GEOG 309 10-52

Group 5ix

The role of land use in the

sedimentation of the Lyttelton Harbour

basin

Prepared for: Lyttelton Harbour/Whakaraup6 Issues Group

By: Justyna Giejsztowt, Melissa Shearer, David Gainsford, Roseanna Dale and James Parsons.

College of Science

Department of Geography

15th October 2010

GEOG309 Project Report Disclaimer

DCWUNIVERSITY OFCANTERBURYTe Whare l~-al1attga 0 WaitahaCHRISTCHURCH l'EW ZEALAl'D

This project report has been completed as part fulfilment of the requirements for GEOG309Research Methods in Geography. This is an undergraduate course offered at the University ofCanterbury in New Zealand.

The intended audience for the project report is GEOG309 academic staff and Community Groupmembers. The report discusses the project aims, methods, key findings and conclusions.

It is important to recognise that the report is the outcome of an undergraduate learning experiencerather than a piece of professional consultancy.

As such, the following clauses apply to the project report and its use:

(a) the project report is offered in good faith but, reflecting its origins in a student learning exercise,no responsibility can be taken for any errors of fact or interpretation herein, nor for any loss ordamage arising from use or interpretation ofthe project material;

(b) the views presented are not the official position of the University of Canterbury;

(c) the project report is not intended or suitable for use as evidence in planning or other forms oflegal negotiations;

(e) the project report and its component parts are not, under any circumstances, to be reproducedin any form for commercial gain;

(f) if paper copies of the report or its parts are made for particular purposes by the community groupfor which it is intended, then a copy of this disclaimer page (or an appropriate summary thereof)should be included with any excerpts.

Table of Contents

Executive Summary

1.0 Introduction

2.0 Methodology

3.0 Results

4.0 Discussion

5.0 Limitations

6.0 Recommendations

7.0 Conclusion

8.0 Acknowledgements

References

1

3

7

11

15

21

22

24

25

25

"

Executive Summary

• Since European settlement, drastic changes in land uses have occurred

resulting in accelerated rates of sediment accumulation within the harbour.

This is affecting shipping access to the port and degrades the harbour as a

habitat for aquatic life. Therefore an assessment of the sources of sediment

from different land uses is required to enable targeted mitigation of the rate

of sedimentation.

• Research questions posed for this project were: which combinations of land

uses within the Lyttelton Harbour catchment produce and deliver the

greatest amount of suspended sediment into the harbour? What practical

management changes can the Lyttelton Harbour Issues Group implement to

decelerate and mitigate sedimentation?

• Research on sediment flow into Lyttelton Harbour from catchments

incorporating a variety of land uses was undertaken, in order to investigate

the relationship of these land uses to sediment outputs from streams.

Quantitative measurements of sediment outputs were obtained using water

sample and turbidity analysis. Stream and catchment characteristics including

land uses were obtained qualitatively through personal observations and

digital analysis. Land uses within catchments were classified as urban,

agriculturat quarrying and forest.

• We have determined there are noticeable differences in the sediment

outputs from different catchments within Lyttelton Harbour. Te Wharau

stream which drains a predominantly agricultural catchment is providing the

highest amount of sediment to the harbour followed by Foleys Stream which

is adjacent to an active quarry site. Forested and urban catchments are

contributing the least amount of sediment into the harbour. Weather within

the harbour has a major control over characteristics of streams.

1

"

• Our research indicates a future remedial measure for high sediment flows

into streams could be the utilisation of riparian buffer zones, as these result

in a function similar to the natural forested condition of the harbour.

• A major limitation for this research were the difficulties in sampling stream

flow rates during storm events, which would gauge the level of sediment

outputs during high flow conditions which are most significant for erosion

rates. In addition, sampling of ephemeral streams, which flow during storm

events, would strengthen our findings. As this work was only completed over

a small time frame, the small sample sizes obtained pose difficulties in

making reliable inferences about the harbour as a whole.

• Future research should therefore focus on sediment outputs during high

rainfall events in both permanent and ephemeral stream networks. A wider

range of streams should also be sampled over a longer time period to gain a

better understanding of the system as a whole.

2

1.0 Introduction

This report aims to explore and quantify the sources of sediment to into Lyttelton

Harbour attributing to the sedimentation issues. Research questions and objectives

we intend to answer are:

• Which combinations of land uses within the Lyttelton Harbour catchment

produce and deliver the greatest amount of suspended sediment into the

harbour?

• What practical management changes can the Lyttelton Harbour Issues Group

implement to decelerate and mitigate sedimentation?

Starting with a literature review, our methods and results are then presented and

discussed, followed up by our recommendations to Lyttelton Harbour Issues Group.

Rapidly increasing rates of sediment transport from the land into coastal systems via

fluvial networks (streams and rivers) is a global phenomenon which has been

increasing in magnitude since the advent of permanent agricultural practice (Walling

2006). The processes of sediment erosion and delivery have pronounced effects on

the Earth's surface and the well being of its inhabitants. The destabilisation of soils is

an issue globally; however solutions must be tailored to local conditions (Scheurer et

al. 2009). Physical considerations such as the climate and regional soil and

vegetation types must be considered together with the economic and social values

of land (Stutter et al. 2007). Therefore, when we looked to combat sedimentation

issues in a local harbour, a holistic approach had to be adopted.

