onsite conditions varyenormously and preclude the use of overly-prescriptive guidelines. When seeking to stabiliseriverbanks you must match erosion process andvegetation. Moreover, you must appreciate that boththe erosion process and the effect of the vegetationwill not remain static but will change over time. Findout what the dominant erosion process is, determinewhere on the riverbank it acts, measure or estimatethe rate of bank retreat, determine the regionalcontext (if any) for the problem (Section 5) - thenchoose species and planting strategies specifically foryour local conditions, bearing in mind the minimumplantation widths.

Figure 2:Decision tree for determining minimum riparian zone widths for bank stability (definitions are given in the text)


7

2.2 Decision tree

The ‘decision tree’ shown in Figure 2 illustrates theprocess of determining riparian zone widths. Thedecision tree begins with minimum requirements forthe width of a vegetated riparian zone, and thenprogressively adds to that minimum for variouscircumstances. Note that the minimum requirementshould in no way be seen as a licence to degrade ariparian zone that exceeds that minimum. In terms ofbank stability, more vegetation is almost always betterthan less.

The following notes refer to different scenarios thatcould be followed through the decision tree; the list isnot exhaustive. Supplementary information is detailedin Sections 5 and 6.

a) Stable bank - green/yellow vegetationAny vegetated riparian zone less than 5 m wide isunlikely to be sustainable. Hence, the basic allowanceprovides a manageable zone for even small streamsthat have a trivial depth. As channels become larger,however, some account must be made for thedecreased stability of their banks. The heightallowance increases the width of the plantation toreflect the larger channel size. Stable banks withyellow vegetation require gaps in the cover to beplanted out. Even stable banks with green vegetationrequire ongoing maintenance - occasionally monitorbank and vegetation condition.

b) Stable bank - red vegetationIn this case, there is no vegetation but the banksappear stable. Planting is still required because thereis no guarantee that a stable bank will remain thatway. Stable bank sections are in fact the ideal place toestablish a plantation (from scratch) as the plants willhave time to grow and provide increasing protectionfrom future erosion pressures.

c) Unstable bank - red vegetation - revegetationwill provide adequate stability

In this case, bank erosion is likely to be the result of alocally depleted riparian zone with nearby vegetatedsections remaining stable. Along with the basicallowance and height allowance, an establishmentallowance should also be included in the riparian zonedesign. The establishment allowance will create spacefor further short-term adjustment withoutcompromising the long-term goals of therevegetation.

d) Unstable bank - green/yellow - revegetation will slow erosion rate

What is the rule if banks with good vegetation coverare still eroding? Clearly, the erosion forces areexceeding the resistance of the vegetation. Typically,this means that the erosion is related to major reachscale processes, particularly deepening, or increasedrates of scour at the bank toe. In these circumstances,even if vegetation cannot reduce the instability toacceptable levels, the erosion rate is likely to beconsiderably slower than would be the case forcleared banks. Regardless of vegetation condition, itis important to establish and maintain an appropriateestablishment allowance along with the basic andheight allowances. Follow-up plantings may berequired. It is often worth persisting with vegetationin these conditions as the processes of bank erosioncontinually shift and change, previously unstablebanks will always stabilise with time. Wherepersistence with vegetation cannot be warranted dueto continued high erosion rates, over long periods,artificial stabilisation techniques may need to beconsidered (Section 8).

e) Unstable bank - red vegetation - revegetationwill not slow erosion rate

This scenario is one of serious riparian degradation.Perhaps the erosion rate is too high for an adequatewidth of riparian zone to be given over to managingchannel stability. Perhaps unchecked bed degradationhas heightened the banks beyond some critical valueand tree roots are unable to penetrate deeply enoughto reinforce failure planes. Whatever the case, ifvegetation cannot be relied upon to stabilise the banksand bank stability is desired, alternative measures willneed to be sought. Remember, though, that evenyoung vegetation will tend to slow bank erosion if itcan be effectively established. However, work hereand overseas (e.g. Hemphill and Bramley, 1989;TRCRD, 1997) has demonstrated that structuralreinforcement at the bank toe combined withvegetation on the upper portions of the bank providesa cheap and viable alternative to fully engineeredoptions. With the toe fixed by structure theestablishment allowance may be relaxed with theriparian zone width comprised of the basic and heightallowance only.


8

2.3 Worked example

A landholder owns one side of the 4 km reach ofmeandering river shown in Figure 3. The channel ismoderately sized: about 30 m wide, with 5 m outer-banks. Remember, the landholder is only assessingthe condition of his or her side of the stream. Thereach consists of four bends with all erosion types -subaerial, scour, mass failure- active along the banks.Based on an assessment with the decision tree, thefarmer decides that the reach should be divided intoseven sub-reaches (A to G) for riparian management.

Mature vegetation that will control erosion isexpected to take 25 years to establish. The farmerknows that sub-reach A has eroded 15 metres towardsa tank his father built 30 years ago. Hence theestablishment allowance for this reach is 25 ( 0.5 =12.5 metres. This is about the same answer that hewould have got using the 1.6% of width per yearestimate for bends.

With reference to Figure 3:

Sub-reach A

Description: Steep outside bend, 5 m high.Stability: Scour at the toe of the bank leads tocontinuing retreat of a steep bank face, with slumpingseemingly un-controlled by the vegetation.Vegetation: Green/yellow.Action: Stabilise toe with rock riprap and plant outgaps in the riparian zone to give continuous cover atminimum width through the sub-reach. Because therock stabilisation will prevent further scour at the toe,there is no need to include an establishmentallowance in the final riparian zone width.Riparian width: Basic plus height allowance (5 + 5 =10 m).

Sub-reach B

Description: Inflection, 3 m high.Stability: Banks are laid back and appear to be stable.Vegetation: Yellow.Action: Improve riparian zone to green vegetationcondition. Riparian width: Basic plus height allowance(5 + 3 = 8 m).

Sub-reach C

Description: Pointbar,

The remaining sections provide background andsupplementary material to the guidelines sketched outin this section. Much of the following material is alsosummarised in the Assessment Tables that areappended to the end of the document. It is importantto emphasise that along with the many causes of bankerosion and the many different riparian species, thereare many different sources of information. Often,material is locally based, incorporating valuable localknowledge that this document cannot provide. InQueensland, for example, QDNR have produced aseries of river fact sheets that provide valuableinformation (available at http://www.dnr.qld.gov.au/fact_sheets/riverfacts.html). Other supportinginformation, that is directly relevant to Queenslandstreams can be found in publications such asO’Donnell (1998), or Kapitzke et al. (1998).


9

perhaps negating the need for structural intervention.After riparian planting, farmer will maintain watchingbrief and reassess the need for structure from time totime.Riparian width: Basic plus height plus establishmentallowance (5 + 4 + 15 = 24 m).