1.1 Study Site

Lyttelton Harbour, located on Bank's Peninsula, New Zealand is a topographical

feature with many contrasting uses, both culturally and economically. A ring of steep

hills feed numerous streams which transport large quantities of sediment into the

harbour. European settlement in c. 1830 AD resulted in the peninsula undergoing

3

drastic changes in land use (Goff 2005). This long history of modification to the land

surface and subsequent destabilization has led to the present day sedimentation

issues within the harbour. Historically, Banks Peninsula was covered in dense native

podocarp forest, which has been all but completely felled. The development of road

networks, small settlements, farming blocks and a quarry further compounded the

fundamental changes to Banks Peninsula and Lyttelton Harbour. Goff (2005)

completed a preliminary core study of the upper Lyttelton Harbour and found a clear

correlation between European settlement and an increase in sediment accumulation

rates. Prior to human settlement, accumulation of sediment was occurring at

0.08cm/year. A subsequent acceleration in accumulation rates occurred with a peak

of 0.85cm/year in the early 1900's. During this time, $$ transition occurred from the

sediments being less than 20% silt to nearly 100% silt and enriched with organic

material. The present accumulation rates are similar to that of the initial stages of

colonisation (Goff 2005). It is now widely accepted that the enhanced sedimentation

is not attributed to natural depositional processes but the modification to land use

(Hart 2004, Goff 2005, ECan 2007). It has been estimated that 44300ta-1 (tonnes per

annum) of soil is eroded (Goff 2005) which over time has resulted in 47 metres of

sediment deposited within the harbour and the creation of the mudflats in the upper

harbour which cover an area of llkm 2 (ECan 2007).

Sedimentation is a persistent issue within this shallow harbour as it alters

bathymetric water flow within the basin (Jaffe et al. 2007) and hinders shipping

access to the port. Moreover, rapid sedimentation is having drastic effects on

harbour ecology, altering community structure by smothering benthos and

degrading habitat (Walling 2006). Little is understood about long- term erosional and

depositional patterns in Lyttelton Harbour, making effective management difficult to

implement (Jaffe et al. 2007). The Lyttelton Harbour Issues Group, a local community

group, has identified a need to discern the processes by which streams contribute to

harbour sedimentation. More specifically, stakeholders must have a grasp of the

effects that a number of current land uses are having on sediment outputs from

streams.

4

1.2 Effects of land uses on erosional processes

Erosional processes have a profound effect on the way in which land uses contribute

to sedimentation. Soil type, soil moisture content, and underlying geology are pivotal

in determining susceptibility (Russell et al. 2001). Carling et al. (2001) states that "the

vulnerability of soils to erosion depends not only on such factors as climate,

topography and soil characteristics, but also on the type and intensity of land use".

Water is a particularly dominant transport agent. A protective vegetative cover can

decrease erosion rates by lowering the water table, reducing surface water velocity

and intercepting raindrop impact. Impermeable urban surfaces will also decrease

sediment erosion due to a lack of ground infiltration; however they may result in

larger peak discharges (Nelson and Booth 2002). The transport of eroded sediment

into deposition zones leads to sedimentation which is problematic for aquatic

systems. Benthic organisms and fish are suffocated by mud layers being laid down on

the sea bed and delicate gills and mouthparts of fauna are damaged by sediment

remaining in suspension (Niyogi et al. 2007).

Connectivity of the eroding land to fluvial systems is fundamental to sediment

delivery (Outeiro et al. 2010). Small catchments, like those on Bank's Peninsula, have

particular sediment export issues because of high inputs from steep slopes and no

floodplains which normally act as a sediment sink. Trends in erosion are complicated

by climatic factors. Soil concentrations in runoff vary with rain intensity (Mwendera

and Saleem 1997) and frequency (Kreutzweiser et al. 2009) therefore storm events

can have profound impacts. Variations in climate between the bays of Lyttelton

Harbour could therefore have an influence on the amount of sediment produced

within each catchment.

Suspended sediment is a diffuse rather than point source type pollutant, meaning

that to remedy the issue the root must be dealt with (Stutter et al. 2007), as land use

is the root of the issue in Lyttelton Harbour, its effects must be mitigated. Land use

5

within Lyttelton Harbour is a mixture of extensive agriculture, urban settlements,

forested blocks and a small quarry. Each land use has fundamentally different

surface characteristics which results in clear differences between the rates of

sediment output from these land uses.

1.3 Literature Review

The literature concerning erosion and sedimentation as a consequence of land use

change is extensive. Initial literature searches were performed to obtain an idea of

knowledge about the role of land use in sedimentation in a range of New Zealand

and international examples. The results of searches into urban, agricultural, forested

and quarrying land uses have been incorporated into the discussion section below.

Knowledge about the erosional process, ephemeral streams and how weather

variations can affect erosion was also obtained from a wide range of sources.

Detailed information specific to each landuse type is elaborated further under the

Discussion section in relation to our results. Answering our research questions

encompassed using this existing knowledge in addition to our findings in the field.

6

2.0 Methodology

2.1 Theoretical Analysis

Desk studies for quantifying and classifying the catchments in the harbours consisted

of aerial photo and topographic map interpretation, which were our first steps in

focussing our research. Reviewing and analysing literature of relevance was also

important in order to gain a grasp of concepts being dealt with in the project topics

and research process.

2.2 Field Methods

Five streams were chosen around the inner harbour area (Fig. 1), which were chosen

as representative of typical catchments in that region of the harbour.

Figure 1: Locations of

samples taken in streams in

the upper harbour basin.

The colours of the symbols

denote the stream

catchment name and are

also used in Table 1. Full

size image is in Appendix.

7

We aimed to encompass the range of land uses which we had previously determined

by map and aerial photo interpretation to be predominant in Lyttelton Harbour,

these being limited to agriculture, forest, urban and quarry (Table 1). Each of the five

streams was sampled in two locations, with a site in both the upper and lower

catchment. Sites were chosen preferentially at land use boundaries, so as to

encompass the effects of different land uses. A site specific description was noted at

each sampling site for purposes of individual comparison and repeatability.

Table 1: The five streams sampled during this study and the land uses present within each of their

catchments broken down into local land use and predominant catchment land use. Site 1 was in the

lower catchment while site 2 was in the upper catchment.~ ....................,... ,........................ ~............................... ..................................................................................................

Sample : Predominant Site 1 local land use : Site 2 local land useStream iCatchment land use (lower catchment) (upper catchment)

!

, Corsair Bay i Forest Urban Forest ,Stream

! Rapaki Agriculture Urban AgricultureStream

!i , ...

i Foleys : Agriculture Quarry Agriculturei Stream ,

!Te Wharau ! Agriculture Orton Bradley Agriculture: Stream facilities

.".....".".

Church Gully Forest Forest Forest .Stream

...............

A water sample was taken from the stream at each site in a jar for subsequent

laboratory analysis. A turbidity meter was used multiple times at each site, giving

instant indications of turbidity (clarity) of the water. Care was taken to ensure no

light from outside sources entered the test chamber and that the machine was

calibrated correctly at the start of each test. Turbidity measurements and water

samples were taken first to ensure we had not disturbed the stream bed.

A flow probe was used to measure the water velocity; the convention in using this is

8

to move it throughout the velocity profile (Fig. 2). When the water level was too

shallow to immerse the propeller, averaged timed floats over a set distance were

used to obtain the velocity.

UpstreamRiver Right /

I

/ Ri,erLe"

Downstream

Lamina Plowa fayeted Row of w.alerwhich the slower l~ay~'f:S

push again!it lhe banks andboltom The Fastest layllrs at-Il on top in mid siloam

'''albl ~lGVl:

ACorkscteW trli:l~OI1

downstream

Figure 2: A generic stream cross section indicating the variation of water velocity through the stream profile. NEEDS SOURCE

Cross-sectional analysis was based on measuring the wetted perimeter. Data

obtained allowed the group to investigate and plot trends of turbidity, stream

discharge and suspended sediment concentration.

2.3 Laboratory Methods

Lab work consisted of analysing water samples where sample jar volume was noted

to obtain a measure of sediment per volume of water for each site. Samples were

passed through pre weighed 250llg filters using a vacuum pump. Each filter was

placed in an oven at 40·C to dry for a minimum of 24 hours. Dried filters were

reweighed to obtain a weight of suspended sediment from each site, per litre of

discharge. Thus, sediment concentrations were standardised so that catchment area

9

"

did not confound the results.

2.3 Digital Tools

Extensive use was made of geospatial information systems (GIS). Proprietary

programs and tools from the ESRI ArcGIS™ and TopoMapTM range were used on

satellite images obtained from Environment Canterbury and NASA. A summary of

the techniques used included:

• Multi-band Radiometric Analysis and Remote Sensing land Classification,

which measured the electromagnetic radiation signature of land cover, and is

a measure of the free water content. This allowed different land use types to

be accurately identified (Martinez 2004, Deloach et al. 2010).

• Stream Profiles -long sections drawn as the course of a stream to illustrate

elevation and topography - were created from an inbuilt tool in the

proprietary TopoMapTM software.

2.4 Statistical Analysis

Statistical analysis was completed with Minitab lS™ software, and involved multiple

regressions, analyses of variance, and analyses of covariance. All data was tested

with inbuilt tools for normality and equality of variances.

10

3.0 Results

3.1 Influences of Water Level and Microclimate Patterns

The actual day that sampling took place was a significant determinant of suspended

sediment concentrations as well as total sediment output in fluvial systems. Day Two

samples - which were collected after a period of persistent rainfall - had the highest

suspended sediment averages (F= 8.96; P=O.002; df=24). Data collected from a

number of weather stations in different locations within the harbour showed that

annual weather patterns differed substantially over small scales (Fig. 3).

Microclimate data showed that precipitation highs were greater in catchments in the

southern half of the basin (such as Diamond Harbour) than those in the northern or

eastern harbour (Lyttelton Port and Charteris BaYI respectively). AdditionallYI

Diamond Harbour meteorological station data indicated higher variability of

precipitation maxima.

- Charteris Bay

- Lyttelton Harbour

- Diamond Harbour

180 -I160 -1---------1--+-----

E 140 +---------11----+---­

.§. 120

§ 100'';:;.~ 80 -J--\--------+-:I---Vl---+--+-­Q,'y 60 -J-.-\----,I-=.........----+l---'I<-I-=----\-­Qlct 40 +-~_I_---'lt----::--tH__"\.--_H'_PIl---

20 I~~~~~~-'W_~ro +--,---,-.....,.

Figure 3: Variation in monthly precipitation highs in the Lyttelton Harbour Basin area from October

2009 to September 2010

11

Agriculture Fo est

3.2 Influence of Land Use on Sediment Transport

The analysis of variance results proved, as expected, that the predominant

catchment land use had a strong and significant effect on suspended sediment

weights (F=12.84; P= 0.001; df=24). Substantially higher totals were recorded in

streams located in catchments with predominantly agricultural management such as

Foleys, Rapaki and Te Wharau (mean= 0.92 grams of sediment per litre discharge, se

= 1.16) than those from primarily forested catchments such as Corsair Bay and

Church Gully (mean = 0.09{g/L), se= 0.18) (Fig. 4).