Sub-reach E

Description: Outside bend, 4 m high.Stability: Eroding, but vegetation will reinforceagainst small slumps and reduce scour.Vegetation: Red.Action: Establish riparian zone.Riparian width: Basic plus height plus establishmentallowance (5 + 4 + 15 = 24 m).

Sub-reach F

Description: Pointbar, ~1 m high.Stability: Essentially the same as Sub-reach C exceptin better condition.Vegetation: Red.Action: Revegetate and improve vegetation to greencondition. The farmer has decided to extend this zonesomewhat to build on the previous work of Sub-reachF and to match up with the wider requirement of Sub-reach E.Riparian width: Extra plus basic plus heightallowance (>5 + 1 = >6 m).

Sub-reach G

Description: Straight, 3 m high.Stability: Stable. This is the farmer’s best bit offrontage. Remnants, and additional plantings fromsome years ago, fenced and well established.Vegetation: Green.Action: Enjoy the view.Riparian width: Basic plus height allowance(5 + 3 = 8 m).

The result of the above procedure is that the requiredriparian zone widths around the reach of river willvary from 5 m to about 24 m. The minimum riparianzone widths recommended here are intended forerosion control only. However, there may be manyother good reasons to vegetate a wider riparian zone.It is also important to emphasise that this exercisewill be much more useful if the landholder on theopposite side of the river adopts a similarmethodology and revegetation occurs on both sides ofthe river.


10


11

3 Riparian forest structure

The density and type of riparian vegetation coverstrongly influence all aspects of riverbank erosion.Riparian forests are typically composed of overstorey,understorey, groundcover and macrophyte species. Insome environments, some of the structural elementsmay be missing. This might occur in situations wherefor example stream velocities do not allowmacrophyte establishment or where the groundcoveris shaded out by the higher elements. The age andhealth of all plants also affect the stabilisingeffectiveness of any riparian forest. Structuralelements are described here in terms of their bankstabilising attributes. We pass no comment on anyecological, biological, establishment or maintenanceconcerns.

3.1 Overstorey

The overstorey consists of emergent trees which,typically, grow to a height anywhere between 5 and20 m (depending on species composition and localconditions). Tree-root networks are as variable as theabove ground parts and are, thus, difficult tocharacterise. Most riparian species typically develop acentral rootball or rootplate of dense roots that canusually be considered as half a sphere below thesurface that has a diameter of about five times thediameter of the trunk. Root density declines rapidlybeyond the rootball, with both lateral distance andwith depth. For reinforcement purposes, there areusually few roots beyond the canopy dripline orbelow about 2 m under the bank surface. High watertables and/or heavy textured sediment affect theextent of the root network and give rise to shallowroot systems as the roots grow laterally to avoid thesaturated and/or hard material.

3.2 Understorey

The riparian understorey ranges from nothing, to acomplex array of shrub species. For the purposes ofbank stability, the shrubs of the understorey act thesame as small trees. Generally, the understorey growsto a height of between 1 m and 5 m, with a rootingdepth less than that of trees but still down to over ametre. As with trees, the lateral extent of the rootmass is about that of the dripline and the root densityquickly diminishes with depth and distance from thetrunk.

3.3 Groundcover

Riparian groundcover is typically less than 1 m high.Groundcover species can include prostrate shrubs,grasses, sedges and forbs. Groundcover is usuallyquick to establish on a bank, but susceptible totrampling and other grazing pressures. Although theroots of grasses can be seen at depths of over a metreon exposed bank profiles, their reinforcementpotential is negligible at depth. For the purposes ofbank reinforcement, the maximum zone of influenceof groundcover is probably constrained to about thetop 30 cm. Regardless of the exact depth, theeffective rooting depth of grasses will be fairlyshallow and may vary between species and betweensites. The main advantages of groundcover are that itdensely covers the bank surface (except for verticalbanks) and has a dense (if shallow) root mat. Adisadvantage of many groundcovers is that they willnot grow below the low-flow waterline.

3.4 Macrophytes

Emergent aquatic macrophytes, such as some sedges,rushes and reeds, are shallow-rooted species whichgrow at the margins of the mean water level. Theyreadily colonise wet areas where terrestrial plants donot establish. Macrophytes will generally not survivefor long periods of time in water that is more than 0.5m deep. They flourish in conditions of low velocity(about 0.2 m/s) but will withstand the short periods ofinundation and high velocity, which occur when thestream is in flood.


12


13

4 Riparian vegetation and streambank stability

Alluvial rivers change through time as the channelevolves naturally or responds to the morphologicalimpacts of imposed management. The cause of localbank instability can be difficult to isolate and identify.Problems may result from a wide variety ofgeomorphological processes, some operating locally,others at reach scale, and some associated withcatchment-wide adjustments to changes in hydrologyor sediment yield. Regardless of the widergeomorphological context, the nature and extent oflocal bank erosion at some specific point along ariverbank are controlled by:

a) local discharge;b) channel shape (cross-section and planform);c) location of eroding bank section (e.g. inner or

outer bank);d) bank geometry;e) bank geotechnical properties;f) bank hydrological properties; andg) vegetation.

With the exception of vegetation, each of the abovefactors are very much dependent on local conditions.However, generalised descriptions usefully identifythe broad interactions of vegetation with each of thethree erosion process groups: mass failure, fluvialscour and subaerial preparation.

4.1 Mass failure

Mass failures occur when whole blocks of materialslide or topple from the bank into the water. Theshape and extent of mass failures are a function of thegeometry of the bank section, the physical propertiesof the bank material, and the type and density ofvegetation. Trees, in particular, have been attributed anumber of influences over slope processes (Gray andLeiser, 1982; Bache and MacAskill, 1984; Barker,1986; Greenway, 1987; Coppin and Richards, 1990;Styczen and Morgan, 1995; Gray and Sotir, 1996).Reinforcement of the bank sediment by roots,transpiration and improved bank drainage all act toenhance the riverbank stability. On the other hand,windloading and the additional weight of ripariantrees have often been implicated in the failureprocesses of bank sections within vegetated reaches.The destabilising effect of the weight, or surcharge, oftrees is dependent on local bank geometry, bankgeotechnical properties and the tree position. That all

of the effects vary seasonally and with plantdevelopment makes it difficult to incorporate theminto bank stability analyses.

Root reinforcementProbably the most obvious and important way thattrees affect bank stability is by increasing the strengthof bank material with their roots. Plant roots tend tobind banks together and act in much the same way assteel reinforcement in concrete. Groundcover speciesdo not generally contribute to the mass stability ofbanks because of their limited root depth.