1.6~ 1.4S_ 1.2c(I) 1E" 0.8(I)

III 0.6

"~ 0.45i 0.2Q.III 0;:,

en -0.2

-0.4

-0.6

Predominant catchment land use

Figure 4: Disparity of suspended sediment concentration between 5 sampled streams in catchments

of predominantly agricultural and forested land use in LyUelton Harbour Basin, 2010.

Likewise, land use directly upstream of the sample site, irrespective of the area

covered, was a good predictor of suspended sediment concentrations in streams

(F=5.72; P= 0.006; df=18) (Fig. 5). Particularly of note was the large contribution of

the quarry in the Foley's Stream catchment which was below the upper catchment,

but above our lower catchment sample site as seen in Fig. 1. Te Wharau stream

(which lies in an agricultural catchment) had a higher sediment load in the lower

catchment, which has an extensive road network. Corsair Bay Stream, the upper

12

catchment of which is a mixture of plantation forest and native grassland, drops

some of its sediment load as it moves through an established urban setting.

1.8-...JC, 1.6-- 1.4t:(1)

E 1.2'0 1(1)til'0 0.8(1)

'0 0.6t:

8. 0.4til~ 0.2

o

•

Upper Catchment

•

Lower Catchment

-+-Corsair

~Foleys

Church Gully

Te Wharau

-Rapaki

Figure 5: Variation in suspended sediment concentrations as a result of different land uses in the

upper and lower catchment in 5 sampled streams in Lyttelton Harbour Basin, 2010.

Generally, agriculture or quarry land use directly upstream of a sample site resulted

in higher sediment concentration than those sections of a stream which were placed

within urban or forested land - even if the urban or forested land was relatively small

in area (Fig. 6).

Figure 6: Variation in average suspended sediment concentrations as a result of different local land

uses in 5 sampled streams in Lyttelton Harbour Basin, 2010.

13

Predominant catchment land use and day of sampling interacted significantly in an

analysis of covariance (F=3.64; P=0.042; df=24). On sampling days with higher

discharge, erosion susceptible land uses such as agriculture had proportionally larger

increases in concentrations of suspended sediment than forested catchments (Fig.7).

- 2.5..J-C)-- 2s::::CI)

E1.5"0

CI)tn"0 1CI)"0s:::: 0.5CI)Q,tn::]

0UJ

~Agriculture

Forest

None Very little Moderatelevel of precipitation in preceding days

Figure 7: Level of precipitation in days preceding sampling had a variable effect on suspended

sediment concentrations, dependant on predominant catchment land use in 5 sampled streams in

Lyttelton Harbour Basin, 2010.

14

,,

4.0 Discussion

4.1 Agricultural Land Use

Agriculture is a particularly dominant land use type in Lyttelton Harbour. All levels of

grazing pressure increase soil and solute runoff (Mwendera and Saleem 1997),

however impacts increase with the intensity of agriculture such as increased

livestock numbers (Outeiro et al. 2010). This is because grazing reduces vegetative

cover (Russell et al. 2001) and vegetation complexity (Greenwood and McKenzie

2001). Livestock also cause higher erosion because they trample the soil and

decrease infiltration, particularly in fine textured soils such as loess (Mwendera and

Saleem 1997) as found in Lyttelton Harbour. In addition to stock impacts, human

construction works like roads, drains and tracks, can result in a large input of

sediment (Outeiro et al. 2010). Our results have indicated that predominantly

agricultural catchments contribute the highest levels of suspended sediments into

the harbour. Te Wharau stream runs through a typically agricultural catchment,

Orton Bradley Park. It has a mixture of uses including extensive grazing of sheep and

cattle, small native bush remnants and recreational facilities such as picnic areas and

a golf course. Te Wharau appears relatively clean; despite this our results indicate

high suspended sediment loads, which are probably a result of runoff from the

sparsely vegetated hillsides. It has been illustrated that the highest erosion rates

occur where surface runoff results in structures such as sheet and rill/gully erosion

(Walling et al. 2002; Capra et al. 2009; Valentin et al. 2005) and there is clear

evidence of this in Te Wharau catchment and other agricultural catchments of

Lyttelton Harbour such as Rapaki (Fig. 8A and 8B). Consequently, Rapaki Stream also

had the one of the highest suspended sediment discharges. In comparison with Te

Wharau Stream, there was ample evidence within the Rapaki catchment to indicate

extensive soil erosion such as the slumping and gullying on the hillsides and thick silt

deposits on the streams edges as the result of a recent storm event (Fig. 8C).

15

A

Figure 8: Evidence of larges-scale soilerosion in agricultural catchments inLyttelton Harbour 2010. (A) Depictsrifling and (B) slumping. (C) is showingsediment deposition on stream banks.Photographer: Melissa Shearer26/8/2010

4.1.1. Forests

Lyttelton Harbour contains a mixture of both native forest and plantation forestry

blocks. Generally, forest cover is important in limiting erosion, mass wasting, soil

degradation and sediment deposition. Afforestation increases rainfall interception,

adds strength to erosion-prone land (Carlton 2001), and reduces overland flow thus

reducing the potential for sediment erosion and transport (Trimble 2003). In

plantations however, there is a large short term erosion risk during and after harvest

until canopy closure, and heavy machinery and road networks can result in

sedimentation of adjacent streams (Boothroyd et al. 2004). This effect can take

decades to be remedied by reforestation, depending on individual tree growth rates

(Parkyn 2004). The presence or absence of riparian buffer zones is also vital in

controlling the amount of sediment reaching streams in a production forest. This will

be especially noticeable on the headland between the Head of the Bay and Charteris

Bay where forestry blocks extend to the shoreline (Fig. 9).

16

sediment when being harvested if a riparian buffer were not established. Photographer David

Gainsford 26/9/2010.

Church Gully stream which is a small ravine type catchment is representative of the

natural forested condition of Bank's Peninsula. Suspended sediment contributions

from Church Gully were minute in comparison to agricultural catchments. It is clear

that the dense vegetation within this catchment is successful at minimising transport

of sediment into the stream channel. This highlights the importance and

effectiveness of riparian buffer zones. The upper reaches are bordered by farmland,

and from anecdotal evidence it is clear when cattle are being grazed in proximity of

the stream as the outflow into Church Bay is noticeably tainted with sediment (pers.

obs.). However maintaining a riparian buffer zone between the paddocks and the

stream would minimise this effect.

4.2 Quarrying

Mining and quarrying activities within a river or harbour catchment can have

significant sediment contributions (Nelson and Booth 2002). There are two quarries

present in Lyttelton Harbour (one active and one converted to a landfill) (ECan

2007), and although they take up a minor proportion of the land area, the active

17

road metals quarry site may provide significant sediment contributions because of its

close proximity to Foleys Stream. The Foleys stream catchment was also dominated

by extensive agriculture, but possibly more significant sediment inputs are derived

from the quarry. This stream was the second highest producer of suspended

sediment in our study, and considerably more would be carried during flood events.

There were profound differences between the stream morphology and suspended

sediment loads above and below the quarry. Above the quarry, the stream was

relatively rapidly flowing with pool and riffle complexes. Below the quarry, the water

was lethargic, highly turbid and the stream banks and bed were comprised of thick

silt and mud (Fig. 10). It can therefore be assumed that a significant amount of

sediment is being introduced to the stream and therefore the harbour from this

quarry. Past studies on the sources of sediment into the Harbour have indicated that

the quarry is of high risk and priority for management (ECan 2007).

Figure 10: The impact of quarry sediment outputs on morphology in Foleys Stream (lyttelton

Harbour, 2010) is visible in these photos. (A) is a photo taken above the quarry in a stream section

which is not impacted by quarry outputs, conversely, (B) shows downstream impact. Photographer:

Melissa Shearer 26/8/2010

4.3 Urban Areas

Urban land surfaces include an impervious stormwater drainage network which is

designed to intercept and rapidly discharge any excess water, which can mean

greater erosion of the stream channel receiving the stormwater discharged during

18

these events (Corbett et al. 1997). However/ the greatest potential for sediment

discharge is during the construction phase of urban development (Nelson and Booth

2002; Wotling and Bouvier 2002; Chin 2006)/ as they often consist of unprotected

soils (Roberts and Pierce 1974). There are currently 10 subdivision developments in

various stages of completion/ such as at Cass Bay. In addition to these current

subdivision developments/ proposed changes to land use zoning allowing for further

development represents the potential for enhancing sedimentation rates in the near

future (ECan 2007). After the construction phase is complete the impervious nature

of urban surfaces such as roofs/ car parks and roads mean sediment levels are

actually relatively low (Nelson and Booth 2002). Catchments which contain urban

settlements such as Corsair Bay have been found to contribute very small amounts

of sediment in our study.

Due to the lack of infiltration/ little or no pollutant filtering occurs before water is

discharged into streams (Corbett et al. 1997). Hydrocarbons and heavy metals are

abundant in urban runoff and are detrimental to marine and freshwater aquatic

communities (Corbett et al. 1997; Walker et al. 1999; Goodwin et al. 2003; Rodriguez

et al. 2003). Riparian Buffer Zones or engineering works such as mesh fences and

settling ponds and grassed swales can improve water quality in urban streams in

terms of both suspended sediment and chemical composition (Parkyn 2004).

4.4 Weather Influences

Variations in climate between the bays of Lyttelton Harbour can have an influence

on the amount of sediment produced within each catchment. As we have illustrated

there is variation in the amount of rain/ and also wind direction and speed. This

variation is due to different aspects of the bays and the prevalent direction of

incoming weather systems and their entailing wind and precipitation patterns.

Higher exposure of vulnerable land surface to erosive agents in the form of adverse

weather increases erosion potential. Both wind and rain are erosive agents/

therefore catchments with higher precipitation such as those in Diamond Harbour

may be more vulnerable. The intensity and frequency of storm events will have

19

significance for the production of sediment within the different catchments

(Kreutzweiser et al. 2009) fhj

The temporal variation in weather patterns therefore influences the sediment

output of streams. We have found that discharge varies significantly from day to day,

as a result of the amount of rainfall received in the catchment within the few days

prior to sampling. We have also found that this variation has resulted in a significant

difference in sediment levels between sampling days. In relatively low rainfall

periods, very little water occupied the area available in stream channels in our study

(the 'wetted perimeter'). This means that there is less kinetic energy to erode the

bed sediments and then transport them downstream. Conversely, periods of high

flow (such as during rain storms) have the potential for greater erosion as higher

water levels mean a greater proportion of the stream channel is subject to erosion.