The extent to which vegetation acts as reinforcementdepends on a number of root properties. Probably themost important are the root tensile strength and rootdensity. Root strength depends on the species, size,age, and condition of the root. Generally, smallerroots are the main contributors to additional soilstrength because many small roots provide moreeffective reinforcement than one or two large ones.This can be thought of as following the sameprinciple as ropes: many small fibres provide a muchstronger structure that a single strand of like diameter.

Bank material strength is a function of its internalangle of friction, and cohesion. The effect of smallroots is to increase the effective cohesion of thesediment. Work overseas suggests that small roots ofNorthern Hemisphere species can increase cohesionby an average of 20%, although this can be up to 50%(Greenway, 1987; Coppin and Richards, 1990;Shields and Gray, 1992). Our work suggests that theeffect of tree roots may be even greater than this, withperhaps up to a ten-fold increase in cohesion close tothe trunks of riparian trees, falling to about a two-foldincrease under the dripline. Longer and more firmlyanchored roots provide greater reinforcement than dotheir shorter and loosely anchored counterparts.Mature trees thus provide more reinforcement thanyounger trees.

Roots also enhance bank stability by inhibiting thedevelopment of tension cracks (Thorne and Lewin,1979). Tension cracks that run parallel to theriverbank are often observed on the surface beforefailure. The eventual failure surface will normallyfollow such cracks, and their position will indicate theapproximate extent of any potential failure (Hemphilland Bramley, 1989). In cases where the potentialdepth of the tension crack is large in proportion to thetotal depth of the slope, the failure surface may besignificantly affected by the tension crack.


14

Bank hydrology modificationDrier banks are more stable than wet ones because theweight of the soil mass is lower, and the soil’scohesion is higher. Vegetation keeps banks drier byintercepting precipitation, transpiration, and increaseddrainage through the soil. Annual evaporation fromeucalypt plantations can be up to seven times thatfrom surrounding grazed pastures when there is agood water supply present in or near the root zone(Greenwood et al., 1985). Furthermore, well-vegetated banks are likely to be better drained thantheir cleared counterparts due to an increasedincidence of organic matter and a higher level ofbiological activity within the soil.

SurchargeThe extra weight, or surcharge, of trees on a riverbankis often thought to encourage the banks to collapse.However, the location of trees on a bank will affectthe degree to which they influence the balance offorces (Gray, 1995; Gray and Sotir, 1996). Thesurcharge of trees growing at the bottom of a bankprone to rotational failure mechanisms is likely toincrease, not decrease, overall bank stability (Coppinand Richards, 1990). However, the weight of trees atthe top of the same bank profile might destabilise thebank sections.

Trees are often restricted to the floodplain beyond thebank crest on actively eroding riverbanks, as they arenot able to establish successfully on an activelyeroding bank face. Tree surcharge on the top of thebank is especially pronounced when the trees leanover the channel as a result of growth asymmetry,grazing on the bankward side only, or wind loading(Thorne, 1990). However, our work on the LatrobeRiver (Abernethy, 1999) shows that the effect of treesurcharge on bank stability, even those that appear tobe detrimental, is marginal.

WindloadingStyczen and Morgan (1995) demonstrated that thepressure exerted on a tree by wind can be transmittedto a bank as increased loading. This reduces thebank’s resistance to failure. The transmission of theforces to the soil is by virtue of the root system withroots held in tension or compression. The stronger thesoil and root-soil bond and the greater the root surfacearea, the larger the uprooting force that can beresisted (Coutts, 1983; Fitter and Ennos, 1989; Ennos,1993).

4.2 Fluvial scour

Scour occurs when the force applied to a bank byflowing water exceeds the resistance of the banksurface to withstand those forces. Vegetation on thebank face reduces scour by directly strengthening thebanks. Dense root mats protect the bank face with thefiner roots particularly useful in holding the bankmaterial together. The root mats of species such asShe-oak (Casuarina cunninghamiana), Bottlebrushes(Callistemon spp.), Paperbarks (Melaleuca spp.) andTea trees (Leptospermum spp.) have been found to beuseful in this regard.

Vegetation also creates backwaters that slow flowagainst the bank face and weaken secondarycirculation in bends (Thorne and Furbish, 1995).Slowing the flow against the bank greatly reduces theability of the flow to scour bank material. At lowdischarges, the high flow resistance associated withgrasses and smaller shrubs standing rigid andunsubmerged often reduces the velocity below thatrequired for bank erosion. At higher discharges,submerged grasses and shrubs often benddownstream, forming a flattened layer which,although having low flow resistance, protects thebank from scour (Kouwen, 1988).

Trees are not as effective as grasses and shrubs atretarding flow velocities near the bank when the flowis slow. At high velocities, however, the much stiffertrunks of trees are useful to retard the flow close tothe bank. The Japanese Technology Research Centrefor Riverfront Development (TRCRD, 1997) hasconducted extensive research on many of thehydraulic aspects of riparian trees. They find that flowwithin stands of trees is markedly slower than flowagainst unprotected banks. However, for effectivereductions in flow attack on the bank, plantingdensities where the crowns begin to intermesh are theminimum for hydraulic effect. Isolated clumps oftrees on banks should also be avoided in revegetationstrategies as they can act as hardpoints that can beoutflanked by the flow in some circumstances.

Another form of bank scour is that due to waveaction. Reed-beds are particularly useful where waveaction from boat traffic is responsible for bank attackbecause they absorb wave energy. A reed-bank 2 mwide can absorb about two thirds of the wave energygenerated by wash from pleasure craft (Bonham,1980). Additionally, emergent aquatic macrophytesrestrict the near-bank flow velocity and provide somereinforcement to the bank surface through their


15

shallow root mat. Frankenberget al. (1996) creditedreduced erosion rates at some sites on the MurrayRiver, near Albury-Wodonga, to the presence ofPhragmites spp.

4.3 Subaerial preparation

Streambank segments exposed to air are subject tosubaerial preparation from a variety of processes thatare largely external to river processes. These includethe effects of windthrown trees, piping, desiccation,rainsplash, rill erosion, and stock trampling which areall discussed below. Some directly cause erosion,while others render banks more susceptible toerosion. Subaerial processes are active on all exposedbanks but are usually only apparent when scour andmass failure are limited. One way to see if subaerialprocesses are important on a given stream is to look atthe erosion processes that are active on banks isolatedfrom the main flow, such as cutoff meander bends orold abandoned channels.

WindthrowWindthrown trees directly deliver sediment into theflow when their rootballs detach from the bank. Theresultant debris dams often redirect flow against thebank and the scallops formed in the bank after thetrees fall present ideal places for concentratederosion. Windthrow problems are exacerbated whentrees occur in a single line along the top of the bank;hence, a wide stand of trees is preferable in terms oftheir impact on bank stability (Thorne, 1990).