Higher energy levels of the faster flowing water also enable greater erosion and

transport. Storm events also affect overland flow as excess precipitation creates

surface runoff. Soils will hold moisture up to a threshold, above which water will

flow down slope over the land surface entraining and transporting sediment into

stream channels and the harbour (Bryan 2000, Russell et al. 2001).

20

5.0 Limitations

Several limitations are present within our study which may have an effect on the

research undertaken.

5.1 Timing

A major limitation to our study was that we were unable to sample during high

rainfall events. This was due to time constraints and difficulties in organising group

outings on short notice to coincide with peak sediment fluxes. We therefore do not

have suspended sediment data for occasions which are typically the most

responsible for significant total sediment discharge.

5.2 Ephemeral Streams

We intended to quantify the contribution of ephemeral streams to sedimentation, as

they are regarded as important sediment transport agents in literature. As these

characteristically only flow during storm events we were unable to sample

ephemeral stream discharge, so decided to discount this from our overall research.

5.3 Sample Size

Despite several sampling occasions, we still have a relatively small sample size due to

the small number of samples taken at each site within the different streams. This

may result in difficulty in making inferences about the total catchment system of the

harbour. In addition to this, we must extrapolate the results from our small selection

of streams around the inner harbour to the harbour as a whole.

5.4 Currency of Data

A discrepancy existed between 2004 land use maps and current satellite imagery

used to obtain land use by inspection and digital analysis. This however could be

overcome by on site judgment of land use percentages during our field sampling

occasions although this too has inherent interpretive inaccuracies.

21

6.0 Recommendations

6.1 Riparian Buffer Zones

A riparian buffer zone is a strip of trees and other vegetation surrounding water

bodies such as rivers and lakes (Fig. 11). This vegetated stream margin has a number

of important influences on in-stream physical and chemical characteristics such as

water temperature, coarse and fine organic nutrient inputs and turbidity (Boothroyd

et ai, 2004; Lee et al. 2004). These factors can influence the community structure

within the stream system including invertebrates, fish, terrestrial birds and mammals

and aquatic plant life (Boothroyd et ai, 2004; Lee et al. 2004). In addition to this,

vegetation reduces the velocity of the water flowing both overland and within

channels, weakening its power to erode and transport sediment (Langer et al. 2008).

Figure 11: A pastoral example of a riparian buffer zone which could be employed as a sediment trap in

Lytellton Harbour. Source: Agriculture and Agri-Food Canada (2010).

Sparovek et al. (2002) clearly outline the factors that need to be considered when

calculating a forested buffer width. Several authors including Greynoth (1979) have

determined that a riparian buffer zone in a plantation forest, consisting of trees,

ferns and shrubs and with a width of 30 metres was sufficient. New Zealand's native

plants have been recognised as excellent bank stabilisers (Boothroyd et al. 2004,

Langer et al. 2008) making them a favourable option for riparian species.

22

We recommend the use of riparian buffer zones around all streams in urban,

agricultural and forested catchments, including the continued planting of seedlings

around Te Wharau Stream in Orton Bradley Park. Additionally, in Orton Bradley Park

and other agricultural catchments, we suggest shelter belts should be established if

not already present. Russell et al. (2001) illustrated the large potential for sediment

delivery reduction from paddocks as a result of shelter belts intercepting both

subsoil and surface water flow. A faster-acting alternative is possible with geotextile

fine mesh netting fixed across stream beds to 'catch' sediment whilst allowing

filtered water pass through (Elliot 2010).The implementation of riparian buffers

should also be a high priority around the quarry site which is a major contributor of

sediment.

6.2 Ephemeral Streams

Further investigations of ephemeral stream outputs in the context of the lyttelton

Harbour are necessary to quantify their contribution to sedimentation as it can be

assumed they are significant contributors both from overseas work (for example

Reid and laronne 1995) and personal observations of local people.

6.3 Urban Development

Planning for proposed subdivisions needs to include stringent controls on sediment

outputs. This can include the use of grass swales to intercept fine sediment (Deletic

2005) and other development strategies such as riparian buffers and low Impact

Development subdivisions (LID). For more information on LID and its effectiveness,

see Dietz and Clausen (2008).

6.4 Stream Modelling

Further modelling of stream flow rates and sediment inputs needs to be quantified

and mapped to form additional base data to our findings. This will be essential for

future decision making and to understand the behaviour of the catchments under a

23

broader range of conditions over longer time periods. Regular surveys of the

bathymetry of the harbour will also help improve understanding of the issues facing

the harbour as the result of sedimentation. We understand that Environment

Canterbury has taken steps to initiate such work, and this should be supported and

encouraged to help maintain progress and momentum.

24

7.0 Conclusion

The rate of sediment accumulation in water bodies is an increasingly important issue

world-wide, and has been attributed to changes in land use, in particular the

removal of forest cover to make way for agricultural land. The removal of native

forest by European settlers has been correlated with acceleration in sediment

accumulation rates within the Lyttelton Harbour. However, contributions of

different catchments and land use related trends are confounded by microclimate

patterns within the Lyttelton Harbour basin. Sedimentation has significant impacts

on both economic and environmental aspects of the harbour environment as port

access has to be maintained by expensive dredging, while aquatic communities face

habitat destruction and water contamination.

It has been established that this issue is a result of land use change rather than

natural processes, it is therefore important to establish how the different land uses

are contributing to sedimentation. From this study, it has been identified that

agricultural land use is producing the greatest amount of sediment. This is possibly

due to the effects of livestock induced soil compaction and the loss of vegetation

diversity, which reduces soil cohesion and exposes soils to erosion by wind and rain.

The second largest contributor of sediment is the road metals quarry adjacent to

Foleys Stream. The sediment from this source flows directly into the stream with no

intercepting riparian buffer. In contrast to agricultural and quarrying land uses,

forested and urban catchments are producing the smallest quantity of sediment.

However, planned subdivisions have the potential to be significant sediment sources

in the future. We strongly recommend the use of riparian buffer zones around all

streams. These aim to imitate the historic natural forested condition of many of the

streams in Lyttelton Harbour, by forming a barrier between the land and the stream

system and thus reduce the rates of sediment export. This should be a step towards

mitigating the sedimentation issue of Lyttelton Harbour.

25

8.0 Acknowledgements

The group would like to thank Claire Findlay, Justin Harrison, Michele Stevenson,

Paul Pritchett and GEOG309 teaching staff for their advice, assistance and helpful

suggestions.

9.0 References

Agriculture and Agri-Food Canada (2010), 'Riparian Buffer Zone' [Image], retrieved 17/10/2010 from

http://www4.agr.gc.ca/AAFC-AAC/display-afficher.do?id=1235756863870&lang=eng.

Boothroyd, IKG, Quinn, JM, Langer, ER (Lisa), Costley, KJ and Steward, G (2004), 'Riparian buffers

mitigate effects of pine plantation logging on New Zealand streams 1. Riparian vegetation structure,

stream geomorphology and periphyton', Forest Ecology and Management, vol. 194, pp. 199-213.

Bryan, RB (2000), 'Soil erodibility and processes of water erosion on hillslope', Geomorphology, vol.

32, pp. 385-415.

Capra, A, Porto, P, Scicolone, B (2009), 'Relationships between rainfall characteristics and ephemeral

gully erosion in a cultivated catchment in Sicily (Italy) Soil and Tillage Research, vol. 105(1), pp. 77-87.

Carling, PA, Irvine, BJ, Hill, A and Wood, M (2001), 'Reducing sediment into Scottish streams: A review

of the efficacy of soil conservation practices in upland forestry', The Science of the Total Environment,

vol. 265, pp. 209-227.

Carlton, J (2001), The effects offorestry on sediment production in the Coalgate Forest, Malvern Hills,

inland Canterbury. MSc Thesis, University Of Canterbury.

Chin, A (2006) 'Urban transformation of river landscapes in a global context', Geomorphology, vol. 79,

pp.460-487

Corbett, CW, Wahl, M, Porter, DE, Edwards, D and Moise, C (1997), 'Nonpoint source runoff

modelling. A comparison of a forested watershed and an urban watershed on the South Carolina

coast', Journal ofexperimental marine biology and ecology', vol. 213, pp.133-149.

26

Deloach Sand Ellis S (2010), 'Subsurface Soil Moisture with Passive Microwave Radiometry', retrieved

14/10/2010 from http://proceedings.esri.com/library/userconf/proc98/proceed/to850/pap845/p845.htm.

Deletic, A (2005) 'Sediment transport in urban runoff over grassed areas', Journal ofHydrology, vol.

301, pp. 108-122.

Dietz, ME and Clausen, JC (2008) 'Storm water runoff and export changes with development in a

traditional and a low impact subdivision', Journal of Environmental Management, vol. 87, pp. 560­

566.

Elliot, C (2010) [Course notes] GEOG340: Field Based Geomorphic Applications, Department of

Geography, University of Canterbury, Christchurch.

Environment Canterbury (2007), lyttelton Harbour Potential Contaminant Sources Study, Report No.

U08/17

Goff, J (2005), Preliminary core study - Upper lyttelton Harbour, National Institute of Water and

Atmospheric Research ltd., Christchurch, New Zealand.

Goodwin, TH, Young, AR, Holmes, MGR, Old, GH, Hewitt, N, leeks, GJl, Packman, JC and Smith, BPG

(2003), 'The temporal and spatial variability of sediment transport and yields within the Bradford Beck

catchment, West Yorkshire', The Science of the Total Environment, vol. 314-316, pp. 475-494.

Graynoth, E (1979), 'Effects of logging on stream environments and faunas in Nelson', New Zealand

Journal ofMarine and Freshwater Research, vol. 13, pp. 79-109.

Greenwood, Kl and McKenzie, BM (2001), 'Grazing effects on soil physical properties and the

consequences for pastures: A review', Australian Journal of Experimental Agriculture vol. 41(8),

pp.1231-1250.

Hart, DE (2004), 'Mapping of the bathymetry, soft sediments and biota of the seabed of Upper

lytellton Harbour', University of Canterbury Departments of Geography and Biology, Report to

Environment Canterbury and DTEC Consulting ltd.

Jaffe, BE, Smith, RE and Foxgrover, AC (2007), 'Anthropogenic influence on sedimentation and

intertidal mudflat change in San Pablo Bay, California: 1856-1983', Estuarine, Coastal and Shelf

Science, vol. 73, pp. 175-187.

Keim, RF and Schoenholtz, SH (1999), 'Functions and effectiveness of silvicultural streamside

management zones in loessial bluff forests', Forest Ecology and Management, vol. 118, pp. 197-209.

27

Kreutzweiser, 0, Capell, S, Good, Kand Holmes, S (2009), 'Sediment deposition in streams adjacent to

upland c1earcuts and partially harvested riparian buffers in boreal forest catchments', Forest Ecology

and Management, vol. 258, pp. 1578-1585.