PipingA common cause of bank undermining is seepage ofwater through the bank that leads to leaching andsoftening of the bank material or in extreme cases,pipe erosion (Hagerty, 1991). In some areas, irrigationmay contribute to seepage as excess water drainstowards the channel through the bank profile. Tensioncrack development can allow surface flows to draininto the bank, increasing seepage force and furtherreducing the stability of the affected bank (Simonsand Li, 1982). The effects of vegetation on piping areboth positive and negative. The casts left by deadroots may leave pathways for piping through the bankwhile evapotranspiration may delay or mitigate theonset of saturated flow within the bank.

DesiccationDry and cracking bank material is highly erodible. Infact, desiccation sometimes has a greater influence onerodibility than does the composition of the bankmaterial itself (Knighton, 1973). Vegetation canreduce desiccation by binding bank material together,while shade from trees, grass and leaf litter reducedrying.

Rainsplash and rill erosionDuring storms, rainsplash and rill erosion can detachmaterial directly from a degraded bank and wash itinto the channel. Well-established bank vegetationwill reduce the rate of surface erosion by one to twoorders of magnitude (Kirkby and Morgan, 1980).Thorne (1982) maintains that overland flow on well-vegetated banks can be ignored in determining bankstability over engineering time scales of 10 to 100years.

Stock tramplingLivestock reduce the resistance of the riverbank toerosion by reducing the vegetation cover andexposing otherwise protected bank material.Trampling also directly breaks down banks, andtransfers large quantities of bank material straight intothe flow (Trimble and Mendel, 1995).


16


17

5 Assessment of existing riparian condition

The existence and extent of the riparian corridor is animportant indicator of channel condition, sensitivityto change and management status. Knowledge of theexisting vegetation assemblage on the floodplain andits influence on the present hydrological, hydraulicand sediment processes is useful for any assessmentof the potential for instability induced by changes invegetation, land use or floodplain management. Thepurpose of this section is to highlight factors thatshould be considered in any on-site assessment ofriparian condition. The following discussion draws onvarious methodologies that have been developedaround the country (e.g. Raine and Gardiner, 1995;Kapitzke et al., 1998) and oversees (e.g. Thorne,1998).

Because QDNR have already shown a preference forthe Raine and Gardiner (1995) Rivercaremethodology (River Fact Sheet R34) we will beginthe discussion with reference to Raine and Gardiner’s‘management style’ approach. The ‘yellow’vegetation condition is the minimum standard thatcould be expected to provide any stabilising influenceto a riverbank. However, it is difficult (perhapsimpossible at some sites) to establish riparianvegetation where the banks are actively eroding, i.e.the ‘red’ channel condition. At sites where the bank isactively eroding, assessment of the bank conditionshould consider alternatives for stabilisation.Alternatives might include: hard engineering options;wider riparian zones that include some leeway forfuture erosion while the plantation matures; acatchment-wide plan of river rehabilitation (e.g.Rutherfurd et al., 1999); or a combination ofapproaches.

5.1 Reach assessment

To get the feel for the geomorphologicalcharacteristics of the problem, a wider view beyondthe bank section at hand is required. Use should bemade of stream condition surveys, undertaken by thelocal stream-management agency, where they exist.Where they do not exist, a field assessment of thechannel planform, hydraulic geometry and erosion/deposition history of the river reach in the vicinity ofthe site should be conducted. This will assist inidentifying current or future pressures on bankstability at the site.

The reach assessment provides context for aparticular site in terms of a broader reach-lengthperspective. During assessment of the reach, assessorsshould try to get a feel for the form and processes thatcurrently occur in the river. It should always beremembered, however, that all alluvial channel formsare transient and will continue to change. What yousee today may not exist tomorrow. The following areabout the minimum that should be considered duringthis part of the assessment.

a) Historical channel change• Look for bed degradation/aggradation

(bridge piers provide good indicators).• Look at the channel planform - is it stable?

(Air photos will indicate erosion extent andrate).

b) Stream hydrology• How big are the floods? When do they occur?

How long do they last? • Do floods strip or deposit sediment from or

onto the floodplain?c) Channel hydraulics

• Look for channel obstructions, particularlythose that redirect flow against the banks.

• Look for channel bars that concentrate flow (particularly those that are vegetated).

d) Channel form• Look at the bank shape around bends and

along straights. • Measure the width and depth of the channel

at various points to compare with your site.• Look for any incidence of erosion.• Pay particular attention to the size and shape

of failures.• Look for undercutting or other evidence of

scour (e.g. floodplain stripping).• Look for seepage through the bankface

(particularly adjacent to irrigation).

5.2 Bank assessment

Following the broader reach investigation, attentionshould then be focussed on the bank section(s) wherethe work is planned. First, an assessment of thecurrent bank condition should be made. This shouldinclude consideration of (at least) the following.

a) Rate of bank retreat (aerial photographs, if available, or some other means).

b) Bank geometry (cf. other bank sections in adjacent reaches).


18

c) Signs of bank slumping (failed material alongthe toe, tension cracks, steep or undercutprofiles, bare patches of soil on the bank face).

d) Bank material (structure, stratification,strength, hydrology).

e) Desiccation cracks.

f) Windthrown trees.

g) Seepage through the bank face.

h) Stock tracks.

In those circumstances where the bank condition doesnot correspond to at least the yellow score overall,engineering strategies to remedy individual problemsmay need to be adopted in conjunction withrevegetation. If the current bank condition relates to ayellow or better score, and the reach-length surveydid not identify any future pressures, then it isprobable that the planned revegetation can beestablished. Revegetation works should endeavour tomake use of any native vegetation that is alreadygrowing at the site.

5.3 Vegetation assessment

The next part of the process is to assess the conditionof the local riparian community. In general, theassess-ment is conducted by comparing the in situvegetation with nearby remnant stands of riparianforest. The vegetation assessment should consider thefollowing points.

a) Structure• Look for species in all of the structural

elements typical of the bio-geographicalregion, e.g. in some regions groundcover maybe missing due to overstorey shading.

b) Species• Look for species with extensive root

networks. • Look for a diversity of species - this implies a

mature state that is stable and well-established.

c) Density• Continuous covers of all structural elements

are preferable for bank stability.d) Location

• Look for plants established near potential

failure planes either near the bank toe or onthe floodplain some distance back from thebank-crest.

• Plants with deeper and more extensive rootsystems are better in these locations as thepossibility of roots intersecting andstrengthening potential failure planes isincreased.

For those sites that score a green vegetationcondition, no further planting is required. Sites withyellow or red vegetation scores require revegetationworks.


19

6 Riparian plantation designddddddd

The main consideration when designing riparianrevegetation works for bank stability is continuity ofcover. Spacing the overstorey to allow developmentof an extensive and intermeshed root network will domuch to stabilise the bank section. The information inthis section relates specifically to stabilising banksagainst mass failure by planting deep-rooted trees.However, associated planting of understorey,groundcover and macrophyte species will assist incontrolling erosion by other mechanisms and ensurethe ongoing stability of the site. Understorey andgroundcover elements tend to be more important inresisting subaerial processes as they provideprotection much closer to the soil surface. Forprotection against fluvial scour, groundcover andmacrophytes are important as they slow the flowclose to the bank. It should also be remembered thatplants cannot provide instant reinforcement: theirstabilising properties will improve with age.

6.1 Background

The ideas presented below are based on the results ofa physically based computer model of bank erosionprocesses and the effect on bank stability of tree rootreinforcement. Further information on the model andits previous applications can be found in Abernethyand Rutherfurd (1999; subm.-a; subm.-b). The modelhas been successfully applied to and calibrated withbank conditions along the Latrobe River inGippsland. We assessed the effect of vegetation onbank stability by analysing the tree root reinforcementof individual River Red Gums (Eucalyptuscamaldulensis) and stands of Swamp Paperbark(Melaleuca ericifolia). The lateral position of theRiver Red Gum was allowed to vary while theSwamp Paperbark stands were maintained on thebank face - extending from the summer baseflowlevel over the bank-crest to about 1 m onto thefloodplain.

Numerical modelling of bank stability requiresdetailed inputs of bank material strength properties,pore-water pressure distributions, channel drawdownconditions, bank geometry data and an assessment ofthe strength and distribution of root reinforcementthroughout the bank profile. The level of detailrequired to fully parameterise our model is beyond

the scope of the type of stability assessment requiredhere. Hence, the following broad recommendationsare the product of a generalised application of theresults of detailed analyses conducted on the LatrobeRiver.

The size and type of any mass failures observedduring the site assessment phase will indicate theextent of bank instability at the site. We found thatplanting trees at locations where potential failureplanes were likely to intercept the profile surfaceprovided the greatest resistance to failure. Failureplanes typically intercept the profile surfacesomewhere near the toe of the bank and somewhereon the floodplain surface behind the bank crest. Forexample, the location of the failure plane of a deep-seated rotational failure on a 6 m high bank was some5 m behind the bank crest on the Latrobe River. Withno vegetation to reinforce this particular bank profile,our model predicted that the bank section wasinherently unstable. The addition of a single matureRiver Red Gum, located at the failure plane on thefloodplain, effectively stabilised the bank section (thefactor of safety rose from 1 to 1.6).

Our modelling work indicated that many unstabledegraded banks along the lower Latrobe River couldbe stabilised with vegetation. For the range of bankheights along the river, planting vegetation meant thatthe banks would have to become undercut before theyfailed. For example, our modelling showed that a 5 mhigh bank supporting no vegetation would fail whensteepened to 70º by scour at the bank toe. However,further model simulations indicated that the samebank reinforced by mature tree roots would remainstable even when vertical. Undercutting was requiredto destabilise the bank.

Vegetated banks subject to bed-degradation withoutundercutting were stable up to some 2 m higher thanvegetation-degraded bank sections. For example, abare bank standing at 59º is stable until bed scourincreases its height above 6.2 m. Reinforcing the bankwith a mature root system improved stability suchthat with vegetation the bank is stable until its heightexceeds 8.9 m. That there are no bank profiles on thelower Latrobe that are this high indicates theimportance of undercutting and reinforcement byvegetation in controlling bank erosion along thatriver.


20

the erosion rate (Section 1.3). Immature plants doprovide some erosion resistance that increases as theplants grow. This increasing resistance tends to slowthe erosion rate as the bank-line moves through theaging plantation. However, for the guidelines to begenerally applied into an unknown future we stressthat the full allowance (based on contemporaryerosion rates) should be adopted in plantation design.

As an example (see also Table 2), if a plantationmatures in 20 years, and the bank erosion rate is 0.5m/year, then a minimum of 10 m additional riparianzone width is required before vegetation can controlerosion in the long-term. Note that bank migrationrates of 3 m per year have been measured in clearedQueensland rivers (e.g.Russell River). If you do notknow the erosion rate for a migrating outside bend,then a rough rule-of-thumb is 1.6% of channel widthper year. This is the median value from anunpublished data-set of 100 rivers (CooperativeResearch Centre for Catchment Hydrology). Thus, a40 m wide meandering channel, in an environmentwhere the maturation of vegetation takes 50 years,will need 32 m of additional riparian zone (0.64 m/yrfor 50 years).

The bank material on which we based our originalmodelling was a relatively homogenous silty loamwith moderately low strength values when comparedto those of typical cohesive riverbanks. For banksformed in cohesive clays, the riparian widthcalculation is likely to be somewhat conservativebecause the strength of the bank material itself willsupport higher and steeper bank profiles. Theplantation width may be similarly conservative forbanks formed in non-cohesive materials. For thesebank types, groundcover species on the bank face(and macrophytes at the bank toe) are likely toprovide most of the direct protection. However, otherplant types will continue to perform stabilising roles,

6.2 Site considerations

The first step to consider is the environment in whicha plant can or cannot thrive, for without ecologicalstability the mechanical advantages of the plantationcannot be assured. For final choice of species,reference should be made to published works or otherlocal expertise. As a matter of course, plantationsshould mimic the natural lateral-zonation of species.Other aspects of vegetation/erosion interaction shouldalso be considered in the final design. For example, asingle line of trees established along the top of thebank could predispose some banks to failure inducedby windthrow. Hence, planting wider riparian stripswill not allow preferred lines of stress to develop inthe bank.

Vegetation will not prevent all bank erosion fromoccurring. Even with an established and vigorousriparian forest, some bank adjustment will continue tooccur. The riparian forest should be wide enough toallow some channel adjustment to take place withoutinterrupting any agricultural activities occurringbeyond the forest zone. Hence, the width ofrevegetation works must allow for the stabilisation ofthe present bank profile and for future bankadjustments.

Where the site assessment indicates that the bank isactively eroding, riparian zone design should allowfor a period of maturation so that the reinforcing rootnetworks of the dominant species may develop fully.In these circumstances, the riparian zone must bewide enough to allow the vegetation away from thebank line to mature by the time the erosion frontreaches that point. If all of the vegetation is planted atthe one time, the erosion front will meet progressivelymore established vegetation as it migrates back. Theresult will be a progressive decrease in the erosionrate. The width of this zone can be calculated as thetime it takes the plantation to mature, multiplied by

Table 2: Estimating the design establishment width of the riparian zone for a 40 m widechannel with differing maturation rates.

Time for plantation maturity

Twenty years Fifty years

Current erosion rate 0.5 m/year 0.5 m/yearDesign establishment width10 metres 25 metresEstimated erosion rate .016 x 40 = 0.64 m/year.016 x 40 = 0.64 m/yearDesign establishment width13 metres 32 metres


21

such as modifying the bank hydrology to prevent highpore-water pressures.

The procedure outlined in Section 2 suggests aminimum width that can be generally applied.However, the results of our modelling on the LatrobeRiver suggest that individual trees even 15 m fromthe bank can still provide reinforcement to otherwiseunstable bank sections. Hence, even trees at themargins of “over-wide” riparian strips will function asa bank stability agent. Even so, final design choicesshould be coloured by the results of thegeomorphological, geotechnical, hydrological andhydraulic assessments of local conditions and theeffect of these factors on bank overall bank stability.The bank geometry and the nature and rate of bankerosion (if any) will indicate any additional width, orspecial treatment, required and species make-up ofthe planned revegetation work.


22


23

7 Maintenance regimes

There is a growing body of locally based informationon appropriate species for riparian planting and theirmaintenance requirements. Reference should be madeto works such as the QDNR River Facts Sheets,O’Donnell (1998) or Kapitzke et al. (1998).Regardless of their final makeup, riparianrevegetation schemes should consider the followingconstruction and maintenance principles.

a) Consult local tree planting groups for advice on suitable species.

b) Determine availability of suitable local plantstock.

c) Ensure correct zonation of species andstructural and floristic diversity.

d) Plant to allow vegetation to establish beforeseasonal flooding.

e) Plant in high density, if necessary, to maintainriparian zone integrity while immature.

f) Thin selectively or provide secondary planting (as necessary) as plantation matures.

g) Provide short-term stabilisation during earlygrowth, as required.

h) Temporarily irrigate to establish plantation, as required.

i) Weed, as necessary (mulch is commonly used to provide initial weed control and reduce evaporation).

j) Fence and/or manage stock to preserve corridor continuity.

k) Follow suitable guidelines for planting.

8 Alternatives to vegetation

Where high levels of stability are required to protectinfrastructure such as bridges or buildings, or wherethe erosion rate is such that riparian plants cannot beestablished, alternative schemes should be consideredbefore deciding on a bank stabilisation program.Alternative schemes should always be evaluatedagainst the guideline objectives. Kapitzke et al.(1998) detail a number of alternative treatments, someof which include:

a) rock revetment;b) bank profile batteringc) groynes and retards; andd) combination of either or some of the above

with vegetation.


24

9 Bibliography

Abernethy, B., 1999. On the Role of WoodyVegetation in Riverbank Stability. PhD Thesis,Monash University, Melbourne.

Abernethy, B. and I.D. Rutherfurd, 1996. Vegetationand bank stability in relation to changing channelscale. In I.D. Rutherfurd and M. Walker (eds.),Stream Management ’96, Proceedings of the FirstNational Conference on Stream Management inAustralia, Merrijig. Cooperative Research Centre forCatchment Hydrology, Melbourne, 213-9.

Abernethy, B. and I.D. Rutherfurd, 1998a. Scaleanalysis of bank stability: targeting river reaches forriparian revegetation. In H.O. Hansen and B.L.Madsen (eds.), River Restoration ’96 - SessionLectures Proceedings. International ConferenceArranged by the European Centre for RiverRestoration, 9-13 September, 1996, Silkeborg.National Environmental Research Institute, Silkeborg,50-6.

Abernethy, B. and I.D. Rutherfurd, 1998b. Wherealong a river’s length will vegetation most effectivelystabilise stream banks? Geomorphology, 23(1):55-75.

Abernethy, B. and I.D. Rutherfurd, 1999. Riverbankreinforcement by riparian roots. In I.D. Rutherfurdand R. Bartley (eds.), The Challenge of RehabilitatingAustralian Streams, Proceedings of the SecondAustralian Stream Management Conference,Adelaide. Cooperative Research Centre forCatchment Hydrology, Melbourne, 1-7.

Abernethy, B. and I.D. Rutherfurd, in press. Thedistribution and strength of riparian tree roots inrelation to riverbank reinforcement. HydrologicalProcesses.

Abernethy, B. and I.D. Rutherfurd, subm.-b. Theeffect of riparian tree roots on riverbank stability.Earth Surface Processes and Landforms.

Alonso, C.V. and S.T. Combs, 1990. Streambankerosion due to bed degradation - a model concept.Transactions of the American Society of AgriculturalEngineers, 33(4): 1239-48.

Bache, D.H. and I.A. MacAskill, 1984. Vegetation inCivil and Landscape Engineering. GranadaPublishing, London.

Barker, D.H., 1986. Enhancement of slope stabilityby vegetation. Ground Engineering, 19(3): 11-5.

Bonham, A.J., 1980. Bank Protection Using EmergentPlants Against Boatwash in Rivers and Canals,Report IT206. Hydraulics Research Station,Wallingford.

Chang, H.H., 1988. Fluvial Processes in RiverEngineering. Wiley and Sons, New York.

Coppin, N.J. and I.G. Richards, 1990. Use ofVegetation in Civil Engineering. ConstructionIndustry Research and Information Association /Butterworths, London.

Coutts, M.P., 1983. Root architecture and treestability. Plant and Soil, 71: 171-88.

Cummins, K.W., 1993. Ecology of Riparian Zones. InS.E. Bunn, B.J. Pusey and P. Price (eds.), Ecologyand Management of Riparian Zones in Australia.Land and Water Resources Research andDevelopment Corporation, Occasional Paper No.05/93, Canberra, 5-20.

Ennos, A.R., 1993. The scaling of root anchorage.Journal of Theoretical Biology, 161: 61-75.

Fitter, A.H. and A.R. Ennos, 1989. Architecturalconstraints to root system function. Aspects ofApplied Biology, 22: 15-22.

Frankenberg, J., W.K. Tennant and W. Tilleard, 1996.Mechanisms of streambank erosion. In I.D.Rutherfurd and M. Walker (eds.), StreamManagement ’96, Proceedings of the First NationalConference on Stream Management in Australia,Merrijig. Cooperative Research Centre for CatchmentHydrology, Melbourne, 93-8.

Gray, D.H., 1995. Influence of vegetation on thestability of slopes. In D.H. Barker (ed.) Vegetationand slopes: stabilisation, protection and ecology,University Museum, Oxford. Telford, London, 2-25.

Gray, D.H. and A.T. Leiser, 1982. Biotechnical SlopeProtection and Erosion Control. Van NostrandReinhold Company, New York.

Gray, D.H. and R.B. Sotir, 1996. Biotechnical andSoil Bioengineering Slope Stabilisation: a PracticalGuide for Erosion Control. Wiley, New York.

Greenway, D.R., 1987. Vegetation and slope stability.In M.G. Anderson and K.S. Richards (eds.), SlopeStability. Wiley, Chichester, 187-230.


25

Greenwood, E.A.N., L. Klein, J.D. Beresford andG.D. Watson, 1985. Differences in annual evaporationbetween grazed pasture and Eucalyptus species inplantations on a saline farm catchment. Journal ofHydrology, 78: 261-78.

Hagerty, D.J., 1991. Piping/sapping erosion. I Basicconsiderations. Journal of Hydraulic Engineering,117(8): 991-1008.

Hemphill, R.W. and M.E. Bramley, 1989. Protectionof River and Canal Banks, a Guide to Selection andDesign. CIRIA Water Engineering Report,Butterworths, London.

Kapitzke, I.R., R.G. Pearson, S.G. Smithers, M.R.Crees, L.B. Sands, S.D. Skull and A.J. Johnston,1998. Stream Stabilisation for Rehabilitation inNorth-East Queensland. Land and Water ResourcesResearch and Development Corporation, OccasionalPaper 05/98, Canberra.

Karssies, L.E. and I.P. Prosser, 1999. Guidelines forRiparian filter Strips for Queensland Irrigators.CSIRO Land and Water Technical Report, 32/99.CSIRO, Canberra.

Kirkby, M.J. and R.P.C. Morgan, 1980. Soil Erosion.Wiley, Chichester.

Knighton, A.D., 1973. Riverbank erosion in relationto streamflow conditions, River Bollin-Dean,Cheshire. East Midland Geographer, 5: 416-26.

Kouwen, N., 1988. Field estimation of thebiomechanical properties of grass. Journal ofHydraulic Research, 26(5): 559-68.

Little, W.C., C.R. Thorne and J.B. Murphey, 1982.Mass bank failure analysis of selected Yazoo Basinstreams. Transactions of the American Society ofAgricultural Engineers, 25: 1321-8.

O’Donnell, S., 1998. Management of River and CreekBank Plantings in Sub-tropical Coastal RiparianRainforest. Mary River Catchment CoordinatingCommittee, Gympie.

Pen, L.J., 1994. Condition of the Kalgan RiverForeshores 1992/93. Draft Report to the AlbanyWaterways Management Authority, Oyster HarbourCatchment Group and the Department of AgricultureWA. Waterways Commission, Perth.

Raine, A.W. and J.N. Gardiner, 1995. Rivercare:Guidelines for Ecologically Sustainable Managementof Rivers and Riparian Vegetation. Land and WaterResources Research and Development Corporation,Occasional Paper Series No. 03/95, Canberra.

Rutherfurd, I.D., K. Jerrie and N. Marsh, 1999. ARehabilitation Manual for Australian Streams. Online: http://www.lwrrdc.gov.au/v1srm.htm, 1September, 1999. Text and figures.

Shields, F.D. and D.H. Gray, 1992. Effects of woodyvegetation on sandy levee integrity. Water ResourcesBulletin, 28(6): 917-31.

Simons, D.B. and R.-M. Li, 1982. Bank erosion onregulated rivers. In R.D. Hey, J.C. Bathurst and C.R.Thorne (eds.), Gravel-Bed Rivers: Fluvial Processes,Engineering and Management. Wiley, Chichester,717-47.

Styczen, M.E. and R.P.C. Morgan, 1995. Engineeringproperties of vegetation. In R.P.C. Morgan and R.J.Rickson (eds.), Slope Stabilisation and ErosionControl: A Bioengineering Approach. E & FN Spon,London, 5-58.

Technology Research Centre for RiverfrontDevelopment (ed.), 1997. Proposed Guidelines on theClearing and Planting of Trees in Rivers. SankaidoBook Publishing Co. Ltd., Tokyo.

Thorne, C.R., 1982. Processes and mechanisms ofriverbank erosion. In R.D. Hey, J.C. Bathurst andC.R. Thorne (eds.), Gravel-Bed Rivers: FluvialProcesses, Engineering and Management. Wiley,Chichester, 227-71.

Thorne, C.R., 1990. Effects of vegetation onriverbank erosion and stability. In J.B. Thornes (ed.),Vegetation and Erosion. Wiley, Chichester, 125-43.

Thorne, C.R., 1998. Stream ReconnaissanceHandbook. Wiley, Chichester.

Thorne, C.R. and J. Lewin, 1979. Bank processes,bed material movement and planform development ina meandering river. In D.D. Rhodes and G.P.Williams (eds.), Adjustments of the Fluvial System.Kendall/Hunt, Dubuque, 117-37.

Thorne, S.D. and D.J. Furbish, 1995. Influences ofcourse bank roughness on flow within a sharplycurved river bend. Geomorphology, 12(3): 241-57.

Trimble, S.W. and A.C. Mendel, 1995. The cow as ageomorphic agent - a critical review. Geomorphology,13: 233-53.


26

How has landuse in the catchmentaltered?

Has the flow regime changed?

Has the sediment load changed?

Has the channel changed?

Are there any weed sources?

Are there other individuals orgroups that can help?

Urbanisation, forestry, agriculture,irrigation, regulation, extraction

Banks inundated longer, changed floodcycles, aggradation/ degradation

Floodplain stripping, overbankdeposition, gully erosion, sand slugs

Channelisation, extraction, knick-pointmigration

Weed infestations

Federal, State, Local GovernmentLandcare, Greening Australia, etc.Catchment groupsNeighbours

Changed catchment hydrology fromaltered landuse is generally reflected inchannel adjustment. It may be that yoursite has yet to be affected by theseadjustments but your plan should accountfor impacts such as a knick-point movingthrough the site, deepening the channel.Or weed infestation causing maintenanceproblems. The key is to be aware ofcurrent and potential pressures that maybe the result of activities elsewhere in thecatchment.

Don’t try to reinvent the wheel. Look atall the available sources of informationand use whatever is useful or appropriatefor your site. Useful documents mightinclude catchment management planswhere they exist, or other LWMP’s. Youshould also try to inspect other sites thathave undergone stability works.Critically appraise them and adapt/adoptthe successful ideas.

Appendix A: Assessment tables

Table A1: Catchment level assessment

Question Evidence/information source Planning outcome


27

Table A2: Reach level assessment.


What is the catchment context ofyour site?

What is the channel form?

Freedom to migrate?

Sinuosity?

Width/depth?

Gradient?

Bed and bank material?

Vegetation?

What is the channel history?

Has the channel avulsed?

Is there active meander migration?

Is the channel bed degrading, aggrading or stable?

Has channel been directly disturbed by human activities?

Is there any incidence of erosion?

What is the dominant erosion process?

Is it widespread?

Can it be controlled with vegetation?

Has any stability work been undertaken?

Catchment area, discharge, channel size

Channel form relies to some extent onposition in the catchment

Channel constrained by hillslopes orbedrock, or unconstrained by openfloodplain

Inspect planform from maps/air photos

Measure cross-sections in the field

Measured channel length betweencontours on topographic maps

Field inspection – shear strengthincreases from gravel to sand to silt toclay

Assess condition of riparian vegetation– pay particular attention to nativeremnants

Mainly applicable to floodplain reaches

Look for billabongs on maps or on theground

Inspect sequential air photos

Bridge piers can often be used to assesslong-term bed movements

Extraction, meander cutoffs, snagging,riparian clearing, etc.

Only answered by field inspection

Look for slump-blocks, undercutting,seepage on the bankface etc.

Inspect the reach and note where theerosion occurs: outside banks,inflections, inside banks, or onvegetated bank sections

Determine if vegetated sections are asprone to erosion as cleared sections.

Ask your neighbours. Contact yourlocal Council. Go and inspect the sites.

Expect dominant erosion processes tochange with position in the catchment.As catchment area, discharge andchannel size increase, the dominanterosion process will change fromsubaerial to scour to mass failure. Thestabilising effect of vegetation willchange with changing scale.

The object of this exercise is to assess thechannel form around your site. With aclear picture of what the river looks like,you will be able to detect anomalies atyour site. These may indicate specificproblems that should be addressed. Beaware that outside banks naturally erodepreferentially and that the width/ depth ofbends and inflections will differ. Locallysteep channel sections may be moreprone to erosion. Remnant patches ofriparian vegetation will help guide yourchoice of revegetation species.

The object of this exercise is again toprovide a wider context for the pressuresacting on your site. Knowing thepressures helps to predict the likelychannel response and allows forappropriate planning. It is cheaper andmore effective to be proactive rather thanreactive.

A critical evaluation of the nature, typicallocation and extent of all erosionprocesses operating throughout the reachwill help your site evaluationenormously. Pay particular attention toother works in the reach and try todetermine


28

Table A3: Site level assessment.


What are the local channelinfluences?

Planform?

Width/depth

Hydraulic?

Bed?

What is the bank form?

Geometry?

Material?

Bank margins?

What is the bank history?

Stable?

Steepening?

Heightening?

Parallel retreat?

What is the erosion process?

Mass failure?

Fluvial scour?

Subaerial preparation?

What is the riparian condition?

Structural elements?

Species mix?

Percent cover?

Pressures?

What is the landuse at the site?

Bend, inflection, straight

Compare with elsewhere in reach

Bars, snags, tributaries

Bed material, bed control structures

Compare with banks in similarpositions along the reach

Height, angle, shape (undercut, convex,concave)

Cohesive clays, non-cohesive sands

Hillslope, floodplain

Compare with reach assessment

No sign of erosion and air photos donot indicate retreat

Progressive slumping leading to steepbare banks, concentrated scour at baseof bank

Local channel incision, compare withreach

May be hard to detect, examine airphotos carefully

Slump-blocks, tension cracks

Undercutting, scour around trees,slump-block removal

Seepage on the bank face, dry andcracking bank material, fallen trees, rilldevelopment

Compare with some benchmark – eitherremnant stands or seek expert advice

Compare existing elements withbenchmark

Identify useful/harmful species andidentify species zonation

Estimate cover of all structuralelements

Pressures may be grazing, or weeds,etc.

Cropping, irrigation (flood, spray etc.),grazing, residential, shed, etc.

Identifying local channel influences willhighlight immediate potential for futureinstability. Some of these may presentproblems for establishing plants and willneed to be treated prior to any otherwork.

The current bank form will tell you muchabout the immediate short term stabilityof the bank. Look at the bank carefully.

The history of the bank along withknowledge of its present form willindicate much about the processes atwork on the bank. Sometimes, banks thatappear stable are seen to be activelyeroding when their history is reviewed.

Correct identification of the type, extent,and magnitude of the active erosionprocess is paramount to successfullystabilising a riverbank with vegetation.Great care should be taken to assess all ofthe available evidence.

This is the starting point for yourrevegetation strategy. Take care toidentify future pressures on the plants. Ifnecessary, allow for season, recent floodsand droughts etc. when you inspect thesite.

The landuse at the site may call for amodification to the design of the strip. Besure to seek advice before ruling somespecies in and some out. Be aware thatterrestrial activities as well as fluvial canaffect the riparian zone.


29

Table A4: Intervention.


Do the assessments suggest thatrevegetation is required?

Required width of plantation?

Species?

Other considerations?

Do the assessments suggest thatengineered options are required?

Structure?

Placement?

Longevity?

What is the best design for longtermstability?

Cost?

Confidence?

Maintenance?

Cleared banks bare of any vegetation

See Section 2

Seek local advice, look at remnants

Fencing, mulch, access, etc.

Widespread erosion, includingvegetated sites

Cause of the erosion, discharge, floodfrequency/magnitude, bank material,bank geometry

Cause of erosion, channel geometry,channel hydraulics

Design life, erosion rate, hybrid designs

Initial, ongoing

Value of protection – land, other assets

Thinning, weeding, fencing, etc.

If all the preceding assessments indicatethat the site can be stabilised withvegetation care must be taken to ensureestablishment. Pick your planting season,exclude stock and use appropriatespecies. Advice on planting techniquesshould always be sought if you have noexperience. Allow for the appropriatewidth requirements of the plantation. Beaware that floods or droughts early in thelife of the plantation might requireadditional maintenance.

Vegetation will not stabilise all sites.Where the erosion process and rate donot allow for plantation establishment,seek alternatives. Experience throughoutAustralia and overseas suggests thatmany sites can be adequately protectedwith vegetation once the toe has beenstabilised with engineered structure. Thedesign of the structure will depend on thetype and extent of the erosion and otherlocal con-siderations. Wherever possibleincorporate vegetation with structure.Seek advice.

By this stage, you should be fairly clearon the final design. Weigh initial costsagainst on-going costs and account forthe value of the asset you are protecting.Revegetation requires maintenance –plan for it. Good luck!


30

Table A5: Reassessment.


Did you get it right?

Stability?

Plantation survival?

Structure survival?

Are your neighbours impressed?

Measure against benchmarks




Establish benchmarks against whichdifferent facets of the project can beassessed. These might include a knownlocation to measure the bank crestagainst, or survival rate of trees, ornumber of curious visitors to the site.Keep a record of your practices, and howthey may have changed during theproject, so that others might learn fromyour experience. Photographs before,during and after plantation establishmentprovide an easy comparison.

- guidelines for stabilising streambanks with riparian … off... · 2018. 4. 17. · 4 riparian...

Documents