Kreutzweiser, 0, Capell, S, Good, Kand Holmes, S (2009), 'Sediment deposition in streams adjacent to

upland c1earcuts and partially harvested riparian buffers in boreal forest catchments', Forest Ecology

and Management, vol. 258, pp. 1578-1585.85.

langer, ER (lisa), Steward, GA and Kimberley, MO (2008), 'Vegetation structure, composition and

effect of pine plantation harvesting on riparian buffers in New Zealand', Forest Ecology and

Management, vol. 256, pp. 949-957.

lee, P, Smyth, C and Boutin, S(2004), 'Quantitative review of riparian buffer width guidelines from

Canada and the United States', Journal of Environment Management, vol. 70, pp. 165-180.

Martinez, E(2004), 'Remote Sensing Techniques for land use classification of Rio Jauca watershed

using IKONOS images'[ Journal Article], retrieved from http://gers.uprm.edu/geoI6225/research.html

on 12/10/10.

Mwendera, EJ and Saleem, MAM (1997), 'Infiltration rates, surface runoff, and soil loss as influenced

by grazing pressure in the Ethiopian highlands', Soil Use and Management, vol. 13(1), pp. 29-35.

Nelson, EJ & Booth, OB (2002) 'Sediment sources in an urbanizing, mixed land use watershed', Journal

of Hydrology, vol. 264, pp. 51-68.

Niyogi, OK, Koren, M, Arbuckle, CJ and Townsend, CR (2007). 'Stream communities along a catchment

land-use gradient: Subsidy-stress responses to pastoral development.' Environmental Management,

vol. 39(2), pp. 213-225.

Outeiro, l, Ubeda, X and Farguell, J (2010), 'The impact of agriculture on solute and suspended

sediment load on a Mediterranean watershed after intense rainstorms', Earth Surface Processes and

Landforms, vol. 35(5), pp.549-560.

Parkyn, S, (2004). 'Review of Riparian Buffer Zone Effectiveness' [Technical Policy Report]. Wellington:

New Zealand Ministry of Agriculture and Forestry and National Institute of Water and Atmospheric

Research.

Personal Communication. Paul Pritchett, lyttelton Harbour Issues Group Member, 25/09/2010.

28

Reid, I and Laronne, JB (1995). 'Bed load sediment transport in an ephemeral stream and a

comparison with seasonal and perennial counterparts.' Water Resources Research, vol. 31(3), pp. 773­

781.

Roberts, WP and Pierce, JW (1974) 'Sediment yield in the Patuxent River (MD.) undergoing

urbanisation, 1968-1969', Sedimentary Geology, vol. 12, pp. 179-197.

Rodriguez, F, Andrieu, Hand Creutin, J-D (2003), 'Surface runoff in urban catchments: morphological

identification of unit hydrographs from urban databanks', Journal ofHydrology, vol. 283, pp.146-168.

Russell, MA, Walling, DE, Hodgkinson RA (2001), 'Suspended sediment sources in two small lowland

agricultural catchments in the UK, Journal of Hydrology, vol. 252(1-4), pp.1-24.

Scheurer, K, Alewell, C, Banninger D, and Burkhardt-Holm P (2009). 'Climate and land-use changes

affecting river sediment and brown trout in alpine countries-a review.' Environmental Science and

Pollution Research, vol. 16(2), pp. 232-242.

Sparovek, G, Ranieri, SBL, Gassner, A, De Maria, IC, Schnug, E, Ferreira dos Santos, Rand Joubert, A

(2002), 'A conceptual framework for the definition of the optimal width of riparian forests',

Agriculture, Ecosystems and Environment, vol. 90, pp. 169-175.

Stream Cross Section [Image] 2006, 'Rivers', retrieved 27/9/2010 from

http://www.vcsar1.org/images/swiftwater/river cross-Section. jpg.

Stutter, MI, Langan, S, and Demars BOL (2007), 'River sediments provide a link between catchment

pressures and ecological status in a mixed land use Scottish River system', Water Research vol. 41(12),

pp. 2803-2815.

Trimble, SW (2003). Environmental Hydrology Second Edition. Washington, D.C: Lewis Publishers.

Valentin, C, Poesen, J, Yong Li (2005), 'Gully erosion: Impacts, factors and control', Catena, vol. 63,

pp.132-153.

Walker, WJ, McNutt, RP and Maslanka, CK (1999), 'The potential contribution of urban runoff to

surface sediments of the Passaic river: Sources and chemical characteristics', Chemosphere, vol. 38,

pp.363-377

Walling, DE (2006), 'Human impact on land-ocean sediment transfer by the world's rivers',

Geomorphology, vol. 79(3-4), pp. 192-216.

29

Walling, DE, Russell, MA, Hodgkinson RA, Zhang, Y (2002), 'Establishing sediment budgets for two

small lowland agricultural catchments in the UK', Catena, vol. 47(4), pp.323-353.

Weather data from professional and amateur weather stations in various locations in the Lytellton

Harbour basin [Primary Data] 2008-2010, retrieved throughout September and October 2010 from

http://www.wunderground.com/cgi-

bin/fmdweather/getForecast?query=Christchurch,%20New%20Zealand&wuSelect=WEATRER.

Wotling, Grvdrr and Bouvier CH (2002), 'Impact of urbanization on suspended sediment and organic

matter fluxes from small catchments in Tahiti', Hydrological Processes, vol. 16, pp.1745-175690

